Visible-Light-Active Photocatalysis: Nanostructured Catalyst Design, Mechanisms, and Applications

Visible-Light-Active Photocatalysis:Nanostructured Catalyst Design,Mechanisms, and Applications

Visible-Light-Active Photocatalysis:Nanostructured Catalyst Design,Mechanisms, and Applications

Edited by Srabanti Ghosh

Editor

Dr. Srabanti GhoshCSIR - Central Glass and CeramicResearch InstituteFuel Cell & Battery Division196, Raja S. C. Mullick Road700 032 KolkataIndia

CoverBackground image Fotolia:Dudarev Mikhail

All books published by Wiley-VCH arecarefully produced. Nevertheless, authors,editors, and publisher do not warrant theinformation contained in these books,including this book, to be free of errors.Readers are advised to keep in mind thatstatements, data, illustrations, proceduraldetails or other items may inadvertentlybe inaccurate.

Library of Congress Card No.: applied for

British Library Cataloguing-in-PublicationDataA catalogue record for this book isavailable from the British Library.

Bibliographic information published bythe Deutsche NationalbibliothekThe Deutsche Nationalbibliotheklists this publication in the DeutscheNationalbibliografie; detailedbibliographic data are available on theInternet at <http://dnb.d-nb.de>.

© 2018 Wiley-VCH Verlag GmbH & Co.KGaA, Boschstr. 12, 69469 Weinheim,Germany

All rights reserved (including those oftranslation into other languages). No partof this book may be reproduced in anyform – by photoprinting, microfilm, orany other means – nor transmitted ortranslated into a machine languagewithout written permission from thepublishers. Registered names, trademarks,etc. used in this book, even when notspecifically marked as such, are not to beconsidered unprotected by law.

Print ISBN: 978-3-527-34293-8ePDF ISBN: 978-3-527-80814-4ePub ISBN: 978-3-527-80815-1Mobi ISBN: 978-3-527-80816-8oBook ISBN: 978-3-527-80817-5

Cover Design Schulz Grafik-Design,Fußgönheim, GermanyTypesetting SPi Global, Chennai, IndiaPrinting and Binding

Printed on acid-free paper

10 9 8 7 6 5 4 3 2 1

I dedicate this book to my HUSBAND and PARENTS.

vii

Contents

Preface xvii

Part I Visible-Light Active Photocatalysis – Research andTechnological Advancements 1

1 Research Frontiers in Solar Light Harvesting 3Srabanti Ghosh

1.1 Introduction 31.2 Visible-Light-Driven Photocatalysis for Environmental Protection 41.3 Photocatalysis for Water Splitting 81.4 Photocatalysis for Organic Transformations 111.5 Mechanistic Studies of Visible-Light-Active Photocatalysis 131.6 Summary 14

References 15

2 Recent Advances on Photocatalysis for Water Detoxificationand CO2 Reduction 27Carlotta Raviola and Stefano Protti

2.1 Introduction 272.2 Photocatalysts for Environmental Remediation and CO2

Reduction 302.2.1 Undoped TiO2 302.2.2 Undoped Metal Oxides Different from TiO2 322.2.3 Carbon Modified Metal Oxides as Photocatalysts 332.2.4 Doped Metal Oxides 342.2.5 Perovskites 352.2.6 Metal Chalcogenides 362.2.7 Other Catalysts 372.3 Photoreactors for Solar Degradation of Organic Pollutants and CO2

Reduction 382.3.1 Non Concentrating (Low Concentration or Low Temperature)

Systems 392.3.2 Medium Concentrating or Medium Temperature Systems 40

viii Contents

2.3.3 High Concentrating or High-Temperature Systems 422.3.4 Parameters of a Solar Reactor 432.4 Conclusion 44

Acknowledgment 44References 45

3 Fundamentals of Photocatalytic Water Splitting (Hydrogen andOxygen Evolution) 53Sanjib Shyamal, Paramita Hajra, Harahari Mandal, Aparajita Bera,Debasis Sariket, and Chinmoy Bhattacharya

3.1 Introduction 533.2 Strategy for Development of Photocatalyst Systems for Water

Splitting 543.3 Electrochemistry of Semiconductors at the Electrolyte Interface 563.4 Effect of Light at the Semiconductor–Electrolyte Interface 583.5 Conversion and Storage of Sunlight 623.6 Electrolysis and Photoelectrolysis 633.7 Development of Photocatalysts for Solar-Driven Water Splitting 653.8 Approaches to Develop Visible-Light-Absorbing Metal Oxides 663.9 Conclusions 68

References 68

4 Photoredox Catalytic Activation of Carbon—Halogen Bonds:C—H Functionalization Reactions under Visible Light 75Javier I. Bardagi and Indrajit Ghosh

4.1 Introduction 754.2 Activation of Alkyl Halides 774.3 Activation of Aryl Halides 914.4 Factors That Determine the Carbon–Halogen Bond Activation of Aryl

Halides 1084.5 Factors That Determine the Yields of the C—H Arylated

Products 1094.6 Achievements and Challenges Ahead 1094.7 Conclusion 110

References 110

Part II Design and Developments of Visible Light ActivePhotocatalysis 115

5 Black TiO2: The New-Generation Photocatalyst 117Sanjay Gopal Ullattil, Soumya B. Narendranath, and Pradeepan Periyat

5.1 Introduction 1175.2 Designing Black TiO2 Nanostructures 1185.3 Black TiO2 as Photocatalyst 1225.4 Conclusions 123

References 123

Contents ix

6 Effect of Modification of TiO2 with Metal Nanoparticles on ItsPhotocatalytic Properties Studied by Time-ResolvedMicrowave Conductivity 129Hynd Remita, María Guadalupe Méndez Medrano, and ChristopheColbeau-Justin

6.1 Introduction 1296.2 Deposition of Metal Nanoparticles by Radiolysis and by

Photodeposition Method 1306.3 Electronic Properties Studied Time-Resolved Microwave

Conductivity 1326.3.1 Surface Modification of Titania with Monometallic

Nanoparticles 1336.3.1.1 Surface Modification of Titania with Pt Clusters 1336.3.1.2 Surface Modification of TiO2 with Pd Nanoparticles 1356.3.1.3 Modification of TiO2 with Ag Nanoparticles 1366.4 Modification of TiO2 with Au Nanoparticles 1386.5 Modification of TiO2 with Bi Clusters 1446.6 Surface Modification of TiO2 with Bimetallic Nanoparticles 1466.6.1 Surface Modification with Au–Cu Nanoparticles 1466.6.2 Surface Modification with Ag and CuO Nanoparticles 1486.6.3 Comodification of TiO2 with Ni and Au Nanoparticles for Hydrogen

Production 1506.6.4 TiO2 Modified with NiPd Nanoalloys for Hydrogen Evolution 1536.7 The Effect of Metal Cluster Deposition Route on Structure and

Photocatalytic Activity of Mono- and Bimetallic NanoparticlesSupported on TiO2 155

6.8 Summary 156References 157

7 Glassy Photocatalysts: New Trend in Solar Photocatalysis 165Bharat B. Kale, Manjiri A. Mahadadalkar, and Ashwini P. Bhirud

7.1 Introduction 1657.2 Fundamentals of H2S Splitting 1667.2.1 General 1667.2.2 Thermodynamics of H2S Splitting 1667.2.3 Role of Photocatalysts 1677.3 Designing the Assembly for H2S Splitting 1687.3.1 Standardization of H2S Splitting Setup 1687.3.2 Interaction of Photocatalyst and Reagent System 1697.4 Chalcogenide Photocatalysts 1707.5 Limitations of Powder Photocatalysts 1707.6 Glassy Photocatalyst: Innovative Approach 1717.6.1 Semiconductor–Glass Nanocomposites and Their Advantages 1717.7 General Methods for Glasses Preparation 1727.7.1 Glass by Melt-Quench Technique 1727.8 Color of the Glass – Bandgap Engineering by Growth of

Semiconductors in Glass 174

x Contents

7.9 CdS–Glass Nanocomposite 1747.10 Bi2S3–Glass Nanocomposite 1787.11 Ag3PO4–Glass Nanocomposite 1797.12 Summary 183

Acknowledgments 184References 184

8 Recent Developments in Heterostructure-Based Catalysts forWater Splitting 191J. A. Savio Moniz

8.1 Introduction 1918.1.1 Band Alignment 1938.2 Visible-Light-Responsive Junctions 1958.2.1 BiVO4-Based Junctions 1958.2.1.1 BiVO4/WO3 1978.2.1.2 BiVO4/ZnO 1978.2.1.3 BiVO4/TiO2 1998.2.1.4 BiVO4/Carbon-Based Materials 1998.2.2 Fe2O3-Based Junctions 1998.2.3 WO3-Based Junctions 2018.2.4 C3N4-Based Junctions 2028.2.5 Cu2O-Based Junctions 2048.3 Visible-Light-Driven Photocatalyst/OEC Junctions 2078.3.1 BiVO4/OEC 2078.3.2 Fe2O3/OEC 2078.3.3 WO3/OEC 2088.4 Observation of Charge Carrier Kinetics in Heterojunction

Structure 2098.4.1 Transient Absorption Spectroscopy 2098.4.2 Electrochemical Impedance Spectroscopy 2118.4.3 Surface Photovoltage Spectroscopy 2138.5 Conclusions 215

References 216

9 Conducting Polymers Nanostructures for Solar-LightHarvesting 227Srabanti Ghosh, Hynd Remita, and Rajendra N. Basu

9.1 Introduction 2279.2 Conducting Polymers as Organic Semiconductor 2289.3 Conducting Polymer-Based Nanostructured Materials 2319.4 Synthesis of Conducting Polymer Nanostructures 2319.4.1 Hard Templates 2329.4.2 Soft Templates 2329.4.3 Template Free 2339.5 Applications of Conducting Polymer 2339.5.1 Conducting Polymer Nanostructures for Organic Pollutant

Degradation 233

Contents xi

9.5.2 Conducting Polymer Nanostructures for Photocatalytic WaterSplitting 237

9.5.3 Conducting Polymer-Based Heterostructures 2429.6 Conclusion 245

References 246

Part III Visible Light Active Photocatalysis for Solar EnergyConversion and Environmental Protection 253

10 Sensitization of TiO2 by Dyes: A Way to Extend the Range ofPhotocatalytic Activity of TiO2 to the Visible Region 255Marta I. Litter, Enrique San Román, the late María A. Grela, Jorge M. Meichtry,and Hernán B. Rodríguez

10.1 Introduction 25510.2 Mechanisms Involved in the Use of Dye-Modified TiO2 Materials for

Transformation of Pollutants and Hydrogen Production under VisibleIrradiation 256

10.3 Use of Dye-Modified TiO2 Materials for Energy Conversion inDye-Sensitized Solar Cells 260

10.4 Self-Sensitized Degradation of Dye Pollutants 26210.5 Use of Dye-Modified TiO2 for Visible-Light-Assisted Degradation of

Colorless Pollutants 26510.6 Water Splitting and Hydrogen Production using Dye-Modified TiO2

Photocatalysts under Visible Light 26910.7 Conclusions 270

Acknowledgement 271References 271

11 Advances in the Development of Novel Photocatalysts forDetoxification 283Ciara Byrne, Michael Nolan, Swagata Banerjee, Honey John, Sheethu Jose,Pradeepan Periyat, and Suresh C. Pillai

11.1 Introduction 28311.2 Theoretical Studies of Photocatalysis 28511.2.1 Doping and Surface Modification of TiO2 for Bandgap

Engineering 28511.2.2 Alignment of Valence and Conduction Band Edges with Water

Oxidation and Reduction Potentials 29111.2.3 Electron and Hole Localization 29311.3 Metal-Doped Photocatalysts for Detoxification 29611.3.1 High-Temperature Stable Anatase TiO2 Photocatalyst 29611.3.2 Main Group Metal Ions on Anatase Stability and Photocatalytic

Activity 29611.3.3 Effect of Transition Metals on Anatase Stability and Photocatalytic

Activity 296

xii Contents

11.3.4 Effect of Rare Earth Metal Ions on Anatase Stability and PhotocatalyticActivity 297

11.4 Graphene-TiO2 Composites for Detoxification 29911.5 Commercial Applications of Photocatalysis in Environmental

Detoxification 30311.5.1 Self-Cleaning Materials 30311.5.2 Bactericidal 30711.5.3 Wastewater Detoxification 30811.6 Conclusions 313

References 313

12 Metal-Free Organic Semiconductors for Visible-Light-ActivePhotocatalytic Water Splitting 329S. T. Nishanthi, Battula Venugopala Rao,and Kamalakannan Kailasam

12.1 Introduction 32912.2 Organic Semiconductors for Photocatalytic Water Splitting and

Emergence of Graphitic Carbon Nitrides 33112.3 Graphitic Carbon Nitrides for Photocatalytic Water Splitting 33212.3.1 Precursor-Derived g-CN 33412.3.2 Nanoporous g-CN by Templating Methods 33612.3.2.1 Hard Templating 33712.3.2.2 Soft Templating 33912.3.2.3 Template-Free 34012.3.3 Heteroatom Doping 34112.3.3.1 Metal Doping 34112.3.3.2 Nonmetal Doping 34212.3.4 Metal Oxides/g-CN Nanocomposites 34412.3.5 Graphene and CNT-Based g-CN Nanocomposites 34512.3.6 Structural Modification with Organic Groups 34512.3.7 Crystalline Carbon Nitrides 34712.3.8 Overall Water Splitting and Large-Scale Hydrogen Production Using

Carbon Nitrides 34812.4 Novel Materials 34912.4.1 Triazine and Heptazine-Based Organic Polymers 34912.4.2 Covalent Organic Frameworks (COFs) and Beyond 35012.5 Conclusions and Perspectives 351

References 352

13 Solar Photochemical Splitting of Water 365Srinivasa Rao Lingampalli and C. N. R. Rao

13.1 Introduction 36513.2 Photocatalytic Water Splitting 36613.2.1 Fundamentals of Water Splitting 36613.2.2 Light-Harvesting Units 36713.2.3 Photocatalytic Activity 36913.2.4 Effect of Size of Nanostructures 369

Contents xiii

13.3 Overall Water Splitting 37113.3.1 One-Step Photocatalytic Process 37113.3.2 Two-Step (Z-Scheme) Photocatalytic Process 37413.4 Oxidation of Water 37613.5 Reduction of Water 38013.5.1 C3N4 and Related Materials 38013.5.2 Semiconductors 38213.5.3 Multicomponent Heterostructures 38313.6 Coupled Reactions 38613.7 Summary and Outlook 387

Acknowledgments 387References 387

14 Recent Developments on Visible-Light Photoredox Catalysis byOrganic Dyes for Organic Synthesis 393Shounak Ray, Partha Kumar Samanta, and Papu Biswas

14.1 Introduction 39314.2 General Mechanism 39314.3 Recent Application of Organic Dyes as Visible-Light Photoredox

Catalysts 39614.3.1 Photocatalysis by Eosin Y 39614.3.1.1 Perfluoroarylation of Arenes 39614.3.1.2 Synthesis of Benzo[b]phosphole Oxides 39714.3.1.3 Direct C—H Arylation of Heteroarenes 39814.3.1.4 Synthesis of 1,2-Diketones from Alkynes 39914.3.1.5 Thiocyanation of Imidazoheterocycles 40114.3.2 Photocatalysis by Rose Bengal 40214.3.2.1 Aerobic Indole C-3 Formylation Reaction 40214.3.2.2 Decarboxylative/Decarbonylative C3-Acylation of Indoles 40414.3.2.3 Oxidative Annulation of Arylamidines 40514.3.2.4 Cross-Dehydrogenative Coupling of Tertiary Amines with Diazo

Compounds 40614.3.2.5 C—H Functionalization and Cross-Dehydrogenative Coupling

Reactions 40714.3.2.6 Oxidative Cross-Coupling of Thiols with P(O)H

Compounds 40814.3.3 Photocatalysis by Methylene Blue 40914.3.3.1 Oxidative Hydroxylation of Arylboronic Acids 40914.3.3.2 Radical Trifluoromethylation 41014.3.4 Photocatalysis by 3,6-Di(pyridin-2-yl)-1,2,4,5-tetrazine 41114.3.4.1 Synthesis of 2-Substituted Benzimidazole and Benzothiazole 41114.3.4.2 Oxidation of Alcohols to Carbonyl Derivatives 41314.3.5 Photocatalysis by Phenothiazine Dyes: Oxidative Coupling of Primary

Amines 41414.4 Conclusion 415

Abbreviations 415References 415

xiv Contents

15 Visible-Light Heterogeneous Catalysts for Photocatalytic CO2

Reduction 421Sanyasinaidu Boddu, S.T. Nishanthi, and Kamalakannan Kailasam

15.1 Introduction 42115.2 Basic Principles of Photocatalytic CO2 Reduction 42215.2.1 Thermodynamic Favorability of the Reactions 42315.3 Inorganic Semiconductors 42415.3.1 Metal Oxides 42415.3.2 Sulfides 42815.3.3 Oxynitrides 42915.4 Organic Semiconductors 43015.4.1 Carbon Nitride and their Composites 43015.4.2 Metal Organic Frameworks (MOFs) 43415.4.3 Covalent Organic Frameworks 43515.5 Semiconductor Heterojunctions 43615.6 Conclusion and Perspectives 437

References 438

Part IV Mechanistic Studies of Visible Light ActivePhotocatalysis 447

16 Band-gap Engineering of Photocatalysts: Surface Modificationversus Doping 449Ewa Kowalska, Zhishun Wei, and Marcin Janczarek

16.1 Introduction 44916.2 Doping 45116.2.1 Metal Ion Doping 45116.2.2 Nonmetal Ion Doping 45316.2.3 Codoping 45516.2.4 Self-Doping 45716.3 Surface Modification 45816.3.1 Metals 45816.3.2 Nonmetals 46416.3.3 Organic Compounds (Colorless and Color) 46416.4 Heterojunctions 46816.4.1 Excitation of One Component 46816.4.2 Excitation of Both Components 46916.5 Z-Scheme 47016.6 Hybrid Nanostructures 47116.7 Summary 473

References 473

Contents xv

17 Roles of the Active Species Generated duringPhotocatalysis 485Mats Jonsson

17.1 Introduction 48517.2 Mechanism of Photocatalysis in TiO2/Water Systems 48617.3 Active Species Generated at the Catalyst/Water Interface 48617.4 Oxidative Degradation of Solutes Present in the Aqueous Phase 49017.5 Impact of H2O2 on Oxidative Degradation of Solutes Present in the

Aqueous Phase 49217.6 The Role of Common Anions Present in the Aqueous Phase 49317.7 Summary of Active Species Present in Heterogeneous Photocatalysis

in Water 494References 495

18 Visible-Light-Active Photocatalysis: Nanostructured CatalystDesign, Mechanisms, and Applications 499Ramachandran Vasant Kumar and Michael Coto

18.1 Introduction 49918.2 Historical Background 49918.3 Basic Concepts 50118.4 Structure of TiO2 50418.5 Photocatalytic Reactions 50618.6 Physical Architectures of TiO2 50718.7 Visible-Light Photocatalysis 50918.8 Ion Doping and Ion Implantation 51018.9 Dye Sensitization 51318.10 Noble Metal Loading 51418.11 Coupled Semiconductors 51818.12 Carbon–TiO2 Composites 51818.13 Alternatives to TiO2 52018.14 Conclusions 521

References 522

Part V Challenges and Perspectives of Visible Light ActivePhotocatalysis for Large Scale Applications 527

19 Quantum Dynamics Effects in Photocatalysis 529Abdulrahiman Nijamudheen and Alexey V. Akimov

19.1 Introduction 52919.2 Computational Approaches to Model Adiabatic Processes in

Photocatalysis 531

xvi Contents

19.3 Computational Approaches to Model Nonadiabatic Effects inPhotocatalysis 532

19.4 Quantum Tunneling in Adiabatic and Nonadiabatic Dynamics 53519.5 The Mechanisms of Organic Reactions Catalyzed by Semiconductor

Photocatalysts 54119.5.1 Methanol Photooxidation on Semiconductor Surfaces 54119.5.2 Water-Splitting Reactions on Semiconductor Surfaces 54419.5.3 Carbon Oxide Redox Reactions on Semiconductor Surfaces 54619.6 Conclusions and Outlook 547

References 549

20 An Overview of Solar Photocatalytic Reactor Designs and TheirBroader Impact on the Environment 567Justin D. Glover, Adam C. Hartley, Reid A. Windmiller, Naoma S. Nelsen,and Joel E. Boyd

20.1 Introduction 56720.2 Materials 56820.3 Slurry-Style Photocatalysis 56920.4 Deposited Photocatalysts 56920.5 Applications 57020.5.1 Gas Phase and Self-Cleaning Applications 57020.5.2 Water Purification Applications 57120.5.3 Inclined Plate Collectors 57120.5.4 Parabolic Trough Concentrator 57220.5.5 Compound Parabolic Concentrator Reactor 57320.5.6 The Environmental Impact of Nanoscale Titania 57420.5.7 Detecting and Quantifying Nanoparticles 57420.5.8 Transformation of Nanoparticles in the Environment 57520.5.9 Toxicity of Nanoparticles 57620.6 Conclusion 577

References 577

21 Conclusions and Future Work 585Srabanti Ghosh

Index 589

xvii

Preface

In the last decades, photocatalysis has been demonstrated to be one of the mostpromising approaches to environmental protection, solar energy conversion, aswell as in the sustainable production of fuels from water and carbon dioxide.Visible-light-induced photocatalysis is relatively a new area of material science,but the major problem remains as poor solar energy conversion efficiency.The development of novel nanoscale structures as visible-light-responsivephotocatalysts causes a dramatic improvement in energy conversion and gen-eration. This book includes the visible-light-active photocatalysis to cover theentire field, focusing on fundamentals, size and shape tunable nanostructures,and the evaluation of their effectiveness as well as perspectives, technologies,applications, and the latest developments, including pollutants degradationby oxidative or reductive processes, organic transformations, CO2 reductionto produce low-carbon fuels, water electrolysis for hydrogen generation, andphotoelectrochemistry for water splitting to produce hydrogen and oxygen andput forward future directions in solar light harvesting.

The book begins with a brief introduction of visible-light-induced photo-catalysis by various nanomaterials in chapter 1, followed by chapters 2–15dealing with the organic pollutants degradation, water detoxification, organictransformations, water splitting, and CO2 reduction. There are chapters 2, 5–9,12 devoted to metal-oxide-based photocatalysts, plasmonic catalysts, hetero-geneous inorganic semiconducting materials such as metal oxides, nitrides,sulfides, oxynitrides, etc., heterostructures-based catalysts, conducting poly-mers nanostructures, organic polymeric semiconductors, and metal–organiccomplex. Effects of bandgap engineering of photocatalysts, mechanistic studies,particularly, roles of the active species on photocatalysis are covered in a separatechapter 16, 17, 18. Chapter 19 is dedicated to the computational modeling ofphotocatalysis, with an emphasis on reactive dynamics and quantum effects.This book also promotes the idea about solar photocatalytic reactor designsand their broader impact on the environment for large-scale applications inchapter 20. Finally, the last chapter 21 outlines a brief summary of the workand puts forward future directions in perspective of the solar light harvesting.

xviii Preface

In order to make each contribution complete in itself, there is some unavoidableoverlap among the chapters.

We believe this book endows with essential reads for university students,researchers, and engineers and allows them to find the latest information onvisible-light-active photocatalysis, fundamentals, and applications.

Kolkata, 2018 Srabanti Ghosh

1

Part I

Visible-Light Active Photocatalysis – Research and TechnologicalAdvancements

3

1

Research Frontiers in Solar Light HarvestingSrabanti Ghosh

CSIR – Central Glass and Ceramic Research Institute, Fuel Cell and Battery Division, 196, Raja S.C. Mullick Road,Kolkata 700032, India

1.1 Introduction

In continuously growing technology-driven society, an urgent need forefficient solar light harvesting to achieve sustainable solutions in science andindustry exists [1, 2]. The rapid growth of industries and some unavoidablehuman activities cause environment pollution to be a threat to the society.Solar-energy-mediated advanced oxidation process in water purification is ahighly desirable approach [3]. To use the solar light, energy harvested fromthe sun needs to be efficiently converted into chemical fuel that can be stored,transported, and used upon demand. Over the last few decades, a significanteffort has been made to develop active materials including inorganic, organic,ceramic, polymeric, and carbonaceous, their composites with tunable size andstructures [4–6]. A broad range of materials including metal oxides, chalco-genides, carbides, nitrides, and phosphides of various compositions such asheterogeneous, plasmonic, conjugated polymers, porous carbon-based materi-als, and graphene-based materials has been explored to address/solve energy andenvironment-related research challenges [7–10]. In this context, oxide-basedsemiconductors, in particular, TiO2, have been recognized as efficient and widelyexplored photocatalysts. Semiconductor-oxide-based catalysts is essentiallylimited by low quantum yield which results from the fast charge carrier (e−/h+)recombination, and the necessity to use UV irradiation (5% of total sun energy)having wide bandgap [11, 12]. To overcome these limitations, surface-tuningstrategies and modification of oxides on the nanometer scale have been devel-oped via doping or surface modifications to produce visible-light-responsivephotocatalysts. Indeed, TiO2 doped with N, C, or S or its modification with metalnanoparticles (Ag, Au, Pt, Cu, Bi) has extended its activity toward the visibleregion [13–16]. However, the photocatalytic activity of the modified materialsin the visible light is still not sufficient for commercial applications. Researchefforts are therefore increasingly being carried out to design and develop moreefficient novel visible-light active catalysts for photocatalysis and solar energyconversion. A considerable number of novel synthetic strategies including

Visible-Light-Active Photocatalysis: Nanostructured Catalyst Design, Mechanisms, and Applications,First Edition. Edited by Srabanti Ghosh.© 2018 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2018 by Wiley-VCH Verlag GmbH & Co. KGaA.

4 1 Research Frontiers in Solar Light Harvesting

fabrication of plasmonic-based novel catalysts, heterojunctions, and cocatalysthave been proposed to offer new visible-light-active photocatalytic materials aspotential substitutes of TiO2 for the most relevant photocatalytic applicationssuch as detoxification and disinfection, removal of inorganic pollutants, watersplitting, and organic synthesis [17–19]. In this regard, the loading of cocatalystsor secondary semiconductors, which can act as either electron or hole acceptorsfor improved charge separation, is a promising strategy for enhanced catalyticactivity. A more innovative implementation of this idea would be based on theuse of polymer-based composites, which could allow enhanced charge separationwith respect to the photocatalytic activity of the inorganic component alone. Inthis chapter, the state of the art on development of novel nanostructures andthe concept of heterojunction for efficient visible-light-driven water splitting,organic or inorganic pollutant degradation, and organic transformation havebeen discussed. The structural features of various nanostructured catalysts andtheir correlation are explained in detail. An overview of recent research effortsin the applications of visible-light-active photocatalysts, which include semi-conductor metal oxides (TiO2, Fe2O3, Cu2O, etc.), polymeric graphitic carbonnitride (C3N4), plasmonic nanostructures (Au, Ag, etc.), conducting polymersnanostructure (PEDOT, PANI, PDPB, etc.), heterostructures, and other novelmaterials in degradation of photocatalytic pollutants , hydrogen generation, CO2reduction, and selective redox organic synthesis are summarized.

1.2 Visible-Light-Driven Photocatalysisfor Environmental Protection

Environmental pollution issues prompted the finding of potential solutions toclean up water and environmental detoxification via exploring clean energyroutes through solar-light-induced photocatalysis. Extensive research has beendone in the area of photocatalytic removal of organic, inorganic, and microbialpollutants using semiconductor photocatalysts (e.g., TiO2, ZnO, and CdS) forwastewater purification [20–23]. The key to the success of solar energy conver-sion is the development of high-performance materials of well-matched photoabsorption with solar spectrum (visible-light-harvesting capability), efficientphotoexcited charge separation to prevent electron–hole recombination, andadequate energy of charges that carry out the photodegradation of dye andother toxic molecules. Continuous efforts have been made to generate activephotocatalysts under visible light, but their efficiency is low due to fast chargerecombination [24]. Many excellent reviews have also come up regarding thedevelopment of oxide-based semiconductors, in particular, TiO2, via fine-tuneof several electronic characteristics (e.g., atomic configuration, bandgap energy,band position, and lifetime of electrons and holes) [25–27]. In addition to dyesensitization, doping with metals and nonmetals, formation of heterojunctionshave been extensively used to enhance the visible-light response of TiO2materials and discussed in detail in Chapters 2, 5, 6, 11, and 16. For example,TiO2 doped with N, C, F, or S or its modification with metal nanoparticles has

1.2 Visible-Light-Driven Photocatalysis for Environmental Protection 5

extended its activity toward the visible region [15, 28–30]. Visible-light activitiesarise from the changes of bandgap structure of semiconductor via adsorbedmodifiers (surface modification) or bandgap narrowing (doping). Synthesis ofdifferent materials, such as M/TiO2 (M=Cu, Ag, Au, Pt, Pd, Bi, Ag—Au, Ag—Cu,Au—Cu, Ag—Pt), and the effect of metal modification on the photocatalyticactivity have been discussed in Chapter 6. Moreover, Chen et al. reporteddisordered TiO2 nanophase derived from hydroxylation through hydrogenationtreatment, which marked as black TiO2 and a considerable enhancement invisible-light-induced photocatalytic activity [31]. It has been reported thathydrogenation treatment induced the oxygen vacancies and Ti3+ sites in blackTiO2, resulting in the bandgap narrowing and the separation of photogeneratedelectrons and holes, which enhanced solar absorption and significantly improvedthe photocatalytic activity of TiO2 [32, 33]. A variety of synthetic strategies ofblack TiO2 are outlined, and the structural and chemical features, electronicproperties, and catalytic activity of the black TiO2 nanomaterials are describedin Chapter 5. Furthermore, oxygen-rich layered titanium oxide is also useful forenhanced visible-light photoactivity [34, 35]. Kong et al. reported Ti—O—Ocoordination bond in layered titanium oxide (composed of TiO6 layers, andinterstitial hydrated H+ ions) initiated visible-light-driven photocatalytic activ-ity [36]. Presence of Ti—O—O coordination bonds lowers the bandgap andpromotes the charge separation of the photoinduced electron–hole pairs.

Another important example is combination of nanostructured plasmonicmetals with a oxide-based semiconductor, which significantly enhanced thephotocatalytic activity due to the local surface plasmon resonance (LSPR) effectwith very large absorption and scattering cross sections [28, 29]. In fact, LSPRcauses an optical antenna effect, which efficiently harvests light and localizeselectromagnetic waves at the nanoscale, and the charge carrier formation withefficient separation is obtained at the semiconductor/liquid interface, whichbenefits the photocatalytic reactions [37–40]. A series of reactions have beentested on Ag, Au, and Cu surfaces, illustrating that low-intensity visible-photonillumination can significantly enhance the rates of chemical transformations aswell as control reaction selectivity with different mechanisms as discussed inChapter 6. Direct plasmonic photocatalysis is believed to occur through the tran-sient transfer of energetic electrons to adsorbate orbitals and the nature of theadsorbate may have a significant impact to control selectivity in plasmon-drivenreactions [17]. These heterogeneous oxide-based semiconductor photocatalystshave been also explored for the removal of inorganic wastewater pollutantsincluding cyanide-containing waste and heavy metal pollutants, such as arsenicspecies and hexavalent chromium [41–43]. Notably, due to high toxicity andcarcinogenicity of hexavalent chromium (Cr(VI)), the concentration of Cr(VI) inwastewater should be controlled in acceptable levels before its release in order toprotect potable water supplies [44, 45]. Although, molecular CO2 has a very lowelectron affinity and is chemically inert as well as very stable, photogeneratedenergetic electrons from photocatalysts can reduce CO2 to methane (CH4) andcarbon monoxide (CO). The photocatalytic reduction of CO2 using solar energyhas drawn considerable attention, which mimics the biological photosynthesisin plants [46–48]. It combines the reductive half reaction of CO2 fixation with


a well- matched oxidative half reaction of water oxidation, in order to achievea carbon neutral cycle, which accomplished with the environment protection.Over the last few decades, various semiconductor photocatalysts, includingmetal oxide, sulfide, and oxynitride, have been investigated [49, 50]. However,the overall efficiency of the CO2 photocatalytic reduction has been limited bythe purification and separation of products. Consequently, efficient and selectiveproduction of highly valuable fuel compounds is a vital issue for practical CO2photoreduction systems. Despite of huge attempts to enhance visible-light activ-ity by narrowing the bandgap of TiO2 through doping, large-scale applicationis limited due to defect-induced charge trapping and recombination sites ofphotoexcited charge carriers. In this regard, plasmon-based photocatalysts havedemonstrated significantly higher photocatalytic performance in comparison toother known visible-light photocatalysts (e.g., N-doped TiO2); however, the poorphotostability of silver salts reduced the photoactivity of the doped TiO2 mate-rial, which limits its extensive use as a visible-light photocatalyst [15, 51–53].Hence, a dopant-free, pure catalyst with a bandgap that matches the visible-lightenergy would be ideal. Numerous efforts have been made for the developmentof new visible-light-induced photocatalysts, and some oxides have shownvisible-light-driven catalytic activity, such as InVO4, BiVO4, Bi2MoO6, WO3,and Bi2WO6. Recently, visible-light-responsive photocatalytic activity of conju-gated polymer nanostructures (CPNs) such as poly(diphenylbutadiyne) (PDPB)nanofibers, poly(3,4-ethylene dioxythiophene) (PEDOT) nanospindle, andpoly(3-hexyl thiophene) (P3HT) nanospheres have been reported for degrada-tion of organic pollutants [54–56]. These CPNs demonstrated high photocatalyticactivity under visible light without the aid of sacrificial reagents or preciousmetal cocatalysts. These novel photocatalytic materials have been proposed aspotential substitutes of TiO2 for the most relevant photocatalytic applications,such as detoxification and disinfection, water splitting, and organic synthesis.

Compared to individual semiconductor photocatalysts, composites of twoor more semiconductor systems, that is, heterostructures, are advantageousin terms of more efficiently facilitating charge separation and charge carriertransfer, thereby substantially improving photocatalytic efficiency. A very largenumber of different semiconductor combinations have been investigated, such asmetal/semiconductor, carbon group materials/semiconductor heterostructures,semiconductor/semiconductor heterostructures with different models includingtype I and type II heterojunctions, p–n heterojunctions, and Z-scheme [57–61].Chapter 8 summarizes the recent strategies to develop such heterostructures andhighlights the most recent developments in the field. For charge carrier separa-tion, TiO2 has been commonly used to form heterostructures with CdS [62, 63],CdSe [64], CuO [65], AgBr [66], PbS [67] for enhanced photocatalytic activities,such as degradation of organic molecules, H2 generation, and CO2 reduction.For example, the integration of a potential semiconductor nanocrystal, ZnO,with a narrow band-gap conducting polymer has also shown to be an effectivemeans of promoting charge carrier separation and improving the utilization ofsolar light [68]. A deep understanding of the charge transfer process throughfundamental studies toward the rational design of heterostructures exhibitinghigh visible-light-harvesting efficiency is addressed in Chapter 9. Similarly, the

1.2 Visible-Light-Driven Photocatalysis for Environmental Protection 7

use of multiple inorganic domains within these heterostructures enables a rapiddissociation of excitons into a spatially separated pair of charges that bears aminimal probability of the backward recombination, with a high extinctioncoefficient in the visible range and a low exciton binding energy, which isbeneficial to photocatalytic applications [69]. In addition to metal oxides, metalsulfides or chalcogenides have been employed for photocatalytic applications[70]. Ganguli et al. reported a type II semiconductor, ZnO/CuS heterostructure,to increase the absorbance in the visible-light region and successful chargeseparation from CuS to ZnO through the hexagonal nanotubes (NTs) of ZnO,leading to enhanced visible-light-induced photocatalysis for the degradation oforganic pollutants due to the efficient separation of photoinduced carriers [71].Wang et al. synthesized mesoporous yolk−shell SnS2−TiO2 and applied themfor the visible-light-driven photocatalytic reduction of Cr(VI) [72].

The two-dimensional (2-D) structure of graphene possessing the large surfacearea can accommodate semiconductor nanoparticles, and the injection of pho-toexcited electrons from the semiconductor particle can readily be transportedalong the graphene surface due to its superior electronic conductivity and highmobility of charge carriers [73–76]. Hence, graphene is a promising componentto create efficient composite photocatalysts for dye degradation, organic transfor-mations, and reduction of carbon dioxide (CO2) [77, 78]. For example, Liang et al.prepared less defective graphene-P25 nanocomposites for the photocatalyticCO2 reduction under visible light [79]. Yu et al. synthesized CdS nanorod/r-GOheterostructures, which demonstrated high catalytic activity for the CO2 reduc-tion with 10 times higher CH4 production rate compared to pure CdS and evenbetter than Pt loaded CdS [80]. Moreover, Meng et al. established the concept ofphotogenerated electron transfer from α-Fe2O3 nanoparticles to the graphenesurface through transient absorption spectroscopy and time-domain terahertzspectroscopy, which increases the lifetime of charge carriers and, consequently,improve the photocatalytic activity [81]. Li et al. showed bandgap engineeringand enhanced interface coupling of graphene–BiFeO3 nanocomposites by theformation of Fe—O—C bonds, which demonstrated enhanced photocatalyticactivity under visible-light illumination [82]. Yang et al. synthesized func-tionalized graphene sheets/ZnO nanocomposites that exhibited an enhancedphotocatalytic activity for the degradation of rhodamine [83]. Zhang et al.showed the excellent performance of CdS–graphene nanocomposite photocata-lyst for selective oxidation of alcohol to corresponding aldehyde [84]. Moreover,Han et al. successfully prepared ternary CdS/ZnO/graphene composite, whichshowed enhanced visible-light-induced photocatalytic activity in comparison tobinary composites and pure ZnO and CdS [85]. Hence, this study highlights thesignificance of charge transport on graphene surface of heterostructures duringcatalysis reaction. Recently, another graphene-like material, layered structuresof MoS2 have been used as a cocatalyst to modify different semiconductorsfor hydrogen production and pollutant removal [86–88]. Zhou et al. preparedfew-layered MoS2 nanosheet-coated TiO2 nanobelt heterostructures to increasethe visible-light absorption ability of TiO2, and MoS2/TiO2 composites showedhigh photocatalytic activity in the degradation of organic dyes [89]. Anotherexample, few-layered MoS2/BiOBr hollow microspheres demonstrated superior


visible-light-response photocatalytic activity for ciprofloxacin and rhodamine Bremoval in comparison to BiOBr alone [90]. The conduction band (CB) edgepotential of MoS2 (−0.09 eV) is more negative than that of BiOBr (0.29 eV), andthe valence band (VB) of BiOBr (3.06 eV) is more positive than that of MoS2(1.81 eV). The energy difference between the CB edge potentials of MoS2 andBiOBr leads to the transfer of the electrons from the CB of MoS2 to that of BiOBr.Hence, the photogenerated electrons can be collected by BiOBr, and holes canbe collected by MoS2, which causes effective charge separation and can bereflected in enhanced photocatalytic activity. Graphitic carbon nitride (g-C3N4)considered as a low-cost photocatalytic system having a graphene-like structureconsisting of two-dimensional frameworks of tri-s-triazine connected via ter-tiary amines with a bandgap of ∼2.7 eV, corresponding to an optical wavelengthof 460 nm in the visible-light range [91, 92]. Hence, two-dimensional g-C3N4nanosheets having a graphene-like structure consisting of two-dimensionalframeworks of tri-s-triazine connected via tertiary amines also offers largesurface area and active sites, which are beneficial for photocatalytic oxygenevolution and CO2 photoreduction [93, 94]. Moreover, transition-metal-basedinorganic compounds have also been coupled with g-C3N4 for the fabricationof noble-metal-free heterostructured photocatalysts. The composites of g-C3N4and metal oxides (e.g., TiO2, ZnO, In2O3, and Bi2WO6) have been investigatedby various research groups for CO2 photoreduction [95–98]. Chapter 15 coversthe current progress of visible-light-induced conversion of CO2 to fuels byheterogeneous photocatalysts over the metal oxides, sulfides, phosphides,oxynitrides, and organic semiconductors as well as highlights the importance ofgraphitic carbon nitrides as emerging photocatalyst.

Another way to extend the range of TiO2 activity to the visible region is mod-ification with visible-light-absorbing dyes such as rose bengal, chlorophyllin,porphyrins, or phthalocyanines [99–101]. Dye-modified TiO2 can be used forvisible-light-assisted photocatalytic degradation of a great variety of organicpollutants from wastewater effluents either by oxidative or reductive processes.Sensitization of TiO2 and other photocatalysts by modification with dyes hasbeen reviewed in Chapter 10, with an emphasis on the physicochemical proper-ties of the modified photocatalysts, the mechanisms involved in the transforma-tion of pollutants, and the possible technological applications. However, the useof organic dyes as sensitizers of semiconductors has the disadvantage of gradualdegradation of organic molecules, which in turn affects the stability of catalysts.

1.3 Photocatalysis for Water Splitting

Solar H2 production by photocatalytic water splitting appears to be an attractiveroute to store solar energy in chemical bonds from renewable resources (waterand sunlight) [102, 103]. However, the complexity of resolving the complete watersplitting problem, structure–property relationships of photocatalysts for the twohalf reactions of water splitting, hydrogen and oxygen evolution reactions in the

1.3 Photocatalysis for Water Splitting 9

presence of sacrificial reagents have been studied extensively [104, 105]. Hence,light-driven water splitting is recognized as one of the major scientific challengesfor hydrogen production. Since the first pioneer report of photocatalytic watersplitting using titanium dioxide by Fujishima and Honda , numerous researchstudies have been conducted on semiconductor materials with proposed mech-anisms of photocatalytic water splitting [47]. A photocatalytic system for thephotoreduction of protons to produce H2 consists of a photosensitizer, a catalyst,and sources of protons and electrons [104]. The reaction is first initiated by pho-ton absorption, which generates numerous electron–hole pairs with sufficientpotentials. The relevant photoreduction processes involve

i) absorption of light by the photosensitizer and subsequent internal chargeseparation

ii) intermolecular charge transfer (i.e., reduction of the catalyst by the photo-sensitizer and reduction of the photosensitizer by direct hole donation froma sacrificial electron donor)

iii) catalytic formation of H2 by the reduced catalyst.

Sacrificial electron acceptors (S2O82−, Ce(SO4)2, FeCl3, Ag+ from AgNO3,

etc.) and donors (ethanol, methanol, triethanolamine, Na2S, Na2S2O3, andNa2SO3) control the production of either hydrogen (electron donor) or oxy-gen (electron acceptor) by combining with the respective charge carrier. Thefundamental aspects of direct photoelectrochemical (PEC) water splittingat semiconductor electrodes are discussed along with recent experimentalprogresses in Chapter 3. The roles of different experimental parameters forsuccessful water-splitting systems are also included. An overview of recentresearch progress in photochemically induced water splitting into hydrogenand oxygen with emphasis on new electrode materials, theoretical advances,and the development of experimental methods for light-driven water-splittingreactions has been discussed in Chapter 13 to identify stable, efficient, andcost–effective light-driven Photocatalytic systems. Now the challenge is tofabricate earth-abundant photoelectrodes and catalyst materials with highefficiency, good durability, and low cost. Recently, new visible-light-responsivephotoelectrodes, including 𝛼-Fe2O3, BiVO4, WO3, CdS, C3N4, and photoanodeshave been tested for water splitting [106–109]. However, severe recombination ofphotogenerated electron–hole pairs on the surface results in poor performanceof photocatalysts. Various attempts have been made to improve the perfor-mance of photocatalysts via doping, loading of cocatalysts, and heterojunctions[19, 57, 110–113]. Alivisatos and co-worker reported the design of multicompo-nent nanoheterostructures composed of platinum- tipped cadmium sulfide rodwith an embedded cadmium selenide tips as highly active catalysts for hydrogenproduction with an apparent quantum yield of 20% at 450 nm [114]. Zhanget al. fabricated two-dimensional titania/cadmium sulfide heterostructuresthrough a controlled sol–gel method with an excellent hydrogen evolutionactivity under visible-light irradiation and an apparent quantum yield of 6.9% at420 nm [115]. Cao et al. developed a highly efficient and robust heterogeneous


photocatalytic material for hydrogen generation (254 000 μmol h−1 g−1 for theinitial 4.5 h) using the CoP/CdS hybrid catalyst in water under solar irradiation[116]. Kozlova et al. synthesized a multiphase photocatalyst Cd1−xZnxS/TiO2with 3D ordered meso-/macroporous structure for H2 evolution reaction fromaqueous solutions of Na2S/Na2SO3 under visible-light irradiation [117]. Zonget al. have employed MoS2/CdS hybrid structure as a catalyst for photocatalyticH2 evolution under visible-light irradiation [118].Wang and co-workers devel-oped shish-kebab-like multiheterostructured metal chalcogenides (CdS—Te,NiS/CdS—Te, and MoS2/CdS—Te) photocatalysts to exhibit enhanced efficiencyand stability toward photocatalytic H2 generation due to intimate interactionsbetween CdS and multicomponent cocatalysts, together with improved sep-aration of photogenerated carriers due to the presence of Te nanotubes andtrace CdTe [119]. A series of MoS2-based heterostructures, such as MoS2/TiO2and MoS2/graphene, have been tested for enhanced visible-light photocatalyticactivities [87, 120]. Shen et al. showed one-dimensional MoS2 nanosheet/porousTiO2 nanowire hybrid nanostructures that facilitated charge separation andenhanced hydrogen generation rate of 16.7 mmol h−1 g−1 in visible light. Changet al. synthesized MoS2/G-CdS composite with an unexpected hydrogen evolu-tion reaction activity. MoS2/G-CdS demonstrated as a promising photocatalystwith high efficiency and low cost for photocatalytic H2 evolution reaction with a1.8 mmol h−1 H2 evolution rate in lactic acid solution corresponding to an appar-ent quantum efficiency (AQE) of 28.1% at 420 nm, which is much higher than thatof Pt/CdS in lactic acid solution. Graphitic carbon nitride, another carbon-based𝜋-conjugated semiconductor material with a planar phase analogous to graphite,is also suitable for photocatalytic hydrogen production from water splittingmade catalytic applications [121, 122]. However, quantum yields under visiblelight for H2 production from water using g-C3N4 is still limited (not exceed4%) due to the high recombination rate of the photoinduced electron–hole pair[91, 93]. Synthesis of porous g-CN, heteroatom-doped g-CN, metal-doped g-CN,structural modification with organic groups, metal oxide-g-CN composites,g-CN-graphene/CNT composites, and g-CN-based Z-scheme with enhancedphotocatalytic activity for either H2 or O2 generation has been discussed inChapter 12. Different nanostructured g-CN materials, such as nanosheets,nanospheres, and quantum dots, covalent organic frameworks (COFs), such ashydrazone COFs, donor–acceptor heptazine systems, and conjugated microp-orous polymers (CMPs) based on pyrene prepared from various C—C couplingreactions for water-splitting applications have also been focused in detail.Excellent performance was realized by hybridization of g-C3N4 with other cocat-alysts. For example, metal sulfides, such as NiS, MoS2, WS2, and hydroxides,such as Ni(OH)2 and Co(OH)2, have been successfully deposited on g-C3N4as cocatalysts for improved photocatalytic hydrogen production [123–126].Meng et al. incorporated g-C3N4 into Ag3PO4, which exhibited an improvedcatalytic activity for the degradation of methylene blue under visible-lightirradiation [127]. The synergic effect between between g-C3N4 and Ag3PO4 ledto structural stability for silver phosphate and high separation efficiency owingto the well- positioned CB and VB and consequently improved photocatalyticactivity. Besides the single-phase and heterostructure-based photocatalysts,

1.4 Photocatalysis for Organic Transformations 11

Z-scheme photocatalyst, which contains two semiconductor photocatalystsystems, is another way to achieve efficient water splitting [128–130]. Eachphotocatalyst is responsible for one half reaction either H2 or O2 production ora new mediator to efficiently transfer charge between two photocatalysts thatinhibit the fast unfavorable recombination of charge. Few examples of Z-schemesemiconductor–metal–semiconductor heterostructures have shown promisingresults for photocatalytic water splitting or CO2 reduction [131, 132]. Notably,the electron storage and transport capabilities of graphene make it an effectivemediator to separate the H2 and O2 evolution on different catalysts, such asZnO, BiVO4, CdS, TiO2-based Z-scheme overall water splitting [133, 134].Various other photocatalysts can be used as a hydrogen evolution photocatalystin a Z-scheme water-splitting system, most notably nitrides and oxynitrides[135–138]. Chapter 13 provides a broad overview on photochemically inducedwater splitting to generate hydrogen and oxygen with various photocatalyticsystems. However, the overall working efficiency in this Z-scheme is limitedby the slow diffusion of redox couple ions and the competitive backwardreactions between them. Moreover, “all-solid-state” Z-scheme based on ternaryheterostructure of semiconductor–metal–semiconductor by using noble metalsas the electron mediator to substitute solution-based redox couples has beenalso proposed [132, 139]. However, there is still a lack of solid evidence to verifythe proposed Z-scheme electron transfer path.

1.4 Photocatalysis for Organic Transformations

The photocatalytic solar energy conversion has attracted increasing attention fororganic transformations in order to develop environmentally friendly and newmethodologies for selective redox organic synthesis at lower cost [140–142].The visible-light-active photocatalysts, such as metal oxides, plasmonic photo-catalysts, and polymeric carbon nitride, has been utilized for the selective redoxorganic transformations are classified as follows [143]:

i) The oxidation of alcohols, amines, alkenes, and alkanes, the hydroxylation ofaromatic compounds with O2

ii) The C𝛼—H bond activation and functionalization with nucleophiles to con-

struct new C—C or C—X (X=O, N, or S) bondsiii) The reduction of nitrobenzenes to corresponding amino benzenes or azoben-

zenes with sacrificial agents under O2-free conditions.

Metal oxides have been widely employed to achieve selective organic transfor-mation under visible-light irradiation. For example, surface-modified TiO2 withcarboxyl group, such as ethylenediaminetetraacetic acid, or phenolic hydroxylgroup could initiate photocatalytic redox reactions under visible-light irradi-ation [144, 145]. Visible-light-induced oxidation of alcohols to correspondingcarbonyl compounds on anatase TiO2 was reported by Higashimoto et al.[146] The selective formation of imines from primary amines was achievedunder visible-light irradiation using high-surface-area anatase TiO2 or Nb2O5


catalysts [147–149]. However, such organic transformation is limited by theadsorption of reactant molecule on the catalyst surface. In fact, activation of asp2 C—H bond in a benzene ring could not be achieved by this methodology.Metal oxides suffer from limited range of visible-light absorption. Furthermore,surface plasmon resonance (SPR)-induced nanostructured Au, Ag, and Cusupported on metal oxides provide an efficient pathway toward such photoredoxorganic transformation. Au/TiO2 can be used as an efficient photocatalyst forthe selective aerobic oxidation of alcohols in toluene or water under visible-lightirradiation [150, 151]. Aromatic alcohols can be selectively transformed intocorresponding aldehydes with O2 in water using Au/CeO2 as catalysts [152].Additionally, a wide range of binary or tertiary metal oxides, such as TiO2, WO3,ZnO, In2O3, and SrTiO3, have been tested as supports to carry out organictransformations under solvent-free conditions with high selectivity. However,the conversion efficiency was very low (0.5%) and plasmonic photocatalystsare more expensive. Alternatively, polymeric graphitic carbon nitride (g-C3N4)can be utilized as a metal-free visible-light photocatalyst for selective redoxreactions at high temperature and O2 pressure in addition to visible light[153, 154]. Aerobic oxidation of both aromatic alcohols and aliphatic alcoholsinto the corresponding aldehydes, selective oxidation of 𝛼-hydroxy ketones to1,2-diketones and the oxidation of primary benzylic amines, and secondary ben-zylic amines to corresponding imines on mesoporous C3N4 have been achievedunder visible-light irradiation at high temperature. It would be ideal to use roomtemperature for visible-light-induced organic synthesis; in fact, selective aerobicoxidation of organic substrates at room temperature has been reported withC3N4 as a photocatalyst [155]. Moreover, g-C3N4 can also act as the supportmaterial for transition metals and play an active role in the visible-light-inducedhydroxylation of benzene with H2O2 [156]. Although g-C3N4 is stable, low-costcatalysts but elevated temperature and high O2 pressure are required toachieve the desired conversions. Alternatively, organic dyes show potentialphotocatalytic activity for selective organic transformations under ambientconditions. The utilization of the organic dyes for visible-light-driven organicsynthesis, which provides an energetically beneficial pathway, has been elab-orated in Chapter 14. The photoredox catalytic methods for the activationof carbon–halogen bonds of both alkyl and aryl halides for systematicallyare described in Chapter 4. In this regard, various interesting strategies havebeen developed for visible-light-induced asymmetric redox catalysis in whichefficient catalytic photochemical processes happened under stereochemicalcontrol and provide chiral molecules in an asymmetric manner for chemicaltransformation [157–161]. Moreover, novel asymmetric photoredox catalystshaving the metal center simultaneously serve as the exclusive source of chirality,the catalytically active Lewis acid center, and the photoredox center, which offernew opportunities for the synthesis of nonracemic chiral molecules by visiblelight [162]. Hence, the development of visible-light-promoted photocatalyticreactions, which enable rapid and efficient synthesis of chemicals, is highlyadvantageous in terms of cost, safety, and environmental friendliness.

1.5 Mechanistic Studies of Visible-Light-Active Photocatalysis 13

1.5 Mechanistic Studies of Visible-Light-ActivePhotocatalysis

In addition to the experimental observations, charge transfer mechanismacross the semiconductor/semiconductor/metal junctions has been proposedand the resultant activity enhancement is also discussed [163]. The varioustechniques, such as transient-state surface photovoltage measurement,transient-state absorption spectra, and time-resolved microwave conductivitymethod (TRMC), are reviewed for photogenerated charge separation duringcatalysis [164]. The role of various active species, such as trapped electron andhole, superoxide radical and hydrogen peroxide (O2

∙− and H2O2), hydroxylradical (OH∙), singlet molecular oxygen (1O2) generated during catalysis, hasbeen discussed in Chapter 17. A number of chemical probes are generallyused to quantify the photocatalytic activity and also to indirectly identifythe primary reactive species. The different theories concerning the natureof the active species are also conferred in view of the experimental resultsin Chapter 17.

Theoretical and computational models can be used to understand the elec-tronic density of states and band structure of semiconductor in order to design arational photocatalyst [165]. With high accuracy and decrease in computationalcosts, high-throughput computational screening has been utilized in order torealize the various aspects of photocatalytic reactions, such as light absorption,electron/hole transport, band edge alignment of semiconductors, and surfacephotoredox chemistry [166–171]. Computational methods are especially help-ful for prediction of impurity states induced by dopants in tuning bandgaps inphotocatalytic systems, such as TiO2 [172]. In this regard, density functional the-ory (DFT) has been employed as a theoretical method to predict and understandthe electronic structure of materials due to high accuracy and predictive power[173–175]. However, the inaccurate prediction of bandgaps is the major draw-back of DFT. Additionally, time-dependent density functional theory (TD-DFT)is not widely used, and few studies implementing these methods are cluster-basedmodels for water-splitting systems [176]. It is crucial to obtain a complete under-standing of electron transfer phenomena to improve the performance of pho-tocatalysts. The theoretical methods are being used to study the dynamics ofcharge separation, diffusion, relaxation, recombination, and related phenomenausing ab initio nonadiabatic molecular dynamics [177]. Chapter 19 is dedicatedto the computational modeling of photocatalysis, with an emphasis on reactivedynamics and quantum effects, such as zero-point energy, tunneling, and nonadi-abatic transitions to predict excited-state electron-nuclear energy redistribution,nuclear dynamics, charge carrier dynamics, carrier recombination, and energyrelaxation pathways in photocatalytic systems. Nevertheless, at present, compu-tational screening studies on photoactive materials are very limited, which havethe potential for the invention of novel materials.

A special chapter (Chapter 20) provides an overview of solar photocatalyticreactor designs on larger scale field and pilot-scale studies utilizing solar


illumination for the purification of water with nanoscale metal-oxide photo-catalysts as well as their broader impacts on the environment. This chapterincludes the design and fabrication of the various reactors with an emphasis onthe barriers to the commercial application of this technology and environmentalnanotoxicology for photocatalytic materials.

1.6 Summary

Design of nanoarchitectures and smart hybridization with specific activematerials has emerged as an interesting platform for light-harvesting andvisible-light-driven photocatalysis. The current status of research on nanos-tructures of common semiconducting materials has been highlighted forphotocatalytic energy conversion and utilization including photocatalytic watersplitting and CO2 reduction, and other photo-assisted reactive applications,such as pollutant degradation, selective conversion of organic compounds, andbiological disinfection. In the past few years, much progress has been made inthe design and synthesis of functional materials based on either metal oxidesor semiconductors to achieve efficient light-harvesting capacity. Research inthis area is continuing to grow with the objective of tuning the intrinsic prop-erties of catalysts, such as excellent light absorption, rapid charge separation,transport, and collection, as well as rapid kinetics of interfacial reactions andmass transport of reactants in nanodimension. The advanced functionalityand improved performance for practical applications of the visible-light-activenanostructured catalysts have been also highlighted. Here, key areas are iden-tified that will need particular attention as the search continues for stable,efficient, and cost–effective light-driven photoelectrolysis systems that exploitelectron/hole separation in semiconductor/electrolyte junctions. Synergisticand cooperative interactions among different functionalities in nanohybridsopen novel strategy for designing molecular materials for photocatalysis,light harvesting, and artificial photosynthesis. Careful selection of a specificcombination of semiconductors to obtain desired band-gap energy woulddemonstrate successful synthesis of high-quality hybrid nanostructures withenhanced photostability and photocatalytic efficiency. It may be concluded thathybrid nanostructures containing diverse functionalities and active materialswill be assembled to harvest solar light for future energy crisis and water splittingor carbon dioxide reduction with energy input from sunlight. Moreover, theefficiency of photocatalytic reduction of CO2 could be deactivated after longirradiation; therefore, it is necessary to pay more attention on the semiconductordeactivation issue in future work. Until now, there have been few reports dealingwith the interpretation and theory background behind the observed collectivephenomena of heterostructures. Theoretical studies would lead to rationalimprovement of band structure and morphological design of photocatalyticmaterials for the discovery of new materials.

References 15

References

1 Xie, W.T., Dai, Y.J., Wang, R.Z., and Sumathy, K. (2011) Concentrated solarenergy applications using Fresnel lenses: a review. Renew. Sustain. EnergyRev., 15, 2588–2606.

2 Turner, J.A. (2004) Sustainable hydrogen production. Science, 305, 972–974.3 Legrini, E.O. and Braun, A.M. (1993) Photochemical processes for water

treatment. Chem. Rev., 93, 671–698.4 Mahaney, O.O.P., Murakami, N., Abe, R., and Ohtani, B. (2009) Correlation

between photocatalytic activities and structural and physical properties oftitanium(IV) oxide powders. Chem. Lett., 38, 238–239.

5 Wang, Z., Chen, C., Ma, W., and Zhao, J. (2013) Sensitization of titaniasemiconductor: a promising strategy to utilize visible light, in Photocataly-sis and Water Purification: From Fundamentals to Recent Applications (ed.P. Pichat), Wiley-VCH, New York.

6 Abe, R. (2013) Z-scheme type water splitting into H2 and O2 under visiblelight through two-step photoexcitation between two different photocata-lysts, in New and Future Developments in Catalysis, Solar Photocatalysis (ed.S.L. Suib), Elsevier, pp. 341–370.

7 Weng, L., Zhang, H., Govorov, A.O., and Ouyang, M. (2014) Hierarchi-cal synthesis of non-centrosymmetric hybrid nanostructures and enabledplasmon-driven photocatalysis. Nat. Commun., 5, 4792–4602.

8 Zhang, N., Zhang, Y., and Xu, Y.-J. (2012) Recent progress on graphene-based photocatalysts: current status and future perspectives. Nanoscale, 4,5792–5813.

9 Li, X., Yu, J., Wageh, S., Al-Ghamdi, A.A., and Xie, J. (2016) Graphene inphotocatalysis: a review. Small, 12, 6640–6696.

10 Coronado, J., Fresno, F., Hernández-Alonso, M.D., and Portela, R. (eds)(2013) Design of Advanced Photocatalytic Materials for Energy and Environ-mental Applications, Green Energy and Technology, Springer, pp. 1–101.

11 Schneider, J., Matsuoka, M., Takeuchi, M., Zhang, J., Horiuchi, Y., Anpo, M.,and Bahnemann, D.W. (2014) Understanding TiO2 photocatalysis: mecha-nisms and materials. Chem. Rev., 114, 9919–9986.

12 Guo, Q., Zhou, C., Ma, Z., Ren, Z., Fan, H., and Yang, X. (2016) Elementaryphotocatalytic chemistry on TiO2 surfaces. Chem. Soc. Rev., 45, 3701–3730.

13 Qu, Y. and Duan, X. (2013) Progress, challenge and perspective of heteroge-neous photocatalysts. Chem. Soc. Rev., 42, 2568–2580.

14 Asahi, R., Morikawa, T., Irie, H., and Ohwaki, T. (2014) Nitrogen-doped tita-nium dioxide as visible-light-sensitive photocatalyst: designs, developments,and prospects. Chem. Rev., 114, 9824–9852.

15 Kowalska, E., Yoshiiri, K., Wei, Z., Zheng, S., Kast, E., Remita, H., Ohtani,B., and Rau, S. (2015) Hybrid photocatalysts composed of titania modifiedwith plasmonic nanoparticles and ruthenium complexes for decompositionof organic compounds. Appl. Catal. B, 178, 133–143.


16 Kouamé, N.A., TahiriAlaoui, O., Herissan, A., Larios, E., José-Yacaman, M.,Etcheberry, A., Colbeau-Justin, C., and Remita, H. (2015) Visiblelight-induced photocatalytic activity of modified titanium(IV) oxide withzero valent bismuth clusters. New J. Chem., 39, 2316–2322.

17 Kale, M.J., Avanesian, T., and Christopher, P. (2014) Direct photocatalysis byplasmonic nanostructures. ACS Catal., 4, 116–128.

18 And, X.J. and Xiao, W.-J. (2012) Visible-light photoredox catalysis. Angew.Chem. Int. Ed., 51, 2–13.

19 Wang, Y., Wang, Q., Zhan, X., Wang, F., Safdar, M., and He, J. (2013) Visiblelight driven type II heterostructures and their enhanced photocatalysis prop-erties: a review. Nanoscale, 5, 8326–8339.

20 Pelaez, M., Nolan, N., Pillai, S., Seery, M., and Falaras, P. (2012) A reviewon the visible light active titanium dioxide photocatalysts for environmentalapplications. Appl. Catal. B, 125, 331–349.

21 Kositzi, M., Poulios, I., Malato, S., Caceres, J., and Campos, A. (2004) Solarphotocatalytic treatment of synthetic municipal wastewater. Water Res., 38,1147–1154.

22 Reddy, P.A.K., Reddy, P.V.L., Kwon, E., Kim, K.-H., Akter, T., and Kalagar, S.(2016) Recent advances in photocatalytic treatment of pollutants in aqueousmedia. Environ. Int., 91, 94–103.

23 Fujishima, A., Zhang, X., and Tryk, D.A. (2007) Heterogeneous photocatal-ysis: From water photolysis to applications in environmental cleanup. Int. J.Hydrogen Energy, 32, 2664–2672.

24 Kubacka, A., Fernandez-Garcia, M., and Colon, G. (2012) Advancednanoarchitectures for solar photocatalytic applications. Chem. Rev., 112,1555–1614.

25 Kamat, P.V. (2012) TiO2 nanostructures: recent physical chemistry advances.J. Phys. Chem. Lett., 3, 663–672.

26 Li, L., Yan, J., Wang, T., Zhao, Z.-J., Zhang, J., Gong, J., and Guan, N. (2015)Sub-10 nm rutile titaniumdioxide nanoparticles for efficient visible-light-driven photocatalytic hydrogen production. Nat. Commun., 6, 5881–5891.

27 Tao, J., Luttrell, T., and Batzill, M. (2011) A two-dimensional phase of TiO2with a reduced bandgap. Nat. Chem., 3, 296–300.

28 Linic, S., Christopher, P., and Ingram, D.B. (2011) Plasmonic-metal nanos-tructures for efficient conversion of solar to chemical energy. Nat. Mater.,10, 911–921.

29 Grabowska, E., Zaleska, A., Sorgues, S., Kunst, M., Etcheberry, A.,Colbeau-Justin, C., and Remita, H. (2013) Modification of titanium(IV)dioxide with small silver nanoparticles: application in photocatalysis. J. Phys.Chem. C, 117, 1955–1962.

30 Hai, Z., Kolli, N.E., Uribe, D.B., Beaunier, P., José-Yacaman, M., Vigneron, J.,Etcheberry, A., Sorgues, S., Colbeau-Justin, C., And, C.J., and Remita, H.(2013) Modification of TiO2 by bimetallic Au—Cu nanoparticles for wastewater treatment. J. Mater. Chem. A., 1, 10829–10835.

31 Chen, X., Liu, L., Yu, P.Y., and Mao, S.S. (2011) Increasing solar absorptionfor photocatalysis with black hydrogenated titanium dioxide nanocrystals.Science, 331, 746–750.

References 17

32 Liu, X., Zhu, G., Wang, X., Yuan, X., Lin, T., and Huang, F. (2016) Progressin black titania: a new material for advanced photocatalysis. Adv. EnergyMater., 6, 1600452.

33 Ullattil, S.G. and Periyat, P. (2015) Green microwave switching from oxygenrich yellow anatase to oxygen vacancy rich black anatase TiO2 solar photo-catalyst using Mn(II) as ‘anatase phase purifier’. Nanoscale, 7, 19184–19192.

34 Etacheri, V., Seery, M.K., Hinder, S.J., and Pillai, S.C. (2011) Oxygen richtitania: a dopant free, high temperature stable, and visible-light activeanatase photocatalyst. Adv. Funct. Mater., 21, 3744–3752.

35 Paek, M.-J., Woo Kim, T., and Hwang, S.-J. (2008) Effects of hydroniumintercalation and cation substitution on the photocatalytic performance oflayered titanium oxide. J. Phys. Chem. Solids, 69, 1444–1446.

36 Kong, X., Zeng, C., Wang, X., Huang, J., Li, C., Fei, J., Li, J., and Feng, Q.(2016) Ti—O—O coordination bond caused visible light photocatalyticproperty of layered titanium oxide. Sci. Rep., 6, 29049.

37 Gao, S.Y., Ueno, K., and Misawa, H. (2011) Plasmonic antenna effects onphotochemical reactions. Acc. Chem. Res., 44, 251–260.

38 Tian, Y. and Tatsuma, T. (2005) Mechanisms and applications ofplasmon-induced charge separation at TiO2 films loaded with gold nanopar-ticles. J. Am. Chem. Soc., 127, 7632–7637.

39 Liu, Z.W., Hou, W.B., Pavaskar, P., Aykol, M., and Cronin, S.B. (2011) Plas-mon resonant enhancement of photocatalytic water splitting under visibleillumination. Nano Lett., 11, 1111–1116.

40 Ingram, D.B. and Linic, S. (2011) Water splitting on compositeplasmonic-metal/semiconductor photoelectrodes: evidence for selectiveplasmon-induced formation of charge carriers near the semiconductorsurface. J. Am. Chem. Soc., 133, 5202–5205.

41 Khalil, L.B., Mourad, W.E., and Rophael, M.W. (1998) Photo-catalytic reduc-tion of environmental pollutant Cr(VI) over some semiconductors underUV/visible light illumination. Appl. Catal. B, 17, 267–273.

42 Garza-Arévalo, J.I., García Montes, I., Hinojosa Reyes, M., Guzmán Mar, J.L.,Rodríguez González, V., and Reyes, L.H. (2016) Fe doped TiO2 photocat-alyst for the removal of As(III) under visible radiation and its potentialapplication on the treatment of As-contaminated ground water. Mater. Res.Bulletin, 73, 145–152.

43 Wu, Q., Zhao, J., Qin, G., Wang, C., Tong, X., and Xue, S. (2013) Photocat-alytic reduction of Cr(VI) with TiO2 film under visible light. Appl. Catal. B,142, 142–148.

44 Cohen, M.D., Kargacin, B., Klein, C.B., and Costa, M. (1993) Mechanisms ofchromium carcinogenicity and toxicity. Crit. Rev. Toxicol., 23, 255–281.

45 Jacobs, J.A. and Testa, S.M. (2005) Overview of chromium(VI) in theenvironment: background and history, in Chromium (VI) Handbook (edsJ. Guertin, J.A. Jacobs, and C.P. Avakian), CRC Press, Boca Raton, FL,pp. 1–22.

46 AlOtaibi, B., Fan, S., Wang, D., Ye, J., and Mi, Z. (2015) Wafer-level artificialphotosynthesis for CO2 reduction into CH4 and CO using GaN nanowires.ACS Catal., 5, 5342–5348.


47 Inoue, T., Fujishima, A., Konishi, S., and Honda, K. (1979) Photoelectrocat-alytic reduction of carbon dioxide in aqueous suspensions of semiconductorpowders. Nature, 277, 637–638.

48 Tu, W., Zhou, Y., and Zou, Z. (2014) Photocatalytic Conversion of CO2 intorenewable hydrocarbon fuels: state-of-the-art accomplishment, challenges,and prospects. Adv. Mater., 26, 4607–4626.

49 Habisreutinger, S.N., Schmidt-Mende, L., and Stolarczyk, J.K. (2013) Pho-tocatalytic reduction of CO2 on TiO2 and other semiconductors. Angew.Chem. Int. Ed., 52, 7372–7408.

50 Kondratenko, E.V., Mul, G., Baltrusaitis, J., Larrazábal, G.O., andPérez-Ramírez, J. (2013) Status and perspectives of CO2 conversion intofuels and chemicals by catalytic, photocatalytic and electrocatalytic pro-cesses. Energy Environ. Sci., 6, 3112–3135.

51 Kowalska, E., Wei, Z., Karabiyik, B., Herissan, A., Janczarek, M., Endo, M.,Markowska-Szczupak, A., Remita, H., and Ohtani, B. (2015) Silver-modifiedtitania with enhanced photocatalytic and antimicrobial properties under UVand visible light irradiation. Catal. Today, 252, 136–142.

52 Awazu, K., Fujimaki, M., Rockstuhl, C., Tominaga, J., Mu- rakami, H.,Ohki, Y., Yoshida, N., and Watanabe, T. (2008) A plasmonic photocata-lyst consisting of silver nanoparticles embedded in titanium dioxide. J. Am.Chem. Soc., 130, 1676–1680.

53 Wang, P., Huang, B. B., Qin, X. Y., Zhang, X. Y., Dai, Y., Wei, J. andWhangbo, M.-H. (2008) Ag@AgCl: A highly efficient and stablephotocatalyst active under visible light. Angew. Chem. Int. Ed., 47,7931–7933.

54 Ghosh, S., Kouamé, N.A., Ramos, L., Remita, S., Dazzi, A.,Deniset-Besseau, A., Beaunier, P., Goubard, F., Aubert, P.-H., and Remita,H. (2015) Conducting polymer nanostructures for photocatalysis undervisible light. Nat. Mater., 14, 505–511.

55 Ghosh, S., Kouame, N.A., Remita, S., Ramos, L., Goubard, F., Aubert, P.-H.,Dazzi, A., Deniset-Besseau, A., and Remita, H. (2015) Visible-light activeconducting polymer nanostructures with superior photocatalytic activity. Sci.Rep., 5, 18002–18011.

56 Floresyona, D., Goubard, F., Aubert, P.-H., Lampr, I., Mathurin, J., Dazzi,A., Ghosh, S., Beaunier, P., Brisset, F., Remita, S., Ramos, L., and Remita, H.(2017) Highly active poly(3-hexylthiophene) nanostructures for photocataly-sis under solar light. Appl. Catal. B, 209, 23–32.

57 Marschall, R. (2014) Semiconductor composites: strategies for enhancingcharge carrier separation to improve photocatalytic activity. Adv. Funct.Mater., 24, 2421–2440.

58 Tran, P.D., Wong, L.H., Barber, J., and Loo, J.S.C. (2012) Recent advancesin hybrid photocatalysts for solar fuel production. Energy Environ. Sci., 5,5902–5918.

59 Yuan, Y.-P., Ruan, L.-W., Barber, J., Chye, S., Loo, J., and Xue, C. (2014)Hetero-nanostructured suspended photocatalysts for solar-to-fuel conver-sion. Energy Environ. Sci., 7, 3934–3951.

References 19

60 Park, Y., McDonald, K.J., and Choi, K.S. (2013) Progress in bismuth vana-date photoanodes for use in solar water oxidation. Chem. Soc. Rev., 42,2321–2337.

61 Li, L., Salvador, P.A., and Rohrer, G.S. (2014) Photocatalysts with internalelectric fields. Nanoscale, 6, 24–42.

62 Zou, J.P., Lei, S.L., Yu, J., Luo, S.L., Luo, X.B., Tang, X.H., Dai, W.L., Sun, J.,Guo, G.C., and Au, C.T. (2014) Highly efficient and stable hydrogen evolu-tion from water with CdS as photosensitizer – a noble-metal-free system.Appl. Catal. B, 150–151, 466–471.

63 Park, H., Choi, W., and Hoffmann, M.R. (2008) Effects of the prepara-tion method of the ternary CdS/TiO2/Pt hybrid photocatalysts on visiblelight-induced hydrogen production. J. Mater. Chem., 18, 2379–2385.

64 Wang, C.J., Thompson, R.L., Baltrus, J., and Matranga, C. (2010) Visiblelight photoreduction of CO2 using CdSe/Pt/TiO2 heterostructured catalysts.J. Phys. Chem. Lett., 1, 48–53.

65 Wang, Z.L., Liu, Y.X., Martin, D.J., Wang, W.D., Tang, J.W., and Huang, W.X.(2013) CuOx—TiO2 junction: what is the active component for photocat-alytic H2 production? Chem. Chem. Phys., 15, 14956–14960.

66 Asi, M.A., He, C., Su, M.H., Xia, D.H., Lin, L., Deng, H.Q., Xiong, Y.,Qiu, R.L., and Li, X.Z. (2011) Photocatalytic reduction of CO2 to hydro-carbons using AgBr/TiO2 nanocomposites under visible light. Catal. Today,175, 256–263.

67 Wang, C.J., Thompson, R.L., Ohodnicki, P., Baltrus, J., and Matranga, C.(2011) Size-dependent photocatalytic reduction of CO2 with PbS quantumdot sensitized TiO2 heterostructured photocatalysts. J. Mater. Chem., 21,13452–13457.

68 Sardar, S., Kar, P., Remita, H., Liu, B., Lemmens, P., Pal, S.K., and Ghosh, S.(2015) Enhanced charge separation and FRET at heterojunctions betweensemiconductor nanoparticles and conducting polymer nanofibers for efficientsolar light harvesting. Sci. Rep., 5, 17313–17327.

69 Razgoniaeva, N., Moroz, P., Lambright, S., and Zamkov, M. (2015) Photo-catalytic applications of colloidal heterostructured nanocrystals: what’s next?J. Phys. Chem. Lett., 6, 4352–4359.

70 Basu, M., Garg, N., and Ganguli, A.K. (2014) A type-II semiconduc-tor (ZnO/CuS heterostructure) for visible light photocatalysis. J. Mater.Chem. A, 2, 7517–7525.

71 Gao, M.-R., Xu, Y.-F., Jiang, J., and Yu, S.-H. (2013) Nanostructured metalchalcogenides: synthesis, modification, and applications in energy conversionand storage devices. Chem. Soc. Rev., 42, 2986–3017.

72 Wang, J.G., Li, X.R., Zhu, J., Li, X., and Li, H. (2013) Mesoporous yolk−shellSnS2—TiO2 visible photocatalysts with enhanced activity and durability inCr(VI) reduction. Nanoscale, 5, 1876–1881.

73 Ghosh, S., Holade, Y., Remita, H., Servat, K., Beaunier, P., Hagège, A.,Kokoh, B. and Napporn, T. W. (2016) Facile synthesis of surfactant-freegraphene supported gold based nanocomposites: Robust catalysts for glucoseoxidation. Electrochim. Acta, 212, 864–875.


74 Sardar, S., Ghosh, S., Remita, H., Kar, P., Liu, B., Lemmens, P., and Pal, S.K.(2016) Reduced graphene oxides–metal composites as counter electrode indye-sensitized solar cells. RSC Adv., 6, 33433–33442.

75 Jiang, X., Nisar, J., Pathak, B., Zhao, J., and Ahuja, R. (2013) Grapheneoxide as a chemically tunable 2-D material for visible-light photocatalystapplications. J. Catal., 299, 204–209.

76 Dreyer, D.R., Park, S., Bielawski, C.W., and Ruoff, R.S. (2010) The chemistryof graphene oxide. Chem. Soc. Rev., 39, 228–240.

77 Low, J., Yu, J., and Ho, W. (2015) Graphene-based photocatalysts for CO2reduction to solar fuel. J. Phys. Chem. Lett., 6, 4244–4251.

78 Yang, M.-Q. and Xu, Y.-J. (2016) Photocatalytic conversion of CO2 overgraphene-based composites: current status and future perspective. NanoscaleHoriz., 1, 185–200.

79 Liang, Y.T., Vijayan, B.K., Gray, K.A., and Hersam, M.C. (2011) Minimiz-ing graphene defects enhances titania nanocomposite-based photocat-alytic reduction of CO2 for improved solar fuel production. Nano Lett.,11, 2865–2870.

80 Yu, J.G., Jin, J., Cheng, B., and Jaroniec, M. (2014) A noble metal-freereduced graphene oxide–CdS nanorod composite for the enhancedvisible-light photocatalytic reduction of CO2 to solar fuel. J. Mater. Chem. A,2, 3407–3416.

81 Meng, F., Li, J., Cushing, S.K., Bright, J., Zhi, M., Rowley, J.D., Hong, Z.,Manivannan, A., Bristow, A.D., and Wu, N. (2013) Photocatalytic wateroxidation by hematite/reduced graphene oxide composites. ACS Catal., 3,746–751.

82 Li, Z., Shen, Y., Guan, Y., Hu, Y., Lin, Y., and Nan, C.W. (2014) Bandgapengineering and enhanced interface coupling of graphene–BiFeO3 nanocom-posites as efficient photocatalysts under visible light. J. Mater. Chem. A, 2,1967–1973.

83 Yang, Y., Ren, L., Zhang, C., Huang, S., and Liu, T. (2011) Facile fabrica-tion of functionalized graphene sheets (FGS)/ZnO nanocomposites withphotocatalytic property. ACS Appl. Mater. Interfaces, 3, 2779–2785.

84 Zhang, N., Zhang, Y., Pan, X., Fu, X., Liu, S., and Xu, Y.J. (2011) Assem-bly of CdS nanoparticles on the two-dimensional graphene scaffold asvisible-light-driven photocatalyst for selective organic transformation underambient conditions. J. Phys. Chem. C, 115, 23501–23511.

85 Han, W., Ren, L., Qi, X., Liu, Y., Wei, X., Huang, Z., and Zhong, J. (2014)Synthesis of CdS/ZnO/graphene composite with high-efficiency photoelec-trochemical activities under solar radiation. Appl. Surf. Sci., 299, 12–18.

86 Zong, X., Yan, H., Wu, G., Ma, G., Wen, F., Wang, L., and Li, C. (2008)Enhancement of photocatalytic H2 evolution on CdS by loading MoS2 ascocatalyst under visible light irradiation. J. Am. Chem. Soc., 130, 7176–7177.

87 Chang, K., Mei, Z., Wang, T., Kang, Q., Ouyang, S., and Ye, J. (2014)MoS2/graphene cocatalyst for efficient photocatalytic H2 evolution undervisible light irradiation. ACS Nano, 8, 7078–7087.

88 Cravanzola, S., Muscuso, L., Cesano, F., Agostini, G., Damin, A., Scarano, D.,and Zecchina, A. (2015) MoS2 nanoparticles decorating titanate-nanotube

References 21

surfaces: combined microscopy, spectroscopy, and catalytic studies.Langmuir, 31, 5469–5478.

89 Zhou, W., Yin, Z., Du, Y., Huang, X., Zeng, Z., Fan, Z., Liu, H., Wang, J.,and Zhang, H. (2013) Synthesis of few-layer MoS2 nanosheet coated TiO2nanobelt heterostructures for enhanced photocatalytic activities. Small, 9,140–147.

90 Xia, J., Ge, Y., Zhao, D., Di, J., Ji, M., Yin, S., Li, H., and Chen, R. (2015)Microwave-assisted synthesis of few-layered MoS2/BiOBr hollow micro-spheres with superior visible-light-response photocatalytic activity forciprofloxacin removal. Cryst.EngComm., 17, 3645–3651.

91 Wang, X., Maeda, K., Thomas, A., Takanabe, K., Xin, G., Carlson, J.M.,Domen, K., and Antonietti, M. (2009) A metal-free polymeric photocata-lyst for hydrogen production from water under visible light. Nat. Mater., 8,76–80.

92 Kroke, E., Schwarz, M., Horath-Bordon, E., Kroll, P., Noll, B., andNorman, A.D. (2002) Tri-s–tri-s-triazine to graphitic C3N4 structures. New J.Chem., 26, 508–512.

93 Zhang, Y., Mon, T., and Ye, J. (2012) Polymeric carbon nitrides: semi-conducting properties and emerging applications in photocatalysis andphotoelectrochemical energy conversion. Sci. Adv. Mater., 4, 282–291.

94 Maeda, K., Kuriki, R., Zhang, M., Wang, X., and Ishitani, O. (2014) Theeffect of the pore-wall structure of carbon nitride on photocatalytic CO2reduction under visible light. J. Mater. Chem. A, 2, 15146–11151.

95 Zhou, S., Liu, Y., Li, J., Wang, Y., Jiang, G., Zhao, Z., Wang, D., Duan, A.,Liu, J., and Wei, Y. (2014) Facile in situ synthesis of graphitic carbon nitride(g-C3N4)—N—TiO2 heterojunction as an efficient photocatalyst for theselective photoreduction of CO2 to CO. Appl. Catal. B, 157–158, 20–29.

96 He, Y., Wang, Y., Zhang, L., Teng, B., and Fan, M. (2015) High-efficiencyconversion of CO2 to fuel over ZnO/g-C3N4 photocatalyst. Appl. Catal. B,167–168, 1–8.

97 Cao, S., Liu, X., Yuan, Y., Zhang, Z., Liao, Y., Fang, J., Loo, S.C.J., Sum, T.C.,and Xue, C. (2014) Solar-to-fuels conversion over In2O3/g-C3N4 hybridphotocatalysts. Appl. Catal., B, 146–147, 940–946.

98 Li, M., Zhang, L., Fan, X., Zhou, Y., Wu, M., and Shi, J. (2015) Highly selec-tive CO2 photoreduction to CO over g-C3N4/Bi2WO6 composites undervisible light. J. Mater. Chem. A, 3, 5189–5196.

99 Vinodgopal, K., Wynkoop, D. E., and Kamat, P. V. (1996) Environmentalphotochemistry on semiconductor surfaces: photosensitized degradation of atextile azo dye, Acid Orange 7, on TiO2 particles using visible light. Environ.Sci. Technol., 30, 1660–1666.

100 Duan, M.-y., Li, J., Mele, G., Wang, C., Lü, X.-f., Vasapollo, G. andZhang, F.-x. (2010) Photocatalytic activity of novel tin porphyrin/TiO2 basedcomposites. J. Phys. Chem. C, 114, 7857−62.

101 Oregan, B. and Gratzel. M. (1991) A low-cost, high-efficiency solar-cellbased on dye-sensitized colloidal TiO2 films. Nature, 353, 737−740.

102 Gratzel, M. (1981) Artificial photosynthesis: water cleavage into hydrogenand oxygen by visible light. Acc. Chem. Res., 14, 376–384.


103 Chen, X.B., Shen, S.H., Guo, L.J., and Mao, S.S. (2010) Semiconductor-basedphotocatalytic hydrogen generation. Chem. Rev., 110, 6503–6570.

104 Kudo, A. and Miseki, Y. (2009) Heterogeneous photocatalyst materials forwater splitting. Chem. Soc. Rev., 38, 253–278.

105 Zou, Z., Ye, J., Sayama, K., and Arakawa, H. (2001) Direct splitting of waterunder visible light irradiation with an oxide semiconductor photocatalyst.Nature, 414, 625–627.

106 Ni, M., Leung, M.K.H., Leung, D.Y.C., and Sumathy, K. (2007) A reviewand recent developments in photocatalytic water-splitting using TiO2 forhydrogen production. Renew. Sustain. Energy Rev., 11, 401–425.

107 Sivula, K. and van de Krol, R. (2016) Semiconducting materials for photo-electrochemical energy conversion. Nat. Rev. Mater., 70, 15010:1–15010:7.

108 Kay, A., Cesar, I., and Gratzel, M. (2006) New benchmark for water pho-tooxidation by nanostructured 𝛼-Fe2O3 Films. J. Am. Chem. Soc., 128,15714–15721.

109 Liang, Y., Tsubota, T., Mooij, L.P.A., and van de Krol, R. (2011) Highlyimproved quantum efficiencies for thin film BiVO4 photoanodes. J. Phys.Chem. C, 115, 17594–17598.

110 Chen, Y., Zhao, S., Wang, X., Peng, Q., Lin, R., Wang, Y., Shen, R., Cao, X.,Zhang, L., Zhou, G., Li, J., Xia, A., and Li, Y. (2016) Synergetic integrationof Cu1.94S—ZnxCd1−xS heteronanorods for enhanced visible-light-drivenphotocatalytic hydrogen production. J. Am. Chem. Soc., 138,4286–4289.

111 Bao, J. (2015) Photoelectrochemical water splitting: a new use for bandgapengineering. Nat. Nanotechnol., 10, 19–20.

112 Li, J., Hoffmann, M.W.G., Shen, H., Fabrega, C., Prades, J.D., Andreu, T.,Hernandez-Ramirez, F., and Mathur, S. (2012) Enhanced photoelectrochem-ical activity of an excitonic staircase in CdS@TiO2 and CdS@anatase@rutileTiO2 heterostructures. J. Mater. Chem., 22, 20472–20476.

113 Qiu, F., Han, Z.P., Jeffrey, J.O., Michael, Y.S., Kelly, L., and Krauss, T.D.(2016) Photocatalytic hydrogen generation by CdSe/CdS nanoparticles. NanoLett., 16, 5347–5352.

114 Amirav, L. and Alivisatos, A.P. (2010) Photocatalytic hydrogen pro-duction with tunable nanorod heterostructures. J. Phys. Chem. Lett., 1,1051–1054.

115 Zhang, J., Zhu, Z., Tang, Y., Müllen, K., and Feng, X. (2014) Titaniananosheet-mediated construction of a two-dimensional titania/cadmiumsulfide heterostructure for high hydrogen evolution activity. Adv Mater., 26,734–738.

116 Cao, S., Chen, Y., Wang, C.-J., Lv, X.-J., and Fu, W.-F. (2015) Spectacu-lar photocatalytic hydrogen evolution using metal-phosphide/CdS hybridcatalysts under sunlight irradiation. Chem. Commun., 51, 8708–8711.

117 Kozlova, E.A., Kurenkova, A.Y., Semeykina, V.S., Parkhomchuk, E.V.,Cherepanova, S.V., Gerasimov, E.Y., Saraev, A.A., Kaichev, V.V., andParmon, V.N. (2015) Effect of titania regular macroporosity on the pho-tocatalytic hydrogen evolution on Cd1−xZnxS/TiO2 catalysts under visiblelight. ChemCatChem., 7, 4108–4117.

References 23

118 Zong, X., Wu, G., Yan, H., Ma, G., Shi, J., Wen, F., Wang, L., and Li, C.(2010) Photocatalytic H2 evolution on MoS2/CdS catalysts under visible lightirradiation. J. Phys. Chem. C, 114, 1963–1972.

119 Hu, J., Liu, A., Jin, H., Ma, D., Yin, D., Ling, P., Wang, S., Lin,Z., and Wang, J. (2015) A versatile strategy for shish-kebab-likemulti-heterostructured chalcogenides and enhanced photocatalytic hydrogenevolution. J. Am. Chem. Soc., 137, 11004–11010.

120 Shen, M., Yan, Z., Yang, L., Du, P., Zhang, J., and Xiang, B. (2014) MoS2nanosheet/TiO2 nanowire hybrid nanostructures for enhanced visible-lightphotocatalytic activities. Chem. Commun., 50, 15447–15449.

121 Thomas, A., Fischer, A., Goettmann, F., Antonietti, M., Muller, J.O.,Schlogl, R., and Carlsson, J.M. (2008) Graphitic carbon nitride materials:variation of structure and morphology and their use as metal-free catalysts.J. Mater. Chem., 18, 4893–4908.

122 Hu, J., Cheng, W., Huang, S., Wu, D., and Xie, Z. (2006) First-principlesmodelling of nonlinear optical properties of C3N4 polymorphs. Appl. Phys.Lett., 89, 261117.

123 Chen, Z., Sun, P., Fan, B., Zhang, Z., and Fang, X. (2014) In Situtemplate-free ion-exchange process to prepare visible-light-activeg-C3N4/NiS hybrid photocatalysts with enhanced hydrogen evolution activ-ity. J. Phys. Chem. C, 118, 7801–7807.

124 Hou, Y., Laursen, A.B., Zhang, J., Zhang, G., Zhu, Y., Wang, X., Dahl, S.,and Chorkendorff, I. (2013) Layered nanojunctions for hydrogen-evolutioncatalysis. Angew. Chem. Int. Ed., 52, 3621–3625.

125 Hou, Y., Zhu, Y., Xu, Y., and Wang, X. (2014) Photocatalytic hydrogen pro-duction over carbon nitride loaded with WS2 as cocatalyst under visiblelight. Appl. Catal., B, 156–157, 122–127.

126 Yu, J., Wang, S., Cheng, B., Lin, Z., and Huang, F. (2013) Noble metal-freeNi(OH)2—g-C3N4 composite photocatalyst with enhanced visible-lightphotocatalytic H2-production activity. Catal. Sci. Technol., 3, 1782–1789.

127 Meng, S., Ning, X., Zhang, T., Chen, S.-F., and Fu, X. (2015) What is thetransfer mechanism of photogenerated carriers for the nanocomposite pho-tocatalyst Ag3PO4/g-C3N4, band–band transfer or a direct Z-scheme? Phys.Chem. Chem. Phys., 17, 11577–11585.

128 Sasaki, Y., Nemoto, H., Saito, K., and Kudo, A. (2009) Solar water splittingusing powdered photocatalysts driven by Z-schematic interparticle electrontransfer without an electron mediator. J. Phys. Chem. C, 113, 17536–17542.

129 Maeda, K. (2013) Z-scheme water splitting using two different semiconduc-tor photocatalysts. ACS Catal., 3, 1486–1503.

130 Wang, X., Liu, G., Wang, L., Chen, Z.G., Lu, G.Q., and Cheng, H.M. (2012)ZnO—CdS@Cd heterostructure for effective photocatalytic hydrogen genera-tion. Adv. Energy Mater., 2, 42–46.

131 Zhou, P., Yu, J., and Jaroniec, M. (2014) All-solid-state Z-scheme photocat-alytic systems. Adv. Mater., 26, 4920–4935.

132 Tada, H., Mitsui, T., Kiyonaga, T., Akita, T., and Tanaka, K. (2006)Solid-state Z-scheme in CdS-Au-TiO2 three-component nanojunctionsystem. Nat. Mater., 5, 782–786.


133 Yeh, T.-F., Cihlá, J., Chang, C.-Y., Cheng, C., and Teng, H. (2013) Roles ofgraphene oxide in photocatalytic water splitting. Mater. Today, 16, 78–84.

134 Iwase, A., Ng, Y.H., Ishiguro, Y., Kudo, A., and Amal, R. (2011) Reducedgraphene oxide as a solid-state electron mediator in Z-scheme photocatalyticwater splitting under visible light. J. Am. Chem. Soc., 133, 11054–11057.

135 Maeda, K., Higashi, M., Lu, D., Abe, R., and Domen, K. (2010) Efficient non-sacrificial water splitting through two-step photoexcitation by visible lightusing a modified oxynitride as a hydrogen evolution photocatalyst. J. Am.Chem. Soc., 132, 5858–5868.

136 Dhanasekaran, P., Salunke, H.G., and Gupta, N.M. (2012)Visible-light-induced photosplitting of water over 𝛾 ′-Fe4N and𝛾′- Fe4N/𝛼-Fe2O3 nanocatalysts. J. Phys. Chem. C, 116, 12156–12164.

137 Abe, R., Higashi, M., and Domen, K. (2011) Overall water splitting undervisible light through a two-step photoexcitation between TaON and WO3 inthe presence of an iodateiodide shuttle redox mediator. ChemSusChem, 4,228–237.

138 Higashi, M., Abe, R., Ishikawa, A., Takata, T., and Ohtani, B. (2008)Z-scheme overall water splitting on modified-TaON photocatalysts undervisible light (500 nm). Chem. Lett., 37, 138–139.

139 Yun, H.J., Lee, H., Kim, N.D., Lee, D.M., Yu, S., and Yi, J. (2011) A com-bination of two visible-light responsive photocatalysts for achieving theZ-Scheme in the solid state. ACS Nano, 5, 4084–4090.

140 Zeitler, K. (2009) Photoredox catalysis with visible light. Angew. Chem. Int.Ed., 48, 9785–9789.

141 Narayanam, J.M.R. and Stephenson, C.R.J. (2011) Visible light photoredoxcatalysis: applications in organic synthesis. Chem. Soc. Rev., 40, 102–113.

142 Lang, X.C. and Zhao, J. (2014) Heterogeneous visible light photocatalysis forselective organic transformations. Chem. Soc. Rev., 43, 473–486.

143 Ravelli, D., Fagnoni, M., and Albini, A. (2013) Photoorganocatalysis. Whatfor? Chem. Soc. Rev., 42, 97–113.

144 Kim, G. and Choi, W. (2010) Charge-transfer surface complex ofEDTA–TiO2 and its effect on photocatalysis under visible light. Appl.Catal. B, 100, 77–83.

145 Park, Y., Singh, N.J., Kim, K.S., Tachikawa, T., Majima, T., and Choi, W.(2009) Fullerol–Titania charge-transfer-mediated photocatalysis workingunder visible light. Chem.-Eur. J., 15, 10843–10850.

146 Higashimoto, S., Kitao, N., Yoshida, N., Sakura, T., Azuma, M., Ohue, H.,and Sakata, Y. (2009) Selective photocatalytic oxidation of benzyl alcoholand its derivatives into corresponding aldehydes by molecular oxygen ontitanium dioxide under visible light irradiation. J. Catal., 266, 279–285.

147 Lang, X.J., Ma, W.H., Zhao, Y.B., Chen, C.C., Ji, H.W., and Zhao, J.C. (2012)Visible-light-induced selective photocatalytic aerobic oxidation of aminesinto imines on TiO2. Chem.-Eur. J., 18, 2624–2631.

148 Furukawa, S., Shishido, T., Teramura, K., and Tanaka, T. (2011) Photocat-alytic oxidation of alcohols over TiO2 covered with Nb2O5. ACS Catal., 2,175–179.

References 25

149 Sarina, S., Zhu, H.Y., Zheng, Z.F., Bottle, S., Chang, J., Ke, X.B., Zhao, J.C.,Huang, Y.N., Sutrisno, A., Willans, M., and Li, G.R. (2012) Driving selectiveaerobic oxidation of alkyl aromatics by sunlight on alcohol grafted metalhydroxides. Chem. Sci., 3, 2138–2146.

150 Tsukamoto, D., Shiraishi, Y., Sugano, Y., Ichikawa, S., Tanaka, S., andHirai, T. (2012) Gold nanoparticles located at the interface of anatase/rutileTiO2 particles as active plasmonic photocatalysts for aerobic oxidation. J.Am. Chem. Soc., 134, 6309–6315.

151 Naya, S., Inoue, A., and Tada, H. (2010) Self-assembled heterosupramolec-ular visible light photocatalyst consisting of gold nanoparticle-loadedtitanium(IV) dioxide and surfactant. J. Am. Chem. Soc., 132, 6292–6293.

152 Tanaka, A., Hashimoto, K., and Kominami, H. (2011) Selective photocatalyticoxidation of aromatic alcohols to aldehydes in an aqueous suspension ofgold nanoparticles supported on cerium(IV) oxide under irradiation of greenlight. Chem. Commun., 47, 10446–10448.

153 Su, F. Z., Mathew, S. C., Lipner, G., Fu, X. Z., Antonietti, M., Blechert, S.and Wang, X. C. (2010) mpg-C3N4-catalyzed selective oxidation of alcoholsusing O2 and visible light. J. Am. Chem. Soc., 2010, 132, 16299–16301.

154 Zheng, Z.S. and Zhou, X.S. (2012) Metal-free oxidation of 𝛼-hydroxy ketonesto 1,2-diketones catalyzed by mesoporous carbon nitride with visible light.Chin. J. Chem., 30, 1683–1686.

155 Möhlmann, L., Baar, M., Rieß, J., Antonietti, M., Wang, X.C., and Blechert,S. (2012) Carbon nitride-catalyzed photoredox C—C bond formation withn-aryltetrahydroisoquinolines. Adv. Synth. Catal., 354, 1909–1913.

156 Zhang, P.F., Wang, Y., Li, H.R., and Antonietti, M. (2012) Metal-free oxi-dation of sulfides by carbon nitride with visible light illumination at roomtemperature. Green Chem., 14, 1904–1908.

157 Nicewicz, D.A. and MacMillan, D.W.C. (2008) Merging photoredox catalysiswith organocatalysis: the direct asymmetric alkylation of aldehydes. Science,322, 77–80.

158 Shih, H.-W., Vander Wal, M.N., Grange, R.L., and MacMillan, D.W.C. (2010)Enantio selective 𝛼-benzylation of aldehydes via photoredox organocatalysis.J. Am. Chem. Soc., 132, 13600–13603.

159 Neumann, M., Füldner, S., König, B., and Zeitler, K. (2011) Metal-free, coop-erative asymmetric organophotoredox catalysis with visible light. Angew.Chem. Int. Ed., 50, 951–954.

160 Bauer, A., Westkämper, F., Grimme, S., and Bach, T. (2005) Catalytic enan-tioselective reactions driven by photoinduced electron transfer. Nature, 436,1139–1140.

161 Müller, C., Bauer, A., and Bach, T. (2009) Light-driven enantioselectiveorganocatalysis. Angew. Chem. Int. Ed., 48, 6640–6642.

162 Huo, H., Shen, X., Wang, C., Zhang, L., Röse, P., Chen, L.-A., Harms, K.,Marsch, M., Hilt, G., and Meggers, E. (2014) Nature, 515, 100–103.

163 Banerjee, S., Pillai, S.C., Falaras, P., O’Shea, K.E., Byrne, J.A., andDionysiou, D.D. (2014) New insights into the mechanism of visible lightphotocatalysis. J. Phys. Chem. Lett., 5, 2543–2554.


164 Nosaka, Y., Nishikawa, M., and Nosaka, A.Y. (2014) Spectroscopic investiga-tion of the mechanism of photocatalysis. Molecules, 19, 18248–18267.

165 Walsh, A., Yan, Y., Huda, M.N., Al-Jassim, M.M., and Wei, S.-H. (2009) Bandedge electronic structure of BiVO4: elucidating the role of the Bi s and V dorbitals. Chem. Mater., 21, 547–551.

166 Nazeeruddin, M.K., De Angelis, F., Fantacci, S., Selloni, A., Viscardi, G.,Liska, P., Ito, S., Takeru, B., and Grätzel, M. (2005) Combined experimen-tal and DFT–TDDFT computational study of photoelectrochemical cellruthenium sensitizers. J. Am. Chem. Soc., 127, 16835–16847.

167 Nayak, P.K. and Periasamy, N. (2009) Calculation of electron affinity, ion-ization potential, transport gap, optical band gap and exciton bindingenergy of organic solids using “solvation” model and DFT. Org. Electron.,10, 1396–1400.

168 Chai, S., Wen, S.-H., Huang, J.-D., and Han, K.-L. (2011) Density functionaltheory study on electron and hole transport properties of organic pentacenederivatives with electron-withdrawing substituent. J. Comput. Chem., 32,3218–3225.

169 De Angelis, F., Fantacci, S., and Selloni, A. (2008) Alignment of the dye’smolecular levels with the TiO2 band edges in dye-sensitized solar cells:a DFT–TDDFT study. Nanotechnology, 19, 424002–424008.

170 Wendt, S., Seitsonen, A., Kim, Y., Knapp, M., Idriss, H., and Over, H. (2002)Complex redox chemistry on the RuO2 (100) surface: experiment and the-ory. Surf. Sci., 505, 137–152.

171 Chen, J., Zhang, H., Tomov, I.V., Wolfsberg, M., Ding, X., andRentzepis, P.M. (2007) Transient structures and kinetics of the ferrioxalateredox reaction studied by time-resolved EXAFS, optical spectroscopy, andDFT. J. Phys. Chem. A, 111, 9326–9335.

172 Etacheri, V., Valentin, C., Schneider, J., Bahnemann, D., and Pillai, S.C.(2015) Visible-light activation of TiO2 photocatalysts: advances in theory andexperiments. J. Photochem. Photobiol. C Photochem. Rev., 25, 1–29.

173 Kohn, W. and Sham, L.J. (1965) Self-consistent equations including exchangeand correlation effects. Phys. Rev., 140, A1133–A1138.

174 Hohenberg, P. (1964) Inhomogeneous electron gas. Phys. Rev., 136,B864–B871.

175 Burke, K. (2012) Perspective on density functional theory. J. Chem. Phys.,136, 150901.

176 Moniz, S., Shevlin, S.A., Martin, D., Guo, Z., and Tang, J. (2015) Visible-lightdriven heterojunction photocatalysts for water splitting – a critical review.Energy Environ. Sci., 8, 731–759.

177 Akimov, A.V., Neukirch, A.J., and Prezhdo, O.V. (2013) Theoretical insightsinto photoinduced charge transfer and catalysis at oxide interfaces. Chem.Rev., 113, 4496–4565.

27

2

Recent Advances on Photocatalysis for WaterDetoxification and CO2 ReductionCarlotta Raviola and Stefano Protti

University of Pavia, PhotoGreen Lab, Department of Chemistry, vialeTaramelli 12, Pavia 27100, Italy

2.1 Introduction

The Rio Declaration on Environment and Development in 1992 [1] makesthe development of measures devoted to climate change mitigation and waterresources preservation a challenge to scientific research. Indeed, the other sideof the coin is benefits afforded by urban and industrial activities. In particular,waste disposal often leads to the contamination of water resources (e.g., lakes,groundwater aquifers) with different noxious pollutants, including pesticides,pharmaceutics, textile dyes, detergents and heavy metals [2]. Advanced oxidationprocesses (AOPs) have emerged in the last decades as promising methods forthe elimination of contaminants from different matrices including wastewaters[3], contaminated soil [4], and air [5]. In such approaches, a chemical (e.g., O3or Fe2+) or a physical (UV light) trigger is responsible for the generation (fromdifferent oxygen sources such as H2O2, ozone, or even water) of reactive oxygenspecies (ROS) including hydroxyl radical or superoxide radical ions. Such inter-mediates then induce the oxidative degradation of organic pollutants. Despitetheir high efficiency in the oxidation of usually refractory compounds, large-scaleapplications of AOPs are limited by the huge amount of chemicals needed andthe significant economic impact. A remarkable exception is sunlight-drivenheterogeneous photocatalysis [6]. In such processes, the absorption of a photonwith energy greater than the bandgap of a semiconductor photocatalyst resultsin the promotion of an electron from the valence band (VB) to the conductionband (CB) and subsequent charge separation (Figure 2.1, Eq. (1)).

Charges can either recombine while dissipating the absorbed energy or migrateto the catalyst surface and initiate a redox process with the adsorbed molecules.Oxidation of water by positive holes at the surface results in the generation of OH∙

radicals (Eq. (2)), whereas trapping of e−(CB) by molecular oxygen leads to super-oxide radical anion (O2

∙−), which further evolves to hydroperoxyl radical (HOO∙)and H2O2 (Eq. (3)). All of these ROS contribute to the oxidative degradation oforganic pollutants present in solution.


28 2 Recent Advances on Photocatalysis for Water Detoxification and CO2 Reduction

PC + hν h+(VB) + e–

(CB) (1)

h+(VB) + H2O OH

. + H+ (2)

e–(CB) + O2 O2

.– H+

HOO. H2O2 (3)

CB

VB

hν

e–(CB)

h+(VB)

Figure 2.1 Photocatalytic generation of ROS.

The advantages of these methods are apparent since the process takes place atroom temperature leading to the complete mineralization of the organic com-pound even at low concentration, and the oxygen required for the generation ofROS is obtained from atmosphere. Furthermore, the photocatalyst involved isusually cheap, nontoxic, recyclable and can be easily immobilized on inert matri-ces, and sunlight is used as the elective light source [7].

In addition to removal of organic contaminants, solar photocatalysis was alsoapplied successfully to water disinfection [8–10]. At the same time, due to its sig-nificant radiative forcing, carbon dioxide [11] is considered the largest contribu-tor to global warming—temperature predicted is predicted to be in the 1.0–3.7 ∘Crange by the late twenty-first century depending on future greenhouse gas emis-sions [12]. Accordingly, a reduction in CO2 gas emitted from combustion of fossilfuels (currently more than 31 billion tons of CO2 per year are emitted) as well asthe adoption of processes for CO2 capture, storage (CCS), and conversion intovaluable useful products (chemicals, fuels) is a key research challenge both inacademics and industries [13–16]. Starting from the seminal work of Inoue et al.[17], several semiconductors have been tested as photocatalysts and successfullyapplied to CO2 reduction [18–20].

Degradation of pollutants involves three main steps: (i) the photogenerationof charge carriers (electron–hole pairs) followed by (ii) migration of chargecarriers to the surface, and (iii) subsequent reduction in CO2 with the photogen-erated electron (in this case water can act as the sacrificial electron donor; seeFigure 2.2). However, CO2 is a thermodynamically inert compound (standard

(4)

(5)

(6)

CB

VB

hν

e–(CB)

h+(VB)

CO2

CO2.–

H2O

1/2O2 + 2H+

CO2 + 2H+ + 2e– E° = –0.61 VHCOOH

CO2 + 2H+ + 2e– CO + H2O E° = –0.53 V

CO2 + 4H+ + 4e– HCHO + H2O E° = –0.48 V

CO2 + 6H+ + 6e– CH3OH + H2O E° = –0.38 V

CO2 + 8H+ + 8e– CH4 + H2O E° = –0.24 VE = 0.81 V

E = –1.90 V

2H+ + 2e– E° = –0.42 VH2

(7)

(8)

(9)

Figure 2.2 Photoexcitation of a semiconductor photocatalyst and electron transfer processesinvolved in the reduction of CO2. Redox potentials (E∘) of half cell reactions (4)–(9) areexpressed versus NHE and calculated at neutral pH.

2.1 Introduction 29

enthalpy of formation of −393.5 kJ mol−1 at 298 K), the first one-electronreduction to the corresponding radical anion CO2

∙− being the rate limitingstep, because of the negative electrochemical reduction potential involved [21].Half-cell reactions depicted in Figure 2.2 illustrate the thermodynamic feasibilityfor the (multielectronic) production of carbon-based solar fuels such as CO(Eq. (2)), formaldehyde (Eq. (3)), methanol (Eq. (4)), and methane (Eq. (5)) fromCO2. Clearly, the process is pH dependent, since CO2 can be present also ascarbonate anion (CO3

2−) at pH of above 5, and production of hydrogen (Eq. (6))can also occur as competing reaction. The poor solubility of CO2 moleculesin water (0.033 M at 100 kPa at room temperature) and the occurrence of backreactions are also critical issues [22].

Reduction of carbon dioxide was investigated in either solid–gas systems (a gasmixture of the reactants reacted with the solid photocatalyst) or aqueous disper-sions (the photocatalyst particles were dispersed in an aqueous solution) [23].As recently highlighted by Habisreutinger et al. [15], the most widespread pho-tocatalytic systems used in such processes are metal oxides (TiO2, InTaO4) orsulfides (ZnS). Other materials, including sensitizing agents [24], were employedto enhance the absorption of light in the visible range. As discussed above, therecombination of the charge carriers on both surface or in the bulk of the semi-conductor (volume recombination) is detrimental to the efficiency of the photo-catalytic process. The presence of noble metals (with Fermi level lying below theCB of the used semiconductor) or, more rarely, metal oxide (NiO, RuO2) nanopar-ticles (NPs) as cocatalysts allows for an efficient separation of photogeneratedholes and electrons by trapping the latter and limiting the recombination sidepathway [25]. In this case, reduction of CO2 thus takes place on the surface of thecocatalyst (Figure 2.3).

In the last decades, an impressive number of papers describing the envi-ronmental application of solar photocatalysis have been reported in literature[26–28]. Most of these works were devoted to the development of selective,stable, and cheap photocatalysts as well as to the assembling of efficient solarphotoreactors. Accordingly, this chapter consists of two sections. In the firstsection, the use of semiconductor photocatalysts are discussed on the basis of thechemical structure, photophysical properties (UV absorption spectra, bandgapenergy), and the efficiency in both photocatalyzed wastewater treatment and in

Figure 2.3 Transfer of aphotogenerated electron fromthe CB of a semiconductor tothe Fermi level of the metalcocatalyst. CB

VB

hν

e–(CB)

h+(VB)

CO2

CO2.–

H2O

1/2O2 + 2H+

E = 0.81 V

EF

Co-catalyst


CO2 reduction in aqueous dispersion systems. In the second section, differentclasses of both batch and continuous flow solar photoreactors are brieflydescribed.

2.2 Photocatalysts for Environmental Remediation andCO2 Reduction

In order to ensure an efficient photocatalytic event, an ideal photocatalyst shouldexhibit different properties including high stability to photocorrosion, a signif-icant overlap of absorption cross section with sunlight emission spectrum (as aconsequence of the bandgap involved; see Figure 2.4), low cost and satisfactoryefficiency in charge carrier separation and high quantum yield of photoreaction[29].

2.2.1 Undoped TiO2

Due to its water insolubility, low cost and high stability against photocorrosion,TiO2 is mainly used, and early reports on the dechlorination of polychlorinatedbiphenyls in contaminated water were dated back to 1976 [30]. Such a photocat-alyst exists mainly in four different polymorphs, namely rutile, anatase, brookite,and monoclinic TiO2. Of these crystalline forms, anatase is the most active [31,32], while rutile is the most thermodynamically stable. The widest diffuse tita-nia photocatalyst is, however, commercially available Degussa-Evonik P25, which

0

–1

–2

1

2

3

3.2 eV

TiO2

(Anatase)

3.0 eV

TiO2

(Rutile)

2.8 eV

WO3

3.0 eV

ZnO

3.2 eV

SrTiO3

2.4 eV

BiVO4

388 nm413 nm 388 nm 443 nm

480 nm

388 nm

2.0 – 2.2 eV

Cu2O

560 nm

2.4 eV

CdS

480 nm

Figure 2.4 Band levels and wavelength thresholds of simple semiconductors photocatalysts.

2.2 Photocatalysts for Environmental Remediation and CO2 Reduction 31

is a mixture of anatase and rutile in a ratio of about 3 : 1 (surface area: up to56 m2 g−1, average size: 25 nm). As an example, anthraquinonic (Alizarin S), azoic(Congo Red), and heteropolyaromatic (methylene blue (MB)) dyes undergo min-eralization when irradiated in an aqueous suspension of titania. Interestingly,nitrogen atoms of amino groups were slowly oxidized into NO3

− ions (via NH4+

ion), whereas the central —N=N— azo-group in azoic dyes was converted ingaseous dinitrogen. This photocatalytic behavior makes TiO2 an ideal materialfor the elimination of nitrogen-containing pollutants [33]. The photodegrada-tion of several organophosphorous pesticides such as malathion, phorate, anddiazinon was successfully conducted in a TiO2(P25)/H2O2 system [34]. Apartfrom P25, a plethora of titania samples with different shape, size, and surfacecharacteristics have been easily synthesized starting from titanium alkoxides asprecursors, via either sol–gel methods followed by aging, drying and calcination,or via hydrothermal processes. Since photocatalytic events take place at activesites on the semiconductor surface, morphology is a key issue, and a maximumexposition of the catalyst surface is needed to both the environment to be treatedand the light source [35, 36].

A reduction in the primary particle size down to nanodimension leads to a veryhigh surface-to-volume ratio along with a reduction in charge recombinationdue to the short charge carrier diffusion distances. Thus, the combination of thetwo factors resulted in an overall increase in the performance of the photocat-alysts. As an example, TiO2 nanoparticles showed an enhanced photocatalyticactivity toward different contaminants including persistent herbicide Butachlor[37] and polynuclear aromatic hydrocarbons [38]. In recent decades, the depen-dence of photocatalytic efficiency of TiO2 on morphology was explored and aplethora of nanostructured materials including one-dimensional nanorods [39],two-dimensional nanosheets [40] and hierarchical structures were investigated.A one-step hydrothermal process carried out in an ethanol–glycerol mixture(titanyl sulfate as the precursor) was exploited for the synthesis of porousTiO2 characterized by a surface area of 350–450 m2 g−1 and a pore volumeof 0.2–0.3 m3 g−1. The resulting photocatalyst efficiently led to mineralizationof methyl orange under UV irradiation while producing small amounts ofhydrogen (up to 200 μmol h−1 g cat−1) [41]. For the photocatalytic conversionof carbon dioxide, irradiation of a CO2 saturated aqueous solution in the pres-ence of Degussa P25 as the photocatalyst with high power (100 W) UV-LEDs(355–385 nm, light intensity reaching the catalyst surface: 120 mW cm−2)afforded CO as the main product (18 μmol h−1 g cat−1) with traces of hydrogen,C1—C4 alkanes, and alcohols [42]. Simulated solar light irradiation of a CO2 sat-urated aqueous suspension of ultrathin TiO2 (thickness down to 1.66 nm) flakesafforded formate with a rate value (1.9 μmol h−1 g cat−1) 450 times higher thanthat of bulk titania [43]. As hinted above, noble metals as well as metal oxides ascocatalyst are often used for minimization of charge recombination. Thin filmsconstituted by Au–Cu alloy nanoparticles loaded on commercially available P25were applied to the reduction of CO2 by water under simulated solar light, thusproducing methane in high yields (2000 μmol g cat−1 h−1) with a selectivity of CBelectrons that achieved a 97% value [44]. A WO3 (2 wt.%)-modified TiO2 catalystwas prepared via the sol–gel method. Interestingly, when UV light is chosen as


irradiation sources, both WO3 and TiO2 sites can be excited. In contrast, undervisible-light exposition, the excitation process involved exclusively WO3. chargeseparation remained more efficient than that observed of bare TiO2 in bothcases. The oxidation pathway involved is wavelength dependent; thus superoxideion (O2

∙−) and H2O2 were generated under UV light irradiation, whereas the useof visible light led to the formation of hydroxyl radicals OH∙.

The prepared catalyst (1 g l−1) has been exploited in the natural sunlight-induced mineralization of pesticide malathion in water, which was achievedafter only 2 h irradiation [45].

2.2.2 Undoped Metal Oxides Different from TiO2

Other semiconductors such as ZnO, WO3, and mixed oxides (BiVO4, Nb and Ta)have been also successfully investigated. Bulk ZnO exhibited an efficiency com-parable to that of TiO2 in the degradation of phenol under concentrated sunlight[46]. Although photocorrosion occurs to decrease the activity during irradiation,ZnO nanoparticles were found superior in the degradation of chloroaromaticsand aromatics to commercially available bulk ZnO and TiO2 (Degussa) [47]. Thedegradation of various endocrine-disrupting compounds, including resorcinol,bisphenol, and methylparaben, was successfully achieved under UV irradiationby means of ZnO micro-/nanospheres. Interestingly, the catalyst was obtainedby simply mixing a ZnO precursor Zn(NO3)2 6H2O and NaOH. The initiallyformed ZnO nucleated to multinuclei aggregates and then to nanosheets thatfinally self-assembled to hierarchical structures illustrated in Figure 2.5 [48].

ZrO2 was found to reduce CO2 with the simultaneous oxidation of waterto hydrogen, though with a bandgap in the 4–5 eV range [49]. Highly porousgallium oxide was tested in the selective photocatalytic conversion of CO2into CH4, without any sacrificial reagent except water. Notably, the presenceof mesopores and macropores in the photocatalyst resulted in impressiveincrease in both CO2 adsorption capacity and the surface area, allowing for a400% higher conversion rate compared to the bulk Ga2O3 nanoparticles. CH4is thus generated with a quantum yield value of 3.993% [50]. Semiconductorsexhibiting lower bandgap materials such as cuprous oxide Cu2O are consideredpromising catalyst for CO2 reduction, though photocorrosion affected theirwide application. A Hematite−cuprous oxide (𝛼-Fe2O3/Cu2O, 1 : 1 molar ratio)

Nucleation

ZnO precursor ZnO nuclei

ZnO nanosheets

Self-assemblyGrowth

Figure 2.5 Growth diagram of ZnO micro-/nanospheres employed for the mineralization ofdifferent endocrine-disrupting chemicals. (Reproduced with permission from Ref. [48].Copyright 2013, Elsevier.)


nanocomposite Z-scheme photocatalyst has been synthesized and applied tothe visible-light-induced conversion of CO2 to CO with a maximum CO yieldattaining 5.0 μmol g cat−1 [51]. In any case, however, the photocatalytic activitydescribed for undoped semiconductors in this field is still very low due to thevacant oxygen space on the NiO surface that allows interactions with CO2molecules and enables the transportation of the photogenerated species. Thisphotocatalyst exhibits a higher efficiency than other single metal oxides such asZnO, WO3, and Cu2O and CH3OH was produced via a six-electron reductionprocess with a rate of 170 μmol g cat−1 h−1) [52].

Mixed oxides have been also tested as photocatalysts for water purification.Bismuth tungstate (Bi2WO6) with corner-shared WO6 octahedral layered struc-ture, which exhibited a 2.69 eV bandgap, was found effective in mineralizing bothCHCl3 and CH3CHO contaminants upon visible-light exposition [25]. Analo-gously, mesoporous Bi2WO6-based photocatalyst was recently found effective inthe solar-light-driven mineralization of Bisphenol A [53]. Notably, shifting to a50% Bi2WO6/BiOBr system resulted in a photocatalytic activity toward this con-taminant 26.6 times higher than that of the bare bismuth tungstate [54]. Bi2WO6with a hierarchical “flake-ball” shape has been found to efficiently decomposeacetic acid in solution and gaseous acetaldehyde [55].

2.2.3 Carbon Modified Metal Oxides as Photocatalysts

Mixing mechanically titania and activated carbon (AC) led to the formation of aninterface between the solid phases. Organic pollutants initially interacted withhighly adsorbing AC and the subsequent mass transfer (via spillover throughthe contact surface) to the photoactive titania enabled the photodegradationprocesses. Importantly, the AC–TiO2 catalytic system can be recycled withoutany loss in efficiency. This approach has been applied to the abatement of severalorganic compounds including phenols and chlorinated compounds [56, 57].TiO2/carbon composites have been also used with success. This is the case oftitania-coated AC that was prepared via sol–gel method from AC and titaniumalkoxylate and employed in the degradation of organic dyes [58, 59] as wellas of toxin Microcystin-LR [60]. Carbon nanotubes (CNTs)–TiO2 systems(including CNTs–TiO2 composites) have been applied to the mineralization ofa wide range of organic pollutants such as phenols and dyes as well as in waterdisinfection [61].

Analogously, graphene/semiconductor nanocomposites (GSNs) have beenenvisaged as a promising class of heterogeneous photocatalysts for the treatmentof industrial wastewaters [62, 63], which in most cases have been investigated inthe oxidative degradation of MB as model compound [64]. As a recent example,(P25)-graphene nanocomposite photocatalyst (Figure 2.6) was obtained in aone-step hydrothermal method from graphene oxide and commercially availableP25. The carbon platform plays a key role in the mineralization of organiccompounds (MB used as model pollutant) due to the increased catalyst adsorp-tivity, the extended light absorption (since the chemical bonds of Ti—O—Calong with the significant transparency of graphene render a red shift in thephotoresponding range) and the minimization of charge recombination (because


H3C

CH3H3C

H3CS

NN N

H3C

H3C

H3C

H3C

CH3

CH3

CH3

CO2, H2O and

other mineralization 200 nm

(a) (b)

e–

h+

CH3

TiO2

S

H3C

CH3H3C

H3CNS

N

N

N ++

N

N+

N

N

NS

ho

Figure 2.6 (a) Tentative process of the degradation of methylene blue by P25 dispersed onthe graphene support (P25-GR). (b) TEM image of P25-GR. (Reproduced with permission fromRef. [64]. Copyright 2010, American Chemical Society.)

of the role of graphene as the carrier of the photogenerated electrons by P25 andfast charge transportation) [65].

2.2.4 Doped Metal Oxides

The main drawback with the use of TiO2 is, however, the large energy bandgap(3.0–3.2 V; see Figure 2.4), which requires an excitation wavelength fallingin the UV region, limiting the use of solar light (its spectrum contains lessthan 5% of UV-A) as the energy source [65]. Doping is probably the mostapplied technique to extend the light absorption range of a semiconductorinto the visible-light range. Indeed, the presence of foreign atoms in the hostlattice of the oxide induced a physical strain in the lattice itself. The intrin-sic red-shift of the semiconductor absorption band was attributed to eitherbandgap narrowing of the semiconductor or to the introduction of oxygenvacancies with the consequent formation of color centers [22]. Basically, twoapproaches are proposed for doping titania, namely metal doping and nonmetaldoping [66, 67]. In the former approach, metal atoms such as Zr, Hf, Cu and Insubstitute the titanium atoms in the crystal lattice, thus generating an emptyenergy state below the CB of TiO2 that is responsible for the visible-lightresponse of the photocatalyst. For example, nanosized 1% copper-doped titaniasol–gel catalysts were efficiently employed in the quantitative mineralization ofpesticide 2,4-dichlorophenoxyacetic acid under UV irradiation [68]. Anotherdoped catalyst, 3% CuO/TiO2 was prepared by doping Cu(NO3)2 into TiO2Degussa-P25. The resulting material exhibited a 2.88 eV bandgap and led to thealmost selective conversion of CO2 into methanol with a quantum yield valueof 0.1923 (442.5 mmol g cat−1 h−1) [69]. Self-doped TiO2 was generally preparedby heating titania in a hydrogen atmosphere or via solvothermal reaction froma mixture of titanium tetraisopropoxide and TiCl3. The presence of Ti3+ sites


resulted in a new energy level located at about 0.8 eV below the CB. As anexample, Zhou et al. described a self-doped anatase–rutile system showingan enhanced photocatalytic activity in the degradation of Rhodamine B dyethat can be repeatedly used five times without any loss in efficiency [70]. Innonmetal doped metal oxides, nitrogen, fluorine and boron-doped TiO2 samplesare the most used materials. In particular, N-doped anatase was applied tothe degradation of lindane [71] and a range of phenoxy acid–based herbicides,including mecoprop and clopyralid [72]. Apart from TiO2 samples, the dopingapproach is widely used for other semiconductor photocatalysts. This is thecase of one-dimensional wedged N/CuO, where the presence of the dopantatom introduced an intermediate band between the VB and the CB of the metaloxides and increased the carrier mobility. The photocatalyst was employed inthe generation of methanol from carbon dioxide [73].

2.2.5 Perovskites

The term perovskite was initially attributed to a mineral composed of calciumtitanate (CaTiO3) and then extended to compounds exhibiting a chemical for-mula ABX3 (where A and B are two cations and X is an anion such as O that bondsA and B) and a crystal structure with cubic symmetry and Pm3M (see Figure 2.7).Notably, more than 90% of metal elements have been successfully incorporatedinto the perovskite lattice [74–76].

Such materials found application in a plethora of research fields, including thepreparation of dye cells [74], water splitting [75], environmental remediation, andCO2 reduction [76]. In particular, such binary oxides were found effective in thephotodegradation of organic dyes such as Rhodamine B and methyl orange [76].As an example, BaBiO3 with perovskite structure was exploited as photocatalystin the light-induced mineralization of organic contaminants, such as acetalde-hyde and MB, which took place with an initial rate of 7.1× 103 and 81.6 μmol h−1,respectively [77]. Perovskite oxynitride CaTaO2N with a bandgap of 2.5 eV loaded

Figure 2.7 Schematic structure of cubicperovskite ABO3 (dark gray, BO6 units; lightgray, A atoms). (Reproduced with permissionfrom Ref. [74]. Copyright 2012, AmericanChemical Society.)


Semiconductor

particle

O

O

Ru/Ru–

Ru∗/Ru–

Ru′/Ru′–

e–

e–

HOHO

HOHO

PN N NN

Ru

Visible light

D

D+

Visible light

RuN N

N Cl

Cl

CO

CONP

HCOOH

CO2CB

VB

Figure 2.8 Reduction of CO2 to formic acid by perovskite oxynitride CaTaO2N coupled to abinuclear Ru(II) complex. (Reproduced with permission from Ref. [78]. Copyright 2015,American Chemical Society.)

with Ag NPs and conjugated to a binuclear Ru(II) complex produced selectively(>99%) HCOOH via CO2 reduction under visible-light irradiation (Figure 2.8)in the presence of an amine as sacrificial electron donor. Notably, in this casea two-step photoexcitation of both perovskite core and sensitizer unit (a ruthe-nium trisbipyridine core) was involved, whereas CO2 reduction occurred at theruthenium carbonyl moiety [78].

2.2.6 Metal Chalcogenides

Despite the rather small value of bandgap energy (2.4 eV), the use of cadmiumsulfide is limited by its poor stability in aqueous media and photocorrosion ledto the release of Cd2+ ions as well as to the conversion of CdS into CdSO4 [22].On the other hand, copper sulfides (including Cu2S, CuS, and Cu7S4) have alsonarrow bandgaps (in the 1.2–2.2 eV) but again are susceptible to photocorrosion.However, it was recently demonstrated that its encapsulation in a shell of largebandgap transparent material prevent oxidation of copper sulfides. Accordingly,CuS/ZnS core/shell nanocrystals (NCs) have been prepared and exploitedas visible-light-absorbing catalysts in the photodegradation of Rhodamine B.Since the CB of CuS lies about 0.9 eV above that of ZnS, electron migrationfrom photoexcited CuS to the CB of ZnS is feasible and, in the presence ofoxygen, ROS were generated on the catalyst surface. Notably, NCs evolved fromspherical-shaped particles to rods and then “flower”-shaped aggregates duringirradiation [79]. A Bi2S3/CdS photocatalyst was found to reduce CO2 to CH3OHwith a rate of 613 μmol g cat−1 under visible-light irradiation in the presence ofsodium sulfite as sacrificial electron donor [80].

Different AgyInzZnkSm solid solutions were tested in the photoreduction ofCO2, and best results were obtained with Cu0.30Ag0.07In0.34Zn1.31S2 customized


with RuO2 that afforded efficiently CO2 to CH3OH (118.5 μmol h−1 g cat−1) uponvisible-light exposition [81].

2.2.7 Other Catalysts

The selective conversion of carbon dioxide (CO2) into methane and carbonmonoxide was investigated by using gallium nitride (GaN) nanowire arrayscoupled to Rh/Cr2O3 core/shell cocatalyst, presence of which suppressed theformation of CO as the by-product (Figure 2.9). In this case, reduction tookplace at the cocatalyst site [82].

Photogeneration of methanol from CO2 was also observed in the presence ofMg-Doped Ga(In)N nanowire arrays as photocatalyst [83].

Hybrid organic/inorganic systems have been recently applied to both waterdetoxification [84] and CO2 photoreduction [85]. A ReI organic polyoxometalatehybrid complex bearing a phenanthroline ligand decorated with a 15-crown-5ether moiety was found able to catalyze the photoreduction of CO2 to CO withH2 as the reducing agent [86].

Analogously, conversion of CO2 to CO was successfully achieved by means ofa rhenium(I) complex covalently bonded to a TiO2 graphene oxide composite[87]. A Ru(II) binuclear complex coupled to Ag-loaded TaON was exploited toselectively produce HCOOH under neutral conditions using ethylenediaminete-traacetic acid disodium salt as a sacrificial reductant (Figure 2.10), with a maxi-mum turnover number of 750 [88].

(a)

10 nm

(b)

GaN NW

2H2O

Rh/Cr2O3

H2

CH4

CO

CO2

H+

+

+

O2

4H+

Figure 2.9 (a) TEM image of Rh/Cr2O3 core/shell decorated GaN nanowire (b) Scheme of thephotoreduction processes of CO2 on Rh/Cr2O3 GaN nanowires. (Reproduced with permissionfrom Ref. [82]. Copyright 2015, American Chemical Society.)


700

TaON

D

CBe–

e–

HCOOH

CO2

NN

NCl

ClCOCO

NNN

NN

Ru

Ru

H2O3P

Ag

2+

H2O3P

VB

D+

8

7

6

5

4

3

2

1

0

600

500

400

300

200

100

(a) (b)

0

0 5 10

Irradiation time (h)

HCOOH

H2

Turn

ove

r num

be

r (T

ON

)

Am

ou

nt (μ

mo

l)

15 20 25

Figure 2.10 (a) Hybrid powder photocatalyst of the Ru(II) binuclear complex adsorbed onAg-modified TaON. (b) Time courses of HCOOH and H2 formation by visible-light (𝜆> 400 nm)irradiation of the photocatalyst in EDTA⋅2Na aqueous solution (4 ml) in the presence ofNa2CO3 (0.1 M) under a CO2 atmosphere. (Reproduced with permission from Ref. [88].Copyright 2016, The Royal Society of Chemistry.)

2.3 Photoreactors for Solar Degradation of OrganicPollutants and CO2 Reduction

First attempts to perform solar photocatalyzed wastewater treatment and CO2reduction were carried out by exposing a vessel containing reagents and thesuspended photocatalyst to natural sunlight. The main drawbacks of the windowledge approach are the discontinuous intensity of solar light (which in turndepends on both geographic position and weather) and the small amount ofmaterial that can be treated for day. To overcome these drawbacks, elaboratedsolar photocatalytic reactors able to concentrate sunlight for “gram-per-day” or“kg-per-year” have been devised in the last decades [89].

A solar photocatalytic reactor or collector is a device that concentrates solarphotons, brings them into contact with reactants and photocatalyst, and thencollects reaction products. A huge variety of solar photocatalytic reactors thatdiffer in photoreactor design, operating conditions, photocatalyst preparation,and fluctuations in solar intensity are available [90, 91]. All of these systems canoperate in either continuous or batch flow. While in the continuous mode thereaction mixture is continuously loaded in the solar photoreactor and comes outas product, in batch mode all reagents are charged at the same time and convertedinto product; only the reaction waste flows out into a tank and recirculates in thecollector until the reaction is completed (see Figure 2.11). Another design issuefor solar photocatalytic collector is whether to use a suspended or a supportedphotocatalyst. Even if the continuous approach ensures a better availability of thecatalytic surface area for absorption and reaction, the removal of small particlesof the photocatalyst by sedimentation or filtration is troublesome [92].

Solar collectors are classified based on the concentration factor (C) definedas the ratio between the “aperture area” (the area that intercepts the solarirradiation) and the “absorber area” (the area of the component that receives

2.3 Photoreactors for Solar Degradation of Organic Pollutants and CO2 Reduction 39

Catalyst separation

Photoreactor

(a)

(b)

Photoreactor

Figure 2.11 Continuous flow (a) and batch reactor design (b). (Reproduced with permissionfrom Ref. [92]. Copyright 2000, Elsevier.)

sunlight) and closely related to the temperature attainable by the system [93].This classification takes into account only the thermal efficiency of solar reactorssince solar thermal processes are based on the collection and concentration oflarge number of photons from all wavelengths to achieve a certain temperaturerange. In contrast, sunlight-driven photochemical processes are based on thecollection of only a fraction of photons (in general short wavelength photons)useful to promote the photochemical reaction.

Accordingly, three classes of photoreactors can be identified:

– Nonconcentrating (low concentration or low temperature) reactors (up to150 ∘C)

– Medium concentrating or medium temperature reactors (from 150 to 400 ∘C)– High concentrating or high temperature reactors (over 400 ∘C).

The main features of these photoreactors are discussed in the followingsections.

2.3.1 Non Concentrating (Low Concentration or Low Temperature)Systems

Non concentrating reactor (NCC) or inclined plate collector (IPC) (Figure 2.12)is a flat or corrugated plate, upon which the photocatalysts are supported. Theplate is oriented to the equator with a different inclination depending on thegeographic position. The reactant solution flows as a thin film (100–200 μm) onthe photocatalytic surface and the photons reaching the reactor first interact withthe reactants and then with the photocatalyst. The backing plate can be made upof different materials such as glass, metal, or even stone, and the collectors canprofit both of direct (radiation with a known direction that has no interferencewith the atmosphere) and diffuse radiation (radiation that has interferencewith atmosphere and consequently reach the ground with a random direction).


θ

Figure 2.12 NCC photoreactor. (Reproducedwith permission from Ref. [93]. Copyright 2015,Elsevier.)

Table 2.1 Main features of NCCs, PTCs, CPCs (see Ref. [93] for further details).

NCC PTC CPC

C = 1 C = 5–35 suns C ≤ 1.2Direct and diffuse radiation Direct solar beams Direct and diffuse radiationLow cost and simple design High cost (tracking system) Moderate capital cost

This characteristic made such a reactor independent from weather conditions.Moreover, leaving the face of the reactor open to the atmosphere furtherincreases efficiency and avoids interaction of light with the reactor covering andthe potential formation of opaque layer on photocatalyst on its interior surface.Loss of water and volatile chemicals and interferences by atmosphere are obviousdrawbacks. Although NCCs show significant advantages with respect to concen-trating systems, their design is not a trivial issue. Indeed, on one hand, the simplearrangement and cheap materials employed made them (almost) in principle themost economic photocatalytic solar reactor models, but on the other hand, due tothe large surface required, high operating pressures are required to pump the fluid(Table 2.1). Recently, flat photocatalytic reactors were installed on house roofsin order to photoreduce CO2 into CH4 or CH3OH under natural sunlight [8].

2.3.2 Medium Concentrating or Medium Temperature Systems

In order to reduce the reactor volume and thus the quantity of photocatalystrequired, medium concentrating or medium temperature systems have beendeveloped. The most diffuse are parabolic trough collectors (PTCs) (Figure 2.13).They are made of a pipe (through which the reactant fluid flows) located on thefocus of the reflective parabolic surface. The photocatalyst is usually suspendedin the fluid; however, examples of supported photocatalyst have been also


(a) (b)

Figure 2.13 (a, b) PTC photoreactor and PROPHIS loop. ((a) Reproduced with permission fromRef. [93]. Copyright 2015, Elsevier. (b) Reproduced with permission from Ref. [94]. Copyright2005, Royal Society of Chemistry.)

reported in the literature [7]. The reactor’s tube must be both transparent andresistant to UV light, which are fulfilled by fluoropolymers, acrylic polymers,and several types of iron borosilicate glasses. Although quartz could be suitable,it is too expensive for large scale-up. The best material used in building ofreflecting/concentrating panels is aluminum because it is cheap and offers highreflectivity in the UV spectrum. Since only the radiation perpendicular to theparabolic surface is efficiently reflected onto the reactant tube, these reactorsare often equipped with motors that allow to point the device directly at the sunall the day. The need of a tracking mechanism rendered them more expensivecompared to NCCs. The main drawback of PTCs is that they are able to captureexclusively direct radiation making them inefficient on cloudy days. The mostfamous example of PTCs is the PROPHIS (parabolic trough collector for organicphotochemical syntheses in solar light) reactor, located in Cologne (Germany),and employed for the production of fine chemicals (Figure 2.13b) [94].

A good compromise between NCCs and PTCs is represented by compoundparabolic concentrators (CPCs) (Figure 2.14), peculiar types of medium concen-trated solar reactors. These are static systems with a parabolic reflective surfacearound a cylindrical reactor tube. This setup allows for an efficient reflection ofindirect light onto the absorbing tube, thus resulting in efficient collection of bothdirect and diffuse solar light. Furthermore, since a tracking mechanism is notrequired, the cost and complexity of the system are low.

CPCs are the most widespread setup used for both the chemical synthesis anddegradation of organic pollutants. One example is reported by De la Cruz andcoworkers who built a solar photoreactor in Spain for the photolysis of propra-nolol, a sympatholytic nonselective 𝛽 blocker. As shown in Figure 2.15, the devicehas a module of six parallel CPCs made up of polished aluminum. The solutionwith propranolol and the photocatalyst in suspension (TiO2) is in a reservoir tankwith a mechanical stirrer and was continuously recirculated in the collector witha pump. Solar radiation was quantified with a radiometer [95].


Concentrator

Receiver

Axis of

rotation

(a) (b)

Figure 2.14 (a, b) Schematic of CPC photoreactor and CPC used for solar photocatalytic waterdetoxification. ((a) Reproduced with permission from Ref. [89]. Copyright 2016, AmericanChemical Society. (b) Reproduced with permission from Ref. [93]. Copyright 2015, Elsevier.)

Aluminum CPC

Photo-reactors

6 CPCs

Stirrer

Samples

Reservoir

tank

Radiometer

Pump

θa = 90’

CCPC= 1

Quartz photo-reactor

Figure 2.15 CPC photoreactor for photolysis of propranolol. (Reproduced with permissionfrom Ref. [95]. Copyright 2013, Elsevier.)

2.3.3 High Concentrating or High-Temperature Systems

Such reactors are paraboloid with a focus point instead of a linear focus able toconcentrate in the range of 10–10 000 suns thus creating extremely high temper-atures (up to 400 ∘C). Parabolic dishes, central power systems, and solar furnacesfall under this category, which are basically used for power generation from con-centrated solar radiation (Figure 2.16) [96].


(b)

Concentrator

Receiver

Axis of

rotation

(a)

Figure 2.16 (a, b) Schematic of parabolic dish and parabolic dish reactor. ((a) Reproduced withpermission from Ref. [89]. Copyright 2016, American Chemical Society. (b) Reproduced withpermission from Ref. [96]. Copyright 2004, Elsevier.)

2.3.4 Parameters of a Solar Reactor

In order to select the most suitable design for the requirement, efficiency andcapital costs of the setup should be evaluated. In general, the efficiency factor 𝜂 isdefined as the ratio between the radiant power used (Qused, J s−1) and the radiantpower absorbed (Qabsorbed).

𝜂 =Qused

Qabsorbed

For photocatalytic processes, 𝜂 can be expressed using the followingrelationship:

𝜂 =Qused

Qabsorbed− V

V1×

rs × ΔHrs × Wirr

Qabsorbed

where V , V 1, stoichiometric coefficients of consumption of reactive species andsubstrate, respectively; rs, specific reaction weight per unit weight of irradiatedcatalyst (mol g−1 s−1); ΔHrs, enthalpy formation of the reactive species (J mol−1);W irr, irradiated catalysts weight (g).

“Figures of merit” are mainly used to evaluate the cost of a photoreactor. Inparticular, since in a solar-energy-driven system the electrical energy componentis almost absent, the capital cost of a solar reactor is proportional to its area; thus,“figures of merit” based on solar collector area are necessary. In the case of highconcentration range, the collector area per mass (ACM, m2 kg−1) is used, definedas the “collector area required to bring about degradation of a unit mass (e.g., onekilogram, kg) of a contaminant C in polluted water or air in a time to (1 h) whenthe incident solar irradiance is 1000 W m−2(E0

s ) base on AM1.5 standard solarspectrum on a horizontal surface” [97]. In contrast, for low concentration range,the collector area per order is taken into account (ACO, m2/m3-order), definedas the “collector area required to reduce the concentration of a contaminant C inpolluted water or air in a unit volume by one order of magnitude in a time to(1 h)when the incident solar irradiance is 1000 W m−2” [68].


The two parameters can be calculated for both batch operation and flowthrough operation:

Batch operation:

ACM =103 × Ar × t × Es

M × Vt × t0 × E0s × (Ci − Cf )

; ACO =A × Es × t

V × lg (Ci/

Cf)

where, Ar, the real collector area, Es, average solar irradiance over the period t ofthe treatment (W m−2); M, molar mass of the substrate (g mol−1); V t, volume oftreated solution (l); t0, 1 h; E0

s , 1000 W m−2;Ci, Cf, influent and effluent concen-tration of the substrate (M).

Flow through operation:

ACM =103 × Ar × Es

M × F × t0 × E0s × (Ci − Cf )

; ACO =A × Es

F × lg (Ci/

Cf)

where Ar, the real collector area; Es, the average solar irradiance over theresident time (W m−2); M, molar mass of the substrate (g mol−1); F , flow rate(m3 h−1); t0, 1 h; E0

s , 1000 W m−2; Ci, Cf, influent and effluent concentration of thesubstrate (M).

2.4 Conclusion

This chapter pointed out how scientific community currently trusts the potentialapplications of solar chemistry to different environmental issues such as waterpurification and CO2 fixation [98]. Impressive advancements have been achieved,and a plethora of heterogeneous photocatalysts has been proposed for these pur-poses. Most efforts are focused on the development of stable catalysts that areable to absorb within the solar emission spectrum, while efficiently perform-ing the desired catalytic transformation. In the last decades, several more effi-cient alternatives to titanium dioxide have been proposed, including perovskites[74–76], sulfides [64, 65], and hybrid [83] photocatalysts.

On the other hand, the introduction of solar photocatalytic reactors allowedscaling-up these processes concentrating sunlight for “gram-per-day” or“kg-per-year”. As reported, since the design of solar collectors is an ever-evolvingscience able to build ever more efficient devices, it is hoped that in the futuresolar photocatalysis will become the elective method for water detoxificationand CO2 reduction.

Acknowledgment

We are grateful to Cariplo Foundation, Italy, project 2015-0756 “Visible LightGeneration of Reactive Intermediates from Azosulfones.”

References 45

References

1 http://www.unesco.org/education/nfsunesco/pdf/RIO_E.PDF, 20/12/2016(accessed 20 December 2016).

2 Reddy, P.A.K., Reddy, P.V.L., Kwon, E., Kim, K.-H., Akter, T., and Kalagar, S.(2016) Recent advances in photocatalytic treatment of pollutants in aqueousmedia. Environ. Int., 91, 94–103.

3 (a) Gogate, P.R. and Pandit, A.B. (2004) A review of imperative technologiesfor wastewater treatment I: oxidation technologies at ambient conditions.Adv. Environ. Res., 8, 501–505; (b) Comninellis, C., Kapalka, A., Malato, S.,Parsons, S.A., Poulios, I., and Mantzavinos, D. (2008) Advanced oxidationprocesses for water treatment: advances and trends for R&D. J. Chem. Tech-nol. Biotechnol., 83, 769–776.

4 Pera-Titus, M., Garcìa-Molina, V., Baños, M.A., Giménez, J., and Esplugas, S.(2004) Degradation of chlorophenols by means of advanced oxidation pro-cesses: a general review. Appl. Catal., B, 47, 219–256.

5 Oppenländer, T. (2007) Photochemical Purification of Water and Air:Advanced Oxidation Processes (AOPs)-Principles, Reaction Mechanisms,Reactor Concepts, Wiley-VCH, Weinheim.

6 Fujishima, A., Zhang, X., and Tryk, D.A. (2007) Heterogeneous photocatalysis:from water photolysis to applications in environmental cleanup. Int. J. Hydro-gen Energy, 32, 2664–2672.

7 Ibhadon, A.O. and Fitzpatrick, P. (2013) Heterogeneous photocatalysis: recentadvances and applications. Catalysts, 3, 189–218.

8 See for reviews Malato, S., Fernàndez-Ibàñez, P., Maldonado, M.I., Blanco,J., and Gernjak, W. (2009) Decontamination and disinfection of waterby solar photocatalysis: recent overview and trends. Catal. Today, 147,1–59.

9 Byrne, J.A., Morris Dunlop, P.S., Hamilton, J.W.J., Fernández-Ibáñez, P.,Polo-López, I., Sharma, P.K., and Vennard, A.S. (2015) Review of hetero-geneous photocatalysis for water and surface disinfection. Molecules, 20,5574–5615.

10 Reddy, P.V.L., Kavitha, B., Reddy, P.A.K., and Kim, K.-H. (2017) TiO2-basedphotocatalytic disinfection of microbes in aqueous media: a review. Environ.Res., 154, 296–303.

11 de Richter, R.K., Ming, T., and Caillol, S. (2013) Fighting global warming byphotocatalytic reduction of CO2 using giant photocatalytic reactors. Renew-able Sustainable Energy Rev., 19, 82–106.

12 Anderson, T.R., Hawkins, E., and Jones, P.D. (2016) CO2, the greenhouseeffect and global warming: from the pioneering work of Arrhenius and Cal-lendar to today’s earth system m. Endeavour, 40, 178–187.

13 Navalòn, S., Dhakshinamoorthy, A., Alvaro, M., and Garcia, H. (2013) Pho-tocatalytic CO2 reduction using non-titanium metal oxides and sulfides.ChemSusChem, 6, 562–577.


14 Morris, A.J., Meyer, G.J., and Fujita, E. (2009) Molecular approaches to thephotocatalytic reduction of carbon dioxide for solar fuels. Acc. Chem. Res., 42,1983–1994.

15 Habisreutinger, S.N., Schmidt-Mende, L., and Stolarczyk, J.K. (2013) Photo-catalytic reduction of CO2 on TiO2 and other semiconductors. Angew. Chem.Int. Ed., 52, 7372–7408.

16 Kondratenko, E.V., Mul, G., Baltrusaitis, J., Larrazábal, G.O., andPérez-Ramírez, J. (2013) Status and perspectives of CO2 conversion intofuels and chemicals by catalytic, photocatalytic and electrocatalytic processes.Energy Environ. Sci., 6, 3112–3135.


18 Tu, W., Zhou, Y., and Zou, Z. (2014) Photocatalytic conversion of CO2 intorenewable hydrocarbon fuels: state-of-the-art accomplishment, challenges,and prospects. Adv. Mater., 26, 4607–4626.

19 Mao, J., Li, K., and Peng, T. (2013) Recent advances in the photocatalytic CO2reduction over semiconductors. Catal. Sci. Technol., 3, 2481–2498.

20 Roy, S.C., Varghese, O.K., Paulose, M., and Grimes, C.A. (2010) Toward solarfuels: photocatalytic conversion of carbon dioxide to hydrocarbons. ACSNano, 4, 1259–1278.

21 Maeda, K. (2013) Direct splitting of pure water into hydrogen and oxy-gen using rutile titania powder as a photocatalyst. Chem. Commun., 49,8404–8406.

22 Protti, S., Albini, A., and Serpone, N. (2014) Photocatalytic generation of solarfuels from the reduction of H2O and CO2: a look at the patent literature.Phys. Chem. Chem. Phys., 16, 19790–19827.

23 Nikokavoura, A. and Trapalis, C. (2017) Alternative photocatalysts to TiO2 forthe photocatalytic reduction of CO2. Appl. Surf. Sci., 391, 149–174.

24 Won, D.-I., Lee, J.-S., Cheong, H.-Y., Cho, M., Jung, W.-J., Son, H.-J., Pac, C.,and Kang, S.O. (2017) Organic–inorganic hybrid photocatalyst for carbondioxide reduction. Faraday Discuss., 198, 337–351.

25 Tanga, J., Zou, Z., and Ye, J. (2004) Photocatalytic decomposition of organiccontaminants by Bi2WO6 under visible light irradiation. Catal. Lett., 92,53–56.

26 Singh, S., Mahalingam, H., and Singh, P.K. (2013) Polymer-supported tita-nium dioxide photocatalysts for environmental remediation: a review. Appl.Catal., A, 462–463, 178–195.

27 Malato, S., Maldonado, M., Fernández-Ibáñez, P., Oller, I., Polo, I., andSánchez-Moreno, R. (2016) Decontamination and disinfection of water bysolar photocatalysis: the pilot plants of the Plataforma Solar de Almeria.Mater. Sci. Semicond. Process., 42, 15–23.

28 Ong, W.-J., Tan, L.-L., Ng, Y.H., Yong, S.-T., and Chai, S.-P. (2016) Graphiticcarbon nitride (𝛾−C3N4)-based photocatalysts for artificial photosynthesis andenvironmental remediation: are we a step closer to achieving sustainability?Chem. Rev., 116, 7159–7329.

References 47

29 Di Paola, A., García-López, E., Marcìa, G., and Palmisano, L. (2012) A surveyof photocatalytic materials for environmental remediation. J. Hazard. Mater.,211–212, 3–29.

30 Carey, J.H., Lawrence, J., and Tosine, H.M. (1976) Photodechlorination ofPCB’s in the presence of titanium dioxide in aqueous suspensions. Bull.Environ. Contam. Toxicol., 16, 697–701.

31 Luttrell, T., Halpegamage, S., Tao, J., Kramer, A., Sutter, E., and Batzill, M.(2014) Why is anatase a better photocatalyst than rutile? - Model studies onepitaxial TiO2 films. Sci. Rep., 4, 4023.

32 Rajeshwar, K., Chenthamarakshan, C.R., Goeringer, S., and Djukic, M. (2001)Titania-based heterogeneous photocatalysis. Materials, mechanistic issues,and implications for environmental remediation. Pure Appl. Chem., 73,1849–1860.

33 Lachheb, H., Puzenat, E., Houas, A., Ksibi, M., Elaloui, E., Guillard, C., andHerrmann, J.-M. (2002) Photocatalytic degradation of various types of dyes(Alizarin S, Crocein Orange G, Methyl Red, Congo Red, Methylene Blue) inwater by UV-irradiated titania. Appl. Catal., B, 39, 75–90.

34 Doong, R.-A. and Chang, W.-H. (1997) Photoassisted titanium dioxide medi-ated degradation of organophosphorus pesticides by hydrogen peroxide.J. Photochem. Photobiol., A, 107, 239–244.

35 Adán, C., Marugán, J., Sánchez, E., Pablos, C., and van Grieken, R. (2016)Understanding the effect of morphology on the photocatalytic activity ofTiO2 nanotube array electrodes. Electrochim. Acta, 191, 521–529.

36 Testino, A., Bellobono, I.R., Buscaglia, V., Canevali, C., D’Arienzo, M.,Polizzi, F., Scotti, R., and Morazzoni, F. (2007) Optimizing the photocat-alytic properties of hydrothermal TiO2 by the control of phase compositionand particle morphology. A systematic approach. J. Am. Chem. Soc., 129,3564–3575.

37 Mahmoodi, N.M., Arami, M., Limaee, N.Y., Gharanjig, K., andNourmohammadian, F. (2007) Nanophotocatalysis using immobilizedtitanium dioxide nanoparticle: degradation and mineralization of water con-taining organic pollutant: case study of Butachlor. Mater. Res. Bull., 42,797–806.

38 Dass, S., Muneer, M., and Gopidas, K.R. (1994) Photocatalytic degradation ofwastewater pollutants. Titanium-dioxide-mediated oxidation of polynucleararomatic hydrocarbons. J. Photochem. Photobiol., A, 77, 83–88.

39 Wu, J.-M., Zhang, T.-W., Zeng, Y.-W., Hayakawa, S., Tsuru, K., and Osaka, A.(2005) Large-scale preparation of ordered titania nanorods with enhancedphotocatalytic activity. Langmuir, 21, 6995–7002.

40 Han, X., Kuang, Q., Jin, M., Xie, Z., and Zheng, L. (2009) Synthesis of tita-nia nanosheets with a high percentage of exposed (001) facets and relatedphotocatalytic properties. J. Am. Chem. Soc., 131, 3152–3153.

41 Li, X., Zhong, W., Zhang, D., Li, G., and Li, H. (2013)Large-specific-surface-area pure-phase TiO2 photocatalyst and preparationmethod and applications thereof China Patent CN103170319A.

42 Korovin, E., Selishchev, D., and Kozlov, D. (2016) Photocatalytic CO2 reduc-tion on the TiO2 P25 under the high power UV-LED irradiation. Top. Catal.,59, 1292–1296.


43 Qamar, S., Lei, F., Liang, L., Gao, S., Liu, K., Sun, Y., Ni, W., and Xie, Y.(2016) Ultrathin TiO2 flakes optimizing solar light driven CO2 reduction.Nano Energy, 26, 692–698.

44 Neatuea, S., Maciaà-Agullò, J.A., Concepcion, P., and Garcia, H. (2014)Gold–copper nanoalloys supported on TiO2 as photocatalysts for CO2 reduc-tion by water. J. Am. Chem. Soc., 136, 15969–15976.

45 Ramos-Delgado, N.A., Hinojosa-Reyes, L., Guzman-Mar, I.L.,Gracia-Pinilla, M.A., and Hernández-Ramírez, A. (2013) Synthesis by sol–gelof WO3/TiO2 for solar photocatalytic degradation of malathion pesticide.Catal. Today, 209, 35–40.

46 Dindar, B. and Içli, S. (2001) Unusual photoreactivity of zinc oxide irradiatedby concentrated sunlight. J. Photochem. Photobiol., A, 140, 263–268.

47 Hariharan, C. (2006) Photocatalytic degradation of organic contaminants inwater by ZnO nanoparticles: revisited. Appl. Catal., A, 304, 55–61.

48 Sin, J.-C., Lam, S.-M., Lee, K.-T., and Mohamed, A.R. (2013) Self-assemblyfabrication of ZnO hierarchical micro/nanospheres for enhanced photocat-alytic degradation of endocrine-disrupting chemicals. Mater. Sci. Semicond.Process., 16, 1542–1550.

49 Kohno, Y., Tanaka, T., Funabiki, T., and Yoshida, S. (1997) Photoreduction ofcarbon dioxide with hydrogen over ZrO2. Chem. Commun., 841–842.

50 Park, H.-A., Choi, J.H., Choi, K.M., Lee, D.K., and Kang, J.K. (2012) Highlyporous gallium oxide with a high CO2 affinity for the photocatalytic conver-sion of carbon dioxide into methane. J. Mater. Chem., 22, 5304–5307.

51 Wang, J.-C., Zhang, L., Fang, W.-R., Ren, J., Li, Y.-Y., Yao, H.-C., Wang, J.-S.,and Li, Z.-J. (2015) Enhanced photoreduction CO2 activity over directZ-scheme 𝛼-Fe2O3/Cu2O heterostructures under visible light irradiation.ACS Appl. Mater. Interfaces, 7, 8631–8639.

52 Yahaya, A.H., Gondal, M.A., and Hameed, A. (2004) Selective laser enhancedphotocatalytic conversion of CO2 into methanol. Chem. Phys. Lett., 400,206–212.

53 Wang, C., Zhang, H., Li, F., and Zhu, L. (2010) Degradation and mineral-ization of bisphenol A by mesoporous Bi2WO6 under simulated solar lightirradiation. Environ. Sci. Technol., 44, 6843–6848.

54 Xia, J., Di, J., Yin, S., Xu, H., Zhang, J., Xu, Y., Xu, L., Li, H., and Ji, M. (2014)Facile fabrication of the visible-light-driven Bi2WO6/BiOBr composite withenhanced photocatalytic activity. RSC Adv., 4, 82–90.

55 Amano, F., Nogami, K., Abe, R., and Ohtani, B. (2008) Preparation andcharacterization of bismuth tungstate polycrystalline flake-ball particles forphotocatalytic reactions. J. Phys. Chem. C, 112, 9320–9326.

56 See for instance Herrmann, J.-M., Matos, J., Disdier, J., Guillard, C., Laine, J.,Malato, S., and Blanco, J. (1999) Solar photocatalytic degradation of4-chlorophenol using the synergistic effect between titania and activatedcarbon in aqueous suspension. Catal. Today, 54, 255–265.

57 Matos, J., Laine, J., and Herrmann, J.-M. (2001) Effect of the type of activatedcarbons on the photocatalytic degradation of aqueous organic pollutants byUV-irradiated titania. J. Catal., 200, 10–20.

References 49

58 Lia, Y., Li, X., Lic, J., and Yin, J. (2006) Photocatalytic degradation of methylorange by TiO2-coated activated carbon and kinetic study. Water Res., 40,1119–1126.

59 Li, Y., Sun, S., Ma, M., Ouyang, Y., and Yan, W. (2008) Kinetic studyand model of the photocatalytic degradation of rhodamine B (RhB) by aTiO2-coated activated carbon catalyst: effects of initial RhB content, lightintensity and TiO2 content in the catalyst. Chem. Eng. J., 142, 147–155.

60 Lee, D.-K., Kima, S.-C., Kima, S.-J., Chung, I.-S., and Kim, S.-W. (2004) Pho-tocatalytic oxidation of microcystin-LR with TiO2-coated activated carbon.Chem. Eng. J., 102, 93–98.

61 Leary, R. and Westwood, A. (2011) Carbonaceous nanomaterials for theenhancement of TiO2 photocatalysis. Carbon, 49, 741–772.

62 Kemp, K.C., Seema, H., Saleh, M., Le, N.H., Mahesh, K., Chandra, V., andKim, K.S. (2013) Environmental applications using graphene composites:water remediation and gas adsorption. Nanoscale, 5, 3149–3171.

63 Chowdhury, S. and Balasubramanian, R. (2014) Graphene/semiconductornanocomposites (GSNs) for heterogeneous photocatalytic decolorization ofwastewaters contaminated with synthetic dyes: a review. Appl. Catal., B,160–161, 307–324.

64 Zhang, H., Lv, X., Li, Y., Wang, Y., and Li, J. (2010) P25-graphene compositeas a high performance photocatalyst. ACS Nano, 4, 380–386.

65 Kumar, S.G. and Devi, L.G. (2011) Review on modified TiO2 photocatalysisunder UV/visible light: selected results and related mechanisms on interfacialcharge carrier transfer dynamics. J. Phys. Chem. A, 115, 13211–13241.

66 Pelaez, M., Nolan, N.T., Pillai, S.C., Seery, M.K., Falaras, P., Kontos, A.G.,Dunlop, P.S.M., Hamilton, J.W.J., Byrne, J.A., O’Shea, K., Entezari, M.H., andDionysiou, D.D. (2012) A review on the visible light active titanium dioxidephotocatalysts for environmental applications. Appl. Catal., B, 125, 331–349.

67 Zaleska, A. (2008) Doped-TiO2: a review. Recent Pat. Eng., 2, 157–164.68 Lòpez, R., Gòmez, R., and Llanos, M.E. (2009) Photophysical and photocat-

alytic properties of nanosized copper-doped titania sol–gel catalysts. Catal.Today, 148, 103–108.

69 Slamet, H.W., Nasution, E., Purnama, S., and Kosela, J. (2005) Photocat-alytic reduction of CO2 on copper-doped titania catalysts prepared byimproved-impregnation method. Catal. Commun., 6, 313–319.

70 Zhou, Y., Chen, C., Wang, N., Li, Y., and Ding, H. (2016) Stable Ti3+

self-doped anatase-rutile mixed TiO2 with enhanced visible light utilizationand durability. J. Phys. Chem. C, 120, 6116–6124.

71 Senthilnathan, J. and Philip, L. (2010) Photocatalytic degradation of lindaneunder UV and visible light using N-doped TiO2. Chem. Eng. J., 161, 83–92.

72 Sojìc, D.V., Despotovìc, V.N., Abazovìc, N.D., Comor, M.I., andAbramovìc, B.F. (2010) Photocatalytic degradation of selected herbicidesin aqueous suspensions of doped titania under visible light irradiation. J.Hazard. Mater., 179, 49–56.

73 Li, P., Xu, J., Jing, H., Wu, C., Peng, H., Lu, J., and Yin, H. (2014) WedgedN-doped CuO with more negative conductive band and lower overpotential


for high efficiency photoelectric converting CO2 to methanol. Appl. Catal., B,156–157, 134–140.

74 Kubacka, A., Fernàndez-García, M., and Colòn, G. (2012) Advanced nanoar-chitectures for solar photocatalytic applications. Chem. Rev., 112, 1555–1614.

75 Kanhere, P. and Chen, Z. (2014) A review on visible light activeperovskite-based photocatalysts. Molecules, 19, 19995–20022.

76 Wang, W., Tadéa, M.O., and Shao, Z. (2015) Research progress of perovskitematerials in photocatalysis- and photovoltaics-related energy conversion andenvironmental treatment. Chem. Soc. Rev., 44, 5371–5408.

77 Tang, J., Zou, Z., and Ye, J. (2007) Efficient photocatalysis on BaBiO3 drivenby visible light. J. Phys. Chem. C, 111, 12779–12785.

78 Yoshitomi, F., Sekizawa, K., Maeda, K., and Ishitani, O. (2015) Selectiveformic acid production via CO2 reduction with visible light using a hybrid ofa perovskite tantalum oxynitride and a binuclear ruthenium(II) complex. ACSAppl. Mater. Interfaces, 7, 13092–13097.

79 Thuy, U.T.D., Liema, N.Q., Parlett, C.M.A., Lalev, G.M., and Wilson, K. (2014)Synthesis of CuS and CuS/ZnS core/shell nanocrystals for photocatalyticdegradation of dyes under visible light. Catal. Commun., 44, 62–67.

80 Li, X., Chen, J., Li, H., Li, J., Xu, Y., Liu, Y., and Zhou, J. (2011) Photoreduc-tion of CO2 to methanol over Bi2S3/CdS photocatalyst under visible lightirradiation. J. Nat. Gas Chem., 20, 413–417.

81 Liu, J.-Y., Garg, B., and Ling, Y.-C. (2011) CuxAgyInzZnkSm solid solu-tions customized with RuO2 or Rh1.32Cr0.66O3 co-catalyst display visiblelight-driven catalytic activity for CO2 reduction to CH3OH. Green Chem., 13,2029–2031.

82 Al Otaibi, B., Fan, S., Wang, D., Ye, J., and Mi, Z. (2015) Wafer-level artificialphotosynthesis for CO2 reduction into CH4 and CO using GaN nanowires.ACS Catal., 5, 5342–5534.

83 AlOtaibi, B., Kong, X., Vanka, S., Woo, S.Y., Pofelski, A., Oudjedi, F., Fan, S.,Kibria, M.G., Botton, G.A., Ji, W., Guo, H., and Mi, Z. (2016) Photochemi-cal carbon dioxide reduction on Mg-doped Ga(In)N nanowire arrays undervisible light irradiation. ACS Energy Lett., 1, 246–252.

84 Reddy, K.R., Hassan, M., and Gomes, V.G. (2015) Hybrid nanostructuresbased on titanium dioxide for enhancedphotocatalysis. Appl. Catal., A, 489,1–16.

85 Kuriki, R. and Maeda, K. (2017) Development of hybrid photocata-lysts constructed with a metal complex and graphitic carbon nitride forvisible-light-driven CO2 reduction. Phys. Chem. Chem. Phys., 19, 4938–4950.

86 Ettedgui, J., Diskin-Posner, Y., Weiner, L., and Neumann, R. (2011) Photore-duction of carbon dioxide to carbon monoxide with hydrogen catalyzed bya rhenium(I) phenanthroline−polyoxometalate hybrid complex. J. Am. Chem.Soc., 133, 188–190.

87 Cui, S.-C., Sun, X.-Z., and Liu, J.-G. (2016) Photo-reduction of CO2 usinga rhenium complex covalently supported on a graphene/TiO2 composite.ChemSusChem, 9, 1698–1703.

88 Nakada, A., Nakashima, T., Sekizawa, K., Maeda, K., and Ishitani, O. (2016)Visible-light-driven CO2 reduction on a hybrid photocatalyst consisting of a

References 51

Ru(II) binuclear complex and a Ag-loaded TaON in aqueous solutions. Chem.Sci., 7, 4364–4371.

89 Oelgemöller, M. (2016) Solar photochemical synthesis: from the beginnings oforganic photochemistry to the solar manufacturing of commodity chemicals.Chem. Rev., 116, 9664–9682.

90 Malato Rodríguez, S., JGálvez, J.B., Rubio, M.I.M., Fernández-Ibáñez, P.,Padilla, D.A., Pereira, M.C., Mendes, J.F., and de Oliveira, J.C. (2004) Engi-neering of solar photocatalytic collectors. Sol. Energy, 77, 513–524.

91 Abdel Maksound, Y., Iman, E., and Ramadan, A. (2016) TiO2 solarphotocatalytic reactor systems: selection of reactor design forscale-up and commercialization-analytical review. Catalysis, 6, 138.doi: 10.3390/catal6090138

92 Alfano, O.M., Bahnemann, D., Cassano, A.E., Dillert, R., and Goslich, R.(2000) Photocatalysis in water environments using artificial and solar light.Catal. Today, 58, 199–230.

93 Spasiano, D., Marotta, R., Malato, S., Fernandez-Ibañez, P., and Di Somma, I.(2015) Solar photocatalysis: materials, reactors, some commercial, andpre-industrialized applications. A comprehensive approach. Appl. Catal.,B, 170–171, 90–123.

94 Jung, C., Funken, K.-H., and Ortner, J. (2005) PROPHIS: parabolictrough-facility for organic photochemical syntheses in sunlight. Photochem.Photobiol. Sci., 4, 409–411.

95 De la Cruz, N., Dantas, R.F., Giménez, J., and Esplugas, S. (2013) Photolysisand TiO2 photocatalysis of the pharmaceutical propranolol: solar and artificiallight. Appl. Catal., B, 131–140, 249–256.

96 Alp, S., Ertekin, K., Horn, M., and Icli, S. (2004) Photostability studies ofthermo-mesomorphic derivatives of 2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione.Dyes Pigm., 60, 103–110.

97 Bolton, J.R., Bircher, K.G., Tumas, W., and Tolman, C.A. (2001)Figures-of-merit for the technical development and application of advancedoxidation technologies for both electric- and solar-driven systems (IUPACTechnical Report). Pure Appl. Chem., 73, 627–637.

98 Ciamician, G. (1912) The photochemistry of the future. Science, 36, 385–394.

53

3

Fundamentals of Photocatalytic Water Splitting(Hydrogen and Oxygen Evolution)Sanjib Shyamal, Paramita Hajra, Harahari Mandal, Aparajita Bera,Debasis Sariket, and Chinmoy Bhattacharya

Indian Institute of Engineering Science & Technology (IIEST), Shibpur, Department of Chemistry, Howrah711103, West Bengal, India

3.1 Introduction

One of the most important problems facing humanity in the twenty-firstcentury is building an enduring, sustainable energy economy. Although fossilfuels can supply the estimated global energy demand well into the foreseeablefuture, this strategy has catastrophic environmental implications due to carbondioxide (CO2) emissions, a leading contributor to the greenhouse gas effect[1]. The technology of renewable energy also has economic benefits such asthe reduced health and environmental restoration costs, job creation, and theintellectual property [2]. Even though the cost of renewable energy is dropping,it is still the major limitation for the implementation of these technologies.Continuous technological advances are needed to make these energy sourcescost-competitive with fossil fuels and drive the evolution of how we consumeenergy. Sunlight is an ideal energy source because it is, for all practical purposes,completely sustainable and delivers more energy to earth in 1 h than is consumedglobally per annum [3]. Although solar energy holds great promise, its large-scaleintegration requires an efficient conversion of light into storable, usable forms ofenergy. Nowadays, the most popular way to use solar energy is through photo-voltaic cells, which could directly convert solar energy into electricity. However,the electricity must be used immediately or stored in a secondary device, suchas capacitors and batteries. In comparison, the production of chemical fuels viaphotocatalytic processes is a more attractive approach to harness solar energy,which could harvest and store solar energy in the form of molecular bondsthrough a thermodynamic uphill reaction. Energy and environmental issues ata global level are important topics. It is indispensable to construct clean energysystems in order to solve these issues. Hydrogen will play an important role inthe system because it is an ultimate clean energy and can be used in fuel cells.Moreover, hydrogen is used in chemical industries. There is a need for H2 to usein fuel cells, resulting in more efficient and decentralized method of electricityproduction compared to combusting fossil fuels. Finally, H2 is used in industrial


54 3 Fundamentals of Photocatalytic Water Splitting (Hydrogen and Oxygen Evolution)

processes, including the Haber–Bosch cycle and Fischer–Tropsch synthesis,indicating that there is an immediate method for implementing renewablyformed H2. At present, hydrogen is mainly produced from fossil fuels such asnatural gas by steam reforming. Hydrogen has to be produced from water usingnatural energies such as sunlight if one thinks of energy and environmentalissues. There has been substantial research into conversion technologies forcarbon-neutral energy formation such as biomass production from green algae[4, 5], photovoltaic devices for electricity production [6, 7], and water splitting,or the direct photoelectrochemical (PEC) conversion of water into its constituentelements, H2 and O2 [8–13]. All of these methods will need to contribute as wepush toward an energy independent nation which relies solely on sustainablesources. Water splitting, though, will have the farthest reaching impact for thefollowing reasons. The ability to generate H2 from water has vast uses in theenergy sector beginning with its storage in a liquid fuel to replace gasoline.Photochemical water splitting into hydrogen and oxygen using semiconductor(SC) photocatalysts has become a promising strategy for effective capturing ofoptical energy (solar) into clean H2 fuel through a cost-effective, environmen-tally benign route. The overall photocatalytic water-splitting reaction may beconsidered into three steps: (i) absorption of light followed by electron–holeseparation inside the semiconductor, (ii) migration of the charge carriers towardthe surface, and (iii) chemical reactions at the surface of the semiconductormaking H2 or O2 evolution. To meet the target, researchers are interested inbandgap modifications, as well as coupling with narrow bandgap semiconduc-tors, dye sensitization, surface plasmonic effect to improve visible-light activityof the semiconductors [10, 14–17]. The efficient charge carrier separation andtransportation may be achieved through physical modifications such as growthof nanostructured semiconductors with different morphologies for minimumcharge diffusion length, high crystallinity, and fewer defects.

Finally, the third step is achieved by the application of a suitable H2-evolutionor O2-evolution cocatalyst, which could extract photogenerated charge carri-ers, host active sites for catalytic H2 or O2 evolution, and improve the stabilityof photocatalysts by suppressing photocorrosion. Therefore, cocatalysts play asignificant role in improving both activity and stability of semiconductor photo-catalysts.

3.2 Strategy for Development of Photocatalyst Systemsfor Water Splitting

Photocatalytic reactions occurring at the semiconductor–aqueous electrolyticinterface may be represented in Figure 3.1.

In metal oxide semiconductors, metal-based orbitals comprise the conductionband (CB) minima. A CB is analogous to the lowest unoccupied molecular orbital(LUMO). The valence band (VB) of an oxide is typically derived from the filled 2porbital of the oxygen and is analogous to the highest occupied molecular orbital(HOMO). The bandgap (Eg) is the energy gap between the VB and CB. The band

3.2 Strategy for Development of Photocatalyst Systems for Water Splitting 55

Figure 3.1 Band positioning forphotoelectrochemical water splitting.

E

hν

Band gap ≥1.23 eV

CB

e–

2H+

h+

VB

H2

H2O

O2

+

–

Figure 3.2 Energy levels ofdifferent materials: conductor,semiconductor, and insulator.

E

Band gap energy (eV)

VB

Overlap

CB

CB

CB

VB

VB

Fermi level

Conductor

Semiconductor

Insulator

≤3.2 eV

structure of a semiconductor is presented in Figure 3.2. The electronic structureof oxides is distinct from most covalent semiconductors, such as Si, and leadsto many of the desired properties such as stability toward corrosion and passi-vation. In d0 oxides, such as the commonly studied TiO2, this gap is typicallylarge (Eg > 3 eV), and, therefore, these materials are not intrinsic semiconductors,which implies that without an imperfection or suitable doping, these materialsare almost insulating.

For example, oxygen vacancies typically result in some fraction of the metalcomponent in the oxide being reduced, which contributes to the donor level thatexists as a state just below the CB minimum. This results in increase of the Fermilevel energy (EF), which is quite important to describe the solid state materialsand their interfacial properties. This is defined as the energy level at which theprobability of occupation by an electron is half; for example, for an intrinsic SCthe Fermi level lies at the mid-point of the bandgap (Figure 3.3a). For an n-typesemiconductor (Figure 3.3b), the Fermi level lies just below the CB, whereas fora p-type semiconductor lies precisely above the VB (Figure 3.3c). In addition, aswith metal electrodes, the Fermi level of a semiconductor electrode varies with


EF

CB

EC

EV

VB

(a)

EC

EV

EE

ED

VB

CB

(b)

EF

EF

EC

E

VB

CB

(c)

EA

EV

Figure 3.3 Schematic diagram of the energy levels of (a) an intrinsic, (b) an n-typesemiconductor, and (c) a p-type semiconductor. EC is the energy level of the conduction band(CB); EV is the energy level of the VB; EF is the Fermi level energy; EA and ED are the acceptorand donor level energy of the semiconductors, respectively.

the applied potential; for example, moving toward more negative potentials willraise the Fermi level. The donor is typically ionized at room temperature, and theextra electron resides in the CB [18]. These electrons are predominantly responsi-ble for conducting current, and therefore, most oxide semiconductors are consid-ered n-type. In these semiconductors, electrons are the majority carrier, and holesare considered the minority carrier. The donor density (ND), or extra number ofelectrons, controls the electrostatics at the electrode–electrolyte interface and ismeasurable. The promotion of electrons leaves a positively charged vacancy in thevalence, which is referred to as a hole. These holes can be moved through spaceby the transfer to an electron to the vacancy; therefore, holes are considered to bemobile. Electrons can be excited to the CB either thermally or photochemically.However, there is another method for generating charge carriers (i.e., electrons orholes) within a semiconductor, referred to as doping. Doping involves the addi-tion of a different element into the semiconductor and can change the distribu-tion of electrons within the solid, and hence changes the Fermi level. The simplestexample of this involves the introduction of a Group V element (e.g., P/As) or aGroup III element (e.g., B/Al) into a Group IV element (e.g., Si or Ge). The addi-tion of P into Si introduces occupied energy levels into the bandgap close to thelower edge of the CB, thereby allowing facile promotion of electrons to the CB(Figure 3.3b). The addition of Al introduces vacant energy levels into the bandgapclose to the upper edge of the VB, which allows facile promotion of electrons fromthe VB (Figure 3.3c). This leads to the formation of holes in the VB.

3.3 Electrochemistry of Semiconductors at theElectrolyte Interface

The first systematic investigation of the semiconductor surface in contact withan electrolyte was reported by Brattain and Garrett [19]. Later it was further

3.3 Electrochemistry of Semiconductors at the Electrolyte Interface 57

demonstrated by Gerischer, Memming, Pleskov, and Nozik. The basic principlesof semiconductor electrochemistry are described in their review articles andpapers and also included in the textbooks [20–26]. The important equations,fundamental principles, and appropriate terminology have been summarized byInternational Union of Pure and Applied Chemistry [27].

The anisotropic forces at the electrode–electrolyte interface and the chargetransfer process occurring at the interface leads to rearrangement of electrons orholes inside the SC electrode and ions and solvent dipoles inside the electrolyte.The charge distributions on the electrode side are widely different for metals andsemiconductors, in the following way:

1) Electrons and holes are the charge carriers in SC, whereas in metals, electronsare the only charge carriers.

2) The carrier density is low in the case of semiconductors (∼1014–1018/cc) thanin metals (∼1022–1024/cc).

3) For metals, the charges are located at the surface, while for the semiconduc-tors, the excess charge extends into the electrode throughout a significantdistance (100–10 000 Å). This region of potential drop is referred to as thespace charge region.

The nature of the space charge layer depends upon the way in which chargetransfer occurs across the interface. The band edges in the interior of the semi-conductor also vary with the applied potential in the same way as the Fermi level.There are three different situations to be considered:

1) At a certain potential, the Fermi energy lies at the same energy as the solutionredox potential (Figures 3.4a and 3.5a). There is no net transfer of charge, andhence there is no band bending. This potential is referred to as the flat-bandpotential, Efb.

2) Depletion regions arise at potentials positive of the flat-band potential for ann-type semiconductor and at potentials negative of the flat-band potential fora p-type semiconductor (Figures 3.4b and 3.5c).

3) At potentials negative of the flat-band potential for an n-type semiconductor,there is excess of the majority charge carrier (electrons) in the space chargeregion, which is referred to as accumulation region (Figure 3.4c). An accumu-lation region arises in a p-type semiconductor at potentials more positive thanthe flat-band potential (Figure 3.5b).

The charge transfer abilities of a semiconductor electrode depend on whetherthere is an accumulation layer or a depletion layer. If there is an accumulationlayer, the behavior of a semiconductor electrode is similar to that of a metallicelectrode, since there is an excess of the majority of charge carrier available forcharge transfer.

On the contrary, in case of depletion layer, there are few charge carriers avail-able for charge transfer, and electron transfer reactions do not occur sponta-neously. However, if the electrode is exposed to radiation of sufficient energy,electrons can now be promoted to the CB. If this process occurs in the interiorof the semiconductor, recombination of the promoted electron and the resultinghole typically occurs, together with the production of heat. However, if it occurs


(b)

E > Efb

(a)

E = Efb

(c)

E < Efb

Space charge

region (depletion)

Bulk SC

x = 0

EC

EF

EV

Bulk

electrolyte

EV

EF

EC

x = 0

Bulk SC

Accumulation

Bulk

electrolyte

Bulk SC

x = 0

EC

EF

EV

Bulk

electrolyte

Figure 3.4 Effect of varying the appliedpotential (E) on the band edges in theinterior of an n-type semiconductor.

in the space charge region, the electric field in this region will cause the separationof the charge. For example, for an n-type semiconductor at positive potentials,the band edges curve upwards and the hole moves toward the interface, and theelectron moves to the interior of the semiconductor. The hole is a high-energyspecies that can extract an electron from a solution species, so that the n-typesemiconductor electrode acts as a photoanode.

3.4 Effect of Light at the Semiconductor–ElectrolyteInterface

At a certain frequency (or wavelength) of the incident light corresponding tothe energy larger than the bandgap energy, electrons and holes are separated

3.4 Effect of Light at the Semiconductor–Electrolyte Interface 59

Figure 3.5 Effect of varying the appliedpotential (E) on the band edges in theinterior of a p-type semiconductor.

(b)

E > Efb

(a)

E = Efb

(c)

E < Efb

Bulk

electrolyte

EV

EF

EC

x = 0

Bulk SC

Bulk

electrolyte

EV

EF

EC

x = 0

Bulk SC

Accumulation

Bulk

electrolyte

Depletion

Bulk SC

x = 0

EC

EF

EV

and migrated to the conduction and VBs, respectively. Under illumination, thesee−/h+ pairs create an unequal distribution of charge, disrupting the equilibriumat the semiconductor–electrolyte interface, and the formed photovoltage drivesa photocurrent. The photogenerated electrons reduce H2O at the CB to H2 andthe holes oxidize H2O to O2 at the VB, mimicking the natural photosynthesis.For effective photoelectrolysis to occur at the semiconductor–electrolyte inter-face, the bottom of the CB should be more negative than the hydrogen reductionpotential while the top of the VB should be more positive than the water oxida-tion potential:

2H+ + 2e−h𝜈

−−−−→ H2 (E∘ = 0.00 V vs. SHE) (3.1)

2H2O + 4h+ h𝜈−−−−→ O2 + 4H+ (E∘ = 1.23 V vs. SHE) (3.2)


TiO2-ASrTiO3

ZnS

TiO2-R

CdS CdTe CdSeSi

WO3 Fe2O3MoS2

ZnO

ZrO2

3.6

3.2

3.2

3.0

2.4 1.4

1.7 1.1

2.8

2.3

1.7

5

3.2

5.0

–2.0

–1.0

0.0

1.0

2.0

3.0

4.0

E vs NHE (pH = 0) E vs Ag/AgCl (pH = 7)

–2.6

–1.6

–0.6

0.4

1.4

2.4

3.4Water splitting

O2/H2O

H2/H2O

Figure 3.6 Band positions of different semiconductors with respect to the thermodynamicpotentials of water splitting.

Therefore, the minimum bandgap of the semiconductor for water splitting isexpected to be 1.23 eV corresponding to light of wavelength ∼1100 nm. Relativeband positions of different semiconductors are shown in Figure 3.6 [28].

The individual band levels usually shift with the change in pH of the elec-trolytic media according to the Nernst’s equation (0.059 V/pH). Among thedifferent metal oxides, ZrO2, KTaO3, SrTiO3, and TiO2 possess suitable bandstructures for water splitting. These materials with high bandgap energies(suitable under ultraviolet (UV) irradiation) are active for water splitting whenthey are selectively modified with different cocatalysts [29–31]. The bandgap of avisible-light-driven photocatalyst should be narrower than 3.0 eV (𝜆> 415 nm).In order to increase the activity of the photocatalysts under visible light,several approaches are considered, such as incorporation of transition metalsand nonmetals, creating oxygen vacancies, and plasmon interaction [32, 33].Quasi-stable energy states are generated by these processes and the visible-lightphotons are able to create excitons, leading to the photocatalytic reaction.Incorporation of crystalline defects in metal oxide semiconductors in the formof vacancies is an effective way to increase visible-light absorption. On the otherhand, CdS with the bandgap energy of 2.4 eV, that is, visible-light responsive, andindividual band positions suitable for water splitting; however, it is not effectivein the reaction to form H2 and O2, because under illumination, CdS is oxidizedby photogenerated holes rather than H2O.

CdS + 2h+ → Cd2+ + S (3.3)

Similarly, ZnO undergoes photocorrosion under illumination.

ZnO + 2h+ → Zn2+ + 1/2O2 (3.4)

These photocorrosion processes generally limit the applicability of differentsemiconductors in the electrolytic media under illumination. However, withincorporation of suitable charge scavengers in the electrolytic media, these

3.4 Effect of Light at the Semiconductor–Electrolyte Interface 61

materials may behave as effective photocatalyst for water-splitting process.Organic compounds, such as alcohols (methanol, ethanol, isopropanol, etc.),acids (formic acid, acetic acid, etc.), and aldehydes (formaldehyde, acetaldehyde,etc.) have all been used as electron donors for photocatalytic hydrogen genera-tion. Among them, methanol was most widely used for the hydrogen generationprocess. For example, in the presence of a hole scavenger (SO3

2−, methanol, S2−,etc.), CdS can effectively produce H2 from water under visible-light irradiation.Similarly, in the presence of an electron acceptor (Ag+, Fe3+, etc.), WO3 acts asa good photocatalyst for O2 evolution under visible-light irradiation. Thus, foran effective photocatalyst, suitable band engineering is essential for visible-lightabsorptivity.

PEC hydrogen and oxygen evolution reaction using holes and electronscavengers:

MeOH + H2Oh𝜈, Photocatalyst (h+)−−−−−−−−−−−−−−−→HCHO + H2 (3.5a)

HCHO + H2Oh𝜈, Photocatalyst (h+)−−−−−−−−−−−−−−−→ HCOOH + H2 (3.5b)

HCOOH + H2Oh𝜈, Photocatalyst (h+)−−−−−−−−−−−−−−−→ CO2 + H2 (3.5c)

SO2−3 + H2O

h𝜈, Photocatalyst (h+)−−−−−−−−−−−−−−−→ SO2−

4 + H2 (3.5d)

Fe3+ + H2Oh𝜈, Photocatalyst (h+)−−−−−−−−−−−−−−−→ Fe2+ + O2 (3.6a)

Ag+ + H2Oh𝜈, Photocatalyst (h+)−−−−−−−−−−−−−−−→Ag ↓ + O2 (3.6b)

The metal oxides are the most stable materials for PEC water-splitting reaction;however, the conventional metal oxides have VB edges that are much too low inenergy with respect to the water oxidation potential (1.23 V versus normal hydro-gen electrode (NHE), at pH 0), resulting in an enormous over potential (>1.7 V).Consequently, metal oxides with VBs comprised predominantly of O (2p) orbitalhaving the bandgap energy that is only responsive to UV light (<400 nm). Manylarge bandgap semiconductors have been discovered for overall water splitting.To describe the true hydrogen production efficiency of a water-splitting reactionunder sunlight, a term called “solar-to-hydrogen” conversion efficiency (STH) isoften used. The definition of STH conversion efficiency may be represented as

STH = (Jsc × 1.23 × 𝜂F)/

Ptotal (3.7)

where Ptotal represents the power density of the incident simulative sunlight(AM1.5G), J sc is the short-circuit photocurrent density, the thermodynamicvoltage required for water splitting is 1.23 V, and 𝜂F is the faradic efficiency. Inorder to become viable and meet the energy demand, a water-splitting systemshould be ∼10% STH efficient, stable over an extended period of time, andcost–effective to fossil fuels. The solar output in the visible part of the spectrumis significantly higher (∼43%) than the UV (4%). As the PEC reaction relies onabsorbing light, the greater number of photons absorbed, the higher will be thetheoretical efficiency of a system. The target goal of 10% STH efficiency alsonecessitates the use of a material that significantly absorbs the visible portion ofthe solar spectrum.


The charge separation and migration of photogenerated carriers inside thesemiconductor matrix also play a crucial role. Crystalline properties of thesemiconductor influence these processes at the semiconductor–electrolyteinterface. For a highly crystalline semiconductor, there will be a minimumamount of defects in it. The defects act as trapping and recombination sitesbetween photogenerated charge carriers resulting in lowering of PEC activity.With decrease in particle size, the distance traveled by the photogeneratedcarriers to the reaction sites becomes short resulting in a decrease in therecombination probability.

The surface chemical reactions for the PEC water splitting involve surfaceactive sites and effective surface area. The photogenerated electrons and holesmay recombine with each other, in the absence of any active sites for redoxreactions on the surface, although they may possess sufficient thermodynami-cally potentials for water splitting. Different cocatalysts are loaded to introduceactive sites for H2 evolution reaction (HER) as the CB levels of many oxidephotocatalysts are not high enough to reduce water to H2, or due to the slowerkinetics for this reaction. Also, cocatalysts are required for four electron H2Ooxidation to O2 since the VB for the metal oxide photocatalysts may be lowenough for this oxidation reaction.

3.5 Conversion and Storage of Sunlight

Currently, 96% of H2 produced is derived from fossil fuels [34]. PEC water split-ting is the most direct mechanism in which H2 can be produced from H2O, andthis technology can potentially replace natural gas reformation as our major H2source. The products, H2 and O2, can be produced at separate electrodes, thusalleviating the safety concerns with forming large quantities of reactive gasesin the same chamber. Water splitting can be carried out at room temperatureand most importantly, the inorganic materials used to construct water-splittingdevices offer a degree of chemical robustness and durability that is difficult toachieve in organic or biological systems. This is crucial as the mechanism forsustainable energy conversion must be cost-efficient.

Water splitting into H2 and O2 is accompanied by a large positive change in theGibbs free energy (237 kJ mol−1), that is, it is an uphill reaction. It is further com-plicated by the kinetic limitations associated with the four-electron oxidation ofwater, which proceeds through highly corrosive and reactive intermediates [35].Sunlight can provide the necessary energy input. However, that energy needs tobe efficiently collected through absorption processes and transferred into a watermolecule to break its chemical bonds. Classification of the solar energy conver-sion processes using direct solar radiation has been presented in Figure 3.7. It isargued that very sophisticated materials engineering must be used for processingthe materials that will satisfy the species requirements for photoelectrodes.

An important issue in the processing of these materials is the bulk versusinterface properties at the solid–solid interfaces (e.g., grain boundaries) andsolid–liquid interfaces (e.g., electrode–electrolyte interface). Consequently, the

3.6 Electrolysis and Photoelectrolysis 63

Solar energy

IR region(52%)

UV region(< 3%)

Visibleregion (44%)

Solar thermal conversion Solar photoconversion

Photochemical process Photoeffects in semiconductor

Biological Non biological

Photogalvanic Photochemical &photocatalytic

Photochemical

Figure 3.7 Classification of the solar energy conversion processes using directly the solarradiations.

development of PECs with the efficiency required for commercialization requiresthe application of up-to-date materials processing technology. The performanceof PECs is considered in terms of: (i) excitation of electron–hole pair in pho-toelectrodes; (ii) charge separation in photoelectrodes; (iii) electrode processesand related charge transfer within PECs; (iv) generation of the PEC voltagerequired for water decomposition. It is accepted that PEC technology is the mostpromising technology for hydrogen production owing to several reasons:1) PEC technology is based on solar energy, which is a perpetual source of energy,

and water, which is a renewable resource.2) PEC technology is environmentally benign, safe with no undesirable by prod-

ucts.3) PEC technology may be adopted on small as well as large scales.4) PEC technology is relatively less complicated.

3.6 Electrolysis and Photoelectrolysis

In electrolysis of water, an external voltage of approximately 2.0 V is required,which includes 1.23 V thermodynamic energy requirement and polarization


H2

(a) (b)

H2

H2O

H2O

O2

hv

O2

H2O

H2O

M

CB e–

VB h+

Eg

E

T

A

L

M

E

T

A

L

Figure 3.8 Water splitting: (a) electrolysis and (b) photo-assisted electrolysis.

losses but in photo-assisted electrolysis of water, this requirement is reducedbut still the external voltage is required which is called as bias voltage as shownin Figure 3.8. The semiconductor absorbs the light and produces electrons andholes. For an n-type semiconductor, electrons are moved to cathode, whereasholes oxidize water to produce oxygen. More than 100 materials and theirderivatives have been explored till date to catalyze the overall water splitting orcause water oxidation or reduction in presence of an external redox mediator[36–38]. However, none of them are capable of catalyzing overall water splittinghaving quantum efficiency greater than 10% has been reported.

Conduction band

Conduction band

O2

(Reducing agents)

(Oxidizing agents)

H2 evaluation reaction O2 evaluation reaction

Valence band

Valence band

e–

e–

h+

h+

H2

H2O

M+

M

H2O

SO32–

SO42–

hνhν

Figure 3.9 PEC water splitting in the presence of sacrificial reagents.

3.7 Development of Photocatalysts for Solar-Driven Water Splitting 65

Sacrificial reagents are often employed to evaluate the photocatalytic activityfor water splitting (Figure 3.9). When the photocatalytic reaction is carried out inan aqueous solution including a reducing reagent, that is, electron donors or holescavengers, such as alcohol and SO3

2−, photogenerated holes irreversibly oxidizethe reducing reagent instead of water. It enriches electrons in a photocatalyst, anda H2 evolution reaction is enhanced. On the other hand, photogenerated elec-trons in the CB are consumed by oxidizing reagents, that is, electron scavengerssuch as Ag+ and Fe3+ resulting in improved O2 evolution reaction. These reac-tions using sacrificial reagents are studied to evaluate if a certain photocatalystsatisfies the thermodynamic and kinetic potentials for H2 and O2 evolution. Thesereactions are regarded as half reactions of water splitting and are often employedas test reactions of photocatalytic H2 or O2 evolution.

3.7 Development of Photocatalysts for Solar-DrivenWater Splitting

Fujishima and Honda first discovered the water splitting using TiO2 anode andplatinum cathode by UV light [39]. Wrighton et al. showed the hydrogen pro-duced by photoelectrolysis using platinized SrTiO3 [40, 41]. Shi et al. exploredthe concept of photochemical diodes and showed the necessity of a bias poten-tial for rutile photoanodes [42]. Hensel et al. reported the composite effects ofCdSe with nitrogen-dopped TiO2 could narrow its bandgap to the response intothe visible region [43]. Park et al. employed hydrothermal method and used TiO2nanoribbons morphology to narrow the bandgap [44]. Improved solar energyconversion efficiencies for the photocatalytic production of hydrogen via TiO2semiconductor electrodes, heat treatment of Ti metal found to influence perfor-mance [45]. Similarly, highly active oxide photocathodes like p-Cu2O, CuBi2O4have been developed for PEC water reduction [46]. The positions of the CB andVBs with respect to water oxidation and reduction potentials are one require-ment for an efficient water-splitting photoelectrode [47, 48]. The positions of theVB and CB edges, relative to the redox potentials of water, for a variety of semi-conductors has been shown in Figure 3.4. Among the different materials the per-ovskite structures of oxides SrTiO3 [29, 30, 49–52] and BaTiO3 [53, 54] indicatethat both of these materials are better suited than TiO2 for water photoelectrol-ysis due to their CB edges are more negative than the water reduction potential.Unfortunately, their bandgaps are so large (>3.2 eV) that they require majorlyUV light to produce photocurrent resulting in very poor photoconversion effi-ciencies. Smaller bandgap metal oxide semiconductors, notably WO3 [55–57],Fe2O3 [58–60], and Cu2O [61], have also been extensively studied. Although thesematerials harvest a larger fraction of the solar spectrum, their CB edges are oftennot optimally positioned for the production of hydrogen [62, 63]. These materi-als are difficult to dope, resulting in lower electrical conductivity and susceptibleto photocorrosion. Determination of the band structure of different metal oxidesemiconductors indicates that its VB positions are nearly the same and very pos-itive. Holes at such a positive potential can easily oxidize water with expenses


Counter

electrode

Photoelectrode

300W

Xe

Lamp

e–

h+

H2O

H2 O2

Ag/AgCl

electrodee–

Potentiostat

Figure 3.10 Schematic diagram of a prototype photoelectrochemical cell for water splitting.

of significant amount of energy. One idea to both extends the spectral responseof large gap oxide semiconductors and to raise the VB energy is to incorporatesulfur, carbon, or nitrogen into the lattice. A series of these compounds, includ-ing Sm2Ti2S2O5 [64], TaON [65], Y2Ta2O5N2 [66], Ta3N5, and LaTiO2N [67] aswell as incorporating into TiO2 [44, 68, 69] have recently been reported by sev-eral research groups. It has been further demonstrated that the bottom of the CBconsists of empty d-orbitals and depending upon the material, resulting in smallerbandgaps for improved solar energy utilization. However, most of these systemsare reported to function in presence of the “sacrificial reagents” in solution [70].

Photochemical water splitting combines the light harvesting ability of asemiconductor with the chemically active surface of an electrolyzer into a singledevice, leading to lower overall device costs. Currently, many PEC devices atthe research level require the input of electrical energy to drive water-splittingreaction at an appreciable rate; however, this results in a loss of efficiency. Thebasic configuration of a PEC cell is shown in Figure 3.10, which consists ofthe photoactive semiconductor electrodes immersed in electrolytic solution.Under illumination with solar light, electron–hole pairs are generated in thesemiconductor. Then, photogenerated electrons transfer to the metal cathodethrough the external circuit. The photoanode needs to oxidize water to produceoxygen efficiently, while for the metal, a low overpotential for reduction ofprotons to hydrogen is required.

3.8 Approaches to Develop Visible-Light-AbsorbingMetal Oxides

Over the past 40 years, there have been many approaches toward increasingvisible-light absorption in metal oxides. The classic and well-studied methodis through doping into wide bandgap semiconductors. Metal cations are

3.8 Approaches to Develop Visible-Light-Absorbing Metal Oxides 67

substituted into large bandgap semiconductors resulting in colored oxides andthe formation of these colored domains is the result of defect structure in thecatalyst [71]. These states allow for new absorption processes that take placein the visible portion of the solar spectrum. Doped oxides are often successfulas photocatalysts for the degradation of organic dye, mostly because these trapstates are capable of facile electron transfer to organic species in solution [72].However, the significant amount of doping needed for the light absorptionbecomes detrimental to sustaining photocurrent in an electrode material, and inmany cases the native oxide performs better as a photoanode. In general, dopingdoes not result in a band shift, but the formation of new absorption pathwaysin the visible portion of the solar spectrum [47, 73, 74]. Additionally, thesenew absorption events resulting in low absorption property, that is, althoughvisible-light absorption is taking place, it is inefficient. This problem may beavoided through the formation of solid solutions.

CdS with a 2.4 eV bandgap is a well-known metal sulfide photocatalyst that canproduce H2 under visible-light irradiation in the presence of a sacrificial reagent.ZnS with 3.6 eV bandgap is also a well-known photocatalyst for H2 evolutionthough it responds to only UV. It shows high activity without any assistanceof cocatalysts such as Pt. Therefore, ZnS is an attractive host photocatalyst fordoping and preparing solid solutions. Photocatalytic H2 evolution on CuInS2,CuIn5S8, AgGaS2, and AgIn5S8 has been reported in the presence of sacrificialreagents (SO3

2−–S2−) [75]. These metal sulfides consist of elements of Groups11 and 13. NaInS2 with layered structure and ZnIn2S4 with spinel structureare active. Feng and coworkers have reported unique photocatalysts of indiumsulfide compounds with open-framework structure [76].

Recently, the researchers are largely interested in developing TiO2-basedstable photocatalysts suitable for use in visible-light irradiation throughmodifications of electronic and optical properties by incorporating differentdopants such as metals and nonmetals and creation of oxygen vacancies [32, 33,77–81]. In this context, purely UV active TiO2 semiconductor transformed tovisible-light-absorbing material (∼2.8 eV) on addition of a suitable amount ofvanadium (30%) when the composite was developed through simple drop-casttechnique. These materials are known to have improved PEC water oxidationbehavior. The chemically synthesized In2O3 semiconductor, on proper annealingconditions, demonstrates partial visible-light responsiveness, due to the indirectbandgap of ∼2.78 eV, and stable water oxidation behavior of this material wasidentified. Thermochemically synthesized “scheelite” structured BiVO4 and itsbinary component Bi2O3 exhibit efficient oxygen evolution from water underillumination. Another ternary metal vanadate, (Fe—V–oxide) with primarilyvisible-light activity is found to be suitable for PEC water splitting. All the abovematerials are n-type in nature when characterized for semiconductor–electrolyteinterface. For H2 production, the p-type cuprous oxide (Cu2O) is well-knownphotocatalyst with bandgap energy of 1.9–2.2 eV. The sequential modificationswith substrates, dopants, or other added elements for better connectivity orcrystallinity lead to reduced charge carrier recombination toward enhanced H2production in presence of visible light.


3.9 Conclusions

In this chapter, we discuss the PEC water-splitting mechanism, with particularemphasis on visible-light-active semiconductor materials suitable for produc-tion of H2 and O2 from water. In the last 40 years, different combinationsof semiconductor materials and electrocatalysts have been configured forphotoelectrolysis research. The search for earth-abundant materials that canbe used in solar water-splitting cells remains an important goal for affordableand environmentally benign methods for solar energy conversion and storage.Titania, a nontoxic, naturally occurring semiconductor, can be photoexcitedunder visible light aimed at solar-driven water splitting through suitable metalor nonmetal doping, which can be incorporated as substitutional or interstitialstate in the crystal lattice. Photoelectrode stability continues to be a majorchallenge for the development of efficient photocathodes and photoanodes. Assolar fuels research expands, standardizing both research methodologies andcharacterization techniques becomes paramount for accurate reporting andultimately helps to move the field forward into new areas of development anddiscovery. Presently, the available efficiency for overall water-splitting systemsfor simultaneous hydrogen and oxygen production under visible-light irradiationis still quite low due to fast charge recombination and backward reactions. Toachieve enhanced and sustainable hydrogen production, the continual additionof electron donors (sacrificial hole scavengers) is required to make up half ofthe water-splitting reaction to reduce H2O to H2 at the suitably placed CBof the semiconductors. Taking into account the lowering cost for solar-to-H2energy conversion, polluting by-product from industries and low-cost renewablebiomass from animals or plants are preferential sacrificial electron donorsin water-splitting systems. The molecular mechanisms and reaction kineticsneed to be considered carefully when designing such photocatalytic hydrogenproduction systems. Nevertheless, such an admirable goal for the practicalapplication of H2—O2 produced through water-splitting systems is especiallyinteresting in light of worldwide energy and environmental concerns.

References

1 Wigley, T.M., Richels, R., and Edmonds, J.A. (1996) Economic and envi-ronmental choices in the stabilization of atmospheric CO2 concentrations.Nature, 379, 240–243.

2 National Renewable Energy Laboratory (1997) Dollars from Sense: The Eco-nomic Benefits of Renewable Energy, U.S. Department of Energy.

3 Lewis, N.S. and Nocera, D.G. (2006) Powering the planet, chemical challengesin solar utilization. Proc. Natl. Acad. Sci. U.S.A., 103, 15729–15735.

4 Demirbas, M.F. (2011) Biofuels from algae for sustainable development. Appl.Energy, 88, 3473–3480.

5 Koller, M., Salerno, A., Tuffner, P., Koinigg, M., Boechzelt, H., Schober,S., Schnitzer, H., Mittelbach, M., and Barunegg, G. (2011) Efficient

References 69

implementation of micro-algae for sustainable environmental and economicdevelopment. Curr. Top. Biotechnol., 6, 71–82.

6 El Charr, L., Lamont, L.A., and Zein, N.E. (2011) Review of photovoltaictechnologies. Renewable Sustainable Energy Rev., 15, 2165–2175.

7 Tian, B.T., Kempa, J., and Lieber, C.M. (2009) Single nanowire photovoltaics.Chem. Soc. Rev., 38, 16–24.

8 Li, Z., Luo, W., Zhang, M., Feng, J., and Zou, Z. (2013) Photoelectrochemicalcells for solar hydrogen production: current state of promising photoelec-trodes, methods to improve their properties, and outlook. Energy Environ.Sci., 6, 347–370.


10 Kudo, A. (2003) Photocatalyst materials for water splitting. Catal. Surv. Asia,7, 31–38.

11 Takata, T., Tanaka, A., Hara, M., Kondo, J.N., and Domen, K. (1998) Recentprogress of photocatalysts for overall water splitting. Catal. Today, 44, 17–26.

12 Paleocrassas, S.N. (1974) Photocatalytic hydrogen production: a solar energyconversion alternative? Sol. Energy, 16, 45–51.

13 Heller, A. (1981) Conversion of sunlight into electrical power and photoas-sisted electrolysis of water in photoelectrochemical cells. Acc. Chem. Res., 14,154–162.

14 Domen, K., Kondo, J.N., Hara, M., and Takata, T. (2000) Photo- andmechano-catalytic overall water splitting reactions to form hydrogen andoxygen on heterogeneous catalysts. Bull. Chem. Soc. Jpn., 73, 1307–1331.

15 Domen, K., Hara, M., Kondo, J.N., Takata, T., Kudo, A., Kobayashi, H., andInoue, Y. (2001) New aspects of heterogeneous photocatalysts for waterdecomposition. Korean J. Chem. Eng., 18, 862–866.

16 Kudo, A. (2006) Development of photocatalyst materials for water splitting.Int. J. Hydrogen Energy, 31, 197–202.

17 Maeda, K. and Domen, K. (2007) New non-oxide photocatalysts designed foroverall water splitting under visible light. J. Phys. Chem. C, 111, 7851–7861.

18 Dmochowski, J.E. (2007) Donors in semiconductors–are they understood inelectronic era? J. Phys. Conf. Ser., 79, 12010–12014.

19 Brattain, W.H. and Garrett, C.G.B. (1955) Experiments on the interfacebetween germanium and an electrolyte. Bell Syst. Tech. J., 34, 129–176.

20 Nozik, A.Z. (1978) Photoelectrochemistry: applications to solar energy con-version. Annu. Rev. Phys. Chem., 29, 189–222.

21 Rajeshwar, K., Singh, P., and Dubow, J. (1978) Energy conversion in photo-electrochemical systems – a review. Electrochim. Acta, 23, 1117–1144.

22 Memming, R. (1980) Solar energy conversion by photoelectrochemical pro-cesses. Electrochim. Acta, 25, 77–88.

23 Gerischer, H. (1979) in Topics in Applied Physics: Solar Energy Conversion,vol. 31 (ed. B.O. Seraphin), Springer-Verlag, New York, pp. 115–172.

24 Morrison, S.R. (1980) Electrochemistry at Semiconductor and Oxidized MetalElectrodes, Plenum Press, New York.

25 Memming, R. (1988) in Electrochemistry II, Topics in Current Chemistry,vol. 143 (ed. E. Steckhan), Springer-Verlag, Berlin, pp. 79–112.


26 Pleskov, Y.V. (1990) Solar Energy Conversion, A PhotoelectrochemicalApproach, Springer-Verlag, Berlin.

27 Bard, A.J., Memming, R., and Miller, B. (1991) Terminology in semiconductorelectrochemistry and photoelectrochemical energy conversion. Pure Appl.Chem., 63, 569–596.

28 Vinu, R. and Madras, G. (2010) Environmental remediation by photocatalysis.J. Indian Inst. Sci., 90, 189–230.

29 Kuang, Q. and Yang, S. (2013) Template synthesis of single-crystal-like porousSrTiO3 nanocube assemblies and their enhanced photocatalytic hydrogen evo-lution. ACS Appl. Mater. Interfaces, 5, 3683–3690.

30 Lv, M., Xie, Y., Wang, Y., Sun, X., Wu, F., Chen, H., Wang, S., Shen, C.,Chen, Z., Ni, S., Liu, G., and Xu, X. (2015) Bismuth and chromium co-dopedstrontium titanates and their photocatalytic properties under visible lightirradiation. Phys. Chem. Chem. Phys., 17, 26320–26329.

31 Kudo, A., Kato, H., and Nakagawa, S. (2000) Water splitting into H2 and O2on new Sr2M2O7 (M=Nb and Ta) photocatalysts with layered perovskitestructures: factors affecting the activity. J. Phys. Chem. B, 104, 571–575.

32 Chen, X. and Burda, C. (2008) The electronic origin of the visible-lightabsorption properties of C-, N- and S-doped TiO2 nanomaterials. J. Am.Chem. Soc., 130, 5018–5019.

33 Hoang, S., Berglund, S.P., Hahn, N.T., Bard, A.J., and Mullins, C.B. (2012)Enhancing visible light photo-oxidation of water with TiO2 nanowire arraysvia cotreatment with H2 and NH3: synergistic effects between Ti3+ and N.J. Am. Chem. Soc., 134, 3659–3662.

34 National Hydrogen Association (2010) Hydrogen and Fuel Cells: The US Mar-ket Report, National Hydrogen Association.

35 Turner, J.A. (1999) A realizable renewable energy future. Science, 285,687–689.

36 Tachibana, Y., Vayssieres, L., and Durrant, J.R. (2012) Artificial photosynthesisfor solar water-splitting. Nat. Photonics, 6, 511–518.

37 Abe, R., Shinmei, K., Koumura, N., Hara, K., and Ohtani, B. (2013)Visible-light-induced water splitting based on two-step photoexcitationbetween dye-sensitized layered niobate and tungsten oxide photocatalystsin the presence of a triiodide/iodide shuttle redox mediator. J. Am. Chem.Soc., 135, 16872–16884.

38 Osterloh, F.E. (2013) Inorganic nanostructures for photoelectrochemical andphotocatalytic water splitting. Chem. Soc. Rev., 42, 2294–2320.

39 Fujishima, A. and Honda, K. (1972) Electrochemical photolysis of water at asemiconductor electrode. Nature, 238, 37–38.

40 Wrighton, M.S., Wolczanski, P.T., and Ellis, A.B. (1977) Photoelectrolysis ofwater by irradiation of platinized n type semiconducting metal oxides. J. SolidState Chem., 22, 17–29.

41 El Zayat, M.Y., Saed, A.O., and El-Dessouki, M.S. (1998) Photoelectrochemicalproperties of dye-sensitized Zr dopped SrTiO3 electrodes. Int. J. HydrogenEnergy, 23, 259–266.

42 Shi, J., Hara, Y., Sun, C., Anderson, M.A., and Wang, X. (2011)Three-dimensional high-density hierarchical nanowire architecture for

References 71

high-performance photoelectrochemical electrodes. Nano Lett., 11,3413–3419.

43 Hensel, J., Wang, G., Li, Y., and Zhang, J.Z. (2010) Synergistic effect of CdSequantum dot sensitization and nitrogen doping of TiO2 nanostructures forphotoelectrochemical solar hydrogen generation. Nano Lett., 10, 478–483.

44 Park, J.H., Kim, S., and Bard, A.J. (2006) Novel carbon-doped TiO2 nanotubearrays with high aspect ratios for efficient solar water splitting. Nano Lett., 6,24–28.

45 Khaselev, O. and Turner, J.A. (1998) A monolithic photovoltaic-photoelectrochemical device for hydrogen production via water splitting.Science, 280, 425–427.

46 Shyamal, S., Hajra, P., Mandal, H., Singh, J.K., Satpati, A.K., Pande, S., andBhattacharya, C. (2015) Effect of substrates on the photoelectrochemicalreduction of water over cathodically electrodeposited p-type Cu2O thin films.ACS Appl. Mater. Interfaces, 7, 18344–18352.

47 Kudo, A., Kato, H., and Tsuji, I. (2004) Strategies for the developmentof visible-light-driven photocatalysts for water splitting. Chem. Lett., 33,1534–1539.

48 Hou, Y., Zuo, F., Dagg, A., and Feng, P. (2012) Visible light-driven 𝛼-Fe2O3nanorod/graphene/BiV1−xMoxO4 core/shell heterojunction array for efficientphotoelectrochemical water splitting. Nano Lett., 12, 6464–6473.

49 Mavroides, J.G., Kafalas, J.A., and Kolesar, D.F. (1976) Photoelectrolysis ofwater in cells with SrTiO3 anodes. Appl. Phys. Lett., 28, 241–243.

50 Jin, K.X., Li, Y.F., Wang, Z.L., Peng, H.Y., Lin, W.N., Kyaw, A.K.K., Jin, Y.L.,Jin, K.J., Sun, X.W., Soci, C., and Wu, T. (2012) Tunable photovoltaic effectand solar cell performance of self-doped perovskite SrTiO3. AIP Adv., 2,042131–042140.

51 Kawasaki, S., Nakatsuji, K., Yoshinobu, J., Komori, F., Takahashi, R.,Lippmaa, M., Mase, K., and Kudo, A. (2012) Epitaxial Rh-doped SrTiO3thin film photocathode for water splitting under visible light irradiation. Appl.Phys. Lett., 101, 033910–033914.

52 Fu, Q., He, T., Li, J.L., and Yang, G.W. (2012) Band-engineered SrTiO3nanowires for visible light photocatalysis. J. Appl. Phys., 112, 104322–104329.

53 Burbure, N.V., Salvador, P.A., and Rohrer, G.S. (2010) Photochemical reactiv-ity of titania films on BaTiO3 substrates: origin of spatial selectivity. Chem.Mater., 22, 5823–5830.

54 Upadhyay, S., Shrivastava, J., Solanki, A., Choudhary, S., Sharma, V.,Kumar, P., Singh, N., Satsangi, V.R., Shrivastav, R., Waghmare, U.V., and Dass,S. (2011) Enhanced photoelectrochemical response of BaTiO3 with Fe doping:experiments and first-principles analysis. J. Phys. Chem. C, 115, 24373–24380.

55 Ng, C., Ng, Y.H., Iwase, A., and Amal, R. (2013) Influence of annealing tem-perature of WO3 in photoelectrochemical conversion and energy storage forwater splitting. ACS Appl. Mater. Interfaces, 5, 5269–5275.

56 Hong, S.J., Lee, S., Jang, J.S., and Lee, J.S. (2011) Heterojunction BiVO4/WO3electrodes for enhanced photoactivity of water oxidation. Energy Environ. Sci.,4, 1781–1787.


57 Liu, R., Lin, Y., Chou, L.Y., Sheehan, S.W., He, W., Zhang, F., Hou, H.J.M., andWang, D. (2011) Water splitting by tungsten oxide prepared by atomic layerdeposition and decorated with an oxygen-evolving catalyst. Angew. Chem. Int.Ed., 50, 499–502.

58 Shwarsctein, A.K., Hu, Y.S., Forman, A.J., Stucky, G.D., and McFarland, E.W.(2008) Electrodeposition of 𝛼-Fe2O3 doped with Mo or Cr as photoanodes forphotocatalytic water splitting. J. Phys. Chem. C, 112, 15900–15907.

59 Cummings, C.Y., Marken, F., Peter, L.M., Wijayantha, K.G.U., and Tahir, A.A.(2012) New insights into water splitting at mesoporous 𝛼-Fe2O3 films: a studyby modulated transmittance and impedance spectroscopies. J. Am. Chem.Soc., 134, 1228–1234.

60 Cesar, I., Kay, A., Martinez, J.A.G., and Grätzel, M. (2006) Translucentthin film Fe2O3 photoanodes for efficient water splitting by sunlight:nanostructure-directing effect of Si-doping. J. Am. Chem. Soc., 128,4582–4583.

61 Kang, D., Kim, T.W., Kubota, S.R., Cardiel, A.C., Cha, H.G., and Choi, K.S.(2015) Electrochemical synthesis of photoelectrodes and catalysts for use insolar water splitting. Chem. Rev., 115, 12839–12887.

62 Shi, Z., Wen, X., Guan, Z., Cao, D., Luo, W., and Zou, Z. (2015) Recentprogress in photoelectrochemical water splitting for solar hydrogen produc-tion. Ann. Phys., 358, 236–247.

63 Chen, X., Shen, S., Guo, L., and Mao, S.S. (2010) Semiconductor-based pho-tocatalytic hydrogen generation. Chem. Rev., 110, 6503–6570.

64 Ishikawa, A., Takata, T., Kondo, J.N., Hara, M., Kobayashi, H., and Domen, K.(2002) Oxysulfide Sm2Ti2S2O5 as a stable photocatalyst for water oxidationand reduction under visible light irradiation (𝜆≤ 650 nm). J. Am. Chem. Soc.,124, 13547–13553.

65 Higashi, M., Domen, K., and Abe, R. (2012) Highly stable water splitting onoxynitride TaON photoanode system under visible light irradiation. J. Am.Chem. Soc., 134, 6968–6971.

66 Liu, M., You, W., Lei, Z., Zhou, G., Yang, J., Wu, G., Ma, G., Luan, G., Takata,T., Hara, M., Domen, K., and Li, C. (2004) Water reduction and oxidation onPt—Ru/Y2Ta2O5N2 catalyst under visible light irradiation. Chem. Commun.(Camb.), 19, 2192–2193.

67 Matsukawa, M., Ishikawa, R., Hisatomi, T., Moriya, Y., Shibata, N., Kubota, J.,Ikuhara, Y., and Domen, K. (2014) Enhancing photocatalytic activityof LaTiO2N by removal of surface reconstruction layer. Nano Lett., 14,1038–1041.

68 Wang, G., Wang, H., Ling, Y., Tang, Y., Yang, X., Fitzmorris, R.C., Wang, C.,Zhang, J.Z., and Li, Y. (2011) Hydrogen-treated TiO2 nanowire arrays forphotoelectrochemical water splitting. Nano Lett., 11, 3026–3033.

69 Zhao, G., Kozuka, H., Lin, H., and Yoko, T. (1999) Sol–gel preparation ofTi1−xVxO2 solid solution film electrodes with conspicuous photoresponse inthe visible region. Thin Solid Films, 339, 123–128.

70 Bao, X.Q., Petrovykh, D.Y., Alpuim, P., Stroppa, D.G., Guldris, N., Fonseca,H., Costa, M., Gaspar, J., Jin, C., and Liu, L. (2015) Amorphous oxygen-rich

References 73

molybdenum oxysulfide decorated p-type silicon microwire arrays for efficientphotoelectrochemical water reduction. Nano Energy, 16, 130–142.

71 Martyanov, I.N., Uma, S., Rodrigues, S., and Klabunde, K.J. (2004) Structuraldefects cause TiO2-based photocatalysts to be active in visible light. Chem.Commun., 10, 2476–2477.

72 Breault, T.M. and Bartlett, B.M. (2012) Lowering the band gap ofanatase-structured TiO2 by coalloying with Nb and N: electronic structureand photocatalytic degradation of methylene blue dye. J. Phys. Chem. C, 116,5986–5994.

73 Choi, W., Termin, A., and Hoffman, M.R. (1994) The role of metal iondopants in quantum-sized TiO2: correlation between photoreactivity andcharge carrier recombination dynamics. J. Phys. Chem., 98, 13669–13679.

74 Aroutiounian, V.M., Arakelyan, V.M., and Shahnazaryan, G.E. (2005) Metaloxide photoelectrodes for hydrogen generation using solar radiation-drivenwater splitting. Sol. Energy, 78, 581–592.

75 Tsuji, I., Kato, H., and Kudo, A. (2005) Visible-light-induced H2 evo-lution from an aqueous solution containing sulfide and sulfite over aZnS—CuInS2—AgInS2 solid-solution photocatalyst. Angew. Chem., 117,3631–3634.

76 Zheng, N., Bu, X., Vu, H., and Feng, P. (2005) Open-framework chalcogenidesas visible-light photocatalysts for hydrogen generation from water. Angew.Chem., 117, 5433–5437.

77 In, S., Orlov, A., Berg, R., García, F., Pedrosa-Jimenez, S., Tikhov, M.S.,Wright, D.S., and Lambert, R.M. (2007) Effective visible light-activatedB-doped and B,N-codoped TiO2 photocatalysts. J. Am. Chem. Soc., 129,13790–13791.

78 Cong, Y., Zhang, J., Chen, F., and Anpo, M. (2007) Synthesis and characteriza-tion of nitrogen-doped TiO2 nanophotocatalyst with high visible light activity.J. Phys. Chem. C, 111, 6976–6982.

79 Etacheri, V., Seery, M.K., Hinder, S.J., and Pillai, S.C. (2010) Highly visi-ble light active TiO2−xNx heterojunction photocatalysts. Chem. Mater., 22,3843–3853.

80 Kim, S., Hwang, S.J., and Choi, W. (2005) Visible light activeplatinum-ion-doped TiO2 photocatalyst. J. Phys. Chem. B, 109, 24260–24267.

81 Ihara, T., Miyoshi, M., Ando, M., Sugihara, S., and Iriyama, Y. (2001) Prepa-ration of a visible-light-active TiO2 photocatalyst by RF plasma treatment.J. Mater. Sci., 36, 4201–4207.

75

4

Photoredox Catalytic Activation of Carbon—HalogenBonds: C—H Functionalization Reactions underVisible LightJavier I. Bardagi1 and Indrajit Ghosh2

1Universidad Nacional de Córdoba, Ciudad Universitaria, INFIQC-CONICET, Departamento de QuímicaOrgánica, Facultad de Ciencias Químicas., Córdoba X5000HUA, Argentina2University of Regensburg, Institute of Organic Chemistry, D-93040 Regensburg, Germany

4.1 Introduction

Photosynthetic organisms transform light energy into chemical free energy forsimultaneous reduction of NADP+ to NADPH and the oxidation of water tomolecular oxygen in order to sustain life on earth [1, 2]. Visible-light-mediatedphotoredox catalysis, an emerging field in synthetic organic chemistry, similarlytransform visible-light energy into redox energy to activate chemical bonds anddrives synthetically important chemical reactions [3–10]. The most compellingfact about the transduction of visible-light energy into redox energy for chemicalbond activation is that the energy of blue photons (e.g., 455 nm) of 262 kJ/molare insufficient to activate many chemical bonds for synthetic applications (cf.,via direct photoexcitation); however, transduction of the same light energy intoredox equivalence via a photoredox catalytic process allows activation of suchbonds via a simple single electron transfer. In this process, a visible-light-activemolecule (i.e., a photoredox catalyst, see Scheme 4.1 for the chemical structuresof commonly used photocatalysts) absorbs visible light and transforms it intoexcited state. Due to the electronic rearrangements upon photoexcitation, thephotoredox catalysts simultaneously become more oxidizing and reducing [11].In thermodynamic terms, upon photoexcitation the ionization potential of a pho-tocatalyst is reduced and the electron affinity increases allowing easy electrontransfer process to or from the targeted substrates for the activation of chemicalbonds. As a consequence, the chemical redox reactions that are thermodynami-cally or kinetically impossible in the dark become feasible upon simple photoex-citation of the photocatalyst using visible light.

Among others, there are manly two ways to activate chemical bonds via pho-toredox catalytic methods.

1) Oxidative activation in which the substrate transfers an electron to the pho-toredox catalyst

2) Reductive activation in which an electron is transformed from the photoredoxcatalysts to the substrate.


76 4 Photoredox Catalytic Activation of Carbon—Halogen Bonds

S

N

Ru

N

N

N

N

N

N

2+

Ir

N

NN

Ru(bpy)32+ fac-Ir(ppy)3

O

Br

Br

O

Br

O

Br

COO

Eosin Y

N N

O

O

O

O

PDI

O

H3C

NH

H3C

O CH3

CH3

N

CH3

O

HCl

PTH

Rh-6G

E1/2 (M+/M*) = –2.1

E1/2 (M+/M*) = –0.81 E1/2 (M+/M*) = –1.73E1/2 (M/M−) = –2.19

E1/2 (M/M–) = –1.06

E1/2 (M/M–) = ca. –0.4

E1/2 (M/M–) = ca. –1.1

E1/2 (M/M–) = –1.33

Scheme 4.1 Chemical structures of the photocatalysts that are typically used for theactivation of alkyl or aryl halides under visible light. The respective reduction potentials arealso depicted.

The examples of the first category cover, among others, the oxidation ofsubstituted arenes or heteroarenes for direct C—H bond functionalizations[12]. Single electron transfer initiated activation of carbon—halogen bonds inorganohalides are among the examples of the later. Due to the current pace ofdeveloping sustainable visible-light-mediated photoredox catalytic methods,there are several examples reported for the activation of different chemical bonds,both via oxidative and reductive electron transfer processes. Herein, we discussmainly the activation of carbon—halogen bonds both for alkyl and aryl halides forsynthetic applications. For the activation of other chemical bonds using differentphotoredox catalytic approach, we encourage the readers to read the reviewarticles published in the special issues “photoredox catalysis in organic chemistry”(Ref. [13] and the review articles in that issue) and “photocatalysis” (Ref. [14] andthe papers published in that issue) published recently in the journals Accounts ofChemical Research and The Journal of Organic Chemistry, respectively [13, 14].

Common alkyl or aryl halides do not absorb visible light and therefore areinactive under direct visible-light photoexcitation. However, carbon—halogenbonds in these substrates could strategically be activated via a single electrontransfer (SET) as depicted in Scheme 4.2. Notably, the mechanisms of activationof alkyl and aryl halides are slightly different. In particular, SET to aliphatichalides follows a concerted-dissociative pathway (i.e., dissociative electrontransfer, DET) in which the carbon—halogen bond (R—X) breaks when theelectron is being transferred [15, 16]. On the other hand, carbon—halogen bonddissociation of aryl halides (Ar—X)1 via a single electron transfer follows a

1 Single electron transfer assisted generation of aryl radicals from aryl iodides could also follow asingle step processes.

4.2 Activation of Alkyl Halides 77

R

X; X = Cl, Br, I

–X; X = Cl, Br, I–e–, –H+

+ eR R

Trap

Trap

R

X; X = Cl, Br, I

–X; X = Cl, Br, I

R–X+ e

R Product

Scheme 4.2 Schematic representation of the generation of alkyl and aryl radicals from theirrespective halides upon single electron transfer.

stepwise mechanism in which the kinetics of dissociation of the formed radicalanions (Ar—X∙−) as intermediates could play a vital role (see below) [16–18].Note that this is the preferred pathway for the activation of Ar—X, as theyhave a π acceptor with low enough energy to accommodate the extra electron,although some exceptions are reported for electrochemical processes [19].Upon single electron transfer, the halide substrates release a halide anion andgenerate the respective alkyl or aryl radicals. The generated reactive radicalsthen either accept a hydrogen atom from a hydrogen atom donor to form thephotoreduced dehalogenated products or could interact with suitable radicaltrapping reagents for the formation of carbon—carbon or carbon—heteroatombonds for synthetic applications. Herein, we discuss both photoredox cat-alytic dehalogenation reactions and bond-forming reactions using transitionmetal-based complexes or commercially available inexpensive luminescentdyes as photoredox catalysts under visible-light photoirradiation. The chemicalstructures of the commonly used photocatalysts for such transformations alongwith their reduction potentials are depicted in Scheme 4.1.

4.2 Activation of Alkyl Halides

One of the very first synthetic examples for the activation of alkyl halides wasreported by Fukuzumi almost three decades ago (Scheme 4.3). Photoredoxcatalytic reductions of phenacyl halides were obtained by using Ru(bpy)3

2+

as the photoredox catalyst (see Scheme 4.1 for the chemical structure),9,10-dihydro-10-methylacridine as an electron donor, and visible light as theenergy source [20]. Phenacyl bromides are visible light transparent and thuscould not be activated via direct photoexcitation using visible light; however,they possess relatively low reduction potentials (E∘ ≤ 0.5 V vs saturated calomelelectrode, SCE) and as a consequence could be activated easily using Ru(bpy)3

2+

as the photoredox catalyst. Interestingly, the electron transfer from the pho-tocatalyst to the phenacyl halides depends strongly on the reaction condition,especially the presence of an acid (in this case, HClO4). In the absence of any acidexcited *Ru(bpy)3

2+ takes an electron form a sacrificial tertiary amine electrondonor to generate Ru(bpy)3

+ (see Scheme 4.3). The photoredox chemically gen-erated Ru(bpy)3

+ is capable of transferring an electron to 2-bromoacetophenone


Ru(bpy)3Cl2visible light, MeCN

Ph

O

BrPh

O

H

N

H H

Scheme 4.3 Photoredox catalytic reductive dehalogenation of phenacyl bromide.

Ru(bpy)3+

Ru(bpy)32+

XEWG

*Ru(bpy)32+

iPr2NEt

N

HiPr

Et

NiPr

Et

EWG

hν

HEWG

Scheme 4.4 General mechanism of the reductive halogenation of activated alkyl halidesubstrates using Ru(bpy)3Cl2 as a photocatalyst and visible light as the energy source. Notethat N,N-diisopropylethylamine (iPr2NEt or DIPEA) is a typical sacrificial electron donor in mostphotoredox catalytic reductive transformations.

giving the phenacyl radical via a dissociative electron transfer, and closesthe photoredox catalytic cycle by regenerating the neutral photocatalyst (seeScheme 4.4 for the general activation of activated alkyl halides via single electrontransfer). However, in the presence of HClO4, upon photoexcitation, *Ru(bpy)3

2+

directly transfers an electron to the phenacyl halides generating relatively stablephenacyl radicals. Under both reaction conditions, the phenacyl radicals eithertake a hydrogen atom from the surrounding hydrogen atom donors to yieldthe corresponding carbonyl compounds or interact with other molecules forbond-forming reactions (see examples below). Notably, the phenacyl radical isrelatively stable due to the presence of carbonyl group at the α position.

Among other discreet earlier reports on the visible-light photoredox catalytictransformations, MacMillan reported one of the very first C—C bond-formingreactions using alkyl halides as substrates. In this elegant report, MacMillancombined photoredox catalysis and organocatalysis for dual catalytic enan-tioselective alkylation of aldehydes (Scheme 4.5) [21]. Notably, the authorscoupled Fukuzumi’s photoredox catalytic generation of carbon centered radicalvia reductive cleavage of carbon—halogen bonds [20] with organocatalyticallygenerated enamines that react as the radical trapping reagents. The proposedmechanism of the dual catalytic cycle is depicted in Scheme 4.6. It is not clearhow the photoredox catalytic cycle starts, as the proposed mechanism describes


Ru(bpy)3Cl2 CFL (15 W)O

R

R′

O

Br

Z

O

*RR′

O

Z

enantioenriched

+

O

*Ph

O

O

*EtO2C

O

*F3CH2CO

N

NO

2,6-lutidine, DMF

83% yield, 95% ee84% yield, 96% ee 80% yield, 92% ee

HX

H

H

EtO2CO

Scheme 4.5 Selected examples of asymmetric alkylation of aldehydes.

Ru(bpy)3+

–Br–

Ru(bpy)32+

hν

*Ru(bpy)32+

R′

O

*

Z

R

NN

O

R′

O*

Z

R

N

NO

RN

N

O

HN

N

O

O

R

H

O

*RR′

O

Z

R′

O

Br

ZR′

O

Z

H+

OHH

Scheme 4.6 Proposed mechanism of the dual catalytic cycle.

an interaction of the excited-state *Ru(bpy)32+ species with an intermediate

radical species that forms via a radical trapping interaction of the phenacylradical with an enamine that formed in situ, independent of the photoredoxcycle. Be this as it may, although more experimental results are needed, lumi-nescence quenching studies support that the photoexcited *Ru(bpy)3

2+ couldtake an electron from the enamine in order to initiate the photoredox cycle. Thereduced Ru(bpy)3

+ is capable of generating phenacyl radical by transferring anelectron to the phenacyl bromide substrates. The phenacyl radical then adds tothe enamine and enters in the organocatalytic cycle producing a new radicalspecies which is capable of giving an electron to the excited *Ru(bpy)3

2+ forming


Photocatalyst visible lightO

R

R′

O

Br

Z

O

*RR′

O

Z

enantioenriched

+

O

*Ph

O

O

*EtO2C

EtO2CO

*O

N

NO

2,6-lutidine, DMF

85% yield, 88% ee65% yield, 96% ee82% yield, 95% ee

H

H

PbBiO2Br, Blue LED

O2N

Eosin Y, Green LED

HX

Scheme 4.7 Selected examples of asymmetric alkylation of aldehydes using organic dyes (inthis case, Eosin Y) or semiconductors as photocatalysts.

an iminium cation . Upon generation of the organocatalyst from the iminiumcation yields the desired product and closes the organocatalytic cycle.

A couple of years later, Zeitler and König reported that organic dyes such asEosin Y, a commercially available inexpensive dye, are equally efficient for suchtransformations providing an ecologically benign greener alternative catalyticsystem for such photoredox transformations. The proposed mechanism withthe organic dyes follows the mechanism initially proposed by MacMillan (seeScheme 4.6) [22, 23]. Similarly, König also reported such transformations usingsemiconductors as heterogeneous photocatalysts (Scheme 4.7).

It is worth mentioning here that Eosin Y is also useful in reductive dehalo-genation of α-carbonyl bromide substrates. Fluorescein, perylenediimide, nilered, alizarin red S, and rhodamine B were among the other investigated organicdyes(see Scheme 4.1). Although most of them were effective for such photoredoxcatalytic transformation to take place the choice of Eosin Y as the best catalystwas based on the photoreduction yield of 2-bromoacetophenone to acetophe-none. Note that the reduction potential of the investigated photoredox catalysts,ranging from −0.8 to −1.22 V versus SCE, is capable to reducing the substrate(E∘ =−0.5 V vs SCE). Hantzsch ester was used as both electron and H donor,and the mechanism follows a reductive quenching cycle via the formation of theradical anion of dye molecules as intermediate.

Stephenson explored the activation of carbon–halogen bonds in alkyl bro-momalonates for carbon–carbon bond formation reactions via intramolecularcyclization reactions (Scheme 4.8) [24]. Upon photoinduced dissociative singleelectron transfer (i.e., by releasing halide anions) the generated malonate radicalsare trapped intramolecularly by tethered indole, pyrrole [24], and unsaturateddouble bonds [25] to obtain important polycyclic compounds in good yields.Moreover, this photoredox catalytic protocol allows synthesis of complexstructures through cascade radical cyclizations with excellent yields under verymild reaction conditions (Scheme 4.8, right). Note that these photoredox chem-ical transformations could be performed using Ru(bpy)3

2+ as a photocatalyst


14, 60%19, 53% 17, 95%

21, 79%O

O

N

CO2Me

CO2Me

CFL (14 W)

CFL (14 W)

Ru(bpy)3Cl2

NR3N, DMF

Ru(bpy)3Cl2R3N, DMF

CO2Me

Br

CO2Me

CO2Me

CO2Me

Et NN

CO2Me

CO2Me

CO2Me

CO2Me

CO2Me

CO2Me

MeO2C

N NH

H

20 Br

CO2Me

Scheme 4.8 Photoredox catalytic reductive radical cyclization of malonate bromides.

R3N

hν

NCO2Me

CO2Me

NBr

CO2Me

CO2Me

N

CO2Me

CO2MeN

CO2Me

CO2MeH

NCO2Me

CO2MeH

R3N

“e–”

–H+

H

H

Ru(bpy)3+

Ru(bpy)32+

*Ru(bpy)32+

–Br–

Scheme 4.9 Mechanism for C—H substitution in reductive radical cyclization of malonatebromides.

as its reduction potential covers the reduction potential of bromomalonatederivatives. However, Ir-complexes as photoredox catalysts were necessary forthe generation of alkyl radicals from 2-halo-esters or amides as they possesshigher reduction potentials than both excited and ground states of Ru(bpy)3

2+

(see Scheme 4.1 for the respective reduction potential values) [25, 26].The proposed mechanism of the catalytic cycle is shown in Scheme 4.9. Upon

photoexcitation, Ru(bpy)32+ is reductively quenched by an amine donor to form

Ru(bpy)3+ which is capable of transferring an electron to the bromide substrate

generating the stable alkyl radical. The alkyl radical then undergoes a radicalcyclization onto the substituted pyrrole heterocycle to yield a relatively stableradical (i.e., a σ-complex) which upon successive oxidation and proton releaseyields the desired polycyclic compounds [27]. Note that the formation of thecyclized products under this photoredox catalytic condition implied a C—Hsubstitution reaction that follows a homolytic aromatic substitution (HAS) overthe heterocycles [27].

It is worth discussing here that in this mechanistic scenario, which is also truein many other photoredox catalytic systems (see below), a question remainsabout the species that is responsible for the oxidation of the stabilized radical.


It has been proposed that the excited state of *Ru(bpy)32+ or the radical cation

of the amine or the substrate (which will result in initiating a chain reaction)could be responsible for such oxidation, and different studies have shown that allthree options are feasible, however, depends strongly on the substrates, electrondonor, photocatalysts, and on the reaction conditions. For example, electrontransfer directly to the substrate is feasible only when the substrate possessesvery low reduction potential (cf., the reduction potentials of diazonium salts).Single electron transfer to the radical cation of the amine donor is unlikelywhen aliphatic amines are used as their radical cations suffer from hydrogen orproton loss resulting in their relatively short lifetimes. However, triarylamines,which are extensively used in electrosynthesis, form stable radical cation andcould execute the oxidation process. Finally, the oxidation by the excited state of*Ru(bpy)3

2+ is possible when the oxidation potential values allow such electrontransfer process, and the lifetime values of the transient species allow enoughaccumulation of such species for reasonable reaction rates.

Stephenson similarly reported an intermolecular version of the trapping ofmalonate radicals for carbon—carbon bond-forming reactions using differentheterocycles, Ru(bpy)3

2+ as the photoredox catalyst, and N ,N-diphenyl-4-methoxyaniline as the sacrificial amine donor (Scheme 4.10, left) [28]. The use ofN ,N-diphenyl-4-methoxyaniline as the sacrificial electron donor was importantto obtain better product yields by suppressing the formation of the undesiredreduction product which forms upon hydrogen atom abstraction of the malonateradicals. Note that malonate radicals abstract hydrogen atoms from oxidizedalkyl amines when used as sacrificial electron donors. Due to the absence of anyα hydrogen to the nitrogen atom in N ,N-diphenyl-4-methoxyaniline, exclusiveC—C bond formations were observed.

An oxidative catalytic cycle was reported for the generation of tertiary alkylmalonate radicals from their respective bromide precursors. Notably, the authorsselected fac-Ir(ppy)3 as the photocatalyst over Ru(bpy)3

2+ due to the higherreduction potential of tertiary alkyl malonate substrate (Scheme 4.10, right). Theexcited-state reduction potential of fac-Ir(ppy)3 (E1/2

Ir(IV)/Ir(III)* =−1.73 V vs SCE,see Scheme 4.1) is higher than the ground and excited-state reduction potentialsof Ru(bpy)3

2+. Additionally, the use of excited-state reduction potential offac-Ir(ppy)3 eliminates the requirement of an electron donor for the photoredoxcatalytic transformations and thus improves the product yields of the suchtransformations (Scheme 4.10, right) [29]. The additive 2,6-lutidine, however,was necessary as a base in order to neutralize HBr that forms as a by-product.

X

MeO2C CO2Me

X

Br

MeO2C

MeO2C

R

NPh2

MeO

R = H R = alkyl, allyl, Bn

X

MeO2C CO2MeR

X = NR, O , S

+

Blue LED Ru(bpy)3Cl2

DMF

H

Ir(ppy)3 blue LED

2,6-lutidine, MeCN

Scheme 4.10 Intermolecular photoredox catalytic C—H functionalizations with malonates.


RO

RCO2Et

EtO2C

74–94%

22 examples

n = 1, 2

+

O

n

X

X

n

CO2Et

CO2Et

H

Ir(ppy)3 blue LED

K2HPO4, DMF

–H+

–Br–

Br

EtO2C

EtO2C

Ir(ppy)3+ Ir(ppy)3

hν

*Ir(ppy)3

R

Ar

O

n

H

CO2EtEtO2C

R

Ar

O

n

H

CO2EtEtO2C

Scheme 4.11 Synthesis of substituted tetrahydrofurans.

MeO2C R

Br

N

EtO2C FF

Br

MeO2C H

Br

Scheme 4.12 Other α-activated substrates used in C—C bond formations.

In a similar approach, the difunctionalization of alkenes through radicaladdition followed by intramolecular etherification was reported [30]. In thisapproach, the generated intermediate benzyl radical gets oxidize by the Ir(ppy)3

+

species forming a cation intermediate (Scheme 4.11). In this example, the radicalcation then reacts intramolecularly with a nucleophile generating the desiredproduct.

Notably, other α activated bromide substrates (see Scheme 4.12) could also beused in order to generate such radicals using proper photoredox catalysts andvisible light [29, 31].

Yu elegantly showed that isocyanides are useful in trapping photoredoxchemically generated 2-methoxy-2-oxo-propan-2-yl radicals for the synthesisof substituted phenanthridines using biphenyl isocyanides in a net insertionreaction of the isocyanide group to the adjacent phenyl ring (Scheme 4.13) [31].fac-Ir(ppy)3 was used as the photoredox catalysts to obtain satisfactory productyields as α-bromoesters are difficult to activate because of their more negativereduction potentials with respect to malonate bromides, discussed before.Note that this reductive photoredox catalytic protocol does not require sacri-ficial amine donors, however, an inorganic base (Na2HPO4, K2HPO4, K2CO3,NaHCO3) as an inexpensive additive was useful to neutralize HBr, formed in situ.


N

C

+N

39–93%

16 examples

R1 R1

R2 R2

CO2Et

fac-Ir(ppy)3 blue LED

Na2HPO4, DMFH MeO2C H

Br

Scheme 4.13 Synthesis of substituted phenanthridines by a net isocyanide insertion.

Ir(ppy)2(dtb-bpy)PF6 CFL 26 WO

R

O

*R

Rf

enantioenriched

O

*

O

*F3C

Ph

O

*F3C

N

NO

75% yield, 97% ee71 yield, 99% ee 72% yield, 98% ee

TFA

H

H

Rf I +

CF3FF

2,6-lutidine, DMF, –20 °C

F

F3CO

Scheme 4.14 Enantioselective perfluoroalkylation of aldehydes by merging photoredoxcatalysis with organocatalysis.

Following his seminal work on asymmetric alkylation of aldehydes, MacMil-lan in 2009 reported the enantioselective α-trifluoromethylation of aldehydesby merging photoredox catalysis with organocatalysis (Scheme 4.14). Perfluo-roiodo alkanes were used as the precursor of the substituted perfluoroalkyl radi-cal because of their accessible reduction potentials (e.g., the reduction potentialsof CF3I E∘ =−1.22 V vs SCE) using an Ir-complex (Ir(ppy)2(dtb-bpy)PF6) as pho-tocatalyst [32]. Note that the presence of fluorine atoms decreases the reductionpotentials of the iodoalkanes via –I effect, and single electron transfer initiatedbond dissociation allows generation of substituted trifluoromethyl radicals. Theproposed photoredox catalytic mechanism resembles the one discussed for thegeneration of phenacyl radicals from their respective bromide precursors (seeScheme 4.6).

Notably generation of other fluoroalkyl radicals (either from trifluoromethyliodide (i.e., Rf−I) or activated bromide substrates, e.g., Br—CF2CO2Et, and otherrelated substrates) and their addition to double bonds or triple bonds generatingaddition and substitution products (depending on the reaction conditions) arealso reported using visible-light-mediated photoredox catalytic protocols [33].These photoredox catalytic transformations could be performed using Ru, Ir,or Pt complexes, mainly via an atom transfer radical addition (ATRA) reaction[9, 33].

Cho reported the trifluoromethylation of aromatic heterocycles via C—H func-tionalization using CF3I as substrate and Ru(bpy)3 as the photoredox catalyst


X

X = NR, O , S

+ CF3

NH

CF3

90%

OCF3

SCF3

H2N

92%

MeO2C

92%

NH

CF3

81%

OEt

O

Ru(bpy)3Cl2 visible light

TMEDA, MeCNIF3C

XH

Scheme 4.15 Trifluoromethylation of aromatic heterocycles.

(Scheme 4.15) [34]. Upon visible-light excitation, Ru(bpy)32+ is reductively

quenched by a trialkylamine generating Ru(bpy)3+ that is capable of generating

CF3 radical from CF3I via a single electron transfer. The generated CF3 radicalthen adds onto the pyrrole heterocycles forming a stable radical, which uponone electron oxidation and proton loss yields the trifluoromethylated products.Notably the species responsible for oxidizing the intermediate radical is notclear yet, however, the excited *Ru(bpy)3

2+ is capable of oxidizing such speciesunder the catalytic condition (see also the discussion above). Substituted furansor thiophenes are useful trapping reagents of CF3 radical under the describedreaction condition.

Similarly, Cho reported the introduction of arylthiofluoroalkyl groups tounactivated heteroaromatic cycles or alkenes using commercially availablephenylthiofluoroalkyl bromides (Scheme 4.16) in one step, synthetically useful,for example, for late-stage functionalizations [35]. In this case, arylthiofluoroalkylradicals are generated via photoredox catalytic activation of carbon—brominebonds. The generated arylthiofluoroalkyl radicals then react with heteroaro-matics or unsaturated double bonds to obtain the desired products. In the caseof alkenes both substitution and addition products were formed. Reaction ofalkenes with PhSCF2Br results in a tandem radical addition–HAS over thephenyl group, to obtain 2,2-difluorothiochroman, due to an optimal geometryfor a HAS reaction of the resulted radical. Interestingly, depending on the relative

X

RfSPh

X

CF2SPh

Rf = C2F4, C4F8

RRfSPh

R = Aryl, akyl

S

F

F

R

R = alkyl

59–90%

RR

52–78%

70–80%52–74%

X

Blue LED Ru(phen)3Cl2

TMEDA, MeCN


2,6-lutidine, DMF

IPhSF2C

R

X = NR, O , S.

I RfSPh

Blue LED fac-Ir(ppy)3

TMEDA, CH2Cl2


2,6-lutidine, DMF

IPhSF2CI RfSPh

R

Scheme 4.16 Visible-light-induced arylthiofluoroalkylations of heteroaromatics and alkenes.


S

CF3

S

CF3

OI

O

F3C

Umemoto

reagent

Yagupolskii-Umemoto

reagent

Togni

reagent

Hu

reagent

SO

CF2HPh

NTs

Scheme 4.17 Fluoromethylating reagents typically being used in visible-light photoredoxcatalysis.

position of the thiophenyl moiety with respect to the fluoromethyl radical adifferential reactivity was observed. The effect is so prominent that differentoptimized reaction conditions were employed for PhS(CF2)nBr (where n= 2, 3)with respect to PhSCF2Br. When the thiophenyl moiety is present distantly, theradical behaves similarly to CF3 radical (see above) [34].

It is worth mentioning here that although CF3 radical could be generated fromthe respective halide precursors, most examples of trifluoromethylation undervisible- light photoredox catalytic conditions are performed using Unemoto,Yagupoiskii-Unemoto, and Togni reagents (for structures see Scheme 4.17) [33,36, 37]. In a typical reaction, CF3 radicals are obtained by single electron transferto the fluoromethylating reagents in an oxidative quenching cycle. The generatedCF3 radicals then react with the unsaturated compounds generating interme-diate radicals that upon oxidation (by the oxidized state of the photocatalyst)yield carbocation intermediates which either release a proton to yield the triflu-oromethylated unsaturated products or react with suitable nucleophiles, presentin the reaction mixture, yielding bisubstituted saturated products (Scheme 4.18).The latter reaction could be performed with nucleophiles intramolecularly orintermolecularly. Using a double bond as the trapping reagent of the CF3 radicalallows oxy-, amino-, and ketotrifluoromethylations addition reactions in additionto normal substitutions reactions. Using the same approach, difluromethylatedproducts could also be obtained using Hu’s reagent (Scheme 4.18).

Stephenson used tertiary benzyl radicals to perform C—C bond-formingintermolecular reactions, which is best represented in the total synthesis of(+)-gliocladin C. The photoredox catalytic coupling protocol allows a keyconnection of two rather complex subunits (Scheme 4.19) [38]. As shownin Scheme 4.19 the key photocatalytic step is the generation of tertiary alkylradical by activating a carbon—bromine bond via single electron transfer using

R

R1

R2

CF2X

Ph

OH

CF2H

Ph

Ph

R

R1

R2

R

R1

R2

CF2X

Nu

Ph

OH

CF3

Ph

NHAc

CF3

Ph

O

CF3CF3 Ph

CF3

Ph

OMe

O PhO

CF3

Photocatalyst visible light

Nu H

visible light Photocatalyst

LGXF2C

Scheme 4.18 Selected examples of tri- and difluoromethylation of alkenes.


N

N

Br

HBoc

BocN

N

H BocBoc

NH

FG

NFG

FG = H, C2′-addition 52%FG = CO2Me C3′-addition 58%

Ru(bpy)3Cl2 blue LED

Et3N, DMF

H

Scheme 4.19 Coupling of bromopyrroloindoline with indoles to access both C2′- and

C3′-substitutions.

Ru(bpy)32+ as the photoredox catalyst. The feasibility of the reaction was initially

investigated by using N-methylindole as the radical trapping reagent. However,like in the most other cases (see the aryl halide activation section below) as thealkyl radical attacks the two position of the biologically important pyrrole orindole derivatives, a typical regioselectivity problem resulting from the radicaladdition to indoles and other heterocyclic, therefore, functionalization was onlyobtained at the two position. The desired C3—C3

′ connectivity was effectivelyperformed by blocking the C2

′-position of indole with a carboxylate group(Scheme 4.19). It is interesting to note that blocking the two position of indoledid not affect the yields of the coupling products and to demonstrate a useful,simple strategy to obtain regioselective functionalization of these biologicallyimportant nitrogen-containing heterocycles using photoredox catalytic radicaltransformations.

Total synthesis of gliocladin C was performed using 2-carbozaldehyde as thetrapping reagent in the photocatalytic step under slightly different conditions

N

NHBoc

CO2Me

N

N

Br

HBoc

Cbz

NN

H BocCbz

O

NH

N

O O

NH

NN

H

N

NH

O

OO

Gliociadin C

10 step 30% overall yield

N O

Ru(bpy)3Cl2 blue LED

Bu3N, DMF82%

H

HH

H

H

H

Scheme 4.20 Total synthesis of Gliocladin C.


fac-Ir(ppy)3 CFL (15 W)O

R

Ar

Br

O

*R

Ar

enantioenriched

2,6-lutidine, DMF, –20 °C

O

*

O

*

O

*

N

NO

Bn

73% yield, 90% ee75% yield, 91% ee 78% yield, 87% ee

HOTf

H

H

+

N N N

Cl

NO2

O2N

Scheme 4.21 Selective examples of enantioselective α-benzylation of aldehydes.

Ar*R

NN

OBn

Ar

*R

N

NO

Bn

RN

N

OBn

HN

N

OBn

O

R

H

O

*R

Ar

Ar–Br–

Br

hν

Ar

H+

OHH

fac-Ir(ppy)3+

fac-Ir(ppy)3

fac-*Ir(ppy)3

Scheme 4.22 Proposed catalytic cycle for aldehyde R-benzylation.

(Scheme 4.20). As pointed out by the authors, this work clearly demonstrates thepotential of photoredox catalytic methods as mild and robust methods to accessa wide variety of complex molecules.

In another work MacMillan reported enantioalkylation of aldehydes usingbenzyl bromide as the alkylating reagents via a merger of organo- and photore-dox catalysis (Scheme 4.21) [39]. The proposed mechanism involves a directreduction of the bromides by the excited photocatalyst that was supportedwith the quenching studies. Note that the previous studies on the generationof halophenacyl, malonyl, and perfluoroalkyl radicals from their respectivehalides involve a redox reaction of the substrate with the reduced photocatalysts(Scheme 4.22).


O

(R)

Ph

97% yield, 99% ee

87% yield, 97% ee 86% yield, 91% ee

NO2

O2N

N

N

O

(S)

CN

O2N

N

N

O

(R)

Ph

N

NiPr

O

Br

IrN

N

N

X tBu

N

X tBu

CCH3

CCH3

O

(R)

99% yield, 97% ee

NO2

O2N

N

N

S

Λ-Ir

Λ-Ir CFL (14 W)

O

R

EWG

Br

enantioenriched

Na2HPO4, MeOH/THF, 40 °C

H+ N

N

O

R

N

N

GWE *

Scheme 4.23 Photoinduced enantioselective alkylation of acyl imidazoles with acceptorsubstituted benzyl bromides and phenacyl bromides.

An exquisite example of catalyst design for photoredox applications waselegantly showed by Meggers for the enantioselective alkylation of acyl imida-zoles with benzyl bromides [40]. In this dual catalytic approach that followsthe original concept reported by MacMillan [21, 32, 39], the authors use achiral Ir-complex that works as a photocatalyst and simultaneous providechiral environment to obtain the desired products with both excellent yieldsand enantioselectivity (Scheme 4.23). The proposed mechanism starts withthe coordination of 2-acyl imidazoles to the catalyst (Λ-Ir) followed by theformation of an enolate complex (I) (Scheme 4.24). The radical generated byphotoredox SET adds to the electron-rich metal-coordinated enolate doublebond affording an iridium-coordinated ketyl radical. Oxidation of this newradical by single electron transfer regenerates the photocatalyst and providesthe iridium-coordinated ketone that upon exchanges reaction with the substrategenerates the products. Detailed studies on the intermediates support theproposed mechanism, and provide support that the in situ generated complex Iis the actual photocatalyst as depicted in Scheme 4.24.

Although the reduction potential of iodoalkanes is rather high with respectto the typically used photoredox catalysts, a few examples for their activationsare reported in the literature using single electron transfer. Stephenson reportedintramolecular carbon–carbon bond-formation reactions using alky halidesusing highly reducing Ir-complexes (fac-Ir(ppy)3, Scheme 4.25) [41]. Note thatthe use of formic acid or Hantzsch ester is crucial in order to yield the cyclizedproducts.

Brasholz reported dearomatizative tandem [4+2] cyclization to furnishfused dihydroindoles in good yields by activating an alkyl iodide usingIr(ppy)2(dtbbpy)PF6 as a photocatalyst and visible light (Scheme 4.26) [42]. Thereaction proceeds via diastereoselective tandem radical cyclization reactions


Λ-Ir

O

R

N

N

[Ir]

O

R

N

N

[Ir]

EWGhν

EWG

Br

PC+

PC*

PC

EWG

Br

EWG

O

R

HN

N

O

R

N

N

[Ir]

+

H

EWG–Br–

O

R

N

N

[Ir]

Complex I

O

R

N

N

GWE

*

**

Scheme 4.24 Proposed mechanism for the enantioselective alkylation of acyl imidazoles.

N

H

N

I

77–78%

H

BnO

92%

O

O

H 90%

H

O

Br

82%

fac-Ir(ppy)3 visible light

Bu3N, MeCN

N

H

H HEtO2C CO2Et

R I R H

fac-Ir(ppy)3 visible light

Bu3N, HCO2H, MeCN

Ts

Ts

Scheme 4.25 Reductive cyclizations of alkyl iodides using fac-Ir(ppy)3 and visible light.

N

I

R

R = CO2Me, CN, COMe, Me

R1 = CO2Me, SO2Ph, CN, CONMe2

R1

NR1

H

H

31–75%

12 examples

+

Ir(ppy)2(dtbbpy)PF6 blue LED

DIPEA, MeCN

R

Scheme 4.26 Synthesis of hexahydropyridoindoles.

4.3 Activation of Aryl Halides 91

initiated by reductive photoredox generation of alkyl radical intermediatefrom corresponding iodide substrates. Notably, selection of the electron andhydrogen donors (represented by a tertiary amine) was crucial to obtain goodchemoselectivity toward the addition product versus the substituted indole. Themechanism of the reaction resembles one showed in Scheme 4.9.

4.3 Activation of Aryl Halides

As discussed in the Introduction section, photoredox catalytic activation ofcarbon—halogen bonds in substituted aryl halides posses’ different challengesin comparison to the alkyl halides. Their reduction potentials are extremelyhigh and therefore most conventional photoredox catalysts (e.g., Ru(bpy)3

2+ orEosin Y) are not capable of initiating a single electron transfer process to thearyl halide substrate. In addition, the fragmentation kinetics for the generationof aryl radicals from aryl halides could play a crucial role in determining thefeasibility of such photoredox transformations (see below). However, as theradical-based synthesis of arylated arenes or more importantly heteroarenesare useful for different applications for their interesting optical and electronicproperties, there are several methods, other than cross-coupling reactions medi-ated by Pd, are reported. Among others, photoredox catalytic generation of arylradicals from aryl halides are achieved by using strong bases, such as potassiumtert-butoxide [43] at elevated temperature [44, 45] or under photostimulation[43, 46, 47], or in the presence of nucleophiles under ultraviolet (UV) irradiation(cf., SRN1 reactions) [48, 49], or in the presence of an excess of highly reactiveneutral organic reducing agents, such as N2, N2, N12, N12-tetramethyl-7,8-dihydro-6H-dipyrido[1,4]diazepine-2,12-diamine and UV-A (365 nm)irradiation as introduced by Murphy [50].

Photoredox catalytic generation of aryl radicals using visible light is typicallyperformed by using aryl diazonium, iodonium, triarylsulfonium salts, and aryl-sulfonyl chlorides, substrates that possess redox-active functional groups andare easily reduced by typical photocatalysts under visible-light photoirradiation[8]. However, many of these aryl radical precursors are unstable, expensive, andoften not commercially available [8]. Aryl halides, in contrary, are commerciallyavailable, bench stable inexpensive bulk chemicals, and are able to generate arylradicals via single electron transfer that upon trapping with suitable trappingreagents produce arylated products (i.e., C—C bond-formation reactions) orsynthetically useful other chemical bonds (e.g., C—P bonds, see below). In thepast few years, efforts from different research groups have made the generationof aryl radicals from aryl halides possible for C—H arylation reactions and otherreactions using different photoredox catalysts and visible light.

Stephenson reported the dehalogenation of aryl iodides under tin-free con-ditions using fac-Ir(ppy)3 and visible light [41]. In the process, the electrontransfer takes place from the excited-state fac-Ir(ppy)3 to the aryl iodides, whichupon fragmentation release the iodide anion and generate the aryl radical. Thegenerated aryl radical takes a hydrogen atom either from the solvent moleculesor from the radical cation of the amine donors to generate the dehalogenated


Ar–X

PDI• –*

PDI• –

Ar–X•–

Ar •

S •

Ar′–H

S–H

PDI

hν

hνPDI*

Ar–H

NEt2

Et3N•+

Et3N•+Et3N

+

Ar–Ar′X–

Scheme 4.27 Schematic representation of the proposed conPET catalytic cycle using PDI.(Reproduced with permission from Ref. [52]. Copyright 2014, AAAS.)

products. Similarly, Kim reported the activation of aryl iodides using fac-Ir(ppy)3complex [51]. It is worth mentioning here that Stephenson has reported thatthe electron transfer takes place from the excited state of fac-Ir(ppy)3 to the aryliodides, whereas Kim’s mechanistic proposal describes the electron transfer fromthe reduced fac-Ir(ppy)3 to the aryl halides. Note that the reduction potentials ofboth excited state and reduced ground state of fac-Ir(ppy)3 cover the reductionpotentials of aryl iodides (see Scheme 4.1 for the reduction potential values).Although the photoredox catalytic reactions mainly use aryl iodide to generatephotoreduced products, these reports elegantly show that the aryl radicals couldbe generated from respective aryl halides using suitable photoredox catalystsand visible light.

König reported the reduction of aryl halides using consecutive photoinducedelectron transfer (conPET) processes (see Scheme 4.27) [8, 52–56]. The mainadvantage of this process is the use of commercially available organic dyes eitherfor the reduction of carbon—halogen bonds, including aryl chlorides, or for theC—H arylation reaction in the presence of a suitable radical trapping reagent.Note that the reported photoredox catalytic conPET system uses the energy oftwo photons in the same photoredox catalytic cycle and represents a minimal-istic photoredox model of biological photosynthesis (cf., biological Z scheme).Perylenediimides (PDI), a class of organic molecules, form stable radical anionunder nitrogen upon photoirradiation in the presence of suitable sacrificialelectron donor, for example, an amine. The stable colored radical anion couldthen be excited again using visible light and the excited radical anion possessesextremely high reduction potential to transfer an electron to the substitutedaryl halide substrates(aryl-iodides, bromides, and chlorides) yielding the corre-sponding radical anions (Ar—X∙−) (Scheme 4.27). Which, upon fragmentationand by releasing a halide anion, generate the corresponding aryl radicals thatare either trapped by hydrogen atom donors or suitable trapping reagents. Thephotoreduction yields of substituted aryl halides are depicted in Scheme 4.28.

The generated aryl radical has also been shown to be useful for C—Harylation reactions in the presence of suitable (hetero)aryl radical trappingreagents. Notably, the yields of the aryl–(hetero)aryl products depend stronglyon the reactivity of the trapping reagents as well as on the choice of solvent.

N

N

O

N

O

72%

60%

O

N

67%

O

N

52%

X, X = I, Br, Cl

+

PDI (10 mol%), Et3N (8 equiv.)

25–50 equiv.

DMSO, 40 °C, 455 nmTrapping reagent Product

R

O

Br

Br

28%

N

NR R1

R2

R3

R = Me, R1 = H, R2 = H, R3 = H; 70%

R = H, R1 = H, R2 = H, R3 = H; 61%

R = Ph, R1 = H, R2 = H, R3 = H; 74%

R = H, R1 = Me, R2 = H, R3 = Me; 68%

R = H, R1 = Me, R2 = Et, R3 = Me; 71%

O

N R

R = Me; 64%R = H; 54%

(A) Aryl iodide as substrate; (B) Aryl bromide as substrate; (C) Aryl chloride as substrate

(A) (B)

(A)

(C)

(B)(B)(B)

R

PDI (5−10 mol%), Et3N (8 eq.)

DMF, 40 °C , 455 nmR

X; X = I, Br, Cl H

CN

H

98%

94%

H

O

H

O 91%

N

H

CF3

70%

H

RR = CHO; 98%

R = COMe; 63%

R = Me; 77%

R = Br; 45%

H

RR = CHO; 98%R = COMe; 82%

H

R

R = CHO; 35%R = CO2Me; 92%

RH

R = CN; 90%R = CF3; 64%

(D) Aryl iodide as substrate; (E) Aryl bromide as substrate; (F) Aryl chloride as substrate

(D)(E) (F)

(D)(D) (E)

(F)

(F)

Scheme 4.28 PDI catalyzed photoreduction and C—H arylation yields using aryl halides and visible light. (Reproduced with permission from Ref. [8].Copyright 2016, American Chemical Society.)


For example, the C—H arylation reactions were more effective when thephotochemical reactions were performed in dimethyl sulfoxide (DMSO) insteadof dimethylformamide (DMF) (note that DMF is a better hydrogen atom donorthan DMSO). Additionally, hydrogen atom abstraction of the aryl radicalfrom the radical cation of the amine donors, generated in situ via a singleelectron transfer to the excited-state photocatalyst, determines the yields ofthe C—H arylated products (see below). Under this reaction condition, pyrrolenucleus was shown to be an extremely efficient aryl radical trapping reagentand produced biologically important C—H arylated products in good yields asdepicted in Scheme 4.28. Notably, the unprotected pyrrole could also be usedas a trapping reagent. The reactions proceed via radical intermediates [52]. TheconPET concept using PDI as the photoredox catalyst was extended by Zengand coworkers to zeolite networks showing the applicability of such system.Although, the yields, both for the photoreduction and C—H arylation reactions,are very similar to the homogeneous system, the reaction times were reportedto be shorter in the presence of 72.0 equiv. of Et3N in comparison to 8.0 equiv.in homogeneous solution.

A second-generation conPET catalyst is rhodamine 6G (Rh-6G), a widelyapplied fluorescent xanthene dye. Like PDI, Rh-6G yields stable radical anionRh-6G∙− upon photoirradiation under nitrogen with visible light in the presenceof a suitable amine electron donor. The radical anion is relatively stable in theabsence of oxygen, and upon photoexcitation with blue light is able to transferan electron to aryl halides possessing even electron-donating groups (e.g.,4-bromoanisole, see below). Interestingly, the absorption spectra of Rh-6G andRh-6G∙− differ significantly. In detail, Rh-6G absorbs both in the green and blueregions of the visible-light spectrum, whereas, Rh-6G∙− absorbs significantlyonly in the blue region. Such wavelength-dependent access of differently reactivecatalytic species (i.e., Rh-6G∙− or *Rh-6G∙−) allows chemoselective and sequen-tial activations of carbon—halogen bonds for functionalizations using differentwavelengths of visible light for photocatalyst activation (i.e., chromoselectivephotoredox activation). In detail, the excited state of Rh-6G (*Rh-6G) has areduction potential of about –0.8 V versus SCE under visible-light irradiation,irrespective of the excitation wavelength. The ground-state reduction potentialof Rh-6G∙− radical anion, formed upon photoirradiation in the presence of anelectron donor (e.g., N ,N-diisopropylethylamine, DIPEA) under green-lightirradiation, corresponds to about –1.0 V versus SCE. The excited-state reductionpotential of the radical anion *Rh-6G∙− under blue-light irradiation reaches orexceeds a reduction potential value of −2.4 V versus SCE.

Using Rh-6G as the chromoselective photocatalyst, Ghosh and König reportedselective and sequential catalytic carbon—halogen bond activations. Note thatthe main requirement for such selective carbon—halogen bond activations isthat the redox potential of the monofunctionalized product should be higherthan the respective substrate. For example, when 1,3,5-tribromobenzene or1,4-dibromo-2,5-difluorobenzene, aryl bromide substrates with three andtwo equivalent C—Br bonds, respectively, were irradiated in the presence ofRh-6G and DIPEA with green light monosubstituted C—H arylated productswere obtained. Whereas, when the same reaction mixture was irradiated with


blue light two C—Br bonds are activated in sequence (via the formation ofmonosubstituted C—H arylated product as intermediate) to yield disubstitutedC—H functionalized products (Scheme 4.29). Similarly, using different lightsources for photocatalyst excitation 2,4,6-tribromopyrimidine, a biologicallyimportant moiety, could be selectively and sequentially functionalized. Inter-estingly, same or different (hetero)aryl trapping reagents are useful for suchsequential C—H arylation reactions. Note that the sequential C—H arylationreactions using different trapping reagents depend strongly on the irrelativereactivity toward aryl radicals. If the reactivity of the trapping reagent used forthe first C—H arylation reaction is better than the second one, a mixture ofdifferently substituted products could be realized. As the reduction potential ofthe arylated products are more negative than their corresponding aryl halides, aredox potential dependent kinetic control on the sequential C—H arylations isalso possible using this chromoselective conPET catalytic protocol. For example,the conversion of 1,3-dibromobenzene into the corresponding aryl radicalrequires the reduction power of the excited *Rh-6G∙−, but as the activation ofthe second bromide of the resulting compound is kinetically slower owing tothe more negative reduction potential, a stepwise sequential substitution withN-methylpyrrole and pyrrole is possible. Noteworthy, in this case, one must becareful in judging the reaction time as the bi-substituted products are also slowlyformed upon prolonged photoirradiation.

The Rh-6G-based chromoselective photoredox catalytic protocol also allowsselective activation of chemical bonds possessing different fragmentationkinetics following the condition that the reduction potential of the monofunc-tionalized product is more negative than the respective substrate [16, 53]. Forexample, the photoredox reaction with ethyl 2-bromo-(4-bromophenyl)-acetate,which requires a moderate reduction potential (cf., the ground state reductionpotential of Rh-6G∙−), to form the radical in benzylic position (Scheme 4.29)proceeds smoothly under green light photoirradiation yielding the photoreducedproduct ethyl 4-bromophenylacetate. Upon blue-light irradiation, the remain-ing aryl−bromide bond gets activated to undergo C—H arylation reactionswith pyrrole or unsaturated double bonds. Similarly, the diazonium group in4-bromobenzene diazonium tetrafluoroborate could be activated selectivelyusing the reduction potential of *Rh-6G (i.e., in the absence of DIPEA) keepingthe aryl carbon−bromine bond intact. It is worth mentioning here that thediazonium salts are not so stable in the presence of amines. Rh-6G activates theremaining carbon−bromine bond for C—H arylation reactions in the presenceof DIPEA under blue-light photoirradiation.

The applications of rhodamine 6G for the activation of aryl halides for C—Harylation reactions using substituted pyrroles or unsaturated double bonds areshown in Scheme 4.30. The reduction potential of the excited-state *Rh-6G∙−

is reported to be higher than the excited-state reduction potential of PDI∙−. Ascould be seen aryl halide substrates possessing normal electron-withdrawinggroups could be activated easily by using the rhodamine 6G catalyst underblue-light (𝜆Ex = 455 nm) photoirradiation. In addition, aryl halide substratesthat possess electron-donating groups could also be activated using this pho-toredox catalytic protocol. It is worth noting here that the C—H reaction yields

Br

BrBr

Rh-6G, 530 nm light

DIPEA, DMSO, 25 °C+

N

Br

BrN

Br

Rh-6G, 530 nm light


N

Br

BrN

1a, 46% yield 2a, 41% yield

F

F

F

F

Light color guided selective as well as sequential activation of carbon–bromide bonds using rhodamine 6G andgreen or blue light

Examples of selective and sequential C–H arylation reactions

Under 530 nm irradiation Under 455 nm irradiation

5a, 33% yield

F

NH

Br

F

F

NH F

HN

5b, 43% yield

N

N

Br

Br

3a, 67% yield

N

N

Br

HN

4b, 43% yield

N

N

Br

Br

4a, 57% yield

N

N

Br

N

3b, 42% yield

NN

NH

HN

Br

BrBr

Rh-6G, 455 nm light

DIPEA, DMSO, 25 °C +

N

Br

NBr

Rh-6G, 455 nm light

DIPEA, DMSO, 25 °C +

N

(1) (2)

BrN

1b, 48% yield 2b, 63% yield

F

F

F

FN

N

Scheme 4.29 Rhodamine 6G catalyzed chromoselective activations of chemical bonds.

CO2Et

Br

Br CO2Et

Br

Rh-6G, 530 nm light

DIPEA, DMSO, 25 °C

(4)

BrBr

Rh-6G, 455 nm light

DIPEA, DMSO, 25 °C+ N

BrN

Rh-6G, 455 nm light


HN

N

(6a)

Selectivity based on different reaction rates

9a, 56% yield 9b, 43% yield

Selectivity based on difference in redox potentials of functional groups

Light color guided sequential activation of carbon–bromide bonds in one pot

HN

N

N

Br

Br

Br

Rh-6G, 530 nm light

DIPEA, DMSO, 25 °C

Rh-6G, 455 nm light

DIPEA, DMSO, 25 °C

OMe

OMeMeO

+(3)N

N

Br

Br

N+

N

N

Br

N

6b, 21% yield6a, 46% yield

7a, 61% yield

O

O O

O

OO

CO2Et

Br

Rh-6G, 455 nm light


CO2Et

64% yield, 7b:7c = 2.8 : 1

CO2Et

+

7b 7c

Br

O

N2+BF4

–

Br

O

O+

Rh-6G, 530 nm light

DMSO, 25 °C

(5)8a, 54% yield

Br N R

O O

+NR

Rh-6G, 455 nm light

DIPEA, DMSO, 25 °C

R = Me; 8b, 30% yieldR = H; 8c, 24% yield

BrN(6b)

H

Figure 4.29 (Continued)

PhPh

N

N

O

OCH3

Yield 42% Yield 43%

Yield 61%R = Me; yield 59%R = pH; yield 67%

N

F3C

NN

N

R

R1

N

Cl

NH

NN RN

Yield 35%

Yield 51%

N

O

O

Yield 58%

X

X

Br

Y

Starting materials for C–H arylationX = CH, N; Y = substituents;

R = H, Me, Ph

N R

PhPh

R = Me; yield 78%

R = H; yield 64%

R = Ph; yield 77%

R1 = CN; yield 71%

R1 = Me; yield 27%

R1 = COMe; yield 54%

O

N

Yield 25%

R

Br

R

TrapTrap

Rh-6G, DIPEA

DMSO, 25 °C455 nm

Scheme 4.30 Rhodamine 6G catalyzed C—H arylation yields using aryl halide and visible light.


and the reaction kinetics depend strongly on the functional groups present in thesubstrate. The reported reaction times in the presence of electron-withdrawinggroups are normally shorter with respect to the substrates with electron-donatingfunctional groups. Although the yields are moderate to good (due to formationof significant amount of reduction by products) the rhodamine 6G basedphotoredox catalytic system demonstrates for the first time C—H arylationreactions using aryl halides possessing electron-donating substituents.

Similarly, heteroaryl halides could also be used for metal free C—H heteroary-lation reactions [55]. König has shown that heteroaryl halides that possess reduc-tion potentials lower than the reduction potential of the excited-state radicalanion of rhodamine 6G could be activated under blue-light photoirradiation. Dif-ferent C—H heteroarylated products are depicted in Scheme 4.31. It is worthdiscussing here that reduction potential-dependent activation of the (hetero)arylhalide is a successful strategy for the activation of (hetero)aryl halides for C—Hfunctionalization of another halogenated (hetero)arene possessing higher reduc-tion potentials. To illustrate, the reduction potential of bromosubstituted pyrroleis too high (see Ref. [55]) to be activated by using Rh-6G-based catalytic system,however, could be used with ease as a trapping reagent of aryl radicals. This pro-cess allows the activation of one carbon–halogen bond for the synthesis of com-pounds possessing another carbon–halogen bond for further functionalization/s.

Similarly, the same group has demonstrated the functionalizations of uracilsusing respective bromides or chlorides. Notably, such reactions are typicallyperformed using either UV lights [57] or transition metals (mostly Pd). Ele-gantly, the König group has used a simple strategy for the functionalizationof uracils [56]. The C—H arylation reactions have been performed with6-chloro-2,4-dimethoxypyrimidine. The C—H arylated products are thenhydrolyzed to generate the substituted uracils. Interestingly, the C—H arylationreactions are slightly more effective in mixed solvents (1 : 12 H2O:DMSO(v/v))presumably due to the stability of the charged radical pairs and of the radicalanion of the photocatalyst (Scheme 4.32).

The application of the rhodamine 6G catalytic systems spans further for thesynthesis of other interesting compounds. Ghosh and König reported the synthe-sis of pyrrolo[1,2-a]quinolines and ullazines via a photoredox catalytic reductiveannulation processes [54]. N-aryl bromide via a single electron transfer from theexcited-state radical anion of rhodamine 6G forms the respective aryl radical thatreacts intermolecularly with an alkyne to generate a reactive vinyl radical inter-mediate. The vinyl radical upon intramolecular cyclization yields the annulatedproducts after oxidation and rearomatization. The ullazine products follow thesame catalytic mechanism but in this case two activation steps and two annula-tions are involved for the formation of ullazines using dibromo compounds. Thesynthesized pyrrolo[1,2-a]quinolines and ullazines using this catalytic protocolare depicted in Scheme 4.33. Notably due to the two successive electron transfersteps (the ullazines are presumably forms via the substituted quinolone interme-diate) the yields of the ullazine derivatives are relatively low but acceptable if weconsider the formation of four C—C bonds in a single process. Other than thepyrrole heterocycles, N-arylpyrazole could also be used as the radical trapping

S

N

R

S

N

S

N

R1

O

R

S

N

O

N

O

N

N

Boc

N

S N

N

SN

O

NN

S

N

O

O

R = Me; yield 65%

R = H; yield 41%

R = Me; yield 62%

R = H; yield 49%

R = Me, R1 = Me; yield 96%R = H; R1 = Me; yield 84%R = Me, R1 = H; yield 72%R = H; R1 = H; yield 79%R = H, R1 = Ph; yield 91%

Yield 80%

Yield 81%

Yield 49%

Yield 87% Yield 51% Yield 79% Yield 59%

S

N

O

Br

Yield 70%

Y

ZBr +

N

R

Rh-6G, DIPEA

DMSO, 25 °C

455 nmY

Z N

R

Scheme 4.31 Rhodamine 6G catalyzed C—H (hetero)arylation yields using (hetero)aryl bromides and visible light.

N

N

Cl O

O

+N

N

O

O

Rh-6G, DIPEA

DMSO/H2O (12 : 1)25 °C

N

N

O

O

NN

N

O

O

NH

N

N

O

O

NN

N

O

O

NN

N

O

O

O

OO

N

N

O

O

NH

N

N

O

O

Yield 65%Yield 69% Yield 58% Yield 54% Yield 27% Yield 82%Yield 32%

N

N

O

O

NH

NH

O

O

MeOH/HCl (1 : 1)

24 h, reflux

NR

NR

NH

NH

O

O

N

NH

NH

O

O

NNH

NH

O

O

N

Yield 95% Yield 99%Yield 95%

Trap

Trap

Scheme 4.32 C—H arylation reactions of uracil using rhodamine 6G as the conPET catalyst and visible light. Yields of the hydrolysis step are also depicted.

N N N

O O

N N N NN

R

O

R1

F

FF

N

N

R2 R3 R4

R = H, yield 60%R = CH3, yield 51%

R = OCH3, yield 50%

R = F, yield 48%

R = H, yield 56%R = CH3, yield 54%R = OCH3, yield 75%

R = F, yield 50%R = Cl, yield 50%

R = H, yield 52%R = CH3, yield 56%

R = OCH3, yield 41%

R = F, yield 41%

R = H, yield 39%

R = OCH3, yield 35%

R = H, yield 49%

R = CH3, yield 50%

Yield 45% Yield 30% Yield 30%

Y

NX

Br +

R

Rh-6G, DIPEA

DMSO, 25 °C

455 nm

Y

NX

R1R1

R

Scheme 4.33 Synthesis of pyrrolo[1,2-a]quinoline and ullazines using rhodamine 6G as the conPET catalyst and visible light.


P

N

OOEt

OEt

P

OOEt

OEtP

OOEt

OEt

O

OP

OOEt

OEt

FF

F

P

OOEt

OEt

P

OOEt

OEt

O O

P

OOEt

OEt

N

PP O

OEt

OEt

O

OEt

OEt

P

OOEt

OEt

N

P

O

OEt

OEt

N

P

O

OEt

OEt

S

NP

OEt

O

OEt

NP

OEt

O

OEt

Boc

N

P

O

EtO

EtO

S

NP

O

EtO

EtO

N N

P P O

OEt

EtOO

OEt

EtO

S

POOEt

OEt

Yield 76%Yield 58% Yield 54% Yield 27% Yield 70%

Yield 68%

Yield 59% Yield 47%

Yield 94%Yield 92% Yield 52%

Yield 68%Yield 70%Yield 45%Yield 78%Yield 75%

Yield 85%

R = Me; yield 79%R = Ph; yield 62%R = iPr; yield 74%

P

OR

ROO

O

OR

R = Et; yield 76%

R = Me; yield 78%

R

Br

R

P

Rh-6G, DIPEA

DMSO, 25 °C455 nm

O R1

R1P(OR1)3

Scheme 4.34 Yields of the photo-Arbuzov reaction using rhodamine 6G (Rh-6G) as theconPET catalyst and visible light.

heterocycle producing interesting products in useful yields. It is worth mention-ing here that the ullazine products are colored and therefore could compete withthe photocatalyst for visible-light absorption.

König also reported photo-Arbuzov reactions using aryl halides, trialkylphos-phites, and rhodamine 6G as the photocatalyst under blue-light irradiation [58].The proposed mechanism demonstrates that the formed aryl radical reacts withtrialkylphosphites forming a C—P bond and the unstable phosphoranyl radical.Release of an ethyl radical and rearrangement result in the formation of arylphos-phonate in good–excellent yields. As depicted in Scheme 4.34, aryl halides pos-sessing both electron-withdrawing and electron-donating groups could be usedas the precursors of aryl radicals. It is interesting to note that unlike the C—Harylation reactions, when the aryl halide substrates possess two bromine atoms,bisubstituted products are formed. The C—P bond-formation reactions couldalso be performed using aryl triflates as the precursors of aryl radicals.

Recently, Read de Alaniz reported that the activation of aryl halides could bepossible by using 10-phenylphenothiazine (PTH) as the photocatalyst and nearUV light source (𝜆Ex = 380 nm) [59]. The excited-state reduction potential of PTHis −2.1 V versus SCE which is higher than the excited-state reduction potentialof fac-Ir(ppy)3 (−1.7 V vs SCE) and very similar to the reduction potential of theground state fac-Ir(ppy)3. The yields of the photoreduction reactions are depicted


O

H

HO

H

H2N

HH

OHO

O

BnO2C

H

N

H H H H H

H

NH2

MeO2C

CO2Me

MeO2C MeO

CO2Me MeMeO2C OH

SH S

H

EtO

O

H

MeO

OMe

H

H H H H

HOOCNCMeO2CBnO2C

Yield 92% Yield 90% Yield 95% Yield 100% Yield 50%




S

NH

Yield 89%

Aryl iodides

Aryl/alkyl bromides

Aryl chlorides

R

XPTH, Tributylamine

Formic acid, ACNRT, 380 nm

R

H

Scheme 4.35 Photoreduction of aryl or alkyl halides using PTH and visible light.

in Scheme 4.35. Aryl halides, including bromides and chlorides, in the presenceof electron-donating and electron-withdrawing groups could be photoreducedin good–excellent yields using this photoredox catalytic protocol. It is interest-ing to note that the photoreduction reactions also proceed in the presence of airand the reaction could be scaled up to grams. Heteroaryl halides and substitutedalkyl bromides could also be reduced using this catalytic protocol. The photore-dox catalytic reaction proceeds via a radical mechanism as demonstrated by theformation of intramolecular cyclized product.

Recently, the same group has also reported [60] chemoselective radical dehalo-genation of aryl halide substrates possessing multiple carbon–halogen bondsusing both PTH and its derivative (namely, tris-acetyl-PTH) and near UV lightsource (𝜆Ex = 380 nm). Note that due to the presence of electron-withdrawinggroups, the reduction potential of tris-acetyl-PTH is lower (about –1.5 V vsSCE) than PTH. Selective reduction of substrates permits monosubstitutionkeeping the other carbon–halogen bonds intact. The synthetic utility has beendemonstrated through the selective C—H arylation reactions of pyrrole usingaryl halides possessing two functionalizable carbon–bromine bonds.

Jacobi von Wangelin has demonstrated that the reduction potential of thearyl halides could be reached by using the singlet state of 2,5-diphenyloxazole,forms via a triplet–triplet annihilation processes [61]. Upon photoexcitation,


butane-2,3-dione (a sensitizer) transfers its triplet energy to 2,5-diphenyloxazole(a triplet annihilator). Upon triplet–triplet annihilation, 2,5-diphenyloxazole istransferred to its singlet state which possesses enough reduction potential forthe generation of aryl radicals from respective aryl halides. Although a few arylhalide substrates have been investigated, and the use of high-intense laser sourcelimits the application of this photoredox catalytic processes for laboratory-basedsynthetic applications, the system elegantly shows the use of two photons in thesame catalytic cycle for the generation of highly reducing singlet state speciesfor reductive activation of aryl halides for synthetic applications.

Interestingly, the same principle also works in supramolecular gel networksin the presence of oxygen, although different sensitizer and annihilator, namelyplatinum(II) octaethyl-porphyrin and 9,10-diphenylanthracene, respectively,have been used to demonstrate the principle [62]. Although the benefits of usinggel networks to avoid very less amount of solvent and a degassing step (to avoidoxygen for the photoreduction reactions) still need to be proved, the processshows that the use of gel networks inhibits the entrance of oxygen in the reactionmixture.

Among other reports, Weaver reported the generation of azolyl radical fromtheir corresponding bromides for azolylations of arenes and heteroarenes usingfac-Ir(ppy)3 as the photoredox catalyst and visible light [63]. The cross-coupledproducts synthesized using this method are depicted in Scheme 4.36. The sub-strate scopes are relatively broad, and different arenes and heteroarenes are func-tionalized with good to excellent yields using this photoredox catalytic protocol.Importantly, the authors demonstrated that the use of a base is beneficial in orderto suppress the dehalogenated by-products form upon hydrogen atom abstrac-tion. It is also worth mentioning here that like in the other cases, the C—H azoly-lation reactions are not effective with all trapping reagent. For example, the C—Hazolylation reactions do not work with heterocycles such as thiophene or furan.

When the reduction potentials of the aryl halide substrates are relatively low,the activation of aryl halides could also be performed in the presence of catalyststhat possess relatively low reduction potentials. For example, recently Königdemonstrated metal-free perfluoroarylation of arenes and heteroarenes usingEosin Y as the visible-light photoredox catalyst (Scheme 4.37). Triethylamine wasused as a sacrificial electron donor [64]. Due to the low reduction potentials ofthe perfluorinated substrates, Eosin Y could be used as the active photocatalyst.Polyfluorinated biaryls are synthesized using this method. A range of trappingreagents are useful for the trapping of perfluoroaryl radical which could beattributed to the stability of the perfluoroaryl radical under the photoredoxcatalytic reaction condition. Perfluorinated pyridines could be used as substratesfor the photoredox catalytic generation of (hetero)aryl radicals. Interestingly,functionalization of the complex unprotected natural product brucine is possiblealthough surprisingly minor amount of methoxy ipso-substituted product wasalso formed in addition to the desired product. Notably, different haloarenesthat could not be reduced using the low reduction potential of Eosin Y serve astrapping reagents of perfluoroaryl radical under the photoredox catalytic proto-col. In addition, as the reduction potential of partially fluorinated compoundsgo beyond the reduction potential of Eosin Y, such compounds are not reducedunder this photoredox catalytic condition.

S

N

S

N

S

N

tBu

tBu

tBu

Me

S

N

O

O

O

NH

N

S

NO

O

S

N

O

O

O

N

S

N

S

N

O

O

O

S

N

N

N

O

O

O

S

N

NH

N

N

N

S

N

N S

N

S

NO

O

Me

Br

NHN

Br N

Cl Cl N

N

F

F

O

ClO

O

EtO N

N

BocN

N

N N

N

O

O

N

H2N

O

S

N N

S

N

F

FH2N

S

N

H2N

S

N S

O

S

N

N S

N

NH

S

N

NH

F S

N

O

O

HH

Yield 88% Yield 68% Yield 75% Yield 83% Yield 85% Yield 82%


Yield 42%Yield 92%Yield 82% Yield 80% Yield 73% Yield 78%



N

NH

BrN

NH +

N

NH

N

N

fac-Ir(ppy)3, DIPEA

blue LEDs45 °C, MeCN

Scheme 4.36 Photoredox catalytic azolylation of arenes and heteroarenes using fac-Ir(ppy)3 as the photoredox catalyst and visible light.

FF

F

F F

N

FF

F F

FF

F F

F

F

FFF

F F

S

N

F

F

F

F

F

F

F

F

F

FFF

F

F F

FF

Br

F F

N

FF

F F

FF

F F

MeOOMe

MeOOMe

OMe

OMeN

OMe

OMe

R

FF

F F

RF

FF

F

F F

N

FF

F F

F

R = H; yield 85%

R = Me; yield 78%

R = iPr; yield 85%

R = OMe; yield 99%

R = F; yield 62%

R = Cl; yield 76%

R = NO2; yield 0%

Yield 91% Yield 87% Yield 57%

Yield 26%

Yield 82%

Yield 60% Yield 68%


Br

+

R

Eosin Y, Et3N

MeCN, 40 °C

535 nmFn

Fn

R

Scheme 4.37 Perfluoroarylation of arenes and heteroarenes using Eosin Y and visible light.


N F

FF

F N F

FF

F

R

RR MeO

O

N F

FF

F

CN

N F

FF

F

N

N F

FF

F

N

NHO

N F

FF

F

NN

NH

F

FF

F

OMe

OMeMeO

R

R = OMe; yield 70%R = tBu; yield 35%R = Et; yield 55%

R = CF3; yield 54%

R = CN; yield 57%

R = F; yield 50%

R = CO2tBu; yield 58%


Yield 57%

N

O

F F

FF

MeO

OMe

MeO

Yield 52%

F

F

F

CN

NH2

MeO

OMe

OMe

Yield 53%

N F

F

F

MeO

OMe

OMe

NH

O

CO2Et

Yield 48%

F F

FF

MeO

OMe

MeO

EtO2C

EtO2C

F F

F F

Yield 50%

F

FCO2Me

OMe

MeO

MeO

FMeO2C

Yield 52%

N

F

F

FF

F

Trap

fac-Ir(ppy)3, DIPEA

ACN, 45 °Cblue LEDs

N F

FF

F

Trap

Scheme 4.38 Perfluoroarylation of arenes and heteroarenes via C—F bond activation usingfac-Ir(ppy)3 and visible light.

It is worth discussing that the activation of carbon–fluorine bonds, which areless common in Pd-catalyzed C—H arylation reactions due to the very high C—Fbond energy, could also be possible using photoredox catalytic methods for C—Harylation reactions [65]. Although the kinetics of the C—F bond activations havenot been explored properly, the single electron injection allows the activationof C—F bonds as reported by Weaver [66]. Like in other halides, aryl radicalsgenerated from their corresponding aryl fluoride substrates could be used forC—H arylation reactions (see Scheme 4.38) .

4.4 Factors That Determine the Carbon–Halogen BondActivation of Aryl Halides

As motioned earlier, there are two main factors that determine the activationof carbon–halogen bonds: The reduction potentials of (hetero)aryl halidesand the carbon–halogen bond breaking kinetics. The kinetics factor is more

4.6 Achievements and Challenges Ahead 109

important when the reduction potentials of the (hetero)aryl halides are lowand/or the C—X bond strength is very high leading to an extremely slow bondbreaking kinetics (and thus an inefficient activation process). For example, thenitro-substituted aryl halides (e.g., 1-bromo-4-nitrobenzene) accept electronsvery easily from photoredox catalysts; however, the very slow bond breakingkinetics do not allow them to be used as the precursors for aryl radicals [52].However, on the other hand, such fragmentation kinetics allow selective bondfunctionalization of polyhalogenated arenes if the substrate could be reducedselectively without affecting the monosubstituted product [53] .

4.5 Factors That Determine the Yields of the C—HArylated Products

The aryl radical is a highly reactive intermediate. Aryl radicals abstract hydrogenatom from surrounding hydrogen atom donors and determine the yields of theC—H arylated products (see Scheme 4.39). The solvents also play a crucial role indetermining the yields of the C—H arylated products. The reports from the Königgroup show that the formation of the reduction by-product is more in the pres-ence of hydrogen-donating solvents (compare the C—H arylation yields in DMFand in DMSO; note that DMF is a better hydrogen atom donor than DMSO).

4.6 Achievements and Challenges Ahead

Visible-light-mediated photoredox catalytic methods allow activation ofcarbon–halogen bonds including C—F bonds for both photoreduction reactionsand C—H functionalization reactions. Especially, the conPET photoredoxcatalytic systems are based on commercially available organic dyes avoidingtransition metals for synthetically important C—H arylation reactions. In addi-tion to normal bond-forming reactions, regioselectivity and enantioselectivitycould also be achieved using different approaches, in particular by using dualcatalytic protocol or by using chiral photocatalyst. As documented, many thingshave already been achieved; however, still there are challenges ahead. Accordingto the literature reports and sometimes not discussed, although several C—Harylated products, including biologically interesting once, have been synthesizedby using fac-Ir(ppy)3 or conPET photocatalysts; however, the C—H arylationreactions using aryl halides are still not efficient for the functionalization of

NN

N

Ar ArHCatalyst* Catalyst

N++

Scheme 4.39 Hydrogen atom abstraction of aryl radical from the radical cation of DIPEA.


thiophene and furan derivatives, among the heteroaromatics. Additionally, theuse of excess amount of the trapping reagents (up to 20.0 equiv. or more) wasteschemicals in significant amounts. Many reaction conditions require sacrificialelectron donors. In addition, the reaction times are exceptionally long (typicallymany hours to days), which also reflect in very poor quantum yield of thephotochemical reactions.

4.7 Conclusion

In conclusion, this chapter demonstrated the applicability of the photoredoxcatalytic methods for the activation of challenging carbon–halogen bonds ofboth alkyl and aryl halides for synthetically useful organic transformations.Alkylation reactions include, among others, enantioselective synthetic trans-formations via dual-catalytic approaches or by using chiral photocatalysts andsynthetically important perfluoroalkylations. C—H arylations, perfluoroary-lations are also possible under visible light. The discussed conPET catalyticsystems use the energy of two visible-light photons and organic dyes in order toactivate aryl halides bearing even electron-donating groups. The use of excesstrapping reagents in order to capture the aryl or alkyl radicals still makes thediscussed photoredox catalytic processes relatively less competitive; however,the use of inexpensive inorganic bases could make these processes immenselyattractive. We do believe that the use of photoredox catalytic methods for theactivation of organohalides could be potential alternatives to the base promotedor transition-metal-based catalytic activations of organohalides.

References

1 Mirkovic, T., Ostroumov, E.E., Anna, J.M., Van Grondelle, R., Govindjee ,and Scholes, G.D. (2017) Light absorption and energy transfer in the antennacomplexes of photosynthetic organisms. Chem. Rev., 117, 249–293.

2 Blankenship, R.E. (2014) Molecular Mechanisms of Photosynthesis, 2nd edn,Wiley Blackwell.

3 Prier, C.K., Rankic, D.A., and MacMillan, D.W.C. (2013) Visible light pho-toredox catalysis with transition metal complexes: applications in organicsynthesis. Chem. Rev., 113, 5322–5363.

4 Yoon, T.P., Ischay, M.A., and Du, J.N. (2010) Visible light photocatalysis as agreener approach to photochemical synthesis. Nat. Chem., 2, 527–532.



7 Romero, N.A. and Nicewicz, D.A. (2016) Organic photoredox catalysis. Chem.Rev., 116, 10075–10166.

8 Ghosh, I., Marzo, L., Das, A., Shaikh, R., and Konig, B. (2016) Visible lightmediated photoredox catalytic arylation reactions. Acc. Chem. Res., 49,1566–1577.

References 111

9 Reiser, O. (2016) Shining light on copper: unique opportunities forvisible-light-catalyzed atom transfer radical addition reactions and relatedprocesses. Acc. Chem. Res., 49, 1990–1996.

10 Xuan, J. and Xiao, W.J. (2012) Visible-light photoredox catalysis. Angew.Chem. Int. Ed., 51, 6828–6838.

11 Pitre, S.P., McTiernan, C.D., and Scaiano, J.C. (2016) Understanding thekinetics and spectroscopy of photoredox catalysis and transition-metal-freealternatives. Acc. Chem. Res., 49, 1320–1330.

12 Romero, N.A., Margrey, K.A., Tay, N.E., and Nicewicz, D.A. (2015)Site-selective arene C–H amination via photoredox catalysis. Science, 349,1326–1330.

13 Stephenson, C. and Yoon, T. (2016) Enabling chemical synthesis with visiblelight. Acc. Chem. Res., 49, 2059–2060.

14 Kozlowski, M. and Yoon, T. (2016) Editorial for the special issue on photo-catalysis. J. Org. Chem., 81, 6895–6897.

15 Andrieux, C.P., Gallardo, I., Saveant, J.M., and Su, K.B. (1986) Dissociativeelectron-transfer. Homogeneous and heterogeneous reductive cleavage of thecarbon–halogen bond in simple aliphatic halides. J. Am. Chem. Soc., 108,638–647.

16 Houmam, A. (2008) Electron transfer initiated reactions: bond formation landbond dissociation. Chem. Rev., 108, 2180–2237.

17 Pierini, A.B. and Vera, D.M.A. (2003) Ab initio evaluation of intramolecularelectron transfer reactions in halobenzenes and stabilized derivatives. J. Org.Chem., 68, 9191–9199.

18 Costentin, C., Robert, M., and Saveant, J.M. (2004) Fragmentation of arylhalide pi anion radicals. Bending of the cleaving bond and activation vsdriving force relationships. J. Am. Chem. Soc., 126, 16051–16057.

19 Pause, L., Robert, M., and Saveant, J.M. (1999) Can single-electron transferbreak an aromatic carbon–heteroatom bond in one step? A novel example oftransition between stepwise and concerted mechanisms in the reduction ofaromatic iodides. J. Am. Chem. Soc., 121, 7158–7159.

20 Fukuzumi, S., Mochizuki, S., and Tanaka, T. (1990) Photocatalytic reductionof phenacyl halides by 9,10-dihydro-10-methylacridine: control between thereductive and oxidative quenching pathways of tris(bipyridine)rutheniumcomplex utilizing an acid catalysis. J. Phys. Chem., 94, 722–726.

21 Nicewicz, D.A. and MacMillan, D.W.C. (2008) Merging photoredox catalysiswith organocatalysis: the direct asymmetric alkylation of aldehydes. Science,322, 77–80.

22 Neumann, M., Fuldner, S., Konig, B., and Zeitler, K. (2011) Metal-free, coop-erative asymmetric organophotoredox catalysis with visible light. Angew.Chem. Int. Ed., 50, 951–954.

23 Cherevatskaya, M., Neumann, M., Fuldner, S., Harlander, C., Kummel,S., Dankesreiter, S., Pfitzner, A., Zeitler, K., and Konig, B. (2012)Visible-light-promoted stereoselective alkylation by combining heterogeneousphotocatalysis with organocatalysis. Angew. Chem. Int. Ed., 51, 4062–4066.

24 Tucker, J.W., Narayanam, J.M.R., Krabbe, S.W., and Stephenson, C.R.J. (2010)Electron transfer photoredox catalysis: intramolecular radical addition toindoles and pyrroles. Org. Lett., 12, 368–371.


25 Ju, X.H., Liang, Y., Jia, P.J., Li, W.F., and Yu, W. (2012) Synthesis of oxindolesvia visible light photoredox catalysis. Org. Biomol. Chem., 10, 498–501.

26 Tucker, J.W., Nguyen, J.D., Narayanam, J.M.R., Krabbe, S.W., and Stephenson,C.R.J. (2010) Tin-free radical cyclization reactions initiated by visible lightphotoredox catalysis. Chem. Commun., 46, 4985–4987.

27 Bowman, W.R. and Storey, J.M.D. (2007) Synthesis using aromatic homolyticsubstitution - recent advances. Chem. Soc. Rev., 36, 1803–1822.

28 Furst, L., Matsuura, B.S., Narayanam, J.M.R., Tucker, J.W., and Stephenson,C.R.J. (2010) Visible light-mediated intermolecular C–H functionalization ofelectron-rich heterocycles with malonates. Org. Lett., 12, 3104–3107.

29 Swift, E.C., Williams, T.M., and Stephenson, C.R.J. (2016) Intermolecular pho-tocatalytic C–H functionalization of electron-rich heterocycles with tertiaryalkyl halides. Synlett, 27, 754–758.

30 Lin, R., Sun, H.N., Yang, C., Shen, W.B., and Xia, W.J. (2015) Visiblelight-induced difunctionalization of electron-enriched styrenes: synthesisof tetrahydrofurans and tetrahydropyrans. Chem. Commun., 51, 399–401.

31 Jiang, H., Cheng, Y.Z., Wang, R.Z., Zheng, M.M., Zhang, Y., and Yu, S.Y.(2013) Synthesis of 6-alkylated phenanthridine derivatives using photore-dox neutral somophilic isocyanide insertion. Angew. Chem. Int. Ed., 52,13289–13292.

32 Nagib, D.A., Scott, M.E., and MacMillan, D.W.C. (2009) Enantioselectiveα-trifluoromethylation of aldehydes via photoredox organocatalysis. J. Am.Chem. Soc., 131, 10875–10877.

33 Chatterjee, T., Iqbal, N., You, Y., and Cho, E.J. (2016) Controlled fluoroalky-lation reactions by visible-light photoredox catalysis. Acc. Chem. Res., 49,2284–2294.

34 Iqbal, N., Choi, S., Ko, E., and Cho, E.J. (2012) Trifluoromethylation ofheterocycles via visible light photoredox catalysis. Tetrahedron Lett., 53,2005–2008.

35 Choi, Y., Yu, C., Kim, J.S., and Cho, E.J. (2016) Visible-light-induced arylth-iofluoroalkylations of unactivated heteroaromatics and alkenes. Org. Lett., 18,3246–3249.

36 Pan, X.L., Xia, H.G., and Wu, J. (2016) Recent advances in photoinducedtrifluoromethylation and difluoroalkylation. Org. Chem. Front., 3, 1163–1185.

37 Koike, T. and Akita, M. (2016) Fine design of photoredox systems for catalyticfluoromethylation of carbon–carbon multiple bonds. Acc. Chem. Res., 49,1937–1945.

38 Furst, L., Narayanam, J.M.R., and Stephenson, C.R.J. (2011) Total synthesis of(+)-gliocladin C enabled by visible-light photoredox catalysis. Angew. Chem.Int. Ed., 50, 9655–9659.

39 Shih, H.W., Vander Wal, M.N., Grange, R.L., and MacMillan, D.W.C. (2010)Enantioselective α-benzylation of aldehydes via photoredox organocatalysis.J. Am. Chem. Soc., 132, 13600–13603.

40 Huo, H.H., Shen, X.D., Wang, C.Y., Zhang, L.L., Rose, P., Chen, L.A.,Harms, K., Marsch, M., Hilt, G., and Meggers, E. (2014) Asymmetric pho-toredox transition-metal catalysis activated by visible light. Nature, 515,100–103.

References 113

41 Nguyen, J.D., D’Amato, E.M., Narayanam, J.M.R., and Stephenson,C.R.J. (2012) Engaging unactivated alkyl, alkenyl and aryl iodides invisible-light-mediated free radical reactions. Nat. Chem., 4, 854–859.

42 Muhmel, S., Alpers, D., Hoffmann, F., and Brasholz, M. (2015) Iridium(III)photocatalysis: a visible-light-induced dearomatizative tandem 4+2 cyclizationto furnish benzindolizidines. Chem. Eur. J., 21, 12308–12312.

43 Buden, M.E., Guastavino, J.F., and Rossi, R.A. (2013) Room-temperature pho-toinduced direct C–H-arylation via base-promoted homolytic aromatic substi-tution. Org. Lett., 15, 1174–1177.

44 Shirakawa, E. and Hayashi, T. (2012) Transition-metal-free coupling reactionsof aryl halides. Chem. Lett., 41, 130–134.

45 Sun, C.L. and Shi, Z.J. (2014) Transition-metal-free coupling reactions.Chem. Rev., 114, 9219–9280.

46 Xu, Z., Gao, L., Wang, L.L., Gong, M.W., Wang, W.F., and Yuan, R.S. (2015)Visible light photoredox catalyzed biaryl synthesis using nitrogen heterocyclesas promoter. ACS Catal., 5, 45–50.

47 Guastavino, J.F., Buden, M.E., and Rossi, R.A. (2014) Room-temperature andtransition-metal-free mizoroki-heck-type reaction. Synthesis of E-stilbenes byphotoinduced C–H functionalization. J. Org. Chem., 79, 9104–9111.

48 Rossi, R.A., Pierini, A.B., and Penenory, A.B. (2003) Nucleophilic substitutionreactions by electron transfer. Chem. Rev., 103, 71–167.

49 Bardagí, J.I., Vaillard, V.A., and Rossi, R.A. (2012) in The SRN1 Reac-tion: Encyclopedia of Radicals in Chemistry, Biology and Materials (edsC. Chatgilialoglu and A. Studer), John Wiley & Sons, Ltd, Chichester, pp.333–364.

50 Cahard, E., Schoenebeck, F., Garnier, J., Cutulic, S.P.Y., Zhou, S.Z., andMurphy, J.A. (2012) Electron transfer to benzenes by photoactivated neutralorganic electron donor molecules. Angew. Chem. Int. Ed., 51, 3673–3676.

51 Kim, H. and Lee, C. (2012) Visible-light-induced photocatalytic reductivetransformations of organohalides. Angew. Chem. Int. Ed., 51, 12303–12306.

52 Ghosh, I., Ghosh, T., Bardagi, J.I., and Konig, B. (2014) Reduction of arylhalides by consecutive visible light-induced electron transfer processes. Sci-ence, 346, 725–728.

53 Ghosh, I. and Konig, B. (2016) Chromoselective photocatalysis: controlledbond activation through light-color regulation of redox potentials. Angew.Chem. Int. Ed., 55, 7676–7679.

54 Das, A., Ghosh, I., and Konig, B. (2016) Synthesis of pyrrolo 1,2-a quino-lines and ullazines by visible light mediated one- and twofold annulation ofN-arylpyrroles with arylalkynes. Chem. Commun., 52, 8695–8698.

55 Marzo, L., Ghosh, I., Esteban, F., and Konig, B. (2016) Metal-free photocat-alyzed cross coupling of bromoheteroarenes with pyrroles. ACS Catal., 6,6780–6784.

56 Graml, A., Ghosh, I., and König, B. (2017) Synthesis of arylated nucleobasesby visible light photoredox catalysis. J. Org. Chem., 82, 3552–3560.

57 Bardagi, J.I. and Rossi, R.A. (2008) A novel approach to the synthesis of6-substituted uracils in three-step, one-pot reactions. J. Org. Chem., 73,4491–4495.


58 Shaikh, R.S., Dusel, S.J.S., and Konig, B. (2016) Visible-light photo-arbuzovreaction of aryl bromides and trialkyl phosphites yielding aryl phosphonates.ACS Catal., 6, 8410–8414.

59 Discekici, E.H., Treat, N.J., Poelma, S.O., Mattson, K.M., Hudson, Z.M.,Luo, Y.D., Hawker, C.J., and de Alaniz, J.R. (2015) A highly reducingmetal-free photoredox catalyst: design and application in radical dehalo-genations. Chem. Commun., 51, 11705–11708.

60 Poelma, S.O., Burnett, G.L., Discekici, E.H., Mattson, K.M., Treat, N.J.,Luo, Y.D., Hudson, Z.M., Shankel, S.L., Clark, P.G., Kramer, J.W., Hawker,C.J., and de Alaniz, J.R. (2016) Chemoselective radical dehalogenation andC–C bond formation on aryl halide substrates using organic photoredoxcatalysts. J. Org. Chem., 81, 7155–7160.

61 Majek, M., Faltermeier, U., Dick, B., Perez-Ruiz, R., andJacobi von Wangelin, A. (2015) Application of visible-to-UV photon upcon-version to photoredox catalysis: the activation of aryl bromides. Chem. Eur. J.,21, 15496–15501.

62 Haring, M., Perez-Ruiz, R., Jacobi von Wangelin, A., and Diaz, D.D. (2015)Intragel photoreduction of aryl halides by green-to-blue upconversion underaerobic conditions. Chem. Commun., 51, 16848–16851.

63 Arora, A. and Weaver, J.D. (2016) Photocatalytic generation of 2-azolyl rad-icals: intermediates for the azolylation of arenes and heteroarenes via C–Hfunctionalization. Org. Lett., 18, 3996–3999.

64 Meyer, A.U., Slanina, T., Yao, C.J., and Konig, B. (2016) Metal-free perfluo-roarylation by visible light photoredox catalysis. ACS Catal., 6, 369–375.

65 Ahrens, T., Kohlmann, J., Ahrens, M., and Braun, T. (2015) Functionalizationof fluorinated molecules by transition-metal-mediated C–F bond activation toaccess fluorinated building blocks. Chem. Rev., 115, 931–972.

66 Senaweera, S. and Weaver, J.D. (2016) Dual C–F, C–H functionalization viaphotocatalysis: access to multifluorinated biaryls. J. Am. Chem. Soc., 138,2520–2523.

115

Part II

Design and Developments of Visible Light Active Photocatalysis

117

5

Black TiO2: The New-Generation PhotocatalystSanjay Gopal Ullattil1, Soumya B. Narendranath2, and Pradeepan Periyat1*

1University of Calicut, Department of Chemistry, Thenjipalam, Kerala 673 635, India2Central University of Kerala, Department of Chemistry, Periya, Kerala 671314, India

5.1 Introduction

Sun is the universal energy source. The present scenario demands the max-imum utilization of such a powerful renewable energy source. Researchersall over the world are indulged in the invention of technologies that couldeasily harvest the sunlight to tackle energy scarcity in future. TiO2 is the mostpromising sunlight-harvesting material that has been widely investigated andits light harvesting capability has been experimentally proved in the field ofphotocatalysis [1], dye-sensitized solar cells [2], self-cleaning coatings [3], solarwater splitting [4], etc. TiO2 has been unequivocally accepted as a light harvesterbecause of its chemical and thermal stability, nontoxicity, high refractive index,and wide bandgap energy (3.2 eV for anatase and 3 eV for rutile) [5].

Sunlight normally contains 5% UV (200–400 nm), 43% visible light(400–700 nm), and 52% IR (700–2500 nm) energy [6]. A colorless/whitematerial can only absorb energy from the UV region. Since TiO2 is a colorlesssolid in its purely crystalline form, researchers have implemented doping onTiO2 for the improvement of its large wavelength absorption. Metal dopingwas implemented in the early times and the formation of secondary impurities(e.g., Al2TiO5, CeTi4O24, and Ce2Ti2O7) has diminished its crystallinity, which,in turn, reduced the efficiency of the material, in particular, the photocatalyticactivity [7]. After the step back of metal doping, nonmetal doping was cameinto act for better light absorption and it was found that the light harvestingprocess has been successfully implemented without forming any impurity phase.However, the absorption coverage of nonmetal-doped TiO2 structures was notable to absorb IR region, which covers 52% of the solar spectrum [8].

If a material is black colored, it can absorb energy even from the IR region. Theblack TiO2 was reported for the first time by Chen et al. in 2011 [9]. After thediscovery of black TiO2 nanoparticles, the synthesis of black TiO2 nanostructuresfor energy applications is found to be a hot spot in the current environmental

*Corresponding author: [email protected].


118 5 Black TiO2: The New-Generation Photocatalyst

nanoscience. Due to its black color, it can obviously absorb energy even from theIR region of the solar spectrum. Different strategies have been employed for thesynthesis of black TiO2 nanomaterial having various functional applications. Thischapter summarizes the various synthetic pathways that have been carried out forthe preparation of black TiO2 nanomaterial, their properties and photocatalyticapplications, which can pave the way toward an ecofriendly tomorrow.

5.2 Designing Black TiO2 Nanostructures

After the discovery of black TiO2 by Chen et al. [9], it has been widely usedin the light harvesting process due to its wide absorption edge extending toIR region. Hydrogenation is the widely used method for synthesizing blackTiO2 [9–26]. High-/low-pressure hydrogen treatment, ambient hydrogen–argontreatment, ambient hydrogen–nitrogen treatment, ambient argon treatment,and hydrogen plasma treatment are the various hydrogenation methods thathave been carried out up to the date. Apart from these, attempts were madeto synthesize black TiO2 through chemical reduction [22–29], microwaveirradiation [30], electrochemical reduction [4, 31–33], ultrasonication [34], andplasma laser ablation techniques [35].

Chen et al. have synthesized black TiO2 by hydrogenation of pure TiO2nanoparticles with in a high-pressure hydrogen system (20 bar atmosphericpressure) at 200 ∘C for 5 days [9]. Here the long processing has been carried outfor eliminating the possibility of attraction of hydrogen toward dangling bondsand thereby increasing the hydrogen concentration in the disordered layer ofTiO2. The optical bandgap of the material was measured to be 1.54 eV, whichindicates an effective reduction in bandgap of black TiO2 compared to purewhite TiO2. Sun et al. reported the synthesis of black TiO2 by hydrogenationof pure TiO2 nanocrystals with predominant (101) surfaces [10]. Lu et al.implemented high-pressure (35 bar) hydrogenation of commercial P25 TiO2 atroom temperature for more than 15 days to synthesize black TiO2 nanoparticlesand they found that the resultant material has a bandgap of around 1.82 eV [11].Liu et al. hydrogenated TiO2 nanotubes in H2-Ar atmosphere under ambientpressure [12]. In a similar manner, Leshuk et al. successfully synthesized blackTiO2 nanoparticles by the prolonged heating at around 500 ∘C of various TiO2nanomaterials under a relatively low concentration of H2 (H2:Ar= 1:9) [13, 14].Wang et al. showed that rutile TiO2 nanowires on annealing in an ultrahighpure hydrogen atmosphere under low pressure for 3 h at 450 ∘C led to theformation of black TiO2 [15]. Naldoni et al. made an attempt to synthesize blackTiO2 nanoparticles from amorphous TiO2. The method involved was heatingunder H2 atmosphere followed by rapid cooling in an inert environment till thetemperature reaches room temperature [16]. Zhu et al. synthesized Pt loadedblack TiO2 material by hydrogenation of Pt impregnated P25 [17]. The presenceof Pt supports the formation of black TiO2 by a hydrogen spill over from Pt toTiO2 at a temperature range of 160–750 ∘C.

Apart from the methods of hydrogenation of pure TiO2, Myung et al. developeda new synthesis strategy that worked under ambient Ar atmosphere [18]. Thisapproach involved a 600 ∘C annealing of a TiO2 gel formed by adding TiCl4 in

5.2 Designing Black TiO2 Nanostructures 119

aqueous ethanol, HF, and urea. Wang et al. endeavored to develop hydrogenatedblack TiO2 nanoparticles using hydrogen plasma in a thermal plasma furnacemaintained at 500 ∘C [19]. Their absorption extending to IR region pointed tothe formation of black TiO2 nanomaterial. H2 plasma treatment was used alsoby Yan et al. [20]. Teng et al. modified the hydrogen-plasma-assisted preparationmethod of black TiO2 and they additionally employed chemical vapor depositionwith hydrogen as reaction gas [21].

Different types of chemical reduction methods are reported to synthesizeblack TiO2 nanomaterials. Reduction can be achieved with metals such as Al,Zn, and reducing agents such as NaBH4. In a typical chemical reduction processusing aluminum as reducing agent, a double-zone vacuum furnace was used.Here, aluminum and TiO2 were separately (about 800 ∘C for TiO2 and 600 ∘Cfor Al) heated in such a way that TiO2 release oxygen to melted aluminum [22].Wang et al., Cui et al., Zhu et al., and Yang et al. significantly contributed tothe synthesis of black TiO2 nanomaterial synthesis by Al reduction of variousTiO2 morphologies [22–25]. Sinhamahapatra et al. developed a new controlledmagnesiothermic reduction to synthesize reduced black TiO2 under H2/Aratmosphere [26]. This method followed reduction of the commercially availablenano-anatase TiO2 using different concentration of Mg followed by 5% H2/Artreatment. The oxygen vacancies created during reduction significantly improvedoptical absorption in the visible and infrared region.

Interestingly, Lin et al. achieved nonmetal-doping in black TiO2 using alu-minum as a reductant [27]. This method followed a two-step process, that is,creation of oxygen vacancies with aluminum (in a double-zone furnace) followedby incorporation of nonmetal dopants (H, N, S, and I) in the oxygen vacancies.The nonmetal-doped TiO2−x nanoparticles showed significant absorptionfeatures extending up to the near-infrared region. Kang et al. implementeda chemical reduction approach using NaBH4 at room temperature [28]. Thecomplete reduction of TiO2 nanotubes, used in this method, can be achievedwithin 1 h. The resulting black TiO2 nanotube had enhanced absorption charac-teristics. Zhang et al. developed novel Ni2+ and Ti3+ codoped black TiO2 withattractive visible to IR absorbance characteristics contributed by the codopedspecies [29]. The strategy followed here is a combination of sol–gel techniqueand an in situ chemical reduction using NaBH4 followed by calcination.

Chemical oxidation can also lead to the formation of black TiO2. Grabstanow-icz et al. [36] prepared black TiO2 by oxidizing TiH2 powder. H2O2 was used as anoxidizing agent, which formed a yellow gel with TiH2 for prolonged stirring. Thedried gel was converted into rutile black TiO2 by heat treatment. The materialpossessed significant absorbance in the visible region extending up to IR.

Very recently, our group has designed two kinds of synthesis strategies forblack TiO2. The first method used microwave irradiation to get oxygen-rich yel-low anatase with visible light absorption and oxygen vacancy-rich black anataseTiO2 with enhanced NIR absorption [30]. This simple and productive methodinvolved the preparation of a sol by merely using titanium butoxide, manganeseacetate, and water as precursors. During the sol preparation, doping and hydrox-ylation were achieved simultaneously. Then, the sol was subjected to microwaveirradiation at 150 ∘C with a stirring speed of 1200 rpm for 5 min followed by dry-ing at 80 ∘C. The second strategy is called a “one-pot” gel combustion method.


Here calcination of a gel formed from titanium(IV) butoxide and ethyleneglycol at 300 ∘C yielded Ti3+ self-doped black TiO2 with oxygen vacancies[37]. Li et al. designed bicrystalline black TiO2−x nanofibers [38]. Initially, theyprepared hydrated titanate powder from dried K2Ti2O5 paste suspended in anagitated HCl solution to reduce the amount K+ ions. Further calcination at600 ∘C for 2 h under pure H2 atmosphere yielded the black TiO2(B)/anatasebicrystalline TiO2−x.

Attempts to design black TiO2 using a different strategy led to electrochemicalreduction process [4]. This method involves mainly three steps, namely anodiza-tion of TiO2 in ethylene glycol, NH4F, and water mixture to make nanotubes(carbon or platinum is used as the cathode), annealing of the anodized TiO2at 450 ∘C, and finally the electrochemical reduction. The process implementeddoping facilitated by the system of annealed TiO2 nanotubes and Pt, as cathodeand anode, respectively. Aqueous Na2SO4 is used as electrolyte and the dopedTiO2 nanotubes have a black color. Xu et al. developed a method of a two-stepanodization followed by electrochemical reductive doping [4]. Zhang et al. alsodeveloped black TiO2 nanotubes in a similar manner [31]. Li et al. also achievedblack TiO2 nanomaterials by the electrolytic reduction of TiO2 nanotubesprepared by the anodization of Ti foil [32]. Zhou and Zhang synthesized blackTiO2 nanotubes with well-defined layers and appreciable optical absorbance.They used multipulse anodization method to fabricate the TiO2 nanotube films.

Dong et al. developed a method to prepare black TiO2 nanotubes by anodiza-tion followed by annealing, further reduction was not required here [33]. Herealso TiO2 nanotubes were prepared using anodization of Ti foil. The resultingTiO2 nanotubes were removed. After the second anodization, Ti foil was washedwith ethanol and distilled water. The dried powder was sintered at 450 ∘C. Theblack TiO2 was retrieved by discarding the top layer.

Recently, Fan et al. observed the formation of black TiO2 through prolongedultrasonication [34]. They started with the simple procedure of making TiO2 solfollowed by ultrasonication for several hours and drying at 80 ∘C. It was alsonoted that with long duration of ultrasonication, the intensity of black colorincreases. A pulsed laser ablation technique was implemented by Chen et al. forthe synthesis of black TiO2 [35]. In this procedure, an aqueous suspension ofcommercially available TiO2 was irradiated with an Nd:YAG pulsed laser. Thesuspension was taken in a cuvette and irradiation was done in such a way thatone side of the cuvette was irradiated for 1 h and the laser beam was focused toanother side for another 1 h irradiation. Further, the suspension was filtered anddried at 80 ∘C for 12 h and the product had a black color. Important synthesisprocesses reported so far can be summarized from Scheme 5.1.

According to the literature, the crystal defects play a crucial role to improvethe optical and consequently in the enhanced catalytic properties of black TiO2nanomaterials. The crystal defects found in black TiO2 span the disorderedcrystalline structure, existence of Ti3+ ions, oxygen vacancies, and the presenceof Ti—OH and Ti—H fragments. All these factors are suggested to contributeto the shifting of valence band edge capable of absorbing from the visibleto IR region of electromagnetic spectrum. The nature of crystalline defectspresent in black TiO2 nanostructures depends on the methods of synthesis

5.2 Designing Black TiO2 Nanostructures 121

Hydrogenation

Anodization, annealing,

reduction

H2 O

2 on TiH2

One pot gel

combustion

Black TiO2

Microw

ave

irradiation

Ultra

sonic

ation

Che

mical

oxida

tion

Ele

ctro

chem

ical

reductio

n

Ti precursor sol

Irrad

iatio

n at

150

°C fo

r

5 m

in o

n pr

ecur

sor s

ol

Hig

h/a

mbie

nt pre

ssure

,

Hydro

gen p

lasm

a

Calc

ination o

f pre

curs

or

gel of T

i at 300 °

C

Al, Zn,

Mg,

NaB

H 4

Nd:Y

AG

pulsed

laser on TiO2

Pulse la

ser a

blation

Chem

ical reduction

Scheme 5.1 Various synthesis routes to black TiO2 nanomaterials.

adapted. The formation of core–shell black TiO2 nanoparticles was observedwith high and ambient pressure hydrogenation, hydrogen plasma treatment, andaluminum chemical reduction [9, 16, 19, 22]. The existence of disordered surfacelayer in the black TiO2 core–shell nanoparticles was contradicted by the clearsurface formed for black TiO2 nanotubes formed through similar hydrogenationroutes [39]. Such types of disorders can effectively be analyzed using powderX-ray diffraction (PXRD), high-resolution transmission electron microscopy(HRTEM), and Raman spectroscopy. The presence of Ti3+ ions were rarelyreported, but Grabstanowicz et al., Zhang et al., and Ullattil et al. observedexistence of Ti3+ in black TiO2 prepared via chemical oxidation of TiH2, electro-chemical reduction, and a one-pot gel synthesis, respectively, with XPS analysis[31, 36, 37]. Oxygen vacancies are very common in black TiO2 nanomaterials. Allof the above-discussed synthesis methods resulted in oxygen vacancies in blackTiO2. Electron spin resonance (ESR) spectroscopy is very useful in studying theexistence of oxygen vacancy. Fourier transform infra-red (FTIR) spectroscopyrevealed the presence of Ti—OH group in black TiO2 prepared via normalhydrogenation. 1H nuclear magnetic resonance (NMR) spectra of hydrogenplasma treated TiO2 also showed the existence Ti—OH group. Zheng et al. andZhang et al. reported the presence of Ti—H fragment by analyzing the XPSspectra and PXRD patterns, respectively [40, 41]. Apart from the experimentalevidences, theoretical approaches also point to the valence band shift in black


TiO2 nanostructures. The theoretical considerations compared the kineticbarrier of hydrogen migration from different facets of TiO2 to the subsurface lev-els to that of desorption of hydrogen. The kinetic barrier was found to be smallerfor hydrogen migration from anatase (101) surface to the subsurface than that ofdesorption [42]. Raghunath et al. proposed that both H and H2 were migrated tothe subsurface from anatase (101) face, and H2 enclosed in the subsurface levelwas transformed into OH fragments by capturing oxygen from TiO2. Furtherthe OH species was converted into water and the oxygen vacancies were createdin TiO2. The presence of interstitial H, H2, and H2O led to structural disorder,which, in turn, promoted the shifting of band positions [43]. Lu et al. suggestedthat the bandgap was affected by the lattice disordering caused by adsorbedspecies via a strong interaction with the oxygen 2p and Ti 3d orbitals. Also,the electron–hole recombination was significantly suppressed due to separatepathways for the charge carriers through different facets [44]. The experimentalevidences corroborate with theoretical observations of the tuning of bandpositions, which makes these nanostructures suitable for various applications.

5.3 Black TiO2 as Photocatalyst

Various morphologies of pure TiO2 were extensively studied for its advancedphotocatalytic applications, mainly water splitting and mineralization ofpollutants. The state-of-the-art points to the enhanced efficiency of blackTiO2 nanomaterials toward photocatalysis. Chen et al. employed black TiO2nanomaterial synthesized by hydrogenation was an active catalyst towardmineralization of contaminants by taking methylene blue (MB) and phenolas model systems [9]. They also proved that the materials were equally activefor photocatalytic hydrogen generation from aqueous methanol [9]. Ullattilet al. have demonstrated the oxygen vacancy-rich black anatase TiO2 for solarphotocatalysis [30]. Experimentally, they have proved that the black anataseTiO2 that has been synthesized was twofold more active than the commerciallyavailable photocatalyst Degussa-P25 for (MB) degradation and fourfold moreactive than the oxygen-rich yellow anatase TiO2. Xin et al. reported a 30-foldenhancement in visible-light decomposition of MB and four times improvementin the maximal transient photocurrent density compared with P25, using blackTiO2 prepared by solvothermal-assisted method [45]. Zhu et al. fabricated blackplate like brookite TiO2 with core/disordered shell structure (TiO2@TiO2−x)through Al reduction [24]. The black brookite TiO2 facilitated photodegradationof MB and methyl orange (MO) under visible light, achieving a higher selectiv-ity toward MO. Samsudin et al. reported black TiO2, prepared via controlledhydrogenation, could be successfully employed for atrazine dye degradation [46].

The Al reduced black TiO2 was found to have exceptionally higher photocat-alytic degradation rate toward MO and phenol [24]. It was also employed inthe process of hydrogen generation, and it was reported that with increasingAl reduction temperature, the amount of H2 generation also increased. Morespecifically, the amount of hydrogen gas produced was 8.5 times more than thatof pristine TiO2. The same black TiO2 was found excellent photoelectrochemical

References 123

electrode exhibiting 1.7% solar to hydrogen efficiency. Black TiO2 nanotubessynthesized by Liu et al. showed a high open circuit photocatalytic hydro-gen production rate without the presence of a cocatalyst [12]. Zhao et al.demonstrated black rutile nanorods through Zn reduction and the catalyst wasgenerated H2 from the water–methanol system both under UV and visible-lightirradiation [47]. Zhou et al. achieved ordered mesoporous black TiO2 with highthermal stability [48]. These black TiO2 having large pore size and high surfacearea led to high solar-driven hydrogen production rate, which was almosttwice that of pristine mesoporous TiO2. The cycling tests of the photocatalytichydrogen generation under air mass (AM) 1.5 and under visible light werealso conducted to confirm the reusability of the photocatalyst. Lepcha et al.reported electrospun black TiO2 nanofibers by hydrogen plasma treatment[49]. These nanofibers exhibited tenfold more photoelectrochemical perfor-mance than pristine TiO2. Yang et al. reported an excellent H2 production byS doping on core–shell nanostructured black rutile TiO2 [25]. The photocat-alyst exhibited 1.67% solar to hydrogen conversion efficiency. Apart from thephotocatalytic applications, black TiO2 nanoparticles are widely studied in thearea of Li ion batteries [18, 20], supercapacitors [32], fuel cells [41], and fieldemission [50].

5.4 Conclusions

Since the complete utilization of the universal energy source, sunlight, is a tediousone, a material with wide optical absorption features and enhanced crystallinequalities may be expected to fulfill the crisis of present energy demand. In thisscenario, black TiO2 nanostructures will be a “Holy Grail” in the complete devel-opment of harvesting sunlight, in particular, photocatalysis and water-splittingapplications. This chapter introduced the new light-harvesting material blackTiO2, its various synthesis methods, and its photocatalytic applications.

References

1 Schneider, J., Matsuoka, M., Takeuchi, M., Zhang, J., Horiuchi, Y., Anpo, M.,and Bahnemann, D.W. (2014) Understanding TiO2 photocatalysis: mecha-nisms and materials. Chem. Rev., 114 (19), 9919–9986.

2 Tian, J., Zhao, Z., Kumar, A., Boughton, R.I., and Liu, H. (2014) Recentprogress in design, synthesis, and applications of one-dimensional TiO2nanostructured surface heterostructures: a review. Chem. Soc. Rev., 43 (20),6920–6937.

3 Banerjee, S., Dionysiou, D.D., and Pillai, S.C. (2015) Self-cleaning applicationsof TiO2 by photo-induced hydrophilicity and photocatalysis. Appl. Catal., B,176, 396–428.

4 Xu, C., Song, Y., Lu, L., Cheng, C., Liu, D., Fang, X., and Li, D. (2013)Electrochemically hydrogenated TiO2 nanotubes with improved photoelec-trochemical water splitting performance. Nanoscale Res. Lett., 8 (1), 1–7.


5 Etacheri, V., Di Valentin, C., Schneider, J., Bahnemann, D., and Pillai, S.C.(2015) Visible-light activation of TiO2 photocatalysts: advances in theory andexperiments. J. Photochem. Photobiol., C, 25, 1–29.

6 Levinson, R., Berdahl, P., and Akbari, H. (2005) Solar spectral optical prop-erties of pigments-part I: model for deriving scattering and absorptioncoefficients from transmittance and reflectance measurements. Sol. EnergyMater. Sol. Cells, 89 (4), 319–349.

7 Periyat, P., Baiju, K.V., Mukundan, P., Pillai, P.K., and Warrier, K.G.K. (2007)Aqueous colloidal sol–gel route to synthesize nanosized ceria-doped titaniahaving high surface area and increased anatase phase stability. J. Sol–Gel Sci.Technol., 43 (3), 299–304.

8 Chen, X. and Burda, C. (2008) The electronic origin of the visible-lightabsorption properties of C-, N- and S- doped TiO2 nanomaterials. J. Am.Chem. Soc., 130 (15), 5018–5019.

9 Chen, X., Liu, L., Peter, Y.Y., and Mao, S.S. (2011) Increasing solar absorptionfor photo-catalysis with black hydrogenated titanium dioxide nanocrystals.Science, 331 (6018), 746–750.

10 Sun, C., Jia, Y., Yang, X.H., Yang, H.G., Yao, X., Lu, G.Q., and Smith, S.C.(2011) Hydrogen incorporation and storage in well-defined nanocrystals ofanatase titanium dioxide. J. Phys. Chem. C, 115 (51), 25590–25594.

11 Lu, H., Zhao, B., Pan, R., Yao, J., Qiu, J., Luo, L., and Liu, Y. (2014) Safe andfacile hydrogenation of commercial Degussa P25 at room temperature withenhanced photocatalytic activity. RSC Adv., 4 (3), 1128–1132.

12 Liu, N., Schneider, C., Freitag, D., Hartmann, M., Venkatesan, U., Müller, J.,and Schmuki, P. (2014) Black TiO2 nanotubes: cocatalyst-free open-circuithydrogen generation. Nano Lett., 14 (6), 3309–3313.

13 Leshuk, T., Linley, S., and Gu, F. (2013) Hydrogenation processing of TiO2nanoparticles. Can. J. Chem. Eng., 91 (5), 799–807.

14 Leshuk, T., Parviz, R., Everett, P., Krishnakumar, H., Varin, R.A., and Gu, F.(2013) Photocatalytic activity of hydrogenated TiO2. ACS Appl. Mater. Inter-faces, 5 (6), 1892–1895.

15 Wang, G., Wang, H., Ling, Y., Tang, Y., Yang, X., Fitzmorris, R.C., and Li, Y.(2011) Hydrogen-treated TiO2 nanowire arrays for photoelectrochemicalwater splitting. Nano Lett., 11 (7), 3026–3033.

16 Naldoni, A., Allieta, M., Santangelo, S., Marelli, M., Fabbri, F., Cappelli,S., and Dal Santo, V. (2012) Effect of nature and location of defects onbandgap narrowing in black TiO2 nanoparticles. J. Am. Chem. Soc., 134(18), 7600–7603.

17 Zhu, Y., Liu, D., and Meng, M. (2014) H2 spillover enhanced hydrogenationcapability of TiO2 used for photocatalytic splitting of water: a traditionalphenomenon for New applications. Chem. Commun., 50 (45), 6049–6051.

18 Myung, S.T., Kikuchi, M., Yoon, C.S., Yashiro, H., Kim, S.J., Sun, Y.K., andScrosati, B. (2013) Black anatase titania enabling ultra high cycling rates forrechargeable lithium batteries. Energy Environ. Sci., 6 (9), 2609–2614.

References 125

19 Wang, Z., Yang, C., Lin, T., Yin, H., Chen, P., Wan, D., and Jiang, M. (2013)H-doped black titania with very high solar absorption and excellent pho-tocatalysis enhanced by localized surface plasmon resonance. Adv. Funct.Mater., 23 (43), 5444–5450.

20 Yan, Y., Hao, B., Wang, D., Chen, G., Markweg, E., Albrecht, A., and Schaaf, P.(2013) Understanding the fast lithium storage performance of hydrogenatedTiO2 nanoparticles. J. Mater. Chem. A, 1 (46), 14507–14513.

21 Teng, F., Li, M., Gao, C., Zhang, G., Zhang, P., Wang, Y., and Xie, E. (2014)Preparation of black TiO2 by hydrogen plasma assisted chemical vapor depo-sition and its photocatalytic activity. Appl. Catal., B, 148, 339–343.

22 Wang, Z., Yang, C., Lin, T., Yin, H., Chen, P., Wan, D., Xu, F., Huang, F.,Lin, J., Xie, X., and Jiang, M. (2013) Visible-light photocatalytic, solar thermaland photoelectrochemical properties of aluminium-reduced black titania.Energy Environ. Sci., 6, 3007–3014.

23 Cui, H., Zhao, W., Yang, C., Yin, H., Lin, T., Shan, Y., Xie, Y., Gu, H., andHuang, F. (2014) Black TiO2 nanotube arrays for high-efficiency photoelectro-chemical water-splitting. J. Mater. Chem. A, 2 (23), 8612–8616.

24 Zhu, G., Lin, T., Lu, X., Zhao, W., Yang, C., Wang, Z., Yin, H., Liu, Z., Huang,F., and Lin, J. (2013) Black brookite titania with high solar absorption andexcellent photocatalytic performance. J. Mater. Chem. A, 1 (23), 9650–9653.

25 Yang, C., Wang, Z., Lin, T., Yin, H., Lü, X., Wan, D., and Xie, X. (2013)Core–shell nanostructured “black” rutile titania as excellent catalyst forhydrogen production enhanced by sulfur doping. J. Am. Chem. Soc., 135 (47),17831–17838.

26 Sinhamahapatra, A., Jeon, J.P., and Yu, J.S. (2015) A new approach to pre-pare highly active and stable black titania for visible light-assisted hydrogenproduction. Energy Environ. Sci., 8 (12), 3539–3544.

27 Lin, T., Yang, C., Wang, Z., Yin, H., Lü, X., Huang, F., and Jiang, M. (2014)Effective nonmetal incorporation in black titania with enhanced solar energyutilization. Energy Environ. Sci., 7 (3), 967–972.

28 Kang, Q., Cao, J., Zhang, Y., Liu, L., Xu, H., and Ye, J. (2013) Reduced TiO2nanotube arrays for photoelectrochemical water splitting. J. Mater. Chem. A,1 (18), 5766–5774.

29 Zhang, H., Xing, Z., Zhang, Y., Li, Z., Wu, X., Liu, C., Zhu, Q., andZhou, W. (2015) Ni2+ and Ti3+ Co-doped porous black anatase TiO2 withunprecedented-high visible-light driven photocatalytic degradation perfor-mance. RSC Adv., 5 (129), 107150–107157.

30 Ullattil, S.G. and Periyat, P. (2015) Green microwave switching from oxygenrich yellow anatase to oxygen vacancy rich black anatase TiO2 solar photocat-alyst using Mn(II) as ‘anatase phase purifier’. Nanoscale, 7 (45), 19184–19192.

31 Zhang, Z., Hedhili, M.N., Zhu, H., and Wang, P. (2013) Electrochemicalreduction induced self-doping of Ti3+ for efficient water splitting perfor-mance on TiO2 based photoelectrodes. Phys. Chem. Chem. Phys., 15 (37),15637–15644.


32 Li, H., Chen, Z., Tsang, C.K., Li, Z., Ran, X., Lee, C., and Pan, B. (2014)Electrochemical doping of anatase TiO2 in organic electrolytes forhigh-performance supercapacitors and photocatalysts. J. Mater. Chem. A,2 (1), 229–236.

33 Dong, J., Han, J., Liu, Y., Nakajima, A., Matsushita, S., Wei, S., and Gao, W.(2014) Defective black TiO2 synthesized via anodization for visible-lightphotocatalysis. ACS Appl. Mater. Interfaces, 6 (3), 1385–1388.

34 Fan, C., Chen, C., Wang, J., Fu, X., Ren, Z., Qian, G., and Wang, Z. (2015)Black hydroxylated titanium dioxide prepared via ultrasonication withenhanced photocatalytic activity. Sci. Rep., 5, 11712.

35 Chen, X., Zhao, D., Liu, K., Wang, C., Liu, L., Li, B., and Shen, D. (2015)Laser-modified black titanium oxide nanospheres and their photocat-alytic activities under visible light. ACS Appl. Mater. Interfaces, 7 (29),16070–16077.

36 Grabstanowicz, L.R., Gao, S., Li, T., Rickard, R.M., Rajh, T., Liu, D.J., andXu, T. (2013) Facile oxidative conversion of TiH2 to high-concentrationTi3+-self-doped rutile TiO2 with visible-light photoactivity. Inorg. Chem.,52 (7), 3884–3890.

37 Ullattil, S.G. and Periyat, P. (2016) A ‘One Pot’ Gel combustion strategytowards Ti3+ self-doped ‘black’ anatase TiO2−x solar photocatalyst. J. Mater.Chem. A, 4 (16), 5854–5858.

38 Li, L., Shi, K., Tu, R., Qian, Q., Li, D., Yang, Z., and Lu, X. (2015) Black TiO2(B)/anatase bicrystalline TiO2-x nanofibers with enhanced photocatalyticperformance. Chin. J. Catal., 36 (11), 1943–1948.

39 Lu, Z., Yip, C.-T., Wang, L., Huang, H., and Zhou, L. (2012) HydrogenatedTiO2 nanotube arrays as high-rate anodes for lithium-ion microbatteries.Chem-PlusChem, 77 (11), 991–1000.

40 Zheng, Z., Huang, B., Lu, J., Wang, Z., Qin, X., Zhang, X., Dai, Y., andWhangbo, M. (2012) Hydrogenated titania: synergy of surfac modificationand morphology improvement for enhanced photocatalytic activity. Chem.Commun., 48 (46), 5733–5735.

41 Zhang, C., Yu, H., Li, Y., Gao, Y., Zhao, Y., Song, W., and Yi, B. (2013) Sup-ported noble metals on hydrogen-treated TiO2 nanotube arrays as highlyordered electrodes for fuel cells. ChemSusChem, 6 (4), 659–666.

42 Aschauer, U. and Selloni, A. (2012) Hydrogen interaction with the anataseTiO2 (101) surface. Phys. Chem. Chem. Phys., 14 (48), 16595–16602.

43 Raghunath, P., Huang, W.F., and Lin, M.C. (2013) Quantum chemical elucida-tion of the mechanism for hydrogenation of TiO2 anatase crystals. J. Chem.Phys., 138, 154705–154709.

44 Lu, J., Dai, Y., Jin, H., and Huang, B. (2011) Effective increasing of opticalabsorption and energy conversion efficiency of anatase TiO2 nanocrystals byhydrogenation. Phys. Chem. Chem. Phys., 13 (40), 18063–18068.

45 Xin, X., Xu, T., Yin, J., Wang, L., and Wang, C. (2015) Management onthe location and concentration of Ti3+ in anatase TiO2 for defects-inducedvisible-light photocatalysis. Appl. Catal., B, 176, 354–362.

References 127

46 Samsudin, E.M., Hamid, S.B.A., Juan, J.C., Basirun, W.J., and Kandjani, A.E.(2015) Surface modification of mixed-phase hydrogenated TiO2 and corre-sponding photocatalytic response. Appl. Surf. Sci., 359, 883–896.

47 Zhao, Z., Tan, H., Zhao, H., Lv, Y., Zhou, L.J., Song, Y., and Sun, Z. (2014)Reduced TiO2 rutile nanorods with well-defined facets and their visible-lightphotocatalytic activity. Chem. Commun., 50 (21), 2755–2757.

48 Zhou, W., Li, W., Wang, J.Q., Qu, Y., Yang, Y., Xie, Y., and Zhao, D. (2014)Ordered mesoporous black TiO2 as highly efficient hydrogen evolution photo-catalyst. J. Am. Chem. Soc., 136 (26), 9280–9283.

49 Lepcha, A., Maccato, C., Mettenbörger, A., Andreu, T., Mayrhofer, L.,Walter, M., and Meerholz, K. (2015) Electrospun black titania nanofibers:influence of hydrogen plasma-induced disorder on the electronic struc-ture and photoelectrochemical performance. J. Phys. Chem. C, 119 (33),18835–18842.

50 Zhu, W.D., Wang, C.W., Chen, J.B., Li, D.S., Zhou, F., and Zhang, H.L. (2012)Enhanced field emission from hydrogenated TiO2 nanotube arrays. Nanotech-nology, 23 (45), 455204.

129

6

Effect of Modification of TiO2 with Metal Nanoparticleson Its Photocatalytic Properties Studied by Time-ResolvedMicrowave ConductivityHynd Remita, María Guadalupe Méndez Medrano, and ChristopheColbeau-Justin

CNRS UMR 8000, Univ Paris-Sud – Université Paris-Saclay, Laboratoire de Chimie Physique,Bât. 349, 91405 Orsay, France

6.1 Introduction

TiO2 is a very efficient photocatalyst due to its strong oxidation capacity,high photochemical and biological stability, and low cost. Since the discoveryof photoinduced decomposition of water on a TiO2 electrode, TiO2-basedphotocatalysts have attracted wide attention [1].

The limitation in TiO2 application results from low quantum yield due tofast charge-carriers (electron/hole e−/h+) recombination and its activationonly under UV irradiation because of the value of its energy band gap (3.2 eVfor anatase and 3.0 eV for rutile) [2]. UV light constitutes only about 3–4% ofthe solar spectrum impinging on the earth’s surface; therefore, modificationof titania to extent its absorption to the visible domain and to enhance itsactivity is a very active area of research. Doping TiO2 with N, C, or S has beenused to extent its activity toward the visible light [3, 4]. Doping TiO2 withions such as Rh3+ or Bi3+ was also used to enhance its photocatalytic activityunder visible light [5–7]. Surface modification with nanoparticles (NPs) of noblemetals (platinum, palladium, silver, and gold) can result in enhancement ofthe photoconversion quantum yield and may allow the extension of the lightabsorption of wide bandgap semiconductors to visible light [8–10]. In particular,plasmonic photocatalysts have appeared as a very promising way to inducea photocatalytic activity of TiO2 in the visible [11, 12]. Coupling titania withanother semiconductor of a smaller band gap (such as CdSor Bi2S3) is also a wayto enhance the photocatalytic efficiency by decreasing the recombination rateand inducing a photocatalytic activity in the visible range [13]. The applicationsof photocatalysts concern mainly: self-cleaning surfaces, water and air treatment,and solar fuel production. Artificial photosynthesis to recycle CO2 is also a topicof increasing interest.

The photocatalytic activity of TiO2 compounds is related to the creation andthe evolution of charge carriers in the photocatalyst [14]. Thus, the knowledgeof the relation existing between charge-carrier lifetimes and material structural


130 6 Effect of Modification of TiO2 with Metal Nanoparticles

parameters can help to understand the mechanisms leading to the photoactivity.To follow the charge-carrier dynamics in TiO2, the variation of the sample con-ductivity after illumination must be determined.

Time resolved microwave conductivity (TRMC) is a contactless method, basedon the measurement of the change of the microwave power reflected by a sampleinduced by laser pulsed illumination [15, 16]. The TRMC signal allows followingdirectly the decay of the number of electrons and holes after the laser pulse byrecombination or trapping of the charge carriers.

We summarize in this chapter studies on surface modification of TiO2 withmono- and bimetallic nanoparticles (NPs) (mainly synthesized by radiolysis)for photocatalytic applications (water depollution and hydrogen generation).We show here that TRMC is a very powerful method to study charge-carrierdynamics and to understand the effect of semiconductor modification on itsphotocatalytic activity.

In this chapter, we present different examples of modification of TiO2 withmetal nanoparticles for photocatalytic applications: water depollution (mainlyphenol photodegradation) and hydrogen production. The relation betweentitania modification, charge-carrier dynamics (electronic properties) and pho-tocatalytic activity has been investigated. It has been evidenced that one cancorrelate the modification with metal NPs with the change of the photocon-ductivity signal and the photocatalytic activity. A strong influence of structuralparameters on the photoconductivity is observed, and a relation between thephotoconductivity and the photoactivity may be evidenced.

6.2 Deposition of Metal Nanoparticles by Radiolysisand by Photodeposition Method

Noble metal nanoparticles (and in particular plasmonic NPs) have been thesubject of strong interest, because of their catalytic properties [17], and theirability to confine high electromagnetic energy within their small particle sizeowing to the localized surface plasmon (LSP) oscillations of the conduction band(CB) electrons. Gold and silver nanoparticles (AuNPs and AgNPs) have attractedincreasing attention because of their optical, catalytic, and electrocatalyticproperties. Metallization of TiO2 surface with noble metals such as Pt, Ag, andAu has been investigated from the early times of photocatalysis to increasethe photocatalytic activity [1, 11, 18–21]. Different studies have shown thatmetal-doped semiconductor composites exhibit shifts in the Fermi level tomore negative potentials. One important factor that can influence the electronicproperties of the TiO2–metal composite is the size of the metal nanoparticles andthe shift in the Fermi level is size dependent. This shift enhances the efficiencyin the interfacial charge transfer process and improves the energetics of thecomposite system. Different methods have been developed for the synthesis ofmetal nanoparticles on inorganic semiconductors such as TiO2. Among them,the photochemical and radiolytic methods are versatile and powerful methodsto synthesize metal nanoparticles of controlled size, shape and to induce

6.2 Deposition of Metal Nanoparticles by Radiolysis and by Photodeposition Method 131

bimetallic nanoparticles and composite materials [22, 23]. These methods havethe advantage of simple physicochemical conditions (room temperature andabsence of contaminants) and lead to homogeneous reduction and nucleation.

In the case of radiolysis, solvated electrons and reducing radicals are gener-ated by solvent excitation. These reducing species reduce the metal precursors(present in solution), which undergo nucleation and growth. In the photochem-ical approach, the metal precursors (salts or complexes) can be, either directlyexcited by light and then reduced or photochemically generated intermediates,such as excited molecules and radicals, can be used for their reduction [23].

Radiolysis is a powerful method to synthesize nanoparticles of controlled sizeand shape in solution and in heterogeneous media [22]. Solvent radiolysis inducesformation of solvated electrons and radicals, which reduce metal ions homoge-neously in the medium leading to a homogeneous nucleation. Small and relativelymonodisperse nanoparticles can therefore be obtained. Radiolysis presents theadvantage of inducing a homogeneous nucleation and growth in the whole vol-ume of the sample and has been used successfully in order to synthesize variousnoble (such as silver, gold, and platinum) and non-noble (such as nickel, iron, andcobalt) metal nanoparticles in solution or on supports [24].

The primary effects of the interaction of high-energy radiation such as electronor ion beams, X-rays, or gamma photons with a solution of metal ions are theexcitation and the ionization of the solvent. For example, in aqueous solutionsaccording to Eq. (6.1):

H2Oγ-ray−−−−→ e−aq,H3O+

,H∙,HO∙

,H2,H2O2 (6.1)

Solvated electrons e−aq (E0 (H2O/e−aq)=−2.87 VNHE) [25] and alcohol radicals arestrong reducing agents able to reduce metal ions to lower valences and finallyto metal atoms. During the irradiation of deoxygenated water, hydroxyl radi-cals (HO∙), which are very strong oxidative species (E0(HO∙/H2O)=+2.34 VNHEat pH 7) [26], are also formed. To avoid competitive oxidation reactions, whichmay limit or even prevent metal reduction, hydroxyl radical scavengers are addedin solution prior to irradiation. Among these scavengers, primary or secondaryalcohols (such as 2-propanol) molecules or format ions, which also react withhydrogen atoms, are generally used [27]:

CH3CHOHCH3 + OH∙ → CH3C∙OHCH3 + H2O k1 = 1.9 × 109l mol−1 s−1

(6.2)CH3CHOHCH3 + H∙ → CH3C∙OHCH3 + H2 k3 = 7.4 × 107 l mol−1 s−1

(6.3)HCOO− + HO∙ or H∙ → COO∙− + H2O or H2 (6.4)

Due to their redox potentials (E0((CH3)2CO/(CH3)2 C∙OH)=−1.8 VNHEat pH 7 [28] and E0(CO2/COO∙−)=−1.9 VNHE) [29], the radicals formed byreactions (6.3) and (6.4) are almost as powerful reducing agents as H∙ atoms(E0 (H+/H∙)=−2.31 VNHE).

When a solution containing metal ions is in contact with a solid support, theions can diffuse in the pores and can be adsorbed on the surface. Therefore, thepenetration of the ionizing radiation enables in situ reduction of metal ions and


then further coalescence of metal atoms inside the confined volumes of polymericmembranes, mesophases, porous materials, such as zeolites, alumina-silica-gelsor colloidal oxides such as TiO2. Metal nanoparticles can be induced in porousmatrixes (including mesoporous oxides such as SiO2 or TiO2). A preparationprocess of a composite material (metal NPs/porous oxide) consists of impreg-nating a microporous or mesoporous solid material with metal precursors, thenof reducing the impregnated material by radiolysis. Gold nanoparticles weredirectly synthesized by radiolysis on TiO2 for photocatalytic applications [30].

Photocatalytic reduction of metal complexes on semiconductors is also anefficient way to design metal–semiconductor composites [8]. Generally, thephoto-irradiation is carried out for solutions containing metal ions, a semicon-ductor support, and hole scavengers. The photo absorption of semiconductorsgenerate electrons and holes. The metal precursors (ions or complexes) adsorbedon the semiconductor surface can be reduced on the surface by the photogener-ated electrons. This approach has been used to deposit metal nanoparticles onTiO2 to enhance its photocatalytic activity [31, 32] and to extend its absorptionfrom the ultraviolet (UV) to visible range. Gold nanoparticles were photode-posited on different TiO2 surfaces: during the photodeposition process underUV light, methanol was used as a sacrificial hole scavenger resulting in Au/TiO2powders of different colors (violet, pink, and gray) with broad absorptionbands in the wavelength range of ca. 400–700 nm and with a peak maximumat ca. 530–610 nm [19, 33]. Silver nanoparticles can also be easily depositedon TiO2 by photoreduction (using, e.g., benzophenone as photosensitizer)or by photocatalytic deposition (by direct formation on illuminated TiO2surface) [34].

6.3 Electronic Properties Studied Time-ResolvedMicrowave Conductivity

Charge-carrier lifetimes in bare and modified TiO2 after UV illumination werestudied by TRMC method [15, 16].

The TRMC technique is based on the measurement of the relative change ofthe microwave power reflected by a sample (semiconductor), ΔP(t)/P, during itssimultaneous irradiation by a laser pulse. Such relative variation can be correlatedto small perturbation in the sample conductivity, Δ𝜎, as shown in the followingequation:

ΔP(t)P

= AΔ𝜎(t) (6.5)

where A is a time-independent proportionality factor. Because the electronmobility, 𝜇e, in TiO2 is much larger than the hole mobility, Δ𝜎(t) can beattributed to excess electrons:

AΔ𝜎(t) ≈ Δn(t)e𝜇e (6.6)

The signal obtained by this technique displays the evolution of the sample con-ductivity, I(t), (denominated photoconductivity) as a function of time (ns). The

6.3 Electronic Properties Studied Time-Resolved Microwave Conductivity 133

main data provided by TRMC are given by the maximum value of the signal (Imax),which reflects the number of the excess charge carriers created by the laser pulse,and the decay is due to the decrease in the excess electrons (free electrons) [16].To analyze the decay, the signal is divided into two sections: short- and long-rangedecays. The short-range decay, arbitrarily fixed up to 40 ns after the maximum ofthe pulse, is represented by the I40ns/Imax ratio, which reflects the fast processesactive during and just after the pulse. Most probably electron–hole recombina-tion and possibly electron scavenging by metal are responsible for this ratio. Thelong-range decay, here fixed from 200 until 1000 ns, is related to slow processesinvolving trapped species, that is, interfacial charge transfer reactions and decayof excess electrons controlled by the relaxation time of trapped holes. In thisrange, the decay of TRMC signal can be fitted to a power decay according to

I = IDtkD (6.7)

where ID is the intensity of the signal due to charge carriers that recombine after200 ns, and kD is an adimensional parameter related to their lifetime: higher kDvalues correspond to faster decays of the TRMC signal.

The numbers of incident photons in the sample, expressed as nanomole ofphotons also called nano-Einstein (nano-ein), were calculated by the followingequation:

nh𝜈 =E ⋅ 𝜆

h ⋅ c ⋅ NA(6.8)

where E is the excitation energy (J), 𝜆 is the wavelength (nm), h is the Planckconstant (J s), c the speed of light (m s−1) and NA the Avogadro constant (mol−1).Imax/nh𝜈 values for each wavelength were plotted. The obtained graphic is called“TRMC action spectra”.

6.3.1 Surface Modification of Titania with Monometallic Nanoparticles

6.3.1.1 Surface Modification of Titania with Pt ClustersSeveral studies report on visible-light photoactivity of small metal clusters.These clusters formed by a few metal atoms exhibit molecular-like excited-stateproperties with well-defined absorption and emission features [35–38].[Pt3(CO)6]n

2−(n= 3–10) clusters (called Chini clusters) absorb strongly in thevisible domain and their optical and electronic properties can be tuned with size[39]. These clusters can be easily synthesized by radiolysis by reduction of Ptcomplexes in alcohol solution under CO atmosphere. Doping silver halides with[Pt3(CO)6]n

2− (n= 3–10) clusters induces enhancement of the photoconversionyield by inhibition of the electron–hole recombination [40]. Pt clusters/TiO2composites absorb in the visible range due to high absorption of platinum Chiniclusters in this region [21]. [Pt3(CO)6]6

2− are green and present two specificnarrow absorption bands at 430 and 802 nm. TiO2-P25 (P25 is a commercialTiO2 with high activity under visible light; it exhibits a surface area of ca.50 m2 g−1 and consists of a mixture of the crystalline phases anatase (73–85%),rutile (14–17%), and amorphous titania (0–13%) [41] and synthesized by sol–gelmethod) modified with platinum complexes (Pt(II) and Pt(IV) complexes) or


[Pt3(CO)6]62− clusters exhibit higher photocatalytic activity compared to bare

titania. The photocatalytic properties of Pt-modified TiO2 were studied forphenol and rhodamine B (RB) degradation (taken as model pollutants).

Phenol is one of the most employed test molecules. It has been proposed bySerpone et al. [42] as standard test molecule, and presents some advantages:

• It does not undergo degradation by photolysis or catalysis.• It presents an absorption band at 269 nm detectable by UV-Visible

spectroscopy.• Its degradation mechanism is quite identified; the principal intermediates are

benzoquinone, hydroquinone, and catechol [43].• It follows a complete mineralization to CO2 and H2O.• It adsorbs very weakly at the surface of TiO2.• It is a real pollutant of water.

Guo et al. suggested that, in photocatalytic degradation of phenol by TiO2, theattack of ∙OH radicals on phenyl ring is the first stage of photocatalytic processwhich leads to the formation of di- and trihydroxybenzenes and, subsequently,to opening of the phenyl ring and forming maleic acid among other intermediateproducts [44].

Faster degradation of phenol and RB was obtained with Pt-modified TiO2 bothunder UV and visible light ( Figure 6.1). In this work, platinization lowers the sig-nal, but the influence on the decay is variable and can be related to the activityin the case of phenol degradation with UV light. Surface modification by Pt clus-ters slows the I decay, showing a slower charge-carrier recombination, which isbeneficial to the photoactivity. TRMC signals show that under UV irradiationplatinum clusters act as charge scavengers hindering charge-carrier recombina-tion (see Figure 6.2).

Modification of TiO2 with metal nanoparticles or clusters leads to moreefficient electron–hole separation and to enhanced ∙OH and O2

∙− radicalsformation. O2

∙−radicals can be subsequently transformed, via H2O2, into ∙OH

100 100

90

80

70

120

Ph

en

ol a

mo

un

t (%

)

Ph

en

ol a

mo

un

t (%

)80

60

40

20No adsorbateCluster (1%Pt)Salt (1%Pt II)Salt (1%Pt IV)

No adsorbateCluster (1%Pt)Salt (1%Pt II)Salt (1%Pt IV)

0

0 5 10

Time (min)(a) (b) Time (min)

15 20 0 30 60 90

Figure 6.1 Degradation of phenol (2× 10−4 M phenol initial concentration) with pure ormodified titania (1 g l−1 photocatalyst) synthesized by sol–gel method (surface modified andnonmodified with Pt salt (II), PtCl4

2−; Pt salt (IV), PtCl62−; Pt cluster, Pt3(CO)6]6

2−): (a) UnderUV/vis light; (b) under visible light (>450 nm). (Reproduced with permission from Ref. [21].Copyright 2008, American Chemical Society.)


0.001

P25 No adsorbateClusters (1% Pt)Salt (1% Pt II)Salt (1% Pt IV)

TiO2 No adsorbateClusters (1% Pt)Salt (1% Pt II)Salt (1% Pt IV)

0.0001

0.001

0.01

0.1

0.01

0.1I (

mV

)

I (m

V)

t (ns)(a) (b)1.0E + 00 1.0E + 01 1.0E + 02 1.0E + 03 1.0E + 04 1.0E + 05

t (ns)

1.0E + 00 1.0E + 01 1.0E + 02 1.0E + 03 1.0E + 04 1.0E + 05

Figure 6.2 TRMC signals after excitation at 355 nm of pure or modified titania with Pt salt (II),PtCl4

2−; Pt salt (IV), PtCl62−; Pt cluster, [Pt3−(CO)6]6

2−: (a) P25; (b) TiO2 synthesized by sol-geltechnique. (Reproduced with permission from Ref. [21]. Copyright 2008, American ChemicalSociety.)

radicals, which are thought to be the most responsible for photocatalyticdegradation of phenol. ∙OH radicals can be simultaneously generated viadirect oxidation of water molecules by photogenerated holes [45]. Thus, higherefficiency of ∙OH radical generation can lead to higher photoactivity.

6.3.1.2 Surface Modification of TiO2 with Pd NanoparticlesThe surface of four commercial TiO2 compounds (Cristal Global PC-series) hasbeen modified with 3 nm Pd nanoparticles induced by gamma radiolysis [46].Their photocatalytic properties have been studied for phenol and rhodamine Bphotocatalytic degradation in aqueous suspensions under UV and visible light.Their electronic properties have been studied by TRMC method to follow thecharge-carrier dynamics. The experiments evidence a complex behavior of thesurface-Pd. Its influence depends on the pollutant and irradiation. The modifica-tion may be strongly favorable to the photocatalytic activity. The results have beeninterpreted in terms of modification of charge-carrier dynamics with TRMCmeasurements. For phenol photocatalytic degradation, under UV irradiation, thesurface-Pd increases the photocatalytic activity of PC50 and PC10. Those resultscan be directly related to the slowdown of the TRMC decay, proving that thesurface-Pd can help to avoid charge-carrier recombination. For RB photocatalyticdegradation, the surface-Pd always promotes the photocatalytic activity underUV irradiation, especially for PC50, whereas it is without effect under visible irra-diation. The surface-Pd can play a role in charge-carrier separations, leading to animproved photocatalytic activity under UV light. However, the Pd surface modi-fication does not lead to an important change in the absorption properties and toa significant formation of charge carriers at 532 nm to create a real photocatalyticactivity under visible light. Pd NPs play a role in charge-carrier separations,increasing the photocatalytic activity under UV light, but show no effect on theabsorption properties, preventing the creation of an activity under visible light.

In another study, Litter et al. reported heterogeneous photocatalytic reductionof nitrate (which is a pollutant due to human activities, with particular impacton groundwaters and drinking water). (2 mM) in the presence of formic acid


(10 mM) using bare and modified TiO2 samples under UV–vis irradiation(at pH 3) [47]. Commercial samples (Evonik P25 and Cristal Global PC500and PC10) were modified with two noble metal (Ag and Pd) nanoparticles(induced by radiolysis). P25 was modified with Ag (0.5 and 2% w/w), while PC10and PC500 were modified with Pd (1% w/w). The order of the photocatalyticactivity of the materials for NO3

− transformation was 2 Ag-P25>PC500> 0.5Ag-P25≈P25≫ 1 Pd-PC500>PC10> 1 Pd-PC10. Nitrite formation wasobserved in all cases, but at low amounts, and its concentration was negligibleafter complete NO3

− reduction. Ammonium was found as final product andremained in considerable amounts at the end of irradiation. The nitrogen balanceaccounted for a large amount of non-identified nitrogen products formed duringthe photocatalytic reaction, probably N2 or NO; this amount was higher for theP25 and PC500 bare samples. The efficiency on the use of formic acid as donorwas evaluated and PC500 was found to be the most efficient sample in this sense.Radiolytic modification of TiO2 with noble metal nanoparticles such as Ag or Pddoes not always increase the photocatalytic efficiency of NO3

− reduction. Thereasons for this differential behavior are related to the inherent mechanism ofnitrate degradation and the possible side reactions that can be involved, such asH2 generation. The modification of PC samples with Pd NPs deteriorates NO3

−

transformation and decreases significantly the efficiency in the use of the donor,probably because of the competence of H2 evolution.

6.3.1.3 Modification of TiO2 with Ag NanoparticlesSilver nanoparticles attract a lot of interest because of localized surface plas-mon resonance (LSPR)[48–50], their size- and shape-dependent optical prop-erties [51, 52], their catalytic activity, [53–55], and their potential applicationsin chemical and biological sensing based on surface-enhanced Raman scattering(SERS)[56, 57] and metal-enhanced fluorescence (MEF)[58, 59]. TiO2 modifiedwith silver nanoparticles exhibit enhanced photocatalytic activity under UV andvisible light and improved antibacterial properties [60–62]. Ag NPs show a veryintense LSPR absorption band in the near-UV region [63] and this is associatedwith a considerable enhancement of the electric near field in the vicinity of theAg NPs. This enhanced near field can boost the excitation of electron–hole pairsin TiO2 and therefore increases the photocatalytic activity.

Surface of commercial TiO2 compounds (P25 and ST-01) has been modifiedwith Ag nanoparticles induced by radiolysis [64]: An alcohol suspension of TiO2containing Ag+ ions was irradiated (by gamma rays or electron beams) under N2atmosphere. Silver ions are reduced by solvated electrons:

Ag+ + e−s → Ag0 (6.9)Reduction of free silver ions by alcohol radicals proceeds via the formation of

a complex involving metal ions and the alcohol radicals, which act as ligands. Ithas been observed in case of 2-propanol [65]:

Ag+ + (CH3)2C∙OH → [Ag(CH3)2C∙OH]+ (6.10)[Ag(CH3)2C∙OH]+ + Ag+ → Ag+2 + (CH3)2CO∙ + H+ (6.11)

The same reactions probably occur also with methanol radical. However, directreduction by ∙CH2OH of silver cations adsorbed on silver clusters or on TiO2 is


also possible (the reduction potentials of Agn+ and Ag+/TiO2 are more positive

than the one of Ag+):∙CH2OH + Ag+∕TiO2 → HCHO + Ag0∕TiO2 + H+ (6.12)∙CH2OH + Ag+n → HCHO + Agn + H+ (6.13)

Very small Ag nanoparticles of 1–2 nm were synthesized by radiolysis onTiO2 P25 (by irradiation of a suspension of TiO2 in an alcohol solution contain-ing Ag+ ions), while on TiO2TS-01, two populations of Ag nanoparticles wereobtained, small nanoparticles (1–2 nm) and larger ones (mean diameter 7–12 nmdepending on the silver loading from 0.5% to 2% in mass) [64]. Modification ofTiO2-P25 with Ag clusters induced by radiolysis leads to a wide absorption of thephotocatalysts in the visible with two maxima at 410 nm and 540–560 nm (seeFigure 6.3a). In the case of Ag-modified ST01, a wide absorption is also observedwith a maximum at 410 nm (see Figure 6.3b). Silver nanoparticles exhibit a plas-mon band with a maximum at around 400 nm in water and this plasmon band issensitive to the environment and can be shifted depending on the stabilizer or onthe substrate. This plasmon band is blue-shifted when these NPs are supportedon titania because of the coupling between the Ag NPs and TiO2 having a highreflective index (the absorption coefficient and refractive index are for anatasephase 90 cm−1 and 2.19 at a wavelength of 380 nm, respectively) [66, 67]. Surfacemodification with silver nanoparticles induced a modification of the absorptionproperties of the photocatalysts inducing an activity under visible light. The mod-ified TiO2 samples absorb in the 540–560 nm region, and this absorption has beenattributed to small (1–2 nm) Ag clusters [68]. The diffusion reflectance spectra ofthe modified samples show a slight shift in the bandgap transition to longer wave-lengths. The red shift in the bandgap transition revealed by diffuse reflectancespectra can be related to the electronic interaction between metal NPs and TiO2.

The photocatalytic activity of Ag-modified TiO2 (P25 and ST01) is enhancedboth under UV and visible light for phenol degradation. Faster decay of TRMCsignals was obtained with Ag–TiO2 compared to bare titania (see Figure 6.4).TRMC measurements have shown that TiO2 modification with Ag nanoparti-cles plays a role in charge-carrier separations increasing the activity under UVlight and that Ag NPs act as electron scavengers. Titania modification with Ag

300

Ab

so

rba

nce

400

Wavelength (nm)(a) (b)

500

P25

ST-01

Ag-ST01(0.5)

Ag-ST01(2)Ag-ST01(1)Ag-P25(1)

Ag-P25(0.5)

Ag-P25(2)

600 300

Ab

so

rba

nce

400

Wavelength (nm)

500 600

Figure 6.3 Diffuse reflectance spectra of pure and modified TiO2: (a) P25 and (b) ST-01 withdifferent silver loading (0.5–2% in mass) and recorded, respectively, using BaSO4 as reference.(Reproduced with permission from Ref. [64]. Copyright 2013, American Chemical Society.)


20 nm

100

I (m

V) 10

1

1 10

UV

hν > 3.2 eV

AgAgCB

VB

Ag-P25(1)

P25

100

Time (ns)

1000 10000

Figure 6.4 TEM image showing Ag clusters induced by radiolysis on P25 and TRMC signals ofbare and modified with Ag clusters P25 obtained with excitation at 355 nm. Inset: a schemeshowing electron scavenging by Ag clusters decreasing the charge-carrier recombination.(Reproduced with permission from Ref. [64]. Copyright 2013, American Chemical Society.)

accelerates the overall decay of the TRMC signals. The modification of TiO2 withAg nanoparticles causes an increase in the photocatalytic activity. The TRMC sig-nal is mainly related to the electron mobility. The decrease in the TRMC signalsis due to efficient electron scavenging by silver nanoparticles deposited on TiO2.It implies a decrease of the charge-carrier recombination, which is beneficial tothe photoactivity.

Another study also shows that deposition of Ag nanoparticles on titaniaincreases the affinity of the surface to oxygen [69], and this affinity to oxygencan have an influence on the photocatalytic activity. Silver modification of TiO2results also in enhancement of the bactericidal activity of TiO2 under UV lightbecause of the improved microorganism adsorption to the particle surfaceand lower electron–hole recombination [60, 70]. Kowalska et al. reported inanother study on Ag/TiO2 system, action spectra (AS) analysis proving that thephotocatalytic activity under visible-light irradiation is due to LSPR of silverNPs [71]. This composite Ag/TiO2 system showed also antimicrobial propertiesunder visible-light irradiation indicating that not only intrinsic properties ofsilver in the dark but also plasmonic properties of Ag/TiO2 were responsiblefor bacteria killing. The evolution of carbon dioxide indicated mineralization ofbacteria cells, and therefore possible application of silver-modified titania fordecomposition of chemical and biological pollutants.

It has also to be mentioned that modification of titania with silver clusters ofa few atoms such as Ag8 and Au25 induces a photocatalytic activity under visiblelight. These clusters are photochemically reactive and can act as electron donorsunder visible excitation due to their molecular-like properties [21, 39, 40]

6.4 Modification of TiO2 with Au Nanoparticles

In the plasmonic photocatalytic Au/TiO2 system, the presence of Au is essential,due to their LSPR, that is, the oscillation of metal-free electrons in constructiveinterference with the electric field of the incident light. These plasmonic

6.4 Modification of TiO2 with Au Nanoparticles 139

properties can be used to induce photocatalytic activity of the semiconductormaterial under visible light [19, 72–78]. However, the support presents alsoimportant effects on the photocatalytic processes, that is, (i) the electron–holerecombination for anatase and rutile particles has been studied and was foundto be higher for rutile than anatase [16, 41, 79, 80]; (ii) the higher Fermi level ofanatase results in stronger electronic interaction with Au-NPs, which can inhibitthe growth and aggregation of metal nanoparticles [81]; (iii) the difference inenergy band gap of anatase (3.2 eV) and rutile (3.0 eV)[2] could result in activa-tion of TiO2 at longer wavelengths for rutile [82]; and finally, (iv) the dielectricconstant of the support could shift the LSPR of metal nanoparticles toward thered spectral region [11, 83]. For example, it has been shown that crystallinecomposition and surface properties of Au/TiO2 photocatalysts, prepared byphotodeposition of gold on 15 commercial titania, influenced significantlythe resultant photocatalytic activities in a different manner under UV andvisible-light irradiation [19]. Under UV irradiation, the gold presence resultedin a high enhancement of the photocatalytic decomposition of acetic acid (oneto threefold) for all the modified samples. By contrast, under visible-light irra-diation, mainly Au/TiO2 samples possessing large crystallites of rutile exhibitedthe highest level of photocatalytic activity. It was proposed that polydispersityof gold deposits, that is, various sizes and shapes (nanoparticles, nanorods),resulted in broad LSPR, and therefore in higher overall photocatalytic activitythan that of Au/TiO2 of fine gold NPs with narrow LSPR.

Small gold nanoparticles (Au-NPs) around 2–3 nm on the surface of tita-nium dioxide work as visible-light absorbers and thermal redox active centers.Au-NPs were synthesized on commercial TiO2 (P25) by reduction withtetrakis(hydroxymethyl)-phosphonium chloride [78]. The optical propertiesof the modified surface of TiO2-P25 were studied by diffuse reflectance spec-troscopy (DRS). The spectrum of TiO2-P25 shows an absorption edge at around400 nm due to the presence of rutile [82]. The photoabsorption properties ofAu/P25 materials are higher than that of pure TiO2-P25, since the Au-NPs inducea shift of the absorbance toward the visible light attributed to the interactionbetween the metal and the semiconductor, that is, the so-called Schottky barrier[74] and because of the plasmon of gold. The position of the LSPR peak ofthe Au-NPs at 520 nm for 13 nm gold NPs in aqueous solution is sensitive tothe dielectric constant of the surrounding medium [11, 83]. This shift in theplasmon is due to the interaction between the Au-NPs and the semiconductorTiO2-P25. Indeed, this plasmon band is sensitive to the size and environment,and can be shifted depending on the stabilizer or the substrate. Because of thecoupling between Au nanoparticles and TiO2 support, the plasmon band in caseof modified titania is usually redshifted [19, 33, 84, 85]. A weak LSPR band from500 to 650 nm with maximum values at 548, 554, and 560 nm for Au 0.5 wt%,Au 1 wt%, and Au 2 wt%, respectively, is observed due to the plasmon of smallnanoclusters (Figure 6.5) [84, 86]. These absorptions result in a pink-purplecolor of the modified TiO2-P25 samples.

The modification of TiO2-P25, with very small Au nanoparticles (<5 nm) local-ized on the anatase phase and with a low metal loading, induced increase in thephotocatalytic activity under UV and visible light [78].


2.0

1.5

1.0

0.5

0.0

300 400 500

548 nm

554 nm563 nm

Wavelength (nm)

600

P25

Au0.5%/P25

Au1%/P25

Au2%/P25

700

(c)

Ab

so

rba

nce

(a)

10 nm

10 nm

(b)

40353025

Fre

q (

%)

Au

0.5

%/P

25

Au

%1 /P

25201510

50

30

25

Fre

q (

%)

20

15

10

5

0

Size (nm)

1.0 1.5 2.0 2.5 3.0 3.5

Size (nm)

1.0 1.5 2.0 2.5 3.0 3.5

Figure 6.5 TEM images of (a) Au0.5%/P25; (b) Au1%/P25, with the corresponding histogram ofthe size distribution of Au-NPs. (c) Diffuse reflectance spectroscopy (DRS) spectra of Au/P25samples, where the LSPR of Au-NPs is observed between marked lines. (Reproduced withpermission from Ref. [78]. Copyright 2016, American Chemical Society.)

Charge-carrier dynamics in Au–TiO2 system have been studied by TRMC.Figure 6.6 showed the electrons are trapped by Au NPs under UV and theinjection of electrons from Au-NPs into the CB of TiO2 under visible-lightexcitation, as a result of the activation of LSPR of Au-NPs [78].


0.0

(a)

(b)

(c)

0.02

1

0.04

Norm

aliz

ed I

(mV

)

0.06Time (μs)

0.08 0.10

100

10

10.2 0.4

Time (μs)

λ = 365 nm λ = 410 nm

λ = 450 nm

λ = 500 nm

λ = 470 nm

λ = 560 nm

I (m

V)

100

10

1

I (m

V)

100

10

1

I (m

V)

0.6

P25Au0.5%/P25

Au1%/P25

Au2%/P25

P25

Au0.5%/P25

Au1%/P25

Au2%/P25

P25

Au0.5%/P25

Au1%/P25

Au2%/P25

P25

Au0.5%/P25

Au1%/P25

Au2%/P25

P25

Au0.5%/P25

Au1%/P25

Au2%/P25

P25

Au0.5%/P25

Au1%/P25

Au2%/P25

0.8 1.0 0.0 0.2 0.4

Time (μs)

0.6 0.8 1.0

0.0 0.2 0.4

Time (μs)

0.6 0.8 1.0 0.0 0.2 0.4

Time (μs)

0.6 0.8 1.0

0.0 0.2 0.4

Time (μs)

0.6 0.8 1.0 0.0 0.2 0.4

Time (μs)

0.6 0.8 1.0

100

10

1

I (m

V)

100

10

1

I (m

V)

100

10

1

I (m

V)

Figure 6.6 TRMC signals of pristine and modified TiO2-P25 at different excitationwavelengths: (a) 365 and 400 nm UV irradiation, (b) 450 and 470 nm visible irradiation, and(c) 500 and 560 nm (Plasmon excitation). (Reproduced with permission from Ref. [78].Copyright 2016, American Chemical Society.)

The photocatalytic activity of Au/TiO2 was evaluated for degradation of phenol,2-propanol, acetic acid, and for H2 production from aqueous methanol solution[78]. This photocatalytic activity is much higher for the plasmonic TiO2 com-pared to the activity of bare titania. The stability of these plasmonic photocat-alysts was also investigated, and the results showed that the Au/TiO2 systemis a stable photocatalysts and can be reused several times, without appreciablechange in structure, activity, or composition [78]. For this system, the apparentquantum efficiency (AQE) was obtained using action spectra (AS) calculated asthe rate of CO2 evolution from the decomposition of acetic acid versus the fluxof incident photons, assuming that four photons were required. Action spectra(quantum yield per unit of incident photons as a function of the wavelength) cor-relate with the absorption spectra (DRS spectra), Figure 6.7, for pure TiO2-P25and modified TiO2-P25 with Au 0.5 wt%, and an appreciable response is obtained


1.0

0.25

0.20

0.15

0.10

Φapp (%

)Φ

app (%

)

0.05

0.00

0.06

0.05

0.04

0.03

0.02

0.01

0.00

0.8

0.6

0.4

10

0-r

efle

cta

nce

(%

)

0.8

0.7

10

0-r

efle

cta

nce

(%

)

0.2

0.0

300

450 500 550 600 650

400 500

Wavelength (nm)

Wavelength (nm)

(a)

(b)

600

Au0.5%/P25

Au0.5%/P25

P25

ΦAu0.5%/P25

ΦAu0.5%/P25

ΦP25

700

Figure 6.7 (a) Comparison between DRS spectra and the action spectra of TiO2-P25 andmodified TiO2-P25 with Au 0.5 wt%, (b) superposition of action spectrum (in blue) andabsorption spectrum (in red) in the Au plasmon range. (Reproduced with permission fromRef. [78]. Copyright 2016, American Chemical Society.)

under UV and visible light, confirming that the decomposition of acetic acidoccurs by a photocatalytic mechanism and much lower quantum yield undervisible light than under UV irradiation for plasmonic photocatalysts has beenusually observed and reported. Under visible-range excitation, the samples mod-ified with Au-NPs show a higher apparent quantum efficiency (Φapp) in the rangebetween 450 and 650 nm, which is related to LSPR [78].

The modified TiO2-P25 was also found to provide promising results forhydrogen generation under UV and visible light [78]. In the photocatalyticwater splitting (PWS) process by photocatalysis, electrons in the CB reduceprotons into H2 and holes in the valence band oxidize water to O2, but the maindrawback of the PWS process by photocatalysis is low H2 production due tothe fast recombination of the charge-carriers [87–89]. For H2 production testsunder 𝜆= 400 and 470 nm, the addition of Au-NPs activates the titania support[78]. Figure 6.8 shows that a considerable higher amount of H2 is produced


22 0.30

0.25

0.20

0.15

0.10

0.05

0.00

(a) (b)

P25

P25

Au

0.5

%/P

25

Au

0.5

%/P

25

Au

1%

/P25

Au

1%

/P25

Au

2%

/P25

Au

2%

/P25

0.40

17.5 0.23

0.13

0.02

12.2

8.7

Hydro

gen (μ

mol g

–1s

–1)

Hydro

gen (μ

mol g

–1s

–1)

λ = 400 nm λ = 470 nm20181614121086420–

Figure 6.8 Production of H2 under visible light at: (a) 𝜆= 400 nm and (b) 𝜆= 470 nm for pureTiO2-P25 and TiO2-P25 modified with Au-NPs at different loadings. The relative uncertainties at400 and 470 nm are 5% and 7%, respectively. (Reproduced with permission from Ref. [78].Copyright 2016, American Chemical Society.)

with Au-P25 samples, compared to the amount produced with bare-P25.Under excitation at 400 nm (see Figure 6.8a), the activity for H2 productiondecreases with the Au-loading. The excess of Au-NPs on TiO2-P25 can act asrecombination centers of photogenerated charges reducing the photocatalyticactivity in H2 production. Under visible irradiation at 470 nm (Figure 6.8b), onlya small amount of H2 is obtained (0.23 μmol g−1 s−1) with the Au-P25 samples.Similar results have been reported by Joo et al. in the range of 300–600 nmusing methanol as holes scavenger (∼4 μmol g−1 s−1 using Au/P25 and around0.2 μmol g−1 s−1 using Au@SiO2(Thin)/P25)[89]. This activity does not dependon the Au loadings (Au 0.5, 1 and 2 wt%) [78].

Small gold nanoparticles can induce a much higher change of the work func-tion compared to larger particles, and this change is an indication of a largercharge separation and improved reduction potential for the photocatalyst andmetal/TiO2 interface.

The action spectra show that the photocatalytic activity under visible lightis directly correlated to the LSPR, and TRMC signals show that Au-NPs injectelectrons in the CB of the semiconductor due to their surface plasmon resonanceinducing an activity in the visible region [78]. When the samples were irradiatedat 470, 500, and 550 nm, a small TRMC signal is observed for the Au-P25samples, attesting the generation of free electrons in the CB of the Au-modifiedTiO2-P25. This suggests that excess electrons are injected in the CB of TiO2-P25after excitation of the Au nanoparticles at a wavelength very close (or equal) totheir LSPR. This electron injection from excited Au-NPs to the CB of TiO2 wasdemonstrated for the first time by TRMC. Direct proof for electron transfer byelectron paramagnetic resonance (EPR) technique has been reported by Carettiet al. and Priebe et al. [90, 91]. Both trapped electrons (Ti+3) and holes (O−)were observed under UV excitation, but under visible excitation only trappedelectrons were observed, confirming that the electron transfer from the metalnanoparticles to the semiconductor was the main mechanism of plasmonictitania activation [90]. Priebe et al. reported EPR results proving electron transferfrom Au-NPs and their trapping by lattice Ti+4 or surface oxygen vacancies [91].


CB

VB

(a)

(b) (c)

TiO2

UV

Schottkybarrier

e–

AA–

A A–

e–

h+

h+

D+

D

h+

h+

D+

D

e–

CB

VB

TiO2 Schottkybarrier

e–

A

Visible Visible

A–A

A–

e–

e–

h–

D+

D

e–

CB

VB

TiO2 Schottkybarrier

e–

e–

AuAu

Au

Figure 6.9 Mechanism purposed for modified TiO2-P25 modified by Au-NPs (a) UV and(b) visible irradiation by electron, and (c) energy transfer. (Reproduced with permission fromRef. [78]. Copyright 2016, American Chemical Society.)

The electrons are probably injected from the metal NPs to the CB of TiO2 byelectron transfer or by energy transfer on account of the LSPR of Au-NPs, andthis corresponds to a higher photocatalytic activity; see Figure 6.9 [78].

In another study, TiO2 was surface modified with silver, gold, or platinumclusters to improve its photocatalytic activity. The effect of metal content, kind ofdopant, and titanium dioxide source (commercial – P25 and ST-01) used duringpreparation procedure on photoactivity were investigated. The photocatalyticactivity was estimated by measuring the decomposition rate of 0.21 mM phenolaqueous solution under UV-vis and visible (𝜆> 400 nm). The highest photoac-tivity was observed for TiO2 loaded with silver (2% Ag on P25), gold (1% Au onP25), and platinum (0.5% Pt on ST-01) clusters. After 60 min of irradiation underUV light, phenol solution was degraded in 91%, 49%, and 91%, respectively [30] .

6.5 Modification of TiO2 with Bi Clusters

Development of cheap and efficient TiO2-based photocatalysts without noblemetals is a challenge. In this context, small (subnanometer) Bi zero-valent clusters

6.5 Modification of TiO2 with Bi Clusters 145

were synthesized on TiO2-P25 by radiolysis for application in photocatalysis[92]. Surface modification of TiO2 with zero-valent Bi nanocluster-induced highphotocatalytic activity under visible light for rhodamine B and phenol degrada-tion. Very small amounts of Bi (0.5 wt%) can activate titania for photocatalyticapplications under visible light, but the photocatalytic activity under UV lightwith bare and modified TiO2 was very similar. These photocatalysts are shownto be very stable with cycling.

The intensity of the TRMC signals of Bi-modified TiO2 under UV illuminationwas higher compared to the one obtained with bare TiO2 (Figure 6.10), provingthat more electrons are induced in the CB of Bi-modified TiO2. This indicatesthat Bi clusters inject electrons in the CB of TiO2. Bi clusters do not have anyinfluence on the TRMC signal decay. More importantly, TRMC measurementsindicated that under visible irradiation Bi nanoclusters inject electrons into theCB of TiO2 (Figure 6.10). These electrons react with O2, forming oxidizing O2

∙−.These oxidizing radicals are responsible for photocatalytic degradation of phenol(taken as model pollutant) under visible-light irradiation.

100

TR

MC

-sig

nal (m

V)

TR

MC

-sig

nal (m

V)

10

1

0.1

P25

P25

0.1-Bi/TiO2

0.1-Bi/TiO2

0.3-Bi/TiO2

0.3-Bi/TiO2

0.5-Bi/TiO2

0.5-Bi/TiO2

1-Bi/TiO2

1-Bi/TiO2 2-Bi/TiO2

2-Bi/TiO2

100

10

1

0.1

BiCB

e–

e–

e–

VBTiO2

0.1 1

0.1

t (μs)(a)

(b) t (μs)

1

Figure 6.10 TRMC signal of bare and Bi-modified samples obtained by irradiation at 355 nm(a) and 450 nm (b). Inset: A scheme showing electron injection from Bi nanoclusters in theconduction band of TiO2 under visible-light excitation. (Reproduced with permission fromRef. [92]. Copyright 2015, RSC.)


6.6 Surface Modification of TiO2 with BimetallicNanoparticles

6.6.1 Surface Modification with Au–Cu Nanoparticles

Bimetallic nanomaterials exhibit unique catalytic, electrocatalytic, electronic,and magnetic properties, which differ from their monometallic counterparts[93, 94]. In particular, bimetallic nanoparticles often show enhanced catalyticperformances in terms of activity, selectivity, and stability, compared to separatecomponents [95].

Au–Cu nanoalloys of homogeneous size were synthesized on TiO2 (P25)(0.5 wt%) via deposition precipitation method with urea (DPU) followed byradiolytic reduction [96]. This deposition procedure ensured a complete adsorp-tion of Au and Cu ions on TiO2. The alloyed structure of Au–Cu NPs wasconfirmed by high-angle annular dark-field scanning transmission electronmicroscopy (HAADF–STEM), energy-dispersive X-ray (EDX) mapping, X-rayphotoelectron spectroscopy (XPS), and DRS. Because of the plasmon of gold andcopper, the modified titania absorb in the visible spectral range. Modificationwith Au–Cu bimetallic nanoparticles induced an enhancement in the photo-catalytic activity under UV irradiation for photodegradation of methyl orange(MO). The highest photocatalytic activity was obtained with Au–Cu/TiO2(atomic ratio Au:Cu= 1 : 3). Modification of TiO2 with Cu and Au–Cu bimetallicNPs resulted in a decrease in the photoluminescence (PL) emission intensity,indicating less electron–hole recombination rates. Modification of TiO2 withAu–Cu nanoparticles induced a better charge-carrier separation because theNPs act as a sink for electrons and, consequently, leads to an enhancement ofthe photocatalytic activity under UV light.

In another study, Au, Cu, and bimetallic Au–Cu nanoalloys were synthe-sized on the surface of commercial TiO2 compounds (P25) by reduction ofmetal precursors with tetrakis(hydroxymethyl) phosphonium chloride (THPC)(0.5% in weight) [93]. The Au–Cu nanoalloys on TiO2 were characterized byHAADF–STEM, EDX mapping (Figure 6.11), HRTEM, and XPS techniques.Modification with Au–Cu nanoparticles induced an increase in the photocat-alytic activity for phenol and RB photodegradation in aqueous suspensionsunder UV–vis irradiation. The highest photocatalytic activity was obtained withAu–Cu/TiO2 (with the atomic ratio Au:Cu 1 : 3) (see Figure 6.12).

TRMC measurements showed that Au, Cu, and Au–Cu nanoparticles actas a sink for electrons, decreasing the charge-carrier recombination: Indeed,the overall decay of the TRMC signal was accelerated for Au–Cu-modifiedTiO2 compounds (see Figure 6.13). Importantly, TRMC measurements alsoshowed that the bimetallic Au–Cu nanoparticles were more efficient in electronscavenging than the monometallic Au and Cu nanoparticles. This influence oncharge-carrier recombination can be related to the photocatalytic activity underUV light.

This acceleration in the decay has also been observed for the modification ofTiO2 with Ag and Au clusters. It is different from the previous observations madewith Pt- [21] and Pd- [46] modified TiO2, where a slowdown of the overall decay

6.6 Surface Modification of TiO2 with Bimetallic Nanoparticles 147

(a)

(b)

Figure 6.11 (a) Energy-dispersive X-ray spectroscopy line scans across a nanoparticle ofAu–Cu1 : 1/P25 (the profile was taken along the green line, the blue line corresponds to Cu–K,and the red one to the Au–L signal) and corresponding STEM images for the samples.(b) Mapping EDS analysis performed on a nanoparticle of Au–Cu1:1/P25 (left). (Reproducedwith permission from Ref. [93]. Copyright 2013, RSC.)

14

Ka

pp (

10

–3 s

–1)

12

P25 Au/P25

Au–Cu1 : 1/P25

Au–Cu1 : 3/P25

Cu/P25

10

8

6

4

2

0

Figure 6.12 Rate constants of the first-order kinetics of phenol photodegradation by pure andmodified TiO2 photocatalysts under UV–visible illumination. (Reproduced with permissionfrom Ref. [93]. Copyright 2013, RSC.)

was observed. Indeed, contrary to metals such as Pt and Pd, which provide anohmic contact, metals such as Ag, Au, and Cu exhibit capacitive properties.

Furthermore, the modification with copper nanoparticles increased the initialTRMC signal intensity in the case of Cu-P25 (Figure 6.13) [93]. This indicated thatmore electrons were produced under UV illumination in the CB of Cu-modifiedP25. These excess electrons are due to the injection of electrons in the CB of TiO2after excitation of copper nanoparticles, which are more easily oxidized.


0.1P25 AuCu1 : 1/P25

AuCu1 : 3/P25

Cu/P25

Au/P25

0.01

I (m

V)

Time (s)

1E – 3

>3.2 eV

TiO2

Au-Cu CB

VB

e–

h+

–

–

+

Ef

hv

1E – 4

1E – 8 1E – 7 1E – 6

Figure 6.13 Time-resolved microwave conductivity signals of modified P25 photocatalystsprepared by the chemical method with THPC. Inset: scheme depicting the electronscavenging and transfer on the Au–Cu-modified TiO2 surface after the absorption of UVphotons. (Reproduced with permission from Ref. [93]. Copyright 2013, RSC.)

6.6.2 Surface Modification with Ag and CuO Nanoparticles

Ag and CuO nanoparticles (NPs) were synthesized on the surface of commercialTiO2-P25 by radiolytic reduction (the loading of Cu or Ag metal was 0.5 wt%)[48]. These nanoparticles were characterized by HAADF-EDS, HRTEM, XPS,and X-ray absorption near-edge spectroscopy (XANES). In the case of modifica-tion with silver and copper, Ag@CuO nanoparticles (large silver cores decoratedwith small clusters of CuO) were obtained on TiO2-P25 (Figure 6.14).

The photocatalytic properties of bare and modified TiO2 were studied for phe-nol degradation and for acetic acid oxidation under UV and visible irradiation.Modification with Ag nanoparticles or CuO nanoclusters induced an increase in

CuO

(a) (b)

Ag

TiO2

Schottky

barrier

5 nm

Figure 6.14 (a) Representative aberration corrected STEM-HAADF image for Ag@CuO1:1/P25sample, and (b) A schematic morphology of the modified TiO2–P25 with Ag–CuOnanoparticles. (Reproduced with permission from Ref. [48]. Copyright 2016, AmericanChemical Society.)


1.0 1.00

0.95

0.90

0.85

0.80

0.75

0.8

P25CuO/P25Ag@CuO1 : 1/P25Ag/P25

P25CuO/P25Ag@CuO(1 : 1)/P25Ag/P25

0.6

0.4

Phenol concentr

ation (

C/C

o)

Phenol concentr

ation (

C/C

o)

0.2

0.00 5 10

Time (min) Time (min)

15 20 0 50 100 150 200 250

(a) (b)

Figure 6.15 Degradation curves of phenol under (a) UV and (b) visible light (𝜆> 450 nm) ofpure system TiO2-P25 and modified systems with, Ag, Ag@CuO1:1and CuO. (Reproduced withpermission from Ref. [48]. Copyright 2016, American Chemical Society.)

the photocatalytic activity under both UV and visible light. The photocatalyticactivity of Ag@CuO/TiO2 was higher under UV light, but lower under visiblelight compared to the activity of CuO/TiO2 and Ag/TiO2 (see Figure 6.15). TRMCmeasurements showed that surface modification of TiO2 with Ag, CuO, andAg@CuO nanoparticles played a role in charge-carrier separation, increasing theactivity under UV light, and that Ag@CuO NPs were more efficient electron scav-engers than Ag NPs and CuO nanoclusters. The LSPR of Ag NPs and the narrowband gap of CuO induced an activity under visible light. The TRMC signal showedalso responses under visible-light irradiation at different fixed wavelengths indi-cating that electrons are injected from Ag NPs in the CB of TiO2. Under visiblelight, the photocatalytic activity of CuO/P25 was higher than that of plasmonicAg/P25. The study showed that CuO clusters can activate TiO2 in a widerrange of wavelengths under visible-light irradiation, compared to the activationobtained with silver modification. Action spectra correlated with the absorptionspectra for irradiation wavelengths in the range of 350–470 nm proving thatdecomposition of acetic acid was carried out by a photocatalytic mechanism [48].

Another study reports on modification of TiO2 nanotubes with Cu,Agcore/Cushell, and Bi nanoparticles induced by radiolysis (Figure 6.16) [97].Here again, surface modification with metal nanoparticles led to enhanced pho-tocatalytic activity under UV–vis irradiation because of the electron trappingby the NPs, and this effect depended on the amount of deposited metal. Modi-fication of TiO2 with metal nanoparticles leads to more efficient electron–holeseparation and to enhanced ∙OH and O2

∙− radicals formation. The mechanism ofphenol degradation on titania nanotubes is shown in Figure 6.17. The photocat-alytic activity (for phenol degradation) of TiO2 nanotubes modified with AgCunanoparticles was higher compared to monometallic samples modified with thesame amount of Cu. The photoelectrochemical experiments performed under theinfluence of simulated solar light irradiation also confirmed the enhanced pho-toactivity of metal-modified nanotubes. The saturated photocurrent for the mostactive Bi- and AgCu-modified samples was over two times higher than for bareTiO2 nanotubes. The modified TiO2 nanotubes were resistant toward photocor-rosion, and this enables their application for long-term photoinduced processes.


(d) (e)

(b)(a) (c)

Figure 6.16 SEM images of pure TiO2 nanotubes (a–c) and AgCu-NT III sample (d); STEMimage of AgCu-NT III sample (e). (Reproduced with permission from Ref. [97]. Copyright 2016,Elsevier.)

OH

OH

Cu, AgCu or Bi

nanoparticles CB

VB

TiO

2 3

.2 e

V

hυ

e–

h+h+

e–

UV-irradiation

O2

O2–

OHHO

HO·

H+ + HO·

H2O2

H2O

Ring opening

and further

oxidation

H2O + CO2

Figure 6.17 Proposed mechanism of phenol decomposition in the presence of TiO2nanotubes decorated with metal nanoparticles under UV–vis irradiation. (Reproduced withpermission from Ref. [97]. Copyright 2016, Elsevier.)

6.6.3 Comodification of TiO2 with Ni and Au Nanoparticlesfor Hydrogen Production

Au and/or Ni nanoparticles were synthesized by radiolysis on TiO2 (commercialP25) at various compositions (metal content) [98]. The modified photocatalystswere characterized by high-resolution transmission microscopy (HRTEM),energy-dispersive X-ray spectroscopy (EDS), UV–vis DRS, and XPS. Thecharge-carrier mobility was studied by TRMC. The photocatalytic activitieswere tested for H2 production under UV–vis irradiation using polychromaticand monochromatic light (action spectrum analysis of AQE).

According to the characterization results, a segregation of two metals wasobserved. Large Au NPs and Ni nanoclusters (partially oxidized) were obtainedon TiO2. The surface modification of TiO2 with Ni and Au NPs resulted inan increase of the photocatalytic activity for hydrogen production using a


methanol–water solution under UV light. The highest production of hydrogenwas obtained with the NiAu/TiO2 catalysts, which was explained in terms of asynergetic effect by the presence on Au NPs and Ni(O) clusters on TiO2, actingas recombination sites for atomic hydrogen conversion to molecular hydrogen.

It was found that a very small amount of gold associated to nickel (atomic ratioNi:Au 5 : 1 and total metal 0.5–1 at%) can induce a significant increase in H2 for-mation; thus, the costs of photocatalyst preparation are relatively low.

Figure 6.18 displays the AQE determined by the action spectra for each sample.The absorbance obtained by DRS and the Imax/photons obtained by TRMC havealso been plotted to follow the evolution with the wavelength of the three steps ofthe photocatalytic mechanism: photon absorption, charge-carrier creation, andchemical surface reaction.

It can be observed that the AQE of bare TiO2 is very weak. The action spec-trum shows that the maximum amount of hydrogen is obtained at a wavelength of350± 5 nm. It suggests that the highest density of electrons in the CB is obtainedat this energy. This agrees with the TRMC results, where the highest photocon-ductivity was obtained under irradiation at 355± 5 nm.

A detailed analysis of AQE spectra of the three modified compounds suggestsappreciable differences among them; the action spectrum of Au/TiO2 shows alow level and a maximum at 380 nm, while the action spectra of compoundscontaining Ni present higher levels and follow the absorption spectra. Ni/TiO2and NiAu/TiO2 samples show similar profiles, but an enhancement of the AQEis clearly shown for NiAu/TiO2. Considering the small amount of gold, theenhancement in H2 production cannot be explained only by an additional effectof gold, but by a synergistic effect of gold with nickel.

1.0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

AQ

E (%

) or Im

ax /p

hoto

ns

0.2

0.0

300 350

DRS

DRS

DRSDRS

TRMC

TRMC

TRMCTRMC

TiO2 Au/TiO2

Ni/TiO2 NiAu/TiO2

AQE AQE

AQE

AQE

400

Wavelength (nm)

450 500

80

60

40

20

0

550

1.0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

AQ

E (%

) or Im

ax /p

hoto

ns

0.2

0.0

300 350 400

Wavelength (nm)

450 500

80

60

40

20

0

550

1.0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

AQ

E (%

) or Im

ax /p

hoto

ns

0.2

0.0

300 350 400

Wavelength (nm)

450 500

80

60

40

20

0

550

1.0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

AQ

E (%

) or Im

ax /p

hoto

ns

0.2

0.0

300 350 400

Wavelength (nm)

450 500

80

60

40

20

0

550

Figure 6.18 Absorption spectra (photon absorption), TRMC spectra (charge-carrier creation),and action spectra (apparent quantum efficiency) of modified samples, metal loading of 0.5 at%. (Reproduced with permission from Ref. [98]. Copyright 2016, Elsevier.)


P25 is a mixture of anatase (main) and rutile with absorption edge at 380 and410 nm, respectively. Thereby, a shoulder of its absorption spectrum at about400 nm is assigned to rutile phase. The observed action spectra of modifiedcompounds suggest that gold and nickel particles were loaded predominantlyon rutile and anatase particles, respectively. It has been reported that platinumparticles were photodeposited preferentially on rutile in P25 if the number ofplatinum particles was small and the corresponding action spectrum showeda dip in the wavelength region at around 350 nm [41]. This was explained bythe disturbance of rutile photoabsorption by inactive anatase crystallites in therelatively short wavelength region. Thus, anatase and rutile crystallites mainlywork in Ni/TiO2 and Au/TiO2, respectively, even though both crystallites absorblight, and NiAu/TiO2 might show activity higher than the sum of activities ofsingly modified samples because both crystallites work effectively.

For bare TiO2, the low AQE values are associated with high TRMC signal. Incomparison, the modified compounds present higher AQE values correspondingto slightly lower TRMC signal. This point confirms the assumption that the pos-itive effect of the NPs is more effective on the H2 overpotential, that is, its abilityto act as a recombination center of atomic hydrogen, than on the separation ofcharge carriers.

The NiAu/TiO2 samples are much more efficient in photocatalytic hydrogengeneration than the monometallic samples. Clearly, the improvement of the pho-tocatalytic performance was due to a synergetic effect between Au and Ni(O)since it was not a simple additive effect.

A reaction scheme is proposed for hydrogen photo-production on NiAu/TiO2samples (see Figure 6.19). The generation of the electron–hole pair takes place onthe TiO2 and NiO surfaces. The holes, coming from TiO2, oxidize water and/orthe methanol mixture generating protons, which are then reduced at the sur-face of both TiO2 and NiO forming atomic hydrogen. Finally, H∙ recombinationoccurs on the surface of metal NPs forming H2. The improvement of hydrogengeneration compared with that of the monometallic samples is attributed to asynergetic effect between both Ni(O) and Au acting as a better atomic hydrogenrecombination site than the monometallic samples.

H+

H+H+

λ ≥ Eg

λ ≥ Eg

H°

H°

2H°

H2

H2O, CH3OH

a** Recombination due to

electron transfer by Ohmic

contact

Au

synergetic effect

NiO

a**

Ni0

TiO2

+ +

–

–

+ –

Figure 6.19 A proposed mechanism for H2 production on NiAu/TiO2 samples. (Reproducedwith permission from Ref. [98]. Copyright 2016, Elsevier.)


6.6.4 TiO2 Modified with NiPd Nanoalloys for Hydrogen Evolution

A systematic study of surface modification of commercial TiO2 (P25) withmono- and bimetallic (Ni, Pd, and Ni–Pd) NPs synthesized by radiolysis has beenrealized [99]. The photocatalysts were characterized by HRTEM, scanning trans-mission electron microscopy (STEM), X-ray diffraction (XRD), EDS, XPS, andUV–vis DRS. The charge-carrier dynamics was studied by TRMC. The photocat-alytic activity was evaluated for hydrogen generation under UV–vis irradiationusing polychromatic and monochromatic lights (action spectra analysis of AQE).

TiO2 modified with Pd–Ni bimetallic NPs exhibits a high activity for H2 gen-eration, and a synergetic effect of the two metals was obtained. The study of lightabsorption, charge-carrier dynamics, and photocatalytic activity revealed thatthe main role of the metal NPs is to act as catalytic sites for recombination ofatomic hydrogen.

The characterization of the Ni–Pd NPs showed that the NPs size was sensitiveto the Ni:Pd atomic ratio. Large aggregates (30 nm) were observed in the Pd-richsample Ni1Pd10/TiO2, while the metal NPs on the Ni-rich sample Ni10Pd1/TiO2exhibit a small size (3 nm). In Ni10Pd1/TiO2 samples, once Pd NPs were formed,some Ni ions (remaining in solution) were reduced on their surface, leading toPdcore–Nishell nanoparticles. For Ni1Pd10/TiO2 samples, the amount of Ni ions wassmall facilitating their complete adsorption on the support, where Ni and Pd ionswere reduced independently: monometallic nanoparticles of Pd and Ni withoutinteraction were observed on TiO2. In both cases, small amounts of NiO, PdO,and PdOx have been observed.

A systematic study of the three major steps involved in the photocatalyticH2 generation has been realized: (i) light absorption by the sample, (ii) thecharge-carrier dynamics, and (iii) surface reactions for H2 generation. Thesesteps were studied by DRS, TRMC, and action spectra, respectively. InFigure 6.20, the measurements of the three mentioned steps for bare TiO2,0.5-Ni10Pd1/TiO2, and 1-Ni1Pd10/TiO2 samples are presented. As shown inFigure 6.20a, bare TiO2 exhibits a strong light absorption below 410 nm. TheTRMC spectrum reveals that such absorption is coherent with the charge(electron and hole) separation because no significant TRMC signal was detectedwith an excitation wavelength higher than 410 nm. Despite the good efficiencyof bare TiO2 to absorb light and to generate electron–hole pairs, it is inefficientto produce H2. This suggests that bare TiO2 does not efficiently perform thethird step involved in the photocatalytic H2 evolution, though our TRMC studyreveals that it has a large amount of excited electrons to reduce the protons. Bycontrast, a better performance of the third step is observed when titania surfaceis modified with metal NPs. The three involved steps in the H2 evolution usingmodified titania are shown in Figure 6.20b,c. The samples exhibited a UV andvisible absorption attributed to the semiconductor and the metal NPs, respec-tively. Even though the samples absorb in the visible range, a charge separationwas obtained using only excitation wavelengths shorter than 380 nm, as shownin its TRMC profile. The latter implies that TiO2 is the only one involved in theelectron–hole pair generation. The amount of H2 generated by modified sampleis much higher than that obtained using bare TiO2 as their respective AQE


1.080

60

40

20

0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

0.2

0.0

350

(a) (b)400 450 500

AQ

E (%

) or Im

ax /p

hoto

ns

1.080

60

40

20

0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

0.2

0.0

350 400

Wavelength (nm)

450 500

AQ

E (%

) or Im

ax /p

hoto

ns

DRS

DRS

0.5-Ni10Pd1/TiO2

AQEAQE

TRMCTRMC

Bare TiO2

Light absorptionLight absorption

Charge carriers

separation

Charge carriers

separation

H2 production H2 production

(c) Wavelength (nm)

1.0

80

60

40

20

0

0.8

0.6

0.4

Absorb

ance (

a.u

.)

0.2

0.0

350 400 450 500

AQ

E (%

) or Im

ax /p

hoto

ns

DRS

1-Ni1Pd10/TiO2

AQE

TRMC

Light absorption

Charge carriers

separationH2 production

Figure 6.20 Light absorption, charge-carrier separation and H2 evolution of (a) bare TiO2(b) 0.5-Ni10Pd1/TiO2 and (c) 1-Ni1Pd10/TiO2 measured by DRS, TRMC and action spectrarespectively. (Reproduced with permission from Ref. [99]. Copyright 2017, American ChemicalSociety.)

profile shows. The metal NPs are the only difference between bare and modifiedTiO2, suggesting that H2 generation (third step) is controlled by the metal NPs.

Therefore, the enhancement in H2 evolution is achieved because the surfacemetal NPs catalyze H2 generation reaction. So that, the metal NPs act as cat-alytic sites, and this is their main role in H2 generation. Figure 6.21a depicts theproposed photocatalytic mechanism to generate H2.

In the Ni10Pd1/TiO2 samples, two groups of NPs were found: (i) small Ni NPsand (ii) large NPs of Pd covered by Ni NPs forming a kind of core–shell structure.The interaction between Ni and Pd NPs might be the main reason that metal NPsact as a better catalytic site than the monometallic samples (Figure 6.21b).

In the sample, Ni1Pd10/TiO2 large Pd NPs and small Ni NPs (not in contact)were observed. This is very interesting, because even if there is not contactbetween Ni and Pd NPs, their concomitant presence on TiO2 leads to increase inthe H2 rate compared to monometallic samples. The distance between supportedmetal NPs can also influence the reactivity of the cocatalyst in the same wayas their size and shape do it [100]. In the Ni1Pd10/TiO2 sample, the proximitybetween Pd and Ni NPs seems to be enough to enhance photocatalytic activityfor H2 production compared to the monometallic samples. Furthermore, thehighest amount of hydrogen evolution was observed with the Ni1Pd10/TiO2 sam-ple and can also be due to the electron trapping by the Pd-based nanoparticles,avoiding their recombination and consequently favoring the proton reductionreaction (Figure 6.21c).

6.7 The Effect of Metal Cluster Deposition Route on Structure and Photocatalytic Activity 155

CH3OH/

H2O

Oxidation

reaction

Oxidation

reactionOxidation

reaction

H

HH

HReduction

reaction

on TiO2

Reduction

reaction

on TiO2

Ni10Pd1/TiO2 Ni1Pd10/TiO2

(a)

(b) (c)

H+

H

Metal

NPs

TiO2

TiO2TiO2

** Beneficial but not necessary

Direct reduction

over Pd NPs

H. + H

.

H. + CH3OH

H2

H2

H2

H2

H2 + .CH2OH

–

+

+

+ +

+

+

+ +

+– –

–

––

––

+

– ––

+

+

+

–

2H+

H–Reductionreaction

(2)

(1)

Direct reduction

over metal NPs

Electrontrapping **

(2)

(3) Surface

reactions

Surface

reactions

Surface

reactionsSurface

reactions

Catalytic

site*

Pd Pd

Ni

NiNi

*Main role of metal NPs

Figure 6.21 Schematic representation of (a) H2 evolution using metal NPs/TiO2 asphotocatalysts where the proton reduction occurs on the surface of TiO2 while the molecularH2 is generated on the surface of metal NPs, (b) H2 evolution using metal Ni10Pd1/TiO2 asphotocatalysts, and (c) H2 evolution using metal Ni1Pd10/TiO2 as photocatalysts. (Reproducedwith permission from Ref. [99]. Copyright 2017, American Chemical Society.)

6.7 The Effect of Metal Cluster Deposition Routeon Structure and Photocatalytic Activity of Mono-and Bimetallic Nanoparticles Supported on TiO2

Zaleska-Medynska et al. reported the influence of metal deposition methodon metal nanocluster morphology and structure and its impact on TiO2 pho-tocatalytic activity under Vis and UV–vis irradiation [101]. TiO2 (P25) wasmodified with small and relatively monodisperse mono- and bimetallic clusters(Ag, Pd, Pt, Ag/Pd, Ag/Pt, and Pd/Pt) induced by radiolysis to improve itsphotocatalytic activity. The photocatalysts were characterized by X-ray fluores-cence spectrometry (XRF), photoluminescence spectrometry (PL), DRS, X-raypowder diffractometry (XRD), STEM, and Brunauer–Emmett–Teller (BET)theory surface area analysis. Both simultaneous and subsequent depositionof Ag/Pd, Ag/Pt, and Pd/Pt metal pairs resulted in formation of alloy-likestructures. The effect of metal type (mono- and bimetallic modification), as wellas deposition method (simultaneous or subsequent deposition of two metals),on the photocatalytic activity in toluene removal in gas phase under UV–visirradiation (light-emitting diodes (LEDs)) and phenol degradation in liquidphase under visible-light irradiation (𝜆> 420 nm) was investigated. The highestphotoactivity under vis light was observed for TiO2 coloaded with platinum(0.1%) and palladium (0.1%) clusters. Simultaneous addition of metal precursors


resulted in formation of larger metal nanoparticles (15–30 nm) on TiO2 surfaceand enhancement of the photocatalytic activity of Ag/PdTiO2 in the visible rangeup to four times, while the subsequent metal ions addition resulted in formationof metal particles size ranging from 4 to 20 nm. Photocatalysts, where the metalswere introduced sequentially, exhibited higher photocatalytic activity in thetoluene degradation in the gas phase under the UV–vis irradiation, and thephotocatalytic activity was stable after four cycles. Direct electron transfer fromthe bimetallic metal nanoparticles to the CB of the semiconductor is responsiblefor visible-light photoactivity, whereas superoxide radicals (such as O2

∙− and∙OOH) are responsible for pollutants degradation over metal–TiO2 composites.

6.8 Summary

Development of efficient photocatalysts under solar light for water and airtreatment and solar fuel production is a main challenge to solve energy andenvironment issues. Charge-carrier dynamics is a main key in photocatalysis.Therefore, to develop efficient photocatalysts, it is of crucial importance tounderstand the effect of semiconductor modification on charge-carrier dynam-ics and to correlate it with their photocatalytic activity. Time resolved microwaveconductivity is a very powerful technique to study these charge-carrier mobilityand dynamics.

Surface modification of TiO2 with metal nanoparticles is a very efficient wayto enhance its photocatalytic activity under UV and visible light. Radiolysis is avery powerful method to synthesize metal nanoparticles of controlled size andcomposition on semiconductors. Enhancement of the photoactivity of modifiedsemiconductors with metal nanoparticles under UV irradiation originates fromprolongation of lifetime of charge carriers (photogenerated electrons and holes);indeed, noble metal nanoparticles act as an electron sink as proved by TRMCstudies on different systems and thus accelerating the transfer of electrons fromthe semiconductor to substrates. Under visible-light excitation, surface modifi-cation with metal nanoparticles can lead to activation of the semiconductor andinduce a photocatalytic activity in this spectral domain. Increasing attention waspaid to plasmonic photocatalysis: modification of wide band gap semiconductors(in particular TiO2) with plasmonic nanoparticles (mainly Ag, Au, and Cu) leadsto visible-light absorption due to the plasmon resonance and to the photocat-alytic activity under visible light. The LSPR and the Schottky junction propertiesare characteristic of the plasmonic photocatalysts and are the main propertiesresponsible for the enhancement of the photoactivity of the composite system(metal nanoparticle/TiO2). An electric field is created by the LSPR, which mayenhance the generation of more electrons and holes and heat up the surroundingenvironment, inducing an increase in the redox reaction rates and the masstransfer, and also polarization of the nonpolar molecules for better adsorption. Inthe case of Au–TiO2, TRMC studies have shown that hot electrons can be ejectedfrom the excited nanoparticles at their plasmon band to the CB of the SC induc-ing an activity under visible light. Modification with bimetallic nanoparticles can

References 157

lead to enhancement of the photocatalytic activity (water and air depollutionand hydrogen generation) compared to modification with monometallic NPs.A systematic study of light absorption, charge-carrier dynamics, and reactionefficiency (the three main steps involved in H2 evolution) demonstrated that themain role of the metal NPs for hydrogen generation is to act as catalytic sites.TRMC studies also showed that the electron transfer from TiO2 to metal NPs isfavorable, but not an indispensable factor for photocatalytic H2 generation.

References

1 Fujishima, A. and Honda, K. (1972) Electrochemical photolysis of water at asemiconductor electrode. Nature, 238, 37–38. doi: 10.1038/238037a0

2 Wang, G., Xu, L., Zhang, J., Yin, T., Han, D., Wang, G., Xu, L., Zhang, J.,Yin, T., and Han, D. (2012) Enhanced photocatalytic activity of pow-ders (P25) via calcination treatment. Int. J. Photoenergy, 2012, 265760.doi: 10.1155/2012/265760, 10.1155/2012/265760

3 Asahi, R., Morikawa, T., Ohwaki, T., Aoki, K., and Taga, Y. (2001)Visible-light photocatalysis in nitrogen-doped titanium oxides. Science,293, 269–71. doi: 10.1126/science.1061051

4 Sakthivel, S. and Kisch, H. (2003) Photocatalytic and photoelectrochemicalproperties of nitrogen-doped titanium dioxide. ChemPhysChem, 4, 487–490.doi: 10.1002/cphc.200200554

5 Kitano, S., Murakami, N., Ohno, T., Mitani, Y., Nosaka, Y., Asakura, H.,Teramura, K., Tanaka, T., Tada, H., Hashimoto, K., and Kominami, H.(2013) Bifunctionality of Rh3+ modifier on TiO2 and working mechanism ofRh3+/TiO2 photocatalyst under irradiation of visible light. J. Phys. Chem. C,117, 11008–11016. doi: 10.1021/jp311801e

6 Zhu, X., Liu, Z., Fang, J., Wu, S., and Xu, W. (2013) Synthesis and character-ization of mesoporous Bi/TiO2 nanoparticles with high photocatalytic activ-ity under visible light. J. Mater. Res., 28, 1334–1342. doi: 10.1557/jmr.2013.92

7 Murcia-López, S., Hidalgo, M.C., and Navío, J.A. (2011) Synthesis,characterization and photocatalytic activity of Bi-doped TiO2 photo-catalysts under simulated solar irradiation. Appl. Catal., A, 404, 59–67.doi: 10.1016/j.apcata.2011.07.008

8 Kamat, P.V. (2002) Photophysical, Photochemical and PhotocatalyticAspects of Metal Nanoparticles. J. Phys. Chem. B, 106, 7729–7744.doi: 10.1021/jp0209289

9 Kamat, P.V. (2012) Manipulation of charge transfer across semiconductorinterface. A criterion that cannot be ignored in photocatalyst design. J. Phys.Chem. Lett., 3, 663–672. doi: 10.1021/jz201629p

10 Dhakshinamoorthy, A., Navalon, S., Corma, A., and Garcia, H. (2012) Pho-tocatalytic CO2 reduction by TiO2 and related titanium containing solids.Energy Environ. Sci., 5, 9217–9233. doi: 10.1039/C2EE21948D

11 Linic, S., Christopher, P., and Ingram, D.B. (2011) Plasmonic-metal nanos-tructures for efficient conversion of solar to chemical energy. Nat. Mater.,10, 911–921. doi: 10.1038/nmat3151


12 Hou, W. and Cronin, S.B. (2013) A review of surface plasmonresonance-enhanced photocatalysis. Adv. Funct. Mater., 23, 1612–1619.doi: 10.1002/adfm.201202148

13 Pelaez, M., Nolan, N.T., Pillai, S.C., Seery, M.K., Falaras, P., Kontos, A.G.,Dunlop, P.S.M., Hamilton, J.W.J., Byrne, J.A., O’Shea, K., Entezari, M.H., andDionysiou, D.D. (2012) A review on the visible light active titanium dioxidephotocatalysts for environmental applications. Appl. Catal., B, 125, 331–349.doi: 10.1016/j.apcatb.2012.05.036

14 Herrmann, J.M. (1999) Heterogeneous photocatalysis: fundamentals andapplications to the removal of various types of aqueous pollutants. Catal.Today, 53, 115–129, doi: 10.1016/S0920-5861(99)00107-8.

15 Kunst, M. and Beck, G. (1986) The study of charge carrier kinetics in semi-conductors by microwave conductivity measurements. J. Appl. Phys., 60,3558–3566. doi: 10.1063/1.337612

16 Colbeau-Justin, C., Kunst, M., and Huguenin, D. (2003) Structural influ-ence on charge-carrier lifetimes in TiO2 powders studied by microwaveabsorption. J. Mater. Sci., 38, 2429–2437. doi: 10.1023/A:1023905102094

17 Daniel, M.-C. and Astruc, D. (2004) Gold nanoparticles: assembly,supramolecular chemistry, quantum-size-related properties, and applicationstoward biology, catalysis, and nanotechnology. Chem. Rev., 104, 293–346.doi: 10.1021/cr030698+

18 Disdier, J., Herrmann, J.-M., and Pichat, P. (1983) Platinum/titanium diox-ide catalysts. A photoconductivity study of electron transfer from theultraviolet-illuminated support to the metal and of the influence of hydro-gen. J. Chem. Soc., Faraday Trans. 1, 79, 651. doi: 10.1039/f19837900651

19 Kowalska, E., Mahaney, O.O.P., Abe, R., and Ohtani, B. (2010)Visible-light-induced photocatalysis through surface plasmon excita-tion of gold on titania surfaces. Phys. Chem. Chem. Phys., 12, 2344–2355.doi: 10.1039/B917399D

20 Ingram, D.B., Christopher, P., Bauer, J.L., and Linic, S. (2011) Predictivemodel for the design of plasmonic metal/semiconductor composite photo-catalysts. ACS Catal., 1, 1441–1447. doi: 10.1021/cs200320h

21 Kowalska, E., Remita, H., Colbeau-Justin, C., Hupka, J., and Belloni, J.(2008) Modification of titanium dioxide with platinum ions and clus-ters: Application in photocatalysis. J. Phys. Chem. C, 112, 1124–1131.doi: 10.1021/jp077466p

22 Belloni, J., Mostafavi, M., Remita, H., Marignier, J.-L., and Delcourt, M.-O.(1998) Radiation-induced synthesis of mono- and multi-metallic clusters andnanocolloids. New J. Chem., 22, 1239–1255. doi: 10.1039/A801445K

23 Abidi, W. and Remita, H. (2010) Gold based nanoparticles generated byradiolytic and photolytic methods. Recent Patents Eng., 4, 170.

24 Remita, H. and Remita, S. (2010) Recent Trends Radiat. Chem., (WSPC),World Scientific Publishing Company, Singapore.

25 Schwarz, H.A. (1981) Free radicals generated by radiolysis of aqueous solu-tions. J. Chem. Educ., 58, 101. doi: 10.1021/ed058p101

26 Ferradini, C. and Pucheault, J. (1983) Biol. L’action Des Rayonnements Ionis,Masson, Paris.

References 159

27 Belloni, J. (2006) Nucleation, growth and properties of nanoclusters studiedby radiation chemistry: application to catalysis. Catal. Today, 113, 141–156.doi: 10.1016/j.cattod.2005.11.082

28 Elliot, A.J. and Simsons, A.S. (1984) Rate constants for reactions of hydroxylradicals as a function of temperature. Radiat. Phys. Chem., 24, 229–231.doi: 10.1016/0146-5724(84)90056-6

29 Schwarz, H.A. and Dodson, R.W. (1989) Reduction potentials of CO2− and

the alcohol radicals. J. Phys. Chem., 93, 409–414. doi: 10.1021/j100338a07930 Grabowska, E., Remita, H., and Zaleska, A. (2010) Photocatalytic activity

of TiO2 loaded with metal clusters. Physicochem. Probl. Miner. Process,45, 29–38. http://www.minproc.pwr.wroc.pl/journal/pdf/2010_45/FPM 45/PPoMP_45_3.pdf.

31 Hada, H., Yonezawa, Y., and Saikawa, M. (1982) Photoreduction of silverIon in a titanium dioxide suspension. Bull. Chem. Soc. Jpn., 55, 2010–2014.doi: 10.1246/bcsj.55.2010

32 Bamwenda, G.R., Tsubota, S., Nakamura, T., and Haruta, M. (1995) Photoas-sisted hydrogen production from a water–ethanol solution: a comparisonof activities of Au---TiO2 and Pt---TiO2. J. Photochem. Photobiol., A, 89,177–189. http://www.sciencedirect.com/science/article/B6TGY-4032409-V/2/0d8a3e6a61fa0d1e43eb2b2db882b22f.

33 Kowalska, E., Abe, R., and Ohtani, B. (2009) Visible light-induced pho-tocatalytic reaction of gold-modified titanium(IV) oxide particles: actionspectrum analysis. Chem. Commun., 241–243. doi: 10.1039/B815679D

34 Albiter, E., Hai, Z., Alfaro, S., Remita, H., Valenzuela, M.A., andColbeau-Justin, C. (2013) A comparative study of photo-assisted depositionof silver nanoparticles on TiO2. J. Nanosci. Nanotechnol., 13, 4943–4948.doi: 10.1166/jnn.2013.7590

35 Selvakannan, P.R., Lampre, I., Erard, M., and Remita, H. (2008) Platinumcarbonyl clusters: double emitting quantum dots. J. Phys. Chem. C, 112,18722–18726. doi: 10.1021/jp808033q

36 Chen, W.-T., Hsu, Y.-J., and Kamat, P.V. (2012) Realizing visible photoac-tivity of metal nanoparticles: excited-state behavior and electron-transferproperties of silver (Ag8) clusters. J. Phys. Chem. Lett., 3, 2493–2499.doi: 10.1021/jz300940c

37 Devadas, M.S., Kwak, K., Park, J.-W., Choi, J.-H., Jun, C.-H., Sinn, E.,Ramakrishna, G., and Lee, D. (2010) Directional electron transfer inchromophore-labeled quantum-sized Au25 clusters: Au25 as an electrondonor. J. Phys. Chem. Lett., 1, 1497–1503. doi: 10.1021/jz100395p

38 Diez, I. and Ras, R.H.A. (2011) Fluorescent silver nanoclusters. Nanoscale, 3,1963–1970. doi: 10.1039/C1NR00006C

39 Treguer, M., Remita, H., Pernot, P., Khatouri, J., and Belloni, J. (2001) Redoxkinetics of Chini-type platinum carbonyl clusters studied by time-resolvedpulse radiolysis. J. Phys. Chem. A, 105, 6102–6108. doi: 10.1021/jp0037201

40 Tréguer, M., Remita, H., Belloni, J., and de Keyzer, R. (2002) Carbonyl plat-inum clusters as silver halide dopant for photographic latent image forma-tion. J. Imaging Sci. Technol., 46, 193–199. http://www.ingentaconnect.com/content/ist/jist/2002/00000046/00000003/art00002.


41 Ohtani, B., Prieto-Mahaney, O.O., Li, D., and Abe, R. (2010) What isDegussa (Evonik) P25? Crystalline composition analysis, reconstructionfrom isolated pure particles and photocatalytic activity test. J. Photochem.Photobiol., A, 216, 179–182. doi: 10.1016/j.jphotochem.2010.07.024

42 Serpone, N., Sauve, G., Koch, R., Tahiri, H., Pichat, P., Piccinini, P.,Pelizzetti, E., and Hidaka, H. (1996) Standardization protocol of processefficiencies and activation parameters in heterogeneous photocataly-sis: relative photonic efficiencies [zeta]r. J. Photochem. Photobiol., A, 94,191–203. http://www.sciencedirect.com/science/article/B6TGY-3TKMP66-J/2/7cf4139c08301e5cc7970bf76881b67a.

43 Sobczynski, A., Duczmal, L., and Zmudzinski, W. (2004) Phenol destructionby photocatalysis on TiO2: an attempt to solve the reaction mechanism.J. Mol. Catal. A: Chem., 213, 225–230. http://www.sciencedirect.com/science/article/B6TGM-4BM90GK-1/2/dc87de1a7f94344d057d6bf874a20005.

44 Guo, Z., Ma, R., and Li, G. (2006) Degradation of phenol by nanomaterialTiO2 in wastewater. Chem. Eng. J., 119, 55–59. doi: 10.1016/j.cej.2006.01.017

45 Rasalingam, S., Kibombo, H.S., Wu, C.-M., Peng, R., Baltrusaitis, J., andKoodali, R.T. (2014) Competitive role of structural properties of titania–silicamixed oxides and a mechanistic study of the photocatalytic degradation ofphenol. Appl. Catal., B, 148, 394–405. doi: 10.1016/j.apcatb.2013.11.025

46 Tahiri Alaoui, O., Herissan, A., Le Quoc, C., Zekri, M.e.M., Sorgues, S.,Remita, H., and Colbeau-Justin, C. (2012) Elaboration, charge-carrier life-times and activity of Pd-TiO2 photocatalysts obtained by gamma radiolysis.J. Photochem. Photobiol., A, 242, 34–43. doi: 10.1016/j.jphotochem.2012.05.030

47 Hérissan, A., Meichtry, J.M., Remita, H., Colbeau-Justin, C., and Litter, M.I.(2017) Reduction of nitrate by heterogeneous photocatalysis over pure andradiolytically modified TiO2 samples in the presence of formic acid. Catal.Today, 281, 101–108. doi: 10.1016/j.cattod.2016.05.044

48 Méndez-Medrano, M.G., Kowalska, E., Lehoux, A., Herissan, A., Ohtani, B.,Bahena, D., Briois, V., Colbeau-Justin, C., Rodríguez-López, J.L., and Remita,H. (2016) Surface modification of TiO2 with Ag nanoparticles and CuO nan-oclusters for application in photocatalysis. J. Phys. Chem. C, 120, 5143–5154.doi: 10.1021/acs.jpcc.5b10703

49 Malinsky, M.D., Kelly, K.L., Schatz, G.C., and Van Duyne, R.P. (2001)Chain length dependence and sensing capabilities of the localized surfaceplasmon resonance of silver nanoparticles chemically modified with alka-nethiol self-assembled monolayers. J. Am. Chem. Soc., 123, 1471–1482.doi: 10.1021/ja003312a

50 Hutter, E., Fendler, J.H., and Roy, D. (2001) Surface plasmon resonancestudies of gold and silver nanoparticles linked to gold and silver sub-strates by 2-aminoethanethiol and 1,6-hexanedithiol. J. Phys. Chem. B, 105,11159–11168. doi: 10.1021/jp011424y

51 Mulvaney, P. (1996) Surface plasmon spectroscopy of nanosized metal parti-cles. Langmuir, 12, 788, doi: 10.1021/la9502711.

52 El-Sayed, M.A. (2001) Some interesting properties of metals confined intime and nanometer space of different shapes. Acc. Chem. Res., 34, 257–264.doi: 10.1021/ar960016n

References 161

53 Roucoux, A., Schulz, J., and Patin, H. (2002) Reduced transitionmetal colloids: a novel family of reusable catalysts. Chem. Rev., 102,3757, doi:10.1021/cr010350j.

54 Jana, N.R., Sau, T.K., and Pal, T. (1999) Growing small silver particle asredox catalyst. J. Phys. Chem. B, 103, 115–121. doi: 10.1021/jp982731f

55 Gang, L., Anderson, B.G., van Grondelle, J., and van Santen, R.A. (2003)Low temperature selective oxidation of ammonia to nitrogen on silver-basedcatalysts. Appl. Catal., B, 40, 101–110. doi: 10.1016/S0926-3373(02)00129-7

56 Creighton, J.A., Blatchford, C.G., and Albrecht, M.G. (1979) Plasma reso-nance enhancement of Raman scattering by pyridine adsorbed on silver orgold sol particles of size comparable to the excitation wavelength. J. Chem.Soc., Faraday Trans. 2, 75, 790–798. doi: 10.1039/F29797500790

57 Tao, A., Kim, F., Hess, C., Goldberger, J., He, R., Sun, Y., Xia, Y., and Yang, P.(2003) Langmuir−Blodgett silver nanowire monolayers for molecular sens-ing using surface-enhanced Raman spectroscopy. Nano Lett., 3, 1229–1233.doi: 10.1021/nl0344209

58 Selvan, S.T., Hayakawa, T., and Nogami, M. (1999) Remarkable influenceof silver islands on the enhancement of fluorescence from Eu3+ Ion-dopedsilica gels. J. Phys. Chem. B, 103, 7064–7067. doi: 10.1021/jp9902755

59 Geddes, C.D., Cao, H., Gryczynski, I., Gryczynski, Z., Fang, J., andLakowicz, J.R. (2003) Metal-enhanced fluorescence (MEF) due to silvercolloids on a planar surface: potential applications of indocyanine green toin vivo imaging. J. Phys. Chem. A, 107, 3443–3449. doi: 10.1021/jp022040q

60 Arabatzis, I.M., Stergiopoulos, T., Bernard, M.C., Labou, D.,Neophytides, S.G., and Falaras, P. (2003) Silver-modified titanium dioxidethin films for efficient photodegradation of methyl orange. Appl. Catal., B,42, 187–201. doi: 10.1016/S0926-3373(02)00233-3

61 Zhang, D., Li, G., and Yu, J.C. (2010) Inorganic materials for photocatalyticwater disinfection. J. Mater. Chem., 20, 4529–4536. doi: 10.1039/B925342D

62 Ji, Z., Ismail, M.N., Callahan, D.M., Pandowo, E., Cai, Z., Goodrich, T.L.,Ziemer, K.S., Warzywoda, J., and Sacco, A. (2011) The role of silvernanoparticles on silver modified titanosilicate ETS-10 in visible light photo-catalysis. Appl. Catal., B, 102, 323–333. doi: 10.1016/j.apcatb.2010.12.021

63 Kerker, M. (1985) The optics of colloidal silver: something oldand something new. J. Colloid Interface Sci., 105, 297–314. doi:10.1016/0021-9797(85)90304-2

64 Grabowska, E., Zaleska, A., Sorgues, S., Kunst, M., Etcheberry, A.,Colbeau-Justin, C., and Remita, H. (2013) Modification of titanium(IV)dioxide with small silver nanoparticles: application in photocatalysis. J. Phys.Chem. C, 117, 1955–1962. doi: 10.1021/jp3112183

65 Tausch-Treml, R., Henglein, A., and Lilie, J. (1978) Reactivity of silver atomsin aqueous solution II. A pulse radiolysis study. Ber. Bunsen Ges. Phys.Chem., 82, 1335–1343. doi: 10.1002/bbpc.19780821213

66 Tang, H., Prasad, K., Sanjinès, R., Schmid, P.E., and Lévy, F. (1994) Electricaland optical properties of TiO2 anatase thin films. J. Appl. Phys., 75, 2042.doi: 10.1063/1.356306

67 Palik, E. (1985) Handb. Opt. Constants Solids, Academic Press Inc., USA.


68 Díez, I., Pusa, M., Kulmala, S., Jiang, H., Walther, A., Goldmann, A.S.,Müller, A.H.E., Ikkala, O., and Ras, R.H.A. (2009) Color tunability andelectrochemiluminescence of silver nanoclusters. Angew. Chem. Int. Ed., 48,2122–2125. doi: 10.1002/anie.200806210

69 Bannat, I., Wessels, K., Oekermann, T., Rathousky, J., Bahnemann, D., andWark, M. (2009) Improving the photocatalytic performance of mesoporoustitania films by modification with gold nanostructures. Chem. Mater., 21,1645–1653. doi: 10.1021/cm803455k

70 Awazu, K., Fujimaki, M., Rockstuhl, C., Tominaga, J., Murakami, H.,Ohki, Y., Yoshida, N., and Watanabe, T. (2008) A plasmonic photocata-lyst consisting of sliver nanoparticles embedded in titanium dioxide. J. Am.Chem. Soc., 130, 1676–1680, doi:10.1021/ja076503n.

71 Kowalska, E., Wei, Z., Karabiyik, B., Herissan, A., Janczarek, M., Endo, M.,Markowska-Szczupak, A., Remita, H., and Ohtani, B. (2015) Silver-modifiedtitania with enhanced photocatalytic and antimicrobial propertiesunder UV and visible light irradiation. Catal. Today, 252, 136–142.doi: 10.1016/j.cattod.2014.10.038

72 Zhang, X., Chen, Y.L., Liu, R.-S., and Tsai, D.P. (2013) Plasmonic photocatal-ysis. Rep. Prog. Phys., 76, 46401. doi: 10.1088/0034-4885/76/4/046401

73 Zhou, X., Liu, G., Yu, J., and Fan, W. (2012) Surface plasmonresonance-mediated photocatalysis by noble metal-based composites undervisible light. J. Mater. Chem., 22, 21337–21354. doi: 10.1039/C2JM31902K

74 Bian, Z., Tachikawa, T., Zhang, P., Fujitsuka, M., and Majima, T. (2014)Au/TiO2 superstructure-based plasmonic photocatalysts exhibiting effi-cient charge separation and unprecedented activity. J. Am. Chem. Soc., 136,458–465. doi: 10.1021/ja410994f

75 Tanaka, A., Ogino, A., Iwaki, M., Hashimoto, K., Ohnuma, A., Amano, F.,Ohtani, B., and Kominami, H. (2012) Gold–titanium(IV) oxide plasmonicphotocatalysts prepared by a colloid-photodeposition method. Langmuir, 28,13105–13111. doi: 10.1021/la301944b

76 Orlov, A., Jefferson, D.A., Macleod, N., and Lambert, R.M. (2004) Pho-tocatalytic properties of TiO2 modified with gold nanoparticles in thedegradation of 4-chlorophenol in aqueous solution. Catal. Letters., 92,41–47. doi: 10.1023/b:catl.0000011084.43007.80

77 Kowalska, E., Rau, S., and Ohtani, B. (2012) Plasmonic titania photocatalystsactive under UV and visible-light irradiation: influence of gold amount, size,and shape. J. Nanotechnol., 2012, 1–11. doi: 10.1155/2012/361853

78 Méndez-Medrano, M.G., Kowalska, E., Lehoux, A., Herissan, A., Ohtani, B.,Rau, S., Colbeau-Justin, C., Rodriguez-Lopez, J.L., and Remita, H. (2016)Surface modification of TiO2 with Au nanoclusters for efficient water treat-ment and hydrogen generation under visible light. J. Phys. Chem. C, 120,25010–25022. doi: 10.1021/acs.jpcc.6b06854

79 Schindler, K.M. and Kunst, M. (1990) Charge-carrier dynamics in titaniapowders. J. Phys. Chem., 94, 8222–8226. doi: 10.1021/j100384a045

80 Yamakata, A., Ishibashi, T., and Onishi, H. (2007) Time-resolved infraredabsorption study of nine TiO2 photocatalysts. Chem. Phys., 339, 133–137.doi: 10.1016/j.chemphys.2007.05.010

References 163

81 Murdoch, M., Waterhouse, G.I.N., Nadeem, M.A., Metson, J.B., Keane, M.A.,Howe, R.F., Llorca, J., and Idriss, H. (2011) The effect of gold loading andparticle size on photocatalytic hydrogen production from ethanol overAu/TiO2 nanoparticles. Nat. Chem., 3, 489–492. doi: 10.1038/nchem.1048

82 Chiarello, G.L., Selli, E., and Forni, L. (2008) Photocatalytic hydrogen pro-duction over flame spray pyrolysis-synthesised TiO2 and Au/TiO2. Appl.Catal., B, 84, 332–339. doi: 10.1016/j.apcatb.2008.04.012

83 Nishimura, S., Abrams, N., Lewis, B.A., Halaoui, L.I., Mallouk, T.E.,Benkstein, K.D., van de Lagemaat, J., and Frank, A.J. (2003) Standing waveenhancement of red absorbance and photocurrent in dye-sensitized titaniumdioxide photoelectrodes coupled to photonic crystals. J. Am. Chem. Soc.,125, 6306–6310. doi: 10.1021/ja034650p

84 Wen, Y., Liu, B., Zeng, W., and Wang, Y. (2013) Plasmonic photocataly-sis properties of Au nanoparticles precipitated anatase/rutile mixed TiO2nanotubes. Nanoscale, 20, 9739–9746. doi: 10.1039/C3NR03024E

85 Esumi, K., Sarashina, S., and Yoshimura, T. (2004) Synthesis of goldnanoparticles from an organometallic compound in supercritical carbondioxide. Langmuir, 20, 5189–5191. doi: 10.1021/la049415e

86 Tsukamoto, D., Shiraishi, Y., Sugano, Y., Ichikawa, S., Tanaka, S., andHirai, T. (2012) Gold nanoparticles located at the interface of anatase/rutileTiO2 particles as active plasmonic photocatalysts for aerobic oxidation. J.Am. Chem. Soc., 134, 6309–6315. doi: 10.1021/ja2120647

87 Waterhouse, G.I.N., Wahab, A.K., Al-Oufi, M., Jovic, V., Anjum, D.H.,Sun-Waterhouse, D., Llorca, J., and Idriss, H. (2013) Hydrogen produc-tion by tuning the photonic band gap with the electronic band gap of TiO2.Sci. Rep., 3, 2849, 1–5. doi: 10.1038/srep02849

88 Ortega Méndez, J.A., López, C.R., Pulido Melián, E., González Díaz, O.,Doña Rodríguez, J.M., Fernández Hevia, D., and Macías, M. (2014)Production of hydrogen by water photo-splitting over commercialand synthesised Au/TiO2 catalysts. Appl. Catal., B, 147, 439–452.doi: 10.1016/j.apcatb.2013.09.029

89 Joo, J.B., Dillon, R., Lee, I., Yin, Y., Bardeen, C.J., and Zaera, F. (2014) Pro-motion of atomic hydrogen recombination as an alternative to electrontrapping for the role of metals in the photocatalytic production of H2. Proc.Natl. Acad. Sci. U. S. A., 111, 7942–7947. doi: 10.1073/pnas.1405365111

90 Caretti, I., Keulemans, M., Verbruggen, S.W., Lenaerts, S., andVan Doorslaer, S. (2015) Light-induced processes in plasmonic gold/TiO2photocatalysts studied by electron paramagnetic resonance. Top. Catal., 58,776–782. doi: 10.1007/s11244-015-0419-4

91 Priebe, J.B., Karnahl, M., Junge, H., Beller, M., Hollmann, D., andBrückner, A. (2013) Water reduction with visible light: synergy betweenoptical transitions and electron transfer in Au-TiO2 catalysts visualizedby in situ EPR spectroscopy. Angew. Chem. Int. Ed., 52, 11420–11424.doi: 10.1002/anie.201306504

92 Kouame, N.A., Alaoui, O.T., Herissan, A., Larios, E., Jose-Yacaman,M., Etcheberry, A., Colbeau-Justin, C., and Remita, H. (2015) Vis-ible light-induced photocatalytic activity of modified titanium(IV)


oxide with zero-valent bismuth clusters. New J. Chem., 39, 2316–2322.doi: 10.1039/C4NJ01979B

93 Hai, Z., El Kolli, N., Uribe, D.B., Beaunier, P., Jose-Yacaman, M., Vigneron, J.,Etcheberry, A., Sorgues, S., Colbeau-Justin, C., Chen, J., and Remita, H.(2013) Modification of TiO2 by bimetallic Au-Cu nanoparticles for wastewa-ter treatment. J. Mater. Chem. A, 1, 10829–10835. doi: 10.1039/C3TA11684K

94 Redjala, T., Remita, H., Apostolescu, G., Mostafavi, M., Thomazeau, C., andUzio, D. (2006) Bimetallic Au-Pd and Ag-Pd clusters synthesised by 𝛾 orelectron beam radiolysis and study of the reactivity/structure relationshipsin the selective hydrogenation of buta-1,3-diene. Oil Gas Sci. Technol. - Rev.IFP., 61, 789–797. doi: 10.2516/ogst:2006019

95 Doherty, R.P., Krafft, J.-M., Méthivier, C., Casale, S., Remita, H., Louis, C.,and Thomas, C. (2012) On the promoting effect of Au on CO oxidationkinetics of Au–Pt bimetallic nanoparticles supported on SiO2: an electroniceffect? J. Catal., 287, 102–113. doi: 10.1016/j.jcat.2011.12.011

96 Hai, Z., Kolli, N.E.L., Chen, J., and Remita, H. (2014) Radiolytic synthe-sis of Au-Cu bimetallic nanoparticles supported on TiO2: application inphotocatalysis. New J. Chem., 38, 5279–5286. doi: 10.1039/C4NJ00883A

97 Nischk, M., Mazierski, P., Wei, Z., Siuzdak, K., Kouame, N.A., Kowalska, E.,Remita, H., and Zaleska-Medynska, A. (2016) Enhanced photocatalytic,electrochemical and photoelectrochemical properties of TiO2 nanotubesarrays modified with Cu, AgCu and Bi nanoparticles obtained via radiolyticreduction. Appl. Surf. Sci., 387, 89–102. doi: 10.1016/j.apsusc.2016.06.066

98 Luna, A.L., Novoseltceva, E., Louarn, E., Beaunier, P., Kowalska, E.,Ohtani, B., Valenzuela, M.A., Remita, H., and Colbeau-Justin, C. (2016)Synergetic effect of Ni and Au nanoparticles synthesized on titania particlesfor efficient photocatalytic hydrogen production. Appl. Catal., B, 191, 18–28.doi: 10.1016/j.apcatb.2016.03.008

99 Luna, A.L., Dragoe, D., Wang, K., Beaunier, P., Kowalska, E.K., Ohtani, B.,Bahena Uribe, D., Valenzuela, M.A., Remita, H., and Colbeau-Justin, C.(2017) Photocatalytic hydrogen evolution using Ni-Pd/TiO2: correlation oflight absorption, Charge-Carrier Dynamics and Quantum Efficiency. J. Phys.Chem. C, 121, 14302–14311, doi: 10.1021/acs.jpcc.7b01167

100 Henry, C.R. (2015) 2D-arrays of nanoparticles as model catalysts. Catal.Lett., 145, 731–749. doi: 10.1007/s10562-014-1402-6

101 Klein, M., Nadolna, J., Gołabiewska, A., Mazierski, P., Klimczuk, T., Remita,H., and Zaleska-Medynska, A. (2016) The effect of metal cluster deposi-tion route on structure and photocatalytic activity of mono- and bimetallicnanoparticles supported on TiO2 by radiolytic method. Appl. Surf. Sci., 378,37–48. doi: 10.1016/j.apsusc.2016.03.191

165

7

Glassy Photocatalysts: New Trend in Solar PhotocatalysisBharat B. Kale, Manjiri A. Mahadadalkar, and Ashwini P. Bhirud

Centre for Materials for Electronic Technology (C-MET), Ministry of Electronics and Information Technology(MeitY), Government of India, Panchwati, off Pashan Road, Pune 411008, India

7.1 Introduction

In this era of depleting fossil fuel, sustainable supply of energy as per escalatingdemand of humankind is a major concern. Exploitation of these nonrenewableenergy sources and increasing industrialization cause major environmentalhazards. Researchers are working on the renewable energy sources such assun, wind, waves, and geothermal vibrations. Hydrogen is also a potentialcandidate for green energy, but the use of fossil fuels for its production restrictsits application on mass scale. In 1972, Fujishima and Honda demonstrated thathydrogen can be generated by photoelectrochemically breaking water moleculeusing semiconductors, that is, titania as photoanode under UV radiation [1].This method was novel, and the use of solar light made it cost–effective. Theprocess of H2 generation is not a single reaction but a series of reactionsfollowing each other. When the semiconductor used as photocatalyst absorbsthe solar radiation, the excited electrons from its conduction band (CB) startthe reduction reaction of water molecule into hydrogen while positively chargedholes in valence band (VB) assist the water oxidation reaction at the active siteson its surface. Same as water, hydrogen sulfide (H2S) can be used as source of H2.

H2S abundantly occurs in natural gas, petroleum, crude oil, industrial, urban,and agricultural sewage. It is a strong corrosion agent and known to be a majorsource of acid rain when oxidized in atmosphere to SO2. It is extremely harmfulfor human health at concentrations as low as 0.5–2 ppb by volume. Solar energycan be used for splitting of H2S to generate hydrogen (H2). In this way, solarenergy can be stored in the form of H2 gas and simultaneously hazardous H2S canbe diminished without generating further secondary pollutant by-products. Cur-rently, the materials required for H2 based technologies are the area of extensiveinterest worldwide [2].


166 7 Glassy Photocatalysts: New Trend in Solar Photocatalysis

7.2 Fundamentals of H2S Splitting

In this chapter, the photocatalytic H2S splitting process for hydrogen generationis considered. It is important to understand H2S splitting process before focusingon materials used as photocatalysts. Therefore, in the next two sections, thermo-dynamics of H2S splitting process and the role of photocatalyst are described indetail.

7.2.1 General

H2S is generated from processes such as petroleum refining, tanning, andeffluent treatment of many other chemical plants. Commercially H2S is decom-posed by the “Claus process.” In this process, the hydrogen in H2S is oxidizedto water, which needs tremendous external energy inputs [3]. Figure 7.1ashows the schematic diagram of Claus process by which elemental sulfur isrecovered from acidic water. This process involves large amount of harmfulwaste products, which are secondary pollutants. Hence, there is a need fordevelopment of a clean and environmentally benign technology to decomposeH2S gas. Interestingly, H2S is a molecule like H2O and can be split to produceH2. Figure 7.1b shows photocatalytic decomposition of H2S gas. According toBaeg et al., photocatalytic H2S splitting is energetically more favorable than thatof water [4]. Thus, splitting of H2S to produce hydrogen is more economical thanwater splitting.

7.2.2 Thermodynamics of H2S Splitting

Thermodynamically, H2S decomposes into H2 and S, which requires energy(ΔG∘ = 33.44 kJ mol−1, where G∘ is the Gibbs energy) [4]. Positive Gibbs freeenergy indicates that the process is endothermic. External energy should be

H2S + O2

Air

Furnace

Claus process Photocatalytic process

Catalytic section

GasH2S

H2S

H2 + S

H2O + S

Liquid sulfur

H2O: ΔG° = 237.19 kJ mol–1

E° = ΔG°/2F = 1.23 eV

H2S: ΔG° = 33.44 kJ mol–1

E° = ΔG°/2F = 0.17 eV

(a) (b)

Figure 7.1 Schematic diagram showing (a) the industrial Claus process and (b) photocatalyticsolar H2 production from H2S decomposition. E∘, standard-state energy of the reaction; G∘,standard-state Gibbs energy; F, Faraday constant.

7.2 Fundamentals of H2S Splitting 167

provided to overcome this barrier formed by positive Gibbs free energy. Energyof photons present in incident solar radiation is utilized for breaking this barrier.Hence, the material competent to absorb photon energy has to be incorporatedfor decomposition of H2S molecule. Semiconductors efficiently absorb photonsand initiate the formation of photoinduced electron and hole pairs, required foroxidation and reduction reactions. Due to this property, semiconductors arewidely used as photocatalyst in splitting of H2S as well as water.

In H2S splitting, the reduction reaction is evolution of H2 from protons similarto water splitting but the oxidative reaction is oxidation of S2−. Hence, the CBedge of the semiconductor should be more negative than the redox potential ofH+/H2, and the valance band edge should be more positive than the redox poten-tial of H2S/S2−. As a result, the semiconductors having less positive valance bandedge can also be used for H2S splitting unlike water splitting. This open all newarena of semiconductors and their composites, which are not suitable for watersplitting, can be exploited as photocatalyst for H2 production by H2S splitting.

7.2.3 Role of Photocatalysts

The role of photocatalysis is to initiate and/or accelerate specific reduction andoxidation reactions on the surface of irradiated semiconductors. Various hetero-geneous semiconductors are used as photocatalysts such as TiO2 [5, 6], ZnO[7, 8], SrTiO3 [9, 10], WO3 [11, 12], Fe2O3 [13, 14], Bi2S3 [15, 16], CdS [17, 18],ZnS [19, 20], CdIn2S4 [4, 21], ZnIn2S4 [22, 23], N-ZnO [24], and N-TiO2 [25]Figure 7.2 shows the band edge positions of some of these semiconductors.

When a semiconducting photocatalyst is irradiated by a photon of sufficientenergy equal to or greater than the bandgap energy of the semiconductor, thefollowing actions occur after light is absorbed by the semiconductor (shown inFigure 7.3). (a) Electrons are excited from the VB to the CB of the semiconductorand a hole (+ve) is left in the VB; (b) holes that reach the surface of the semicon-ductor can oxidize adsorbed donor species Dads; (c) photoexcited electrons that

Vacuum

–2.5 –2.0

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

–3.0

–3.5

–4.0

–4.5

–5.0

–5.5

–6.0

–6.5

–7.0

–7.5

–8.0

–8.5

E NHE

SrTiO3

TiO2

ZnSCdS

CdSe

H+/H2

H2S/S2–

O2/H2O

ZnO

In2O3

Fe2O3

WO3

Bi2O3

Bi2S3

CdIn2S4

Figure 7.2 Band edge position of several semiconductors using the normal hydrogenelectrode (NHE) as a reference.


e+

+ +

– –

+ + + + + + + +

+

–

+

+–

–

– – – – – –

–

Ox–

Ox

c

a

CB

VB b

d

Red

Red+

Dads

hν ≥ Eg

+

d

Aads

Figure 7.3 Schematic representation of the light absorption by a semiconductor.

reach the surface of the semiconductor can reduce adsorbed acceptor speciesAads; generated electrons and holes (d) and (e) changes can recombine either onthe surface or in the bulk.

Thus, the materials used as a photocatalyst must satisfy several prerequire-ments with respect to bandgap energy and electrochemical properties of theprocesses expected to occur under visible-light irradiation. The expected condi-tions are given as follows: (i) suitable solar visible-light absorption capacity, (ii)bandgap potentials appropriate for H2S splitting, (iii) ability to separate photoex-cited electrons from holes, (iv) minimization of recombination of photoexcitedcharges and energy losses related to charge transport, (v) chemical stability tocorrosion and photocorrosion in aqueous environment, (vi) kinetically stableelectron transfer properties from photocatalyst surface to solvent interfaceand hence, both bulk and surface properties of photocatalyst are important.

7.3 Designing the Assembly for H2S Splitting

Taking into account the copious nature of H2S, the assembly for its decompo-sition has to be much more complex and different than that of H2O. Variousfactors related to safety issues of handling H2S gas have to be taken into accountwhile designing this setup. In this section, the unique arrangement of in-housedesigned setup for H2S splitting is explained in great detail along with the inter-action between photocatalyst and reagent system used.

7.3.1 Standardization of H2S Splitting Setup

The photo-reactor system for photocatalytic H2S splitting is shown in Figure 7.4which consists of (a) H2S gas cylinder, (b) empty trap for H2S storage, (c)calibrated water bubbler to know the rate of H2S gas, (d) CaCl2 trap, (e)photo-reactor (quartz) with water jacket, (f ) visible radiation source (xenonlamp or sunlight), (g) two NaOH traps, and (h) H2 gas collection cylinder. Theflow of H2S at the rate of 2.5 ml min−1 is maintained throughout the reaction.

7.3 Designing the Assembly for H2S Splitting 169

(a)(b)

Ar Gas

(c) (d) (e)

(f)

Lam

p

(g) (h)

(i)

(j)S1

H2S

H2S

H2

H2

Figure 7.4 Schematic of H2S splitting. (a) H2S gas generator, (b) empty trap for H2S storage,(c) calibrated water bubbler to know the rate of H2S gas, (d) CaCl2 trap, (e) photo-reactor withwater jacket, (f ) lamp to expose reactor contents, (g) two NaOH traps, (h) H2 gas collectioncylinder, (i) water bath, and (j) H2 measuring cylinder.

The H2S gas from the cylinder (Figure 7.4a) is collected into the empty trapwhich acts as a buffer (Figure 7.4b). Then, it is allowed to pass through thewater bubbler to know the rate of H2S gas (Figure 7.4c). H2S gas is then passedthrough the fused calcium chloride (CaCl2) to absorb moisture, which is shownin Figure 7.4d. This dry H2S gas is bubbled into the catalyst dispersion presentin the photo-reactor (Figure 7.4e), which is designed in our laboratory. Theknown quantity of photocatalyst powder (0.25 g) is dispersed in 200 ml aqueoussolution of 0.5 M KOH using magnetic stirrer. Initially, the whole assembly isflushed with argon gas for 20 min to remove the remains from previous use.Then H2S gas is bubbled at the rate of 3 ml min−1 for about 90 min to saturate thecatalytic dispersion with it. Then, the reactor is exposed to the Xe lamp (300 W,LOT ORIEL GRUPPE, EOROPA and LSH302, Figure 7.4f ).

The photo-reactor temperature is maintained at room temperature by thecontinuous water circulation through the outer jacket surrounding the reac-tor. Under the exposure of visible light, H2S gas is split into H2 and S. Thisgenerated gas is allowed to pass through traps filled with 1 M NaOH solution(Figure 7.4g,h). As H2S gas is soluble in NaOH solution, it ensures the collec-tion of pure H2 gas. This H2 gas is collected using inverted cylinder method(Figure 7.4i). Volume of H2 gas is measured at suitable time intervals (15 min).This gas samples are collected in Tedlar bag for qualitative analysis of collectedH2 gas by using gas chromatography.

7.3.2 Interaction of Photocatalyst and Reagent System

H2S is a weak diprotic acid with pK a values of 7.0 and 11.96 [26]. In 0.5 M KOHsolution (pH 13.5), both the dissociation reactions of H2S occur, yielding hydro-sulfide HS− is in equilibrium with H2S. On bandgap excitation, the photocatalystgenerates CB electrons (e−) and VB holes (h+). The holes oxidize HS− ions toelemental sulfur (S), releasing a proton from HS−, electrons (e−) from CB reduces


H+ ions to molecular hydrogen. The constant flow of H2S into the photoreactorhelps to maintain the equilibrium (Eq. (7.1)) and continuous evolution of H2 gas.

H2S + OH− ↔ HS− + H2O (7.1)Semiconductor → e−CB + h+

VB (7.2)2HS− + 2h+

VB → 2S2− + 2H+ (7.3)2H+ + 2e−CB → H2 (7.4)Net ∶ H2S → H2 + S (7.5)

This is an established mechanism and has been confirmed by many studies ofphotolysis of H2S and sulfide ion [26] .

7.4 Chalcogenide Photocatalysts

Many reports exist on oxide-based photocatalysts. But the sulfide-based photo-catalysts (CdS, CdIn2S4, and ZnIn2S4) are most ideal, as they are highly active,economical, and environmentally compatible. Some of the metal sulfides areattractive visible-light-driven photocatalysts because of their narrow bandgapswith VBs at relatively negative potential. CdS is an attractive material thathas been extensively used as a photocatalyst for hydrogen generation anddegradation of various organic dyes because of its optimum bandgap (2.4 eV)and other thermodynamic and electrochemical properties suitable for pho-tocatalysis [27–31]. But the use of CdS as a photocatalyst is restricted due toits photocorrosion problem; however, to overcome this, different approaches,such as chemically anchoring nanometal sulfide films in inert matrices such asSiO2[32], incorporating CdS particles into the interlayer regimes of layered metaloxides [33], coupling CdS with another stable wide bandgap semiconductor[34], and preparing CdS from an unconventional precursor followed by thermalsulfidation [35] are used. One of the approaches is to prepare solid solutionsuch as CdIn2S4, which has all the desirable qualities of CdS such as appropriateband structure and at the same time it is stable against photocorrosion dueto prolonged exposure. Its marigold-flower-like unique morphology providesporous structure, which enhances H2 production [4]. Further, the modification inmorphology can be obtained using capping agents such as polyvinylpyrrolidone(PVP) and cetyl trimethylammonium bromide (CTAB) to increase H2 produc-tion CdIn2S4 commendably [21]. CdIn2S4/graphene nanocomposite is prepared,which gives enhanced H2 production due to increased surface area as tinyCdIn2S4 nanopetals grow on thin, crumpled, two-dimensional graphene sheets.Graphene acts as an excellent electron transporter, cost–effective cocatalyst aswell as structure-directing agent in CdIn2S4/graphene nanocomposite [36] .

7.5 Limitations of Powder Photocatalysts

In the literature, many semiconductor photocatalysts (e. g., ZnS, ZnO, TiO2,N-TiO2, CdS, CdSe, and CdIn2S4) are reported, which show the excellent

7.6 Glassy Photocatalyst: Innovative Approach 171

photocatalytic activity for H2 generation by water splitting as well as H2Ssplitting [4–8, 17–21, 25]. Nanosized photocatalysts show the higher photo-catalytic activity compared to their bulk counterparts [37]. These nanosizedmaterials are synthesized by various routes such as hydrothermal, microwave,co-precipitation methods to achieve the size reduction as well as differentmorphologies. However, it is very difficult to synthesize nanoparticles havingnarrow size distribution as it shows fascinating optical, electronic, physical,and thermal properties, which are more distinctive from that of their bulkcounterparts. A capping agent is usually a strongly absorbed monolayer oforganic molecules used to aid in stabilization of nanoparticles. Capping agentsare used to control growth of nanoparticles and stabilize them from aggregation.But the use of these capping agents or surfactants reduces the processability ofthe material and sometimes hampers the photocatalytic activity. Recovery ofpowder photocatalyst after its use is also a major difficulty. As the photocatalystis dispersed in solution, it has to be separated by filtration process; however,substantial amount of sample is lost during this process.

To overcome these problems, semiconductors such as CdS [38], CdSSe [39],CdS/CdSSe [40, 41], CdSe [42], and Bi2S3 [43] can be stabilized by incorporatingthem in a borosilicate glass matrix in the form of quantum dots (QDs). Thesesemiconductor–glass nanocomposites are the new class of photocatalysts withmagical structural, morphological, and optical properties.

7.6 Glassy Photocatalyst: Innovative Approach

Glass is highly disordered state of matter and hence is amorphous in nature. It hasno long-range order, that is, unlike crystals there is no regularity in the arrange-ment of its molecular constituents on the scale larger than few times the sizeof these groups [44]. It has widespread practical, technological, and decorativeusage. It is used to give the structural transparency or translucency to the object.Glasses have unique combinations of various physical and chemical propertiessuch as transparency, chemical inertness, thermal stability, corrosion resistance,electrical insulation, extended durability, and biocompatibility. Since glass can bemolded in any shape, it has been used for photonic devices.

In general, properties of glasses are mainly dependent upon their composi-tion and technique used for manufacturing. Traditional glasses have been madeby fusion of inorganic materials such as silica sand, sodium/calcium carbonates,feldspars, borates, and phosphates. However, today we have many exotic varietiessuch as polymer glasses, splat-cooled (very high cooling rate) metallic glasses,electronically conducting glasses, non-oxide and fluoride glasses.

7.6.1 Semiconductor–Glass Nanocomposites and Their Advantages

Semiconductor–glass nanocomposite is an intimate mixture of optically func-tional semiconductor materials within a glass matrix, where the nanoparticlespossess the desirable optical properties and the glass matrix imparts process-ability. The host glasses containing the semiconductor nanoparticles play thesignificant role of providing a stable matrix, preventing the agglomeration


and decay of the nanocrystals apart from providing the confining potential. Inthese systems, very small particle sizes enhance the optical properties, whilethe matrix materials act to stabilize the particle size and growth. By growingsemiconductor nanoparticles into glass matrix, many of their interesting opticalproperties including absorption, fluorescence, luminescence, and nonlinearitycan be altered dramatically. Growing the semiconductor nanoparticles insidethe glass matrix enhances the possibilities of achieving monodispersed particlesand narrow distribution of size.

7.7 General Methods for Glasses Preparation

There are various methods for preparation of glass. In early 1940s, chemical vapordeposition (CVD) method was developed for glass preparation [45]. The processis based on thermal activation of metal halide vapors by homogeneous oxida-tion or hydrolysis to form particulate glass material “soot.” It is then followed byviscous sintering of the soot into glass. These oxidation or hydrolysis reactionsare typically triggered by oxygen plasma or an oxy-hydrogen flame. The specialglasses produced using this process are high-purity silica glasses used for variousoptical and optoelectronic devices [46], TiO2–SiO2 glasses of ultra-low thermalexpansion for telescopes [47], optical fiber for telecommunications [48], and soon. Another method for glass preparation is sol–gel process which is extensivelystudied by H. Dislich in 1971 [49]. This sol–gel process includes formation of asol, that is, colloidal dispersion in a liquid medium. This sol is further poured inmolds to form a gel by coagulation of these colloids. Then the gel is dried and fur-ther sintered at a temperature slightly above the glass transition temperature ofthe final glass [50]. As this is a low-temperature process, thermally unstable com-pounds such as non-oxide semiconductors can be incorporated in glass matrixto achieve glasses with special properties such as high optical non-linearity.

7.7.1 Glass by Melt-Quench Technique

Before the development of CVD and sol–gel processes, the melt quenching tech-nique was the only method by which the bulk glasses for practical applicationshave been produced. The process is based on the fusion of crystalline raw mate-rials into a viscous liquid at high temperature followed by quenching to form aglass in any shape [44]. This method of glass preparation can be distinguishedfrom other methods in many aspects including the available systems, size andshape of the products, and number of components. Glasses are produced fromhigh-quality, chemically pure components or from its less pure minerals depend-ing the need. Regardless of the source of component used to produce a specificglass, the glass batch materials can be divided into following five categories onthe basis of their role.

Glass former is the inevitable component in glass, and each glass has one ormore component serving as primary source of structure, which gives host glassits name. Silica (SiO2), boric oxide (B2O3), phosphoric acid (P2O5), and undercertain circumstances GeO2, Bi2O3, As2O3, Sb2O3, TeO2, Ga2O3, V2O5, BeF2,ZrF4 are some well-known glass formers. Further, Na2O, K2O, PbO, and Li2O areadded known as fluxes. Fluxes reduce the processing temperature of the glasses

7.7 General Methods for Glasses Preparation 173

as well as degrade the glass properties. Hence, alkaline earth metal oxides,transition metal oxides, alumina (Al2O3) are incorporated as property modifiers,which counteract degradation of glass properties. Colorants are used to controlcolor of final glass. Oxides of 3d transition metals, oxides of 4f rare earths, U, Au,and Ag are added as colorants, which also neutralize the effect of decolorantssuch as MnO2, As2O3, Sb2O3, Se. Fining agents such as As2O3, Sb2O3, KnO3,NaNO3, NaCl, CaF2, NaF, Na3AlF6, and number of sulfates (quantity <1%)facilitate removal of bubbles from the melt to give uniform texture and desiredtransparency to the final glass.

A large quantity of glass can be produced commercially by mixing the prede-termined amounts of pulverized crystalline raw materials and placing it in tankfurnace to be fused at high temperature. This allows producing the glass contin-uously round the clock. Production of glass by melting techniques involves foursteps: batch calculations, batch melting, fining, and homogenization as shownin Figure 7.5. In laboratory, glasses having special properties are prepared by

(a) (b)

Figure 7.5 Various steps involved in formation of glass by melt-quench technique.


melting a batch mixture in a crucible with an externally supplied heat. After melt-ing and sufficient soaking, the melted mass is casted into desired shape using themold, pressed if required, and air quenched. Further, the casted glass is usuallyannealed at glass transition temperature to remove the thermal stresses.

High flexibility of the geometry of the glass and large flexibility of compositionare the main advantages of the melt-quenching techniques over other techniquessuch as sol–gel and CVD.

7.8 Color of the Glass – Bandgap Engineeringby Growth of Semiconductors in Glass

When a glass is irradiated by white light, mobile electrons of the outer electronshells of certain ions accept energy of certain wavelength. The remaining light defi-cient in certain wavelength appears as colored light. There are various ways toachieve coloration in glass. Commercial colored glasses contain either 3d tran-sition metal ions or 4f rare earth (lanthanide) ions. In this type of glasses, thecolor is due to the ligand field effect. In semiconductor–glass nanocomposites,the same effect can be achieved by formation of semiconductor nanoparticles or“quantum dots” inside the glass matrix. A number of glasses ranging from yel-low to orange to red to black can be produced by doping the melt with variouscombinations of CdS, CdSe, CdSSe, and/or CdTe. Mostly casted glasses are col-orless and particular color is developed when they are reheated near to its tran-sition temperature for longer period. This phenomenon is known as “striking”of the glass. The most probable explanation for color generation in striked glassinvolves an electronic transition from the occupied VB to the empty CB in semi-conducting chalcogenide crystals in a glass matrix. The striking effect is specifi-cally caused by the variation in the size of the bandgap as a function of crystallitesize; hence, this effect is known as “quantum confinement.” The optical spectra ofthese glasses differ from those of the colloidal metal-colored glasses, with a sharpcut-off of transmission in the visible or near infrared, instead of the absorptionbands observed for glasses colored by gold, silver, or copper colloids. The colorand ultimately, the optical cut-off wavelength can be tailored by doping the glasswith the semiconducting material of particular bandgap. CdS is used as a dopantto achieve yellow glasses having optical cut-off wavelengths ranging from 450 to500 nm. The CdSSe/CdSe are used to achieve the optical cut-off wavelength from500 to 700 nm and CdTe is used to achieve the optical cut-off 700 nm and above.

Crystal growth of semiconductors in a silica glass matrix is a thermodynamicprocess of precipitation from a supersaturated solution and occurs in differentstages such as (i) nucleation of semiconductor clusters, (ii) normal growth coales-cence of nuclei, and (iii) cluster-induced devitrification of bulk glass process [51].

7.9 CdS–Glass Nanocomposite

CdS is well known for its use in photocatalytic water splitting [52]. It is associ-ated with photocorrosion due to prolong exposure to light. To solve this problem

7.9 CdS–Glass Nanocomposite 175

without disturbing its apt band structure, incorporation of CdS nanoparticles instable host glass matrix is the most appropriate solution. It helps to obtain unifor-mity in size of CdS NPs grown in glass matrix and minimizes photocorrosion aswell as reduces the losses during the separation of photocatalyst after use [53, 54].

In a typical procedure, the respective raw materials weighed as per thestoichiometric proportion (SiO2 (49–63%), Na2O (3–14%), K2O (15–22%), ZnO(0.5–13.5%), and B2O3 (1–8%)) are ground along with cadmium sulfide powder(0.5 wt%) in a ball mill for 6 h to obtain a homogeneous mixture. It is then meltedin a recrystallized alumina crucible using an electrically heated muffle furnaceat 1500–1600 ∘C. This glass melt is homogenized at the same temperature for3 hours. It is then quenched in air on a hot brass plate and processed immediatelyfor annealing. In the annealing process, the glass samples are heat-treated in var-ious temperature ranges near its glass transition temperature (Tg = 570–580 ∘C),for several hours in various programmed heating cycles to ensure uniformnanocrystal growth of CdS throughout the glass matrix and is cooled down toroom temperature to remove the stresses. Bright yellow color of the final glass isthe first indication of uniform growth of CdS QDs in glass matrix.

The overall X-ray diffraction (XRD) pattern illustrates (Figure 7.6A) the broadpeak at 2𝜃 ∼ 30∘, which confirms the amorphous nature of the glass. A slightshift was observed in the sample of silica (49%), which is due to the highercontent of ZnO and B2O3. The XRD pattern of host glass did not show any peak,which implies the amorphous or noncrystalline nature of the glass, while theXRD pattern of the CdS–glass nanocomposite showed a broad peak (2𝜃 = 27.9∘

(a) (b) (c) (d)

WHT 575 600 625(C)

3600 100

90

80

70

60

50

40

30

20

10

0

340032003000280026002400220020001800In

tensity

Tra

nsm

itta

nce (

%T

)

1600140012001000

800600400

20

(A) (B)

30 40

Si 49 wt% 0.5 CdS WHT

575 °C for 8 h

600 °C for 8 h

625 °C for 8 h

Si 55 wt% 0.5 CdS

Si 60 wt% 0.5 CdS

Si 60 wt% 0.75 CdS

50

2θ (°) Wavelength (nm)

60 70 80 200 400 600 800

Figure 7.6 (A) XRD of different % of SiO2 and CdS in glass, (B) UV–vis spectra of 60% SiO20.75% CdS glass is heat treated at various temperatures, (C) photograph of host andheat-treated glasses.


corresponding to hexagonal crystallite system (JCPDS PDF # 41-1049))[38]. Thisbroad peak is a clear indication of the presence of CdS QDs in the glass matrix.The UV–vis spectrum (Figure 7.6B) shows the variation in optical cut-off of glassdue to change in different striking temperatures listed in Table 7.1. The bandgapof these glasses decreases with increasing striking temperature, also reflectedin their color which changes from colorless to bright yellow (Figure 7.6C). Dueto the quantum confinement effect of CdS in the glass matrix, this variation inoptical cut-off can be observed. When the particle size is reduced below thebulk exciton diameter, quantum confinement and size dependent coulombicinteractions affect the optical transition energy [55, 56]. There is a small decreasein the bandgap of CdS–glass nanocomposite than that of pure CdS nanoparticles(2.5 eV) of the same size (2.5 nm), which is fairly explicable due to the impactof the dielectric medium such as glass [57, 58]. The steep and smooth nature of

Table 7.1 Semiconductor–glass nanocomposites with growth parameters, optical properties, andvolume of H2 evolved.

Semiconductor –glass nanocomposite

Annealingtemperature(∘C)

Time(h)

Cutoff(nm)

Particlesize(nm)

Bandgap(eV)

Volume ofH2 evolved(𝛍mol h−1)

60% SiO2 0.75% CdS WHT — 370 — 3.35 —60% SiO2 0.75% CdS 575 8 415 3–4 2.98 —60% SiO2 0.75% CdS 600 8 495 4–5 2.48 357060% SiO2 0.75% CdS 625 8 515 5–6 2.40 332052% SiO2 0.5% Bi2S3 WHT — 354 — 3.50 —52% SiO2 0.5% Bi2S3 550 8 370 3–4 3.35 554452% SiO2 0.5% Bi2S3 575 8 387 5—6 3.20 5312.452% SiO2 0.5% Bi2S3 600 8 427 7–10 2.90 5122.852% SiO2 0.6% Bi2S3 WHT — 381 — 3.25 —52% SiO2 0.6% Bi2S3 550 8 400 3–4 3.1 597052% SiO2 0.6% Bi2S3 575 8 510 5–6 2.43 558052% SiO2 0.6% Bi2S3 600 8 551 7–10 2.25 529652% SiO2 0.7% Bi2S3 WHT — 349 — 3.55 —52% SiO2 0.7% Bi2S3 550 8 381 3–4 3.25 6418.852% SiO2 0.7% Bi2S3 575 8 406 5–6 3.05 613852% SiO2 0.7% Bi2S3 600 8 496 7–10 2.50 563660% SiO2 0.4% Ag3PO4 WHT — 330 — 3.86 —60% SiO2 0.4% Ag3PO4 525 8 484 3–4 2.56 367760% SiO2 0.4% Ag3PO4 550 8 500 5–6 2.48 329760% SiO2 0.4% Ag3PO4 575 8 552 8–13 2.25 266560% SiO2 0.4% Ag3PO4 2% NH4H2PO4 510 8 489 3–5 2.53 3920.460% SiO2 0.4% Ag3PO4 2% NH4H2PO4 520 8 500 4–6 2.48 3678.660% SiO2 0.4% Ag3PO4 2% NH4H2PO4 530 8 516 6–7 2.39 3250.460% SiO2 0.4% Ag3PO4 2% NH4H2PO4 540 8 532 6–9 2.33 2914.8

7.9 CdS–Glass Nanocomposite 177

(a)

(d) (e)

1.28 Å (331)

2.05

5 1/nm

10 nm

2.032 Å (220)1.73 Å (311)

(b) (c)

Figure 7.7 TEM images of CdS QDs in CdS–glass nanocomposites (a) without heat treatmentand annealed at (b) 575 ∘C, (c) 600 ∘C, (d) 625 ∘C, and (e) SEAD pattern.

the spectra is due to uniform distribution of CdS QDs, which engage transitionbetween the VB and CB and not any impurity levels [59].

This uniform distribution of CdS QDs in CdS–glass nanocomposite canbe clearly observed in the transmission electron microscopy (TEM) images(Figure 7.7) The TEM images of the CdS in the glass matrix reveal the sphericalmorphology. The particle size of CdS QDs in glasses without heat treatment(Figure 7.7a) and annealed at 575 ∘C (Figure 7.7b), 600 ∘C (Figure 7.7c), 625 ∘C(Figure 7.7d) is observed to be 2–3, 3–4, 4–5, and 5–6 nm, respectively. Surpris-ingly, the unstriked glass shows the particle size of CdS in the range of 2–3 nm inthe glass matrix. This is observed due to the formation of CdS in the glass matrixduring quenching of the glass, itself.

The particle size increases with increasing striking temperature of glasses.The electron diffraction (ED) pattern of striked glasses (Figure 7.7e) shows verygood crystallinity of CdS as compared to unstriked glass. Since, the crystallitesize obtained in unstriked sample is very small (below the Bohr radius); the EDpattern shows very weak diffraction rings. However, the striked samples showsvery well-defined diffraction rings, indicating the growth of CdS QDs. Thed-values obtained show the existence of hexagonal CdS. The optical and TEMstudy showed that at higher temperatures, large-sized CdS QDs were formedthan at lower temperatures. It is quite well known that there is an increase inthe size of nanoparticles due to temperature and time. The crystal growth isaccelerated with temperature as per the Oswald ripening phenomenon.

The hydrogen evolved during photo-decomposition of H2S under visible-lightirradiation. In 0.5 M KOH solution of pH 12.5 (pK a = 7.0), the weak diproticacid H2S (pK a = 11.96) dissociates and maintain an equilibrium with HS− ions.The CdS QDs absorb the visible light and generate electrons (e−) and holes (h+).Due to small size and more surface area, generated e− and h+ easily transportto the surface of the catalyst and readily available for the photocatalytic activity.


The photogenerated h+ from catalyst in VB oxidizes the HS− ion to proton (H+)and disulfide (S2−) ions. The photogenerated e− in CB from the catalyst generatesthe molecular H2 by reducing the proton. In alkaline media, H2S exists in thede-protonated form as HS− and H+. The oxidized products are water solubleand hence do not interfere in the catalytic process. The maximum hydrogenproduction rate achieved was 3570 μmol h−1 (Table 7.1).

7.10 Bi2S3–Glass Nanocomposite

To overcome stability problem of Bi2S3, synthesis of Bi2S3 QDs in glass matrixhas been carried out by melt and quench method [43]. In this study, Kadam et al.developed the Bi2S3 QDs (0.5–0.7%)–glass nanocomposites and tuned the size ofBi2S3 QDs with change in striking temperature. The multicomponent glass com-position, that is, 52% SiO2, 10% Na2O, 6% MgO, 6% B2O3, 12% K2O, 10% ZnO,and 4% TiO2 has been designed. The bulk Bi2S3 synthesized by hydrothermalmethod is used as a source of Bi2S3 QDs for glass nanocomposite. This homoge-neous mixture is melted in a muffle furnace at 1100–1150 ∘C. The glass melt ismechanically homogenized at the same temperature for 2 h. After refining, theglass melt is air quenched and processed immediately for annealing (heat-treatedat 550, 575, and 600 ∘C for 8 h). At a lower concentration of Bi2S3 dissolutionof Bi2S3 in glass matrix is adequate. However, at higher concentrations of Bi2S3doping, dissolution takes longer time at melting condition, which leads todecomposition of Bi2S3 present at the surface to Bi. Hence, lower doping showsonly Bi2S3 and higher doping shows slight formation of Bi, which is quite obvious.Color of as-prepared glass is pale yellow and after heat treatment changes to darkyellow to brown depending on temperature and time of heat treatment [58]. Thecolor of glass is attributed to the growth of Bi2S3 or Bi QDs into the glass matrix.The drastic shift in the bandgap from 3.55 to 2.50 eV is due to a strong quantumconfinement effect of the Bi2S3 QDs. When the particle size is less than the Bohrradius, the materials are in the strong confinement region, and both electronand hole confinement is assumed to be dominant relative to the coulombicinteractions [60, 61]. This results in the splitting of both VB and CB into a seriesof sub-bands, and a bandgap is occurred between the top of the subband ofthe VB and the bottom of the sub-bands of the CB [62]. Room-temperatureRaman spectra of Bi2S3–glass composite show characteristic peaks at ∼98.5,114, 198, 185, and 230 which are in good agreement with that of Bi2S3 [63]. Theintensity of peak located at 114 cm−1 increases with increase in wt% of Bi2S3, i.e.,with increase in dopant percent in glass matrix. The maximum H2 generation,i.e., 6418.8 μmol h−1 g−1 was achieved for the Bi2S3–glass nanocomposite with0.7 wt% of dopant concentration striked at 550 ∘C (Table 7.1). It is observed thatwith increase in the particle size of Bi2S3 QDs, photocatalytic activity decreases,which is quite obvious. The obtained results are much higher than the previouslyreported bulk Bi2S3 semiconductor catalyst [64].

7.11 Ag3PO4–Glass Nanocomposite 179

7.11 Ag3PO4–Glass Nanocomposite

In the preceding sections, we discussed CdS–glass nanocomposites. Earlier theuse of Cd chalcogenide–glass nanocomposites was limited, such as optical andphotonic applications [65, 66]. An important breakthrough was established forCd chalcogenide–glass nanocomposites when used as a photocatalyst for pho-tocatalytic H2 production from H2S splitting [38, 52, 67]. However, due to itstoxicity, the use of cadmium (Cd) has become a major concern for commercialapplications. Moreover, the use of materials consisting of heavy metals such asCd, Hg, and Pb has been banned in many regions of the world [68] and therefore,scientists throughout the world are exploring Cd-free materials.

In this light, Patil et al. have reported semiconductor oxide silver phosphate(Ag3PO4)–glass nanocomposite, which shows similar optical properties to thatof CdS [69]. Ag3PO4 has an indirect bandgap of 2.36 eV and a direct bandgapof 2.43 eV, which make it potential candidate for visible-light-driven photocat-alytic applications in organic pollutant degradation as well as for water splitting[70, 71]. However, this material has some inherent shortcomings such as sensitiv-ity to light and solubility in aqueous solution, which make it photocorrosive [72,73]. Incorporation of Ag3PO4 in glass matrix helps to control the dimensions ofAg3PO4 particles, which can tune its physical as well as optical properties andalso resolve the issues related to photocorrosion, stability, and reusability.

Ag3PO4–glass nanocomposites are prepared by 60% SiO2, 6% Na2O, 15.5%K2O, 8% ZnO, 2% TiO2, 3% B2O3, and 5.5% BaO along with 0.4 wt% Ag3PO4(AP glasses) and APD glasses are prepared by replacing TiO2 by NH4H2PO4.The AP glass nanocomposites were striked at temperatures 525, 550, 575 ∘C for8 h and labeled as AP525, AP550, AP575, respectively (Figure 7.8A). The APDglass nanocomposites striked at temperatures 510, 520, 530, 540 ∘C for 8 h werenamed as APD510, APD520, APD530, and APD540, respectively (Figure 7.8D).

High-resolution transmission electron microscopy (HRTEM) analysis revealsthe formation of around 3–4 nm sized Ag3PO4 nanoparticles in AP525, 5–6 nmin AP550, and 8–13 nm in AP575 glass nanocomposite samples, respectively,while 4–5 nm in APD510 (Figure 7.9a,b), 6 nm in APD520 (Figure 7.9c,d), 7 nm inAPD530 (Figure 7.9e,f ), and 6–9 nm in APD540 (Figure 7.9g,h) glass nanocom-posite samples. The temperature above 575 ∘C conferred heavy precipitation ofAg3PO4 with bigger particle size in glass matrix and diminishes the transparency.

The unstriked glass shows a sharp optical spectrum at 330 nm correspondingto a bandgap of 3.66 eV. During Ag3PO4–glass nanocomposite formationinitially, the semiconductor dopant Ag3PO4 is present in ionic form in the glassmatrix. Ag+ and PO4

3− ions are randomly distributed in the quenched glasssample. When as-synthesized glass is heat-treated near to its glass transitiontemperature, crystal growth of Ag3PO4 occurs. During heat treatment process,Ag+ and PO4

3− ions come closer to form Ag3PO4 nuclei. Hence, red-shift isobserved in the optical cut-off edge with the increase in striking temperature,leading to increase in particle size and shift in the bandgap from 2.56 to 2.25 eV

(A)

(a)

(a)

APD510 APD520 APD530 APD540

Unstriked

glass AP525 AP550 AP575

(b) (c) (d)

80

60

40

Tra

nsm

itta

nce

(%

T)

Inte

nsity (

a.u

.)In

ten

sity (

a.u

.)

20

0

300 400 500

Wavelength (nm) Wavelength (nm)(B)

80

60

40

Tra

nsm

itta

nce

(%

T)

20

0

(E)

(C)

600 700 800

300 400 500

Wavelength (nm)

600 700 800

375 450

Ag3PO4

AP525AP550AP575

Ag3PO4

AP 510AP 520AP 530AP 540

525 600 675

Wavelength (nm)(F)

375 450 525 600 675

Unstriked

Unstriked

AP525

AP550

APD 510

APD 520

APD 530

APD 540

AP575

(b) (c) (d)

(D)

Figure 7.8 AP glasses of unstriked and heat-treated glasses at different temperatures (A) photograph, (B) UV–vis spectra, (C) photoluminescence spectra; APDglasses of unstriked and heat-treated glasses at different temperatures, (D) photograph, (E) UV–vis spectra, (F) photoluminescence spectra. (Reproduced withpermission from Ref. [69]. Copyright 2016, Elsevier.)

(a) 4 nm

5 nm

4 nm

7 nm

6 nm 6 nm

7 nm

9 nm

8 nm

6 nm

5 nm−1

5 nm−1

0.26 nm (210)

0.26 n

m (2

10)

5 nm−1

(220)(210)

(210)(210)

(420)

20 nm

20 nm 20 nm 5 nm

20 nm 2 nm

0.21 nm (220)

0.26 nm (210)

2 nm

20 nm

(b) (c) (d)

0.26

0.21

(e) (f) (g) (h)

Figure 7.9 HRTEM images of (a,b) APD510, (c,d) APD520, (e,f ) APD530, and (g,h) APD540 samples. Insets show the SAED patterns of respective samples.(Reproduced with permission from Ref. [69]. Copyright 2016, Elsevier.)


(Figure 7.8B). Surprisingly, a considerable anomalous hump was observed atthe wavelength of 350 nm for glass nanocomposite samples AP525 and AP550,further disappeared for glass striked at higher temperature for AP575. Thisanomalous hump observed in the spectra can be attributed to the growth ofatomic Ag, which results in deficiency of PO4

2− ions during the growth ofAg3PO4. Although the hump disappears completely at a striking temperatureof 575 ∘C, glass becomes translucent (color changes to blackish red) due to theformation of bigger sized Ag3PO4 nanoparticles in the glass matrix. The growthmechanism of nanoparticles in aqueous solution and glass is quite different. Inglass it takes place just above the transition temperature and below the softeningtemperature of glass. During heat treatment, mobility of ions increased withtemperature and time within the glass regime. The movement of ions is very slowand hence required more time (in hours) for the formation of nuclei and furthernanoparticles. In order to obtain the uniform growth of Ag3PO4 in glass matrixand to maintain high transparency, the composition of host glass was modified(APD glass) by the addition of NH4H2PO4. Homogeneous yellow-coloredtransparent Ag3PO4– glass nanocomposite samples were obtained with 2 wt%of NH4H2PO4. This excess NH4H2PO4 in the present case provides extra PO4

2−

ions to the dissolved ionic Ag+ in the glass matrix, which reduces the intensityof the anomalous hump centered at 350 nm (Figure 7.8E). The absorption edgesfor Ag3PO4–glass nanocomposites APD510, APD520, APD530, and APD540were observed to be 489, 500, 516, and 531 nm, respectively, corresponding tothe bandgap energies of 2.54, 2.49, 2.41, and 2.33 eV, respectively (Figure 7.8E).

The utmost H2 production, that is, 3677 and 3920 μmol h−1 g−1 were obtainedfor AP525 and APD510 photocatalyst samples, respectively (Table 7.1). This indi-cates lower nanoparticle sized glass samples showed higher H2 production rateand as the particle size increases H2 production rate decreases. Hence, the betterH2 production rate obtained for AP525 (3–4 nm) and APD510 (3–5 nm) photo-catalyst samples is quite justifiable.

First reason for this enhanced H2 production is that AP525 and APD510photocatalyst samples have comparatively smaller particle size of Ag3PO4 in therange of 3–4 nm and 3–5 nm, offering more exposed area for light absorption.Therefore, more number of Ag3PO4 nanoparticles would get exposed to lightirradiation owing to the formation of more charge carriers. As evidenced fromthe optical study, presence of plasmonic Ag in Ag3PO4–glass nanocompositeaccelerates the photocatalytic performance since Ag can trap electron and retardthe charge recombination [74, 75]. The photoluminescence study also shows thegreat inhibition of charge carrier recombination in the case of Ag3PO4–glassnanocomposite with lower particle size, which justifies its higher photocatalyticactivity (Figure 7.8C,F). H2 would not be formed over pristine Ag3PO4 sincethe potential of the CB of Ag3PO4 (+0.285 eV VNHE) is insufficient for protonreduction. But, in Ag3PO4–glass nanocomposites, due to the ability of Agto reduce H+, H2 is produced at Ag sites. The expected working mechanisminvolves different steps such as (i) initially, Ag nanoparticles absorb the inci-dent photon through their surface plasmon resonance (SPR) excitation; (ii)meanwhile, the Ag3PO4 also absorbs incident photon and produces electronsin CB and holes in VB; (iii) the photoexcited electrons are transferred from

7.12 Summary 183

(a) (b)

Pure Ag3PO4 (black) Ag3PO4 glass nanocomposite (yellow)

Figure 7.10 Photographs of separated photocatalysts after H2S splitting (a) pure Ag3PO4 and(b) Ag3PO4–glass nanocomposite. (Reproduced with permission from Ref. [69]. Copyright2016, Elsevier.)

CB of Ag3PO4 to the electron-deficient Ag nanoparticles, returning to originalstate [76]. Then, the electrons at the surface of Ag reduce H+, resulting information of H2. Since hydrogen overvoltage of Ag metal (−0.22 V) is morenegative than that of pristine Ag3PO4 (+0.285 eV VNHE), H2 evolution overAg is quite favorable and reasonable [76–78]. Overall, the presence of Ag inAg3PO4–glass nanocomposite keeping the electron potential negative (transferof electron from Ag3PO4 to Ag) is very important and key for H2 formationover Ag3PO4 under visible-light irradiation. When Ag3PO4 bulk powder wasdirectly exposed to light, it became black due to photocorrosion (Figure 7.10a).In contrast, Ag3PO4–glass nanocomposite retained its stability (yellow color)even after exposing to light irradiation for longer time (Figure 7.10b). This mightbe because the glass matrix holds the Ag3PO4 crystals by internal glass formernetwork, which further protects Ag3PO4 and suppresses the photocorrosion.The photocatalytic stability of Ag3PO4–glass nanocomposite was reconfirmedby reusing it for photocatalytic reaction for H2 production.

7.12 Summary

The oxidation of HS− ion to proton (H+) and sulfide (S2−) ions by photo-generated holes and reduction of this proton (H+) to molecular hydrogen byphotogenerated electrons is the main part of H2S splitting process. The choice ofphotocatalyst with appropriate band structure is the key feature of continuoushydrogen production. This chapter presented the mechanism and working ofH2S splitting process and mainly concentrates on the semiconductor–glassnanocomposites used as novel photocatalyst for this process. As alreadydiscussed in this chapter, the host glass matrix provides a stable support forcontrolling the growth of semiconductors QDs without disturbing its bandstructure. The quantum confinement effect is responsible for the magicalstructural, morphological, and optical properties of this new class of glassyphotocatalysts. Semiconductors with bandgap in the visible region, narrow


size distribution of monodispersed particles in the glass matrix will be idealcandidates for photocatalytic applications. In future, glass nanocomposites ofnew earth abundant materials such as graphene, MoS2, CuSx, WS2, NiS as wellas ternary semiconductors such as CdIn2S4, ZnIn2S4, CdxZn1−xS, CuInS2 andmany more can be prepared to utilize as photocatalysts. Also, the photocatalytichydrogen production can be enhanced by incorporation of suitable co-catalystin semiconductor–glass nanocomposite. It would be interesting to studynew charge transfer pathways between co-catalyst and semiconductor–glassnanocomposites. It would be fascinating to replace borosilicate glasses by ger-manate or phosphate glasses used as host glass matrix in semiconductor–glassnanocomposites and their effect on efficiency of overall photocatalytic processcan be studied. Designing suitable reagent systems to improve the stability andreusability of photocatalysts to get uninterrupted H2 production is an emergingarea in photocatalytic H2S as well as water splitting. Harnessing of solar energyin the form of hydrogen is unquestionably best among the most beneficial andmaturing practices. Extreme research in this field gives assurances to resolveenergy crisis faced worldwide.

Acknowledgments

Authors are thankful to Ministry of Electronics and Information Technology(MeitY), Government of India, DST and ISRO for financial support. Authorsare greatful to Dr. S. K. Apte, Dr. R. P. Panmand, Dr. S. S. Patil, Dr. S. R. Kadam,Dr. R. S. Sonawane, Dr. M. V. Kulkarni, Mrs. S. D. Naik, Dr. J. D. Ambekarand project staff, Nanocrystalline Materials and Glasses group, C-MET Pune,without them this would not be possible. Authors are also thankful to DirectorGeneral C-MET Dr N. R. Munirathanam for his encouragement.

References


2 Li, Q., Li, X., Ahmed, S.W., Al-Ghamdi, A., and Yu, J. (2015) CdS/graphenenanocomposite photocatalysts. Adv. Energy Mater., 5, 1500010.

3 Gargurevich, I.A. (2005) Hydrogen sulfide combustion: relevant issues underClaus furnace conditions. Ind. Eng. Chem. Res., 44, 7706–7729.

4 Kale, B.B., Baeg, J.O., Lee, S.M., Chang, S.J., and Lee, C.W. (2006) CdIn2S4nanotubes and marigold nanostructures: a visible-light photocatalyst. Adv.Funct. Mater., 16, 1349–1354.

5 Hernández-Alonso, M., Fresno, F., Suárez, S., and Coronado, J. (2009) Devel-opment of alternative photocatalysts to TiO2: challenges and opportunities.Energy Environ. Sci., 2, 1231–1257.

6 Tada, H., Kiyonaga, T., and Naya, S. (2009) Rational design and applica-tions of highly efficient reaction systems photocatalyzed by noble metalnanoparticle-loaded titanium(IV) dioxide. Chem. Soc. Rev., 38, 1849–1858.

References 185

7 Huang, J., Yin, Z., and Zheng, Q. (2011) Applications of ZnO in organic andhybrid solar cells. Energy Environ. Sci., 4, 3861–3877.

8 Djurišic, A.B., Chen, X., Leung, Y.H., and Ng, A.M.C. (2012) ZnOnanostructures: growth, properties and applications. J. Mater. Chem., 22,6526–6535.

9 Ouyang, S., Tong, H., Umezawa, N., Cao, J., Li, P., Bi, Y., Zhang, Y., and Ye, J.(2012) Surface-alkalinization-induced enhancement of photocatalytic H2 evo-lution over SrTiO3-based photocatalysts. J. Am. Chem. Soc., 134, 1974–1977.

10 Kienzle, D. and Marks, L. (2012) Surface transmission electron diffraction forSrTiO3 surfaces. Cryst. Eng. Commun., 14, 7833–7839.

11 Bignozzi, C., Caramori, S., Cristino, V., Argazzi, R., Meda, L., and Tacca, A.(2013) Nanostructured photoelectrodes based on WO3: applications to pho-tooxidation of aqueous electrolytes. Chem. Soc. Rev., 42, 2228–2246.

12 Liu, X., Wang, F., and Wang, Q. (2012) Nanostructure-based WO3 photoan-odes for photoelectrochemical water splitting. Phys. Chem. Chem. Phys., 14,7894–7911.

13 Barroso, M., Pendlebury, S., Cowan, A., and Durrant, J. (2013) Charge carriertrapping, recombination and transfer in hematite (𝛼-Fe2O3) water splittingphotoanodes. Chem. Sci., 2013 (4), 2724–2734.

14 Wheeler, D., Wang, G., Ling, Y., Li, Y., and Zhang, J. (2012) Nanostructuredhematite: synthesis, characterization, charge carrier dynamics, and photoelec-trochemical properties. Energy Environ. Sci., 5, 6682–6702.

15 Cademartiri, L., Guerin, G., Bishop, K., Winnik, M., and Ozin, G. (2012)Polymer-like conformation and growth kinetics of Bi2S3 nanowires. J. Am.Chem. Soc., 134, 9327–9334.

16 Yang, Q., Hu, C., Wang, S., Xi, Y., and Zhang, K. (2013) Tunable synthe-sis and thermoelectric property of Bi2S3 nanowires. J. Phys. Chem. C, 117,5515–5520.

17 Zhai, T., Fang, X., Li, L., Bando, Y., and Golberg, D. (2010) One-dimensionalCdS nanostructures: synthesis, properties and applications. Nanoscale, 2,168–187.

18 Efrati, A., Yehezkeli, O., Tel-Vered, R., Michaeli, D., Nechushtai, R., andWillner, I. (2012) Electrochemical switching of photoelectrochemical pro-cesses at CdS QDs and photosystem I-modified electrodes. ACS Nano, 6,9258–9266.

19 Feigl, C., Russo, S., and Barnard, A. (2010) Safe, stable and effective nanotech-nology: phase mapping of ZnS nanoparticles. J. Mater. Chem., 20, 4971–4980.

20 Zhu, G., Zhang, S., Xu, Z., Ma, J., and Shen, X. (2011) Ultrathin ZnS singlecrystal nanowires: controlled synthesis and room-temperature ferromagnetismproperties. J. Am. Chem. Soc., 133, 15605–15612.

21 Bhirud, A., Chaudhari, N., Nikam, L., Sonawane, R., Patil, K., Baeg, J.-O., andKale, B. (2011) Surfactant tunable hierarchical nanostructures of CdIn2S4 andtheir photohydrogen production under solar light. Int. J. Hydrogen Energy, 36,11628–11639.

22 Chaudhari, N.S., Bhirud, A.P., Sonawane, R.S., Nikam, L.K., Warule, S.S.,Rane, V.H., and Kale, B.B. (2011) Ecofriendly hydrogen production from


abundant hydrogen sulfide using solar light-driven hierarchical nanostruc-tured ZnIn2S4 photocatalyst. Green Chem., 13, 2500–2506.

23 Kale, B.B., Bhirud, A.P., Baeg, J.-O., and Kulkarni, M.V. (2017) Template freearchitecture of hierarchical nanostructured ZnIn2S4 rose-like flowers for solarhydrogen production. J. Nanosci. Nanotechnol., 17, 1447–1455.

24 Bhirud, A.P., Sathaye, S.D., Waichal, R.P., Nikam, L.K., and Kale, B.B. (2012)An eco-friendly, highly stable and efficient nanostructured p-type N-dopedZnO photocatalyst for environmentally benign solar hydrogen production.Green Chem., 14, 2790–2798.

25 Chaudhari, N.S., Warule, S.S., Dhanmane, S.A., Kulkarni, M.V., Valant, M.,and Kale, B.B. (2013) Nanostructured N-doped TiO2 marigold flowers for anefficient solar hydrogen production from H2S. Nanoscale, 5, 9383–9390.

26 Buehler, N., Meier, K., and Reber, J. (1984) Photochemical hydrogen produc-tion with cadmium sulfide suspensions. J. Phys. Chem., 88, 3261–3268.

27 Serpone, N., Borgarello, E., and Gratzel, M. (1984) Visible light induced gen-eration of hydrogen from H2S in mixed semiconductor dispersions; improvedefficiency through inter-particle electron transfer. J. Chem. Soc., Chem. Com-mun., 6, 342–344.

28 Borrell, L., Cervera-March, S., Gimenez, J., Simarro, R., and Andujar, M.J.(1992) A comparative study of CdS-based semiconductor photocatalysts forsolar hydrogen production from sulphide + sulphite substrates. Sol. EnergyMater. Sol. Cells, 25, 25–39.

29 Linkous, C.A., Mingo, T.E., and Muradov, N.Z. (1994) Aspects of solar hydro-gen production from hydrogen sulfide using semiconductor particulates.Int. J. Hydrogen Energy, 19, 203–208.

30 Murugan, A.V., Sonawane, R.S., Kale, B.B., Apte, S.K., and Kulkarni, A.V.(2001) Changes in OH-Cr-montmorillonite after heating in air and nitrogenatmospheres. Mater. Chem. Phys., 71, 98–102.

31 D. Park, J. O. Baeg, (2001) CdS photocatalyst for hydrogen production, prepa-ration therefor and method for producing hydrogen by use of the same, U. S.Patent 6 300 274 B1.

32 Akdogan, Y., Uzum, C., Dag, O., and Coombs, N. (2006) Synthesis of solidsolutions of Cd1−xZnxS nanocrystals in the channels of mesostructured silicafilms. J. Mater. Chem., 16, 2048–2055.

33 Shangguan, W. and Yoshida, A. (2002) Photocatalytic hydrogen evolutionfrom water on nanocomposites incorporating cadmium sulfide into the inter-layer. J. Phys. Chem., 106, 12227–12230.

34 Jang, J.S., Li, W., Oh, S.H., and Lee, J.S. (2006) Fabrication of CdS/TiO2nano-bulk composite photocatalysts for hydrogen production from aqueousH2S solution under visible light. Chem. Phys. Lett., 425, 278–282.

35 Jing, D. and Guo, L. (2006) A novel method for the preparation of a highlystable and active CdS photocatalyst with a special surface nanostructure.J. Phys. Chem. B, 110, 11139–11145.

36 Mahadadalkar, M.A., Kale, S.B., Kalubarme, R.S., Bhirud, A.P., Gosavi,S.W., Kulkarni, M.V., Park, C.-J., and Kale, B.B. (2016) Architecture of theCdIn2S4/graphene nanoheterostructure for solar hydrogen production andanode for lithium ion battery. RSC Adv., 6, 34724–34736.

References 187

37 Sathish, M. and Vishwanath, R.P. (2007) Photocatalytic generation of hydro-gen over mesoporous CdS nanoparticle: effect of particle size, noble metaland support. Catal. Today, 129, 421–427.

38 Kale, B.B., Baeg, J.-O., Apte, S.K., Sonawane, R.S., Naik, S.D., and Patil, K.R.(2007) Confinement of nano CdS in designated glass: a novel functionalityof quantum dot–glass nanosystems in solar hydrogen production. J. Mater.Chem., 17, 4297–4303.

39 Azhniuk, Y.M., Gomonnai, A.V., Hutych, Y.I., and Zahn, D.R.T. (2009)Evidence for formation of Se molecular clusters during precipitation ofCdSe1−xSx nanoparticles in glass. Appl. Phys. A, 95, 473–477.

40 Apte, S.K., Kale, B.B., Sonawane, R.S., Naik, S.D., Bodhale, S.S., and Das, B.K.(2006) Homogeneous growth of CdS/CdSSe nanoparticles in glass matrix.Mater. Lett., 60, 499–503.

41 Sonawane, R.S., Naik, S.D., Apte, S.K., Kulkarni, M.V., and Kale, B.B. (2008)CdS/CdSSe quantum dots in glass matrix. Bull. Mater. Sci., 31, 495–499.

42 Yan, F. and Parker, J.M. (1989) Quantum size effects in a heat-treatedCdSe-doped glass. J. Non-Crystal. Solids, 112, 277–281.

43 Kadam, S.R., Panmand, R.P., Sonawane, R.S., Gosavi, S.W., and Kale, B.B.(2015) A stable Bi2S3 quantum dot–glass nanosystem: size tuneable photocat-alytic hydrogen production under solar light. RSC Adv., 5, 58485–58490.

44 Vogel, W. (1994) Glass Chemistry, 2nd edn, Springer-Verlag, BerlinHeidelberg, Ch. 2&3.

45 Scherer, G.W. and Brinker, C.J. (1990) Sol–Gel Science, Academic Press,Boston, p. 1.

46 Zachariasen, W.H. (1932) The atomic arrangement in glass. J. Am. Chem. Soc.,54, 3841–3851.

47 Smekal, A. (1951) On the structure of glass. J. Soc. Glass Technol., 35,392–394.

48 Stanworth, J.E. (1950) Physical Properties of Glass, Oxford University Press,London, Ch 2&8.

49 Dislich, H. (1971) New routes to multicomponent oxide glasses. Angew. Chem.Int. Ed. Engl., 10, 363–370.

50 Rawson, H. (1967) Inorganic Glass Forming Systems. Academic Press London,Ch 1–16

51 Yang, J., Gao, M., Zhang, Y., Yang, L., Lang, J., Wang, D., Liu, H., Liu, Y.,and Wang, Y. (2008) Effects of annealing temperature on morphologies andoptical properties of ZnO nanostructures. Superlattices Microstruct., 44,137–142.

52 Apte, S.K., Garaje, S.N., Valant, M., and Kale, B.B. (2012) Eco-friendly solarlight driven hydrogen production from copious waste H2S and organic dyedegradation by stable and efficient orthorhombic CdS quantum dots–GeO2glass photocatalyst. Green Chem., 14, 1455–1462.

53 Li, Q., Guo, B., Yu, J., Ran, J., Zhang, B., Yan, H., and Gong, J.R. (2011)Highly efficient visible-light-driven photocatalytic hydrogen productionof CdS-cluster-decorated graphene nanosheets. J. Am. Chem. Soc., 133,10878–10884.


54 Ma, G., Yan, H., Shi, J., Zong, X., Lei, Z., and Li, C. (2008) Direct splitting ofH2S into H2 and S on CdS-based photocatalyst under visible light irradiation.J. Catal., 260, 134–140.

55 Beydoun, D., Amal, R., Low, G., and McEvoy, S. (1999) Role of nanoparticlesin photocatalysis. J. Nanopart. Res., 1, 439–458.

56 Kamat, P.V. (1995) Tailoring nanostructured thin films. Chem. Tech., 25,22–28.

57 Nehi, C.L., Liao, H., and Hafner, J.H. (2006) Optical properties of star-shapedgold nanoparticles. Nano Lett., 6, 683–688.

58 Kriebig, U., Gartz, M., Hilger, A., and Hovel, H. (1996) Mie-plasmon spec-troscopy: a tool of surface science. Fine Particles Sci. Technol. ed., 12,499–515.

59 Zhou, G., Lu, M., Gu, F., Wang, S., Xiu, Z., and Chang, X. (2004) Controlledsynthesis and optical properties of PbCrO4 nanorods and nanoparticles.J. Cryst. Growth, 270, 283–287.

60 Azhniuk, Y.M., Gomonnai, A.V., Hutych, Y.I., Lopushansky, V.V., Turok, I.I.,Yukhymchuk, V.O., and Zahn, D.R.T. (2010) X-ray irradiation-induced ion-ization of CdS1−xSexCdS1−xSex nanocrystals embedded in borosilicate glass.J. Appl. Phys., 107, 113528–8.

61 Liu, H., Liu, Q., and Zhao, X. (2007) Crystal growth and optical properties ofCdS-doped lead silicate glass. Mater. Charact., 58, 96–100.

62 Azhniuk, Y.M., Gomonnai, A.V., Hutych, Y.I., Lopushansky, V.V., Turok, I.I.,Yukhymchuk, V.O., and Zahn, D.R.T. (2010) Thermal treatment-dependentchemical composition of ternary CdS1−xSex nanocrystals grown in borosilicateglass. J. Cryst. Growth, 312, 1709–1716.

63 Trentelman, K. (2009) A note on the characterization of bismuth black byRaman microspectroscopy. J. Raman Spectrosc., 40, 585–589.

64 Kawade, U.V., Panmand, R.P., Sethi, Y.A., Kulkarni, M.V., Apte, S.K.,Naik, S.D., and Kale, B.B. (2014) Environmentally benign enhancedhydrogen production via lethal H2S under natural sunlight using hierarchicalnanostructured bismuth sulphide. RSC Adv., 4, 49295–49302.

65 Borelli, N.F., Hall, D.W., Holl, H.J., and Smith, D.W. (1987) Quantum confine-ment effects of semiconducting microcrystamtes in glass. J. Appl. Phys., 61,5399–5409.

66 Yukselici, H., Pearsans, P., and Hayes, T. (1995) Optical studies of the growthof Cd1-xZnxS nanocrystals in borosilicate glass. Phys. Rev. B: Condens. Matter,52, 11763–11772.

67 Apte, S.K., Garaje, S.N., Naik, S.D., Waichal, R.P., Baeg, J., and Kale, B.B.(2014) Quantum confinement controlled solar hydrogen production fromhydrogen sulfide using a highly stable CdS0.5Se0.5/CdSe quantum dot–glassnanosystem. Nanoscale, 6, 908–915.

68 Pickett, N.L., and Daniels, S. (2006) Commercial volumes of quantumdots: controlled nanoscale synthesis and micron scale applications. Patent,PCT/GB2006/004003, 1-7.

69 Patil, S.S., Patil, D.R., Apte, S.K., Kulkarni, M.V., Ambekar, J.D., Park, C.-J.,Gosavi, S.W., Kolekar, S.S., and Kale, B.B. (2016) Confinement of Ag3PO4nanoparticles supported by surface plasmon resonance of Ag in glass: efficient

References 189

nanoscale photocatalyst for solar H2 production from waste H2S. Appl. Catal.B: Environ., 190, 75–84.

70 Yingpu, B., Ouyang, S., Umezawa, N., Cao, J., and Ye, J. (2011) Facet effectof single-crystalline Ag3PO4 sub-microcrystals on photocatalytic properties.J. Am. Chem. Soc., 133, 6490–6492.

71 Yi, Z., Ye, J., Kikugawa, N., Kako, T., Ouyang, S., Stuart-Williams, H.,Yang, H., Cao, J., Luo, W., Li, Z., Liu, Y., and Withers, R.L. (2010) Anorthophosphate semiconductor with photooxidation properties undervisible-light irradiation. Nat. Mater., 9, 559–564.

72 Xu, J.W., Gao, Z.D., Han, K., Liu, Y., and Song, Y.Y. (2014) Synthesis ofmagnetically separable Ag3PO4/TiO2/Fe3O4 heterostructure with enhancedphotocatalytic performance under visible light for photoinactivation of bacte-ria. Appl. Mater. Interfaces, 6, 15122–15131.

73 Liu, L., Liu, J., and Sun, D.D. (2012) Graphene oxide enwrapped Ag3PO4composite: towards a highly efficient and stable visible-light-induced photo-catalyst for water purification. Catal. Sci. Technol., 2, 2525–2532.

74 Liu, L.C. and Risbud, S.H. (1990) Quantum-dot size-distribution analysis andprecipitation stages in semiconductor doped glasses. J. Appl. Phys., 68, 28–31.

75 Patil, S.S., Patil, R.H., Kale, S.B., Tamboli, M.S., Ambekar, J.D., Gade, W.N.,Kolekar, S.S., and Kale, B.B. (2014) Nanostructured microspheres of silver @zinc oxide: an excellent impeder of bacterial growth and biofilm. J. Nanopart.Res., 16, 1–11.

76 Tanaka, A., Hashimoto, K., and Kominami, H. (2014) Visible-light-inducedhydrogen and oxygen formation over Pt/Au/WO3 photocatalyst utilizing twotypes of photoabsorption due to surface plasmon resonance and band-gapexcitation. J. Am. Chem. Soc., 136, 586–589.

77 Attia, Y.A., Buceta, D., Blanco-Varela, C., Mohamed, M.B., Barone, G., andArturo López-Quintela, M. (2014) Structure-directing and high-efficiencyphotocatalytic hydrogen production by Ag clusters. J. Am. Chem. Soc., 136,1182–1185.

78 Kominami, H., Furusho, A., Murakami, S.-Y., Inoue, H., Kera, Y., and Ohtani,B. (2001) Effective photocatalytic reduction of nitrate to ammonia in an aque-ous suspension of metal-loaded titanium(IV) oxide particles in the presenceof oxalic acid. Catal. Lett., 76, 31–34.

191

8

Recent Developments in Heterostructure-Based Catalystsfor Water SplittingJ. A. Savio Moniz

University College London, Solar Energy & Advanced Materials Research Group, Department of ChemicalEngineering, Torrington Place, London, WC1E 7JE, UK

8.1 Introduction

With the ever-increasing global reliance on nonrenewable, geopolitically sensi-tive sources of energy, such as natural gas and coal, coupled with highly volatilecrude oil prices, there has never been such an urgency to secure alternativeclean, renewable energy supplies. Nearly 90% of the global energy supply isgenerated from carbon-based fuels, and efforts to develop viable routes to solarfuels are thus of critical importance, that is, via a light-driven electrochemical orphotochemical process. In particular, further understanding of the fundamentalmechanism and kinetics of the processes occurring during artificial photo-synthesis is needed in order to improve practical efficiency. The first reportedsplitting of water under solar irradiation utilized n-type TiO2 coupled with a Ptcounter electrode [1]; however, due to its wide bandgap of 3.2 eV, it can only beexcited by ultraviolet (UV) irradiation, which accounts for just 4% of the solarenergy reaching the earth [2]. Therefore, to utilize efficiently the energy of thesolar spectrum, new visible-light-responsive photocatalysts are required [3]. Theconstraints of choosing suitable photocatalysts for this process are limited tomaterials that not only possess appropriate bandgap positions that straddle theredox potentials of water splitting with a conduction band (CB) more negativethan 0 V (vs NHE (normal hydrogen electrode) at pH 0) and the valence band(VB) more positive than 1.23 V, but also exhibit appropriate surface reactionkinetics and reasonable stability in aqueous solution during irradiation [4].Thermodynamically, the water-splitting reaction is an uphill process, requiringa minimum energy of 1.23 eV because the Gibbs free energy change for thereaction is ΔG∘ = 237.2 kJ mol−1 or 2.46 eV mol−1 of H2O, and therefore requireshigh overpotentials. Nature itself demonstrates an efficient strategy to utilizesolar irradiation (near unity quantum yield) by spatially separating electronsand holes in wireless photosynthesis reactions. The process of water splittingcan be envisaged as two half reactions: water oxidation, and second, protonreduction to hydrogen fuel. Figure 8.1 summarizes the different steps and theirtypical kinetics in water splitting [5]. The four-hole water oxidation process has


192 8 Recent Developments in Heterostructure-Based Catalysts for Water Splitting

et– + O2 et

– + water

e–

O2–

H2

O2

μs μs

ht+ + et

–

4ht+ + water

h+

TiO2

μs

ps

hv

pss

Figure 8.1 Mechanism of solar-driven water splitting on nanocrystalline TiO2. (Reproducedwith permission from Ref. [5]. Copyright 2008, American Chemical Society.)

been shown to be the rate determining step during water cleavage (timescale ofseconds) and competes with recombination, which takes place on the order ofmicroseconds as determined from transient spectroscopy.

Numerous materials have been reported but only for the half reactions; eitherwater reduction or oxidation (e.g., Fe2O3 [6], WO3 [7, 8], and BiVO4) [9, 10] witheven fewer materials able to split water under visible-light irradiation [11]. It iswidely accepted that the efficiencies of suspension-based photocatalytic systemsare too low to develop commercially and there is the challenge of gas separationand suppression of back reaction. One viable option is the development of pho-toelectrochemical cells (PEC) to reduce water to H2, where the anodic reactionwill involve oxidation of water and counter electrode (e.g., Pt) can drive protonreduction to H2 under minimal applied bias [3]. The electrical bias can driveelectrons from photocatalyst to counter electrode and alleviate recombinationeffects and/or photocorrosion. Nevertheless, the current solar-to-hydrogenconversion efficiency (STH) of these systems is improving but still too low todevelop a commercial device for large-scale water splitting, thus much attentionhas been focused on the coupling a solar cell to an electrocatalyst to improveefficiency. Hence a new materials strategy is required to enhance the energyconversion efficiency of PEC water splitting. One of the key limiting factorsaffecting efficiency in artificial photosynthesis is charge recombination. If thecarriers recombine faster than the expected reactions, the energy conversionefficiency will suffer. Thus for optimal efficiency, the charge carriers should beseparated as far and as long as possible. Suppression of charge recombinationcan be attempted in a number of ways:

i) Use of scavengers (sacrificial agents), which can remove either holes or elec-trons in the system so that only one half of the water-splitting reaction canbe studied in isolation (either reduction or oxidation) [12]

ii) Variation of the morphology of the photocatalyst, which has been shown toimprove photocatalytic activity due to the increase in surface area and theshortening of charge carrier diffusion pathways to the surface [13, 14]

8.1 Introduction 193

iii) Creation of a heterojunction, whereby charge carriers are generated in onephotocatalyst and subsequently vectorially transferred to the other materialallowing for long-lived electron–hole pairs, mimicking the mechanism exhib-ited in Photosystem II during artificial photosynthesis [15].

8.1.1 Band Alignment

To describe the various band alignments commonly found in junctions, thethree main types of junction architectures will be described (Figure 8.2). Here,A corresponds to semiconductor/component A and B corresponds to semicon-ductor/component B (one should note that if they are semiconductors, they canbe either n-type or p-type). In the case of type I heterojunctions, they consist of(two) semiconductors whereby the CB of component B is higher than that of A.The VB of B is lower than that of A; therefore, holes and electrons will transferand accumulate on component A.

A type II junction relies on the transfer of photoexcited electrons from B to Adue to the more negative CB position of B. Holes can travel in the opposite direc-tion from the more positive VB of A to B, leading to all-round efficient chargeseparation and enhanced photocatalytic activity. The third type, type III, is iden-tical to type II except for the much more pronounced difference in VB and CBpositions, which gives a higher driving force for charge transfer [16].

In water splitting, when a semiconductor electrode is immersed in an elec-trolyte solution, electron transfer takes place between the semiconductor andthe electrolyte solution, which results in equilibration of the Fermi level (Ef) tothat of the redox potential of the electrolyte and thus is the basis of the semicon-ductor liquid junction (SCLJ) [17]. Electron transfer processes at the interfaceof the semiconductor/electrolyte cause band bending because electron densityis absolute deep within a semiconductor and the band positions are pinned.This band bending is more pronounced for interfaces in intimate contact butless so when the space charge layer width is greater than the particle size, in thecase of nanoparticles. Semiconductor–liquid interfaces are the most commonlyreported type of junction in water splitting, and nanostructuring of the SCLJ hasproven to be an effective approach to shorten the carrier diffusion length [18].Electron transfer is possible between the semiconductor and electrolyte when the

CB

VB

A

CB

VB

B

Type I Type II Type III

CB

VB

A

CB

VB

BCB

VB

A

CB

VB

B

Figure 8.2 Band alignment in type I, II, and III heterojunctions.


Fermi level of the semiconductor is in the appropriate position (more positive/negative) than the potential of the electrolyte to then either accept/donate elec-trons. The space-charge layer contributes to the formation of an electric field; inn-type materials (photoanodes), photoexcited holes accumulate on the surfaceof the semiconductor and are consumed in oxidation reactions, while electronsare transferred to a counter electrode via the back contact and an externalcircuit, and used in reduction reactions, such as proton reduction to H2 [19].

Recently, heterojunctions formed between two solid materials haveattracted more attention, including semiconductor–semiconductor (S–S),semiconductor–metal (S–M), and semiconductor–carbon (S–C) (carbon nan-otubes (CNTs), graphene) heterojunctions [20]. By far the most commonlyemployed heterojunction is based on an S–S architecture, usually between ap-type and an n-type semiconductor in close contact (Figure 8.3a). The result isa space-charge region at the interface and the formation of an electric field fromthe diffusion of charge carriers. This can direct the flow of electrons into the CBof the n-type material and the holes move to the VB of the p-type material, result-ing in more efficient separation, longer charge carrier lifetimes and therefore animprovement in efficiency. For S–M junctions (Figure 8.3b), a Schottky barrieris formed when a semiconductor is in close contact with a metal and the result isFermi level alignment induced by electron flow from the material with the higherFermi level to the lower level (e.g., Pt/TiO2). The metal acts as an electron trapto receive photoelectrons from the semiconductor after excitation, improvingcharge carrier separation and reducing recombination, as charge cannot flow inthe opposite direction (unlike in an ohmic contact). For S–C junctions, severaltypes of carbon species have been utilized; CNTs and graphene are the most com-monly used due to their metallic-like conductivity, high electron mobility, and

n-Typesemiconductor

h+

Ef

(a) (b)

E0

Ef Ef

Electron

VB

CB

Metal

Hole

ϕm

ϕs

p-Typesemiconductor

Semiconductor (n-type)

e–

Figure 8.3 Band bending and alignment in (a) S–S, and (b) S–M (S–C) junctions. In both casesphotogenerated charges are driven in opposite directions due to favorable differences in bandenergies and the formation of an electric field. ((a) Reproduced with permission from Ref. [21].Copyright 2013, The Royal Society of Chemistry. (b) Reproduced with permission fromRef. [22]. Copyright 2012, The Royal Society of Chemistry.)

8.2 Visible-Light-Responsive Junctions 195

high surface area, allowing for facile electron injection from the light-absorbingsemiconductor; thus a similar result to the S–M junction is obtained. The useof graphene-based junctions is described in several recent reviews, so readersare directed to these for more examples [22, 23]. Furthermore, multicomponent“sandwich” junctions (e.g., CdS–Au–TiO2) [24] have been employed to furthereffect charge separation and are more active than the bijunctions CdS/TiO2 andAu/TiO2. These may have unusual charge transfer mechanisms. Many of the indi-vidual materials used for water splitting in both suspension and electrode systemshave been covered in some detail in a number of recent reviews [11, 16, 25].

A heterojunction is a promising strategy to improve the efficiency of pho-tocatalytic water splitting through combining two or more simple materialsthat already possess appreciable visible light absorption, high efficiency, andreasonable stability. Typical materials used include oxides, nitrides, sulfides, andphosphates. Furthermore, there has been some progress in the use of complexoxides (Rh-SrTiO3 [26], PbBi2Nb2O9 [27], and Sr1−xNbO3 [28]; and oxynitridesSrNbO2N [29], LaTiO2N [30], and TiON [31]) for overall water splitting undervisible light; however, it is quite challenging to synthesize these materials. Oneof the most promising, but lesser studied oxynitrides for overall water splittingunder visible light is TaON (CBE at −0.3 V vs NHE, pH 0), which has recentlyreceived much attention through its incorporation into a heterojunction, forexample, with CaFe2O4 [32], N-doped TiO2 [33], and Cu2O [34]. The use of earth-abundant oxygen evolution catalysts (OECs) is also important for sustainablesolar fuel generation. A large overpotential (𝜂), which is the extra potentialneeded to be applied beyond the thermodynamically required value, is alwaysmandatory for fuel production due to the relatively slow kinetics of the oxygenevolution reaction (OER) [35]. An OEC functions through collection of photo-holes from the light-absorbing semiconductor, thus aiding charge separation,and also lowers the activation energy for water oxidation. The OEC usuallyself-assembles on the light absorber and can be regenerated in situ, some evenat low pH [36]. The complicated mechanism of water oxidation on a typicalOEC has been previously investigated [37]. Of the numerous earth-abundantelectrocatalysts identified, the most promising for low-cost, efficient solar fuelssynthesis contains cobalt, iron, and nickel species [38–40].

This chapter summarizes the main visible-light-driven heterojunction photo-catalysts reported within the last 5–10 years that have the greatest potential tobe used for large-scale water splitting. This chapter also summarizes the mainexperimental techniques used to measure important kinetic parameters in thesecandidate materials. The next section primarily talks about oxygen evolution pho-tocatalysts and then goes on to discuss hydrogen evolution photocatalyst-basedjunctions.

8.2 Visible-Light-Responsive Junctions

8.2.1 BiVO4-Based Junctions

BiVO4 has attracted widespread attention as a highly responsive visible-light-driven photocatalyst for water oxidation in both suspension and PEC


systems [12]; it possesses a bandgap of 2.4 eV with the VB edge (VBE) located atabout 2.4 V versus NHE (pH 0), which provides sufficient overpotential for pho-toholes to oxidize water. However, the CB edge (CBE) is located just under thethermodynamic level for proton reduction to H2 [41]. A 4.9% STH efficiency wasreported for a gradient-doped W:BiVO4 photoanode connected to a Si solar cellin tandem configuration for water splitting [42]. Recently, a record photocurrentof about 5.9 mA cm−2 at 1.23 V versus reference hydrogen electrode (RHE) wasobtained from a FeOOH—NiOOH/carbon quantum dots/BiVO4 compositeelectrode [43]. Numerous groups have attempted to improve the STH conversionefficiency of BiVO4, most notably through coupling with other semiconductorsin an S–S junction. Progress in developing BiVO4 photoanodes for watersplitting has recently been covered in a comprehensive review [44], which hasattempted to summarize the recent key results in junctions containing BiVO4(Table 8.1) .

Table 8.1 Summary of recent key advances in BiVO4-based heterojunction photoanodes.

Judd junction Synthetic methodMaximumphotocurrent

IPCE valueat 400 nm References

BiVO4/WO3 Spincoating/solvothermal

∼1.6 mA cm−2

(1 V vs Pt CE)31% [45]

BiVO4/WO3 Polymer-assisteddeposition

1.74 mA cm−2

(0.7 V vs Ag/AgCl)37% [46]

BiVO4/SnO2/WO3 Spin coating 2.5 mA cm−2

(1.23 V vs RHE)∼40% [47]

Co—Pi/BiVO4/WO3 Glancing-angledeposition

∼3 mA cm−2

(1.23 V vs RHE)60% [48]

Co—Pi/W:BiVO4 Spray pyrolysis ∼3.6 mA cm−2

(1.23 V vs RHE)— [42]

FeOOH—NiOOH/(W,Mo)—BiVO4/WO3

Oblique-angledeposition/dropcasting

5.35 mA cm−2

(1.23 V vs RHE)>90% [49]

FeOOH—NiOOH/carbondots/BiVO4

Electrodeposition 5.9 mA cm−2

(1.23 V vs RHE)>80% [43]

Co—Pi/BiVO4/ZnO Hydrothermal/spraypyrolysis

∼3 mA cm−2

(1.23 V vs RHE)∼47% [50]

Co—Pi/Mo—BiVO4 Spin coating 1.1 mA cm−2 (1.1V vs Ag/AgCl)

∼65% [51]

Co—Pi/Mo—BiVO4 Spin coating ∼2.5 mA cm−2

(1.23 V vs RHE)— [52]

FeOOH/BiVO4 Electrodeposition ∼2 mA cm−2

(1.23 V vs RHE)∼45% [53]

FeOOH—NiOOH/BiVO4 Electrodeposition/dropcasting

∼4 mA cm−2

(1.23 V vs RHE)60% [54]

Ni—B/BiVO4 Spin coating ∼1.25 mA cm−2

(1.23 V vs RHE)30% [55]


8.2.1.1 BiVO4/WO3

WO3 is a stable low-cost n-type semiconductor and the position of its CB edge(0.42 V vs RHE) is suitable for accepting electrons from the CB of BiVO4 [52,56]. Coupling BiVO4 with WO3 and inserting an SnO2 layer in-between resultedin enhanced photocurrent (about 2.5 mA cm−2 at 1.23 V vs RHE), significantlyhigher than BiVO4/WO3 and the individual materials when tested in carbonateelectrolyte [47]. The Fermi level of SnO2 is located between those of BiVO4 andWO3, while the difference in the CB positions of these semiconductors allows anelectron cascade pathway from BiVO4 to SnO2 to WO3 to the counter electrode.BiVO4/WO3 nanorod array electrodes grown by Grimes [45] demonstratedimproved IPCE (incident photon to current efficiency) at 420 nm, increasingfrom 9.3% to 31% compared to planar films due to facile electron transfer fromBiVO4 to WO3. Lee et al. [46] have demonstrated that a composite electrodeconsisting of BiVO4 coupled to four WO3 layers exhibited a 74% increase inphotocurrent relative to bare WO3, and 730% relative to bare BiVO4, withalmost a fourfold increase in IPCE at 425 nm (about 37%). This is due to electrontransfer from BiVO4 to WO3 and this strategy overcomes the poor chargetransport observed for BiVO4. Very recently, Domen demonstrated that triplejunction Co—Pi/BiVO4/WO3 nanorod photoanodes produce a photocurrentof about 3 mA cm−2 at 1.23 V versus RHE and 60% IPCE at 400 nm, againdue to electron transfer to WO3, but further enhanced via light trapping byWO3 nanorods and hole transfer to Co—Pi for efficient water oxidation [48].Furthermore, BiVO4/CuWO4 heterojunctions overcome the stability issuesposed by WO3 and can exhibit photocurrents as high as 2 mA cm−2 at 1 V (vsAg/AgCl) [57]. A BiVO4/WO3 “double-deck inverse opal junction” synthesizedby swell-shrinking around polystyrene spheres exhibited a photocurrent densityof about 3.3 mA cm−2 at 1.23 V versus RHE, enhancing the poor charge carriermobility of BiVO4 by combining it with a WO3 skeleton and increasing thesurface area through the inverse opal structure [58]. One of the highest perform-ing BiVO4-based junctions consists of a triple junction FeOOH—NiOOH/(W,Mo)-BiVO4/WO3 helix nanostructure that achieved 5.35 mA cm−2 at 1.23 Vversus RHE and near 100% IPCE at 420 nm, caused by a synergy between lighttrapping, enhanced charge separation, and high surface area of the WO3 helix[49]. Diffusion of tungsten from the WO3 layer into BiVO4 creates a gradientdoping of W, beneficial for charge separation, and additionally could introduce asmoother interface between BiVO4 and WO3.

8.2.1.2 BiVO4/ZnODue to its intrinsic high electron mobility, ZnO has been employed as an electronacceptor in a wide range of heterojunctions for water splitting, and its couplingwith BiVO4 has resulted in several mechanisms for enhanced activity. We firstdemonstrated that nanoparticulate BiVO4/ZnO nanowire photoanodes couldexhibit high photocurrents under visible-light irradiation (about 2 mA cm−2

at 1.23 V vs RHE) and the introduction of a Co—Pi surface OEC can improvethe photocurrent to about 3 mA cm−2 and IPCE to 47% at 410 nm (Figure 8.4a)[50]. For this triple junction, we proposed that electron transfer from BiVO4


D

BiVO4

Ev

Ec

V vs NHE

(a)

(b)

A

Ec′

Ev′

ZnOA–

e

h

A:acceptor, D:donor

0 eV

2.16 eV

–0.94 eV (400 nm)

–0.16 eV (~530 nm)

D+

h

hv

hv hv

hv hv

Co–Pi

BiVO4

ZnO

High-energy

Visible light

h+

e– e– e–

h+h+

e

: –O–Si–O–

Figure 8.4 (a) Design strategy of a Co—Pi/BiVO4/ZnO heterojunction by Moniz et al. [50]involving (i) increased light absorption and charge generation in both BiVO4 and ZnO inconjunction with light-trapping effect of the nanorods, (ii) electron injection into ZnOnanorods followed by prompt electron transport along ZnO nanorods, and (iii) simultaneoushole transfer to Co—Pi for efficient water oxidation; (b) charge transfer mechanism proposedby Fu et al. [59] involving spatial transfer of visible-light-excited high-energy electrons fromBiVO4 to ZnO. (Reproduced with permission from Ref.[59]. Copyright 2014, American ChemicalSociety.)

to ZnO rods followed by hole transfer to Co—Pi is the most likely mechanism,as the flat-band position (Efb) of BiVO4 was found to be more negative thanZnO rods and hence provides a sufficient driving force. The efficiency wasfurther improved due to the light-trapping effect of vertically aligned ZnO rods(Figure 8.4).


Fu et al. [59] suggested that the mechanism of charge transfer in BiVO4/ZnOjunctions is even more complex. In fact, they report spatial transfer ofvisible-light-excited high-energy electrons from BiVO4 to ZnO on the basis ofEPR and photocurrent action spectra, while the insertion of a silicate bridgeimproved charge transfer between the individual materials (Figure 8.4b). Here,the energy level of the high-energy electrons is higher than that of the protonreduction potential for H2 production from water, and represents a quite unusualelectron transfer process in heterojunction photocatalysts.

8.2.1.3 BiVO4/TiO2

Very recently, BiVO4/TiO2 junctions for water splitting are receiving more inter-est, possibly due to the ambiguity of the CB position of BiVO4. We reported thatspin-coated BiVO4/TiO2 junctions show a fourfold enhancement in activity forwater oxidation compared to the pure BiVO4 due to improved charge transferfrom BiVO4 to TiO2 and the unique ultrafine morphology of our materials [60].Fu et al. [61] reported visible-light-excited high-energy electrons in BiVO4 cantransfer to the CB of TiO2, probed using PEC measurements and surface pho-tovoltage spectroscopy (SPS). Ta-doped TiO2 was also found to improve chargecarrier collection in BiVO4, resulting in a photocurrent density of 2.1 mA cm−2

at 1.23 V versus RHE [62].

8.2.1.4 BiVO4/Carbon-Based MaterialsThere are many reports on the coupling of BiVO4 with carbon-based materials(nanotubes, graphene) to enhance their activity for water splitting. For instance,Kudo et al. [63] reported the coupling of reduced graphene oxide (RGO) toBiVO4, which yielded a near tenfold enhancement in PEC activity compared withpure BiVO4 under visible-light illumination. This improvement was attributed tothe longer electron lifetime of excited BiVO4 as the electrons are injected to RGOinstantly at the site of generation, leading to a significant reduction in chargerecombination. W-doped BiVO4 with graphene surface modification displayedenhanced activity for PEC water splitting, at about 1 mA cm−2 at 1.23 V versusRHE, due to fast hole transfer to the graphene surface, which catalyzes wateroxidation [64]. Nevertheless, the total photocurrents in these systems are muchlower than the aforementioned BiVO4/metal oxide junctions and suggest morework is needed to improve their activity. However, most carbon-based materialsare extremely cheap in comparison to most semiconductors and precious metals,and therefore if made more efficient could be considered a viable option.

8.2.2 Fe2O3-Based Junctions

Hematite (α-Fe2O3) has a relatively narrow bandgap of about 2.1 eV that canabsorb a large portion of visible photons from sunlight. Furthermore, it ischeap, earth-abundant, nontoxic, and stable [65]. Although its VBE potential is∼1.2 V more positive than required for water oxidation thermodynamically, itsperformance is still hindered by high charge recombination in the bulk, low con-ductivity, poor kinetics for water oxidation at its surface, and short hole diffusionlengths [66, 67]. As such it has almost never been used in suspension-based


water-splitting systems. A review of hematite-based photoanodes has recentlybeen published and readers are directed to this for current progress in thefield [65]. Coupling α-Fe2O3 to other metal oxide semiconductors has seldombeen reported; however, Sivula et al. [68] reported the use of a WO3 hostscaffold to improve light absorption and increase the surface area of α-Fe2O3.Higher activity for water oxidation was exhibited by the α-Fe2O3/WO3 electrodebecause more of the α-Fe2O3 is closer to the hematite/electrolyte interface,allowing a greater fraction of the photogenerated holes to transfer to the SCLJand participate in water oxidation.

Ti-doped Fe2O3/SnO2 junction photoelectrodes exhibited a twofold increasein electron lifetime at 0.13 V compared to Ti:Fe2O3 [69]. In a separate study,α-Fe2O3/ZnFe2O4 composite electrodes were grown through surface treatmentof Fe2O3 with Zn2+ ions and exhibited enhanced photocurrents caused byelectron transfer from ZnFe2O4 to α-Fe2O3 and hole transfer in the oppositedirection [70]. We have recently observed that coupling α-Fe2O3 nanoparticlesonto TiO2 results in enhanced water oxidation due to facile electron transferfrom TiO2, supported by DFT calculations, which revealed a resultant increasedsurface reactivity on TiO2 [71]. Coupling α-Fe2O3 to graphene nanoplates(GNP, 0.2 wt%) allowed for efficient water oxidation (2.5 mA cm−2 at 0.75 V vsSCE) under visible-light irradiation, the mechanism of which was attributed toefficient charge transfer at the semiconductor/electrolyte junction, a redshiftin the absorption spectra of the Fe2O3–GNP compared to pristine α-Fe2O3,and improved conductivity of α-Fe2O3 due to the introduction of conductivegraphene [72]. Indeed, this relatively simple, low-cost strategy of coupling Fe2O3to carbon-based materials reveals great potential to increase the efficiency forwater-splitting applications. The coupling of α-Fe2O3 to other metals/metallicspecies as charge collectors has been less reported on account of the rise ofcarbon-based conductors [73].

However, Wang et al. [74] synthesized an α-Fe2O3/TiSi2 nanonet core–shellheterojunction for water splitting (Figure 8.5). The work functions predictohmic contact between n-type Fe2O3 and highly conductive metallic TiSi2,which led to excellent performance for PEC water splitting compared to planarα-Fe2O3 – almost 50% IPCE at 400 nm. Likewise, the introduction of Al-dopedzinc oxide (AZO) into α-Fe2O3 to form a Fe2O3–AZO junction improvedelectron collection to the counter electrode while increasing the surface areaof the photoanode, resulting in almost double IPCE at 400 nm. However, AZOwas found to have poor stability under the experimental condition without theALD-grown α-Fe2O3 layer [75]. Furthermore, by forming a simple p–n Fe2O3homojunction via atomic layer deposition (ALD), the onset of photocurrentwas shifted by almost 200 mV compared to bare n-type Fe2O3 [76], whilean Si/α-Fe2O3 dual-absorbing heterojunction exhibited an even lower onsetpotential of 0.6 V versus RHE, representing a cathodic shift of approximately400 mV, achieved in part, by the utilization of low-energy photons by the Sinanowires [77].


TiS

i2 n

anonet

Fe2O3e–

h+

H2O

L < Dτ(a)

(b) (c)

5 nm200 nm

TiSi2

Fe2O3

TiSi2

(330)–

(120)(110)

(120)

–

Fe2O3

0.250 nm

0.145 nm

Figure 8.5 (a) Schematic illustration of the TiSi2/α-Fe2O3 nanonet photoanode. Efficientcharge collection is achieved when the hematite thickness is smaller than the charge-diffusiondistance. (b) TEM image of the TiSi2 core/hematite shell nature. (c) High-resolution (HR) TEMimages of the junction. (Reproduced with permission from Ref. [74]. Copyright 2011, AmericanChemical Society.)

8.2.3 WO3-Based Junctions

WO3, akin to Fe2O3, has received considerable attention as a potential photoan-ode material for PEC water splitting as it possesses a bandgap in the visible range(2.4 eV), a long-hole diffusion length (150 nm) compared to α-Fe2O3 (∼4 nm) andhas a VB position sufficiently more positive than the potential for water oxidation[78]. However, similar to hematite, the onset potential for water oxidation is rel-atively high (about 0.4 V) compared to other photoanodes. There are numerousreports of its use as an efficient electron collector when coupled to other semicon-ductors that have a more negative CB position, but relatively few reports of its usein junctions where WO3 is the main light absorber. One of the main drawbacksof using WO3 as a photoanode is its thermodynamic instability toward anodicphotocorrosion and the formation of peroxo species on its surface that competeswith O2 production. Recently, the combination of a 5 nm ALD-deposited Al2O3


overlayer with WO3 was shown to suppress the formation of surface peroxospecies through decreasing electron trapping while promoting hole trapping,facilitating water photooxidation and retarding the recombination process [79].The use of WO3 coupled with RGO has been shown to increase the PEC activityfor water splitting by lowering charge carrier recombination at the particleinterface of WO3, facilitated by the highly conductive RGO substrate [80].A photocurrent of 1.1 mA cm−2 at 1 V versus Ag/AgCl was observed; how-ever, the onset potential (about 0.3 V vs Ag/AgCl) did not shift. Furthermore,electrochemical impedance spectroscopy (EIS) revealed that in the low biasregion (0.4 V), the RGO does not improve charge separation; it is only athigher potentials (>1 V) that charge transfer is improved. Domen showed thatPtOx/WO3 in a suspension system was able to readily evolve oxygen undervisible-light irradiation but the addition of small amounts of MnOx, CoOx,RuO2, or IrO2 as secondary cocatalysts resulted in better activity; evidenced byan apparent quantum yield of 14.4% at 420 nm with RuO2 [81]. Recently, a layerof vertically aligned WO3 nanorods coated with a conformal layer of TiO2 wasfound to exhibit an unusual electron transfer process, where photogeneratedelectrons move from the WO3 layer into TiO2 [82]. The findings, supportedby computational studies, could open the way for more studies into junctionswith WO3.

8.2.4 C3N4-Based Junctions

Metal-free graphitic carbon nitride (g-C3N4) is the most stable allotrope of car-bon nitride and has attracted much attention in recent years for photocatalytichydrogen evolution from water in suspension systems after the breakthroughreport by Wang et al. [83] It has the appropriate electronic structure: a bandgapin the visible region (∼2.7 eV) and a CB position sufficiently negative to driveproton reduction to hydrogen. In the presence of sacrificial reagents, it canreduce or oxidize water with and without cocatalysts; very recently we reported ag-C3N4 synthesized from urea which exhibited a hydrogen evolution rate (HER)of nearly 20,000 μmol h−1 g−1 under full-arc irradiation and an internal quantumyield of 26.5% at 400 nm, a direct result of its more negative CB position andimproved exciton distribution over its structure [14]. Numerous cocatalysts havebeen incorporated with g-C3N4 to achieve better performance, such as RuOx,Rh, Ir, Pt, Au, Pd, as well as through doping with fluorine and sulfur [84]. Oxygenevolution is more challenging because the two electron oxidation to peroxideis more favorable on its surface. Several heterojunctions incorporating C3N4have been reported for H2 evolution, for example, g-C3N4–SrTiO3:Rh evolved223.3 μmol h−1 of H2 under visible-light irradiation, over three times that ofSrTiO2:Rh [85]. Carbon-based electron acceptors also facilitate more efficientcharge separation [86], for example, g-C3N4/graphene composites exhibiteda threefold enhancement in photocurrent and H2 production under visiblelight compared to bare g-C3N4 [87]. Here, graphene sheets act as conductivechannels to efficiently separate the photogenerated charge carriers, while asimilar mechanism was proposed in a red phosphorus/C3N4 junction [88].As a visible-light absorber, C3N4 has been coupled to many wide-bandgap


350

200

150

100

50

0

1st

0 1 day 5 days 30 days 200 days

8.4

H2

O2 O2 O2 O2

8

6

4

2

0

H2 H2 H2

μmol h−1

4.1μmol h−1

5th

Cycle number Wavelength (nm)

Gases e

volu

tion (

μmol)

Absorb

ance (

a.u

.)

Absorb

ance (

a.u

.)

Quantu

m e

ffic

iency (

%)

30th 200th 400 500 600 700 800

400


(c) (d)

E v

s E

v (e

V) V

vs V

RH

E (V)

450 500 550 600 650

C3N4

CDots-C3N4

H2O2/H2O

H+/H2

O2/ H2O

–4.0 –0.5

0.0

0.5

1.0

1.5

2.0

2.5

–4.5

–5.0

–5.5

–6.0

–6.5

–7.0VB

CB

CDots-C3N4

Figure 8.6 (a) UV–vis absorption spectra of C3N4 (black curve) and CDots-C3N4 (red curve)catalysts; (b) Band structure diagram for CDots-C3N4; (c) H2 and O2 production from waterunder visible-light irradiation (𝜆> 420 nm) catalyzed by CDots-C3N4; (d) Wavelength-dependent quantum yield (red dots) of water splitting by CDots-C3N4. (Reproduced withpermission from Ref. [99]. Copyright 2015, AAAS.)

semiconductors to improve solar harvesting. A 50 wt% C3N4/TiO2 junctionwas found to double H2 evolution compared to pure C3N4 under visible-lightirradiation [89]. Since the CBE of g-C3N4 (−1.12 eV) is more negative than thatof TiO2 (−0.29 eV) the photoinduced electrons on g-C3N4 transfer easily to TiO2before recombination [90]. Furthermore, g-C3N4-coated SrTiO3 also displayeda high HER of 440 μmol h−1 g−1 under visible-light irradiation as a result ofelectron transfer from the CB of C3N4 to that of SrTiO3 followed by migration tothe Pt cocatalyst [91]. A ZnO photocatalyst hybridized with graphite-like C3N4via a monolayer-dispersed method exhibited a fivefold increase in photocurrentunder UV irradiation and visible-light-driven photocurrent, along with sup-pression of ZnO photocorrosion [92]. The enhancement under UV irradiationwas due to high separation efficiency of photoinduced holes from ZnO to theHOMO of C3N4, while under visible-light irradiation, the electron excited fromthe HOMO to the LUMO of C3N4 may directly inject into the CB of ZnO. About1.1% NiS/C3N4 composites have also been shown to have appreciable activity forH2 production from water [93, 94], while similarly, the use of NiS2 as a cocatalystfor H2 production on C3N4 has recently been reported [95]. The coupling ofhigh-surface-area C3N4 with Ta3N5 also resulted in enhanced visible-light-drivenH2 evolution [96] and g-C3N4-CdS QDs composites improved the hydrogen


production over bare g-C3N4 by over nine times due to in situ electron transferto CdS; however, the photocurrent recorded was actually quite small [97], with asimilar mechanism reported for C3N4/CdS core–shell nanowires [98]. By far themost significant recent research involving carbon nitride composites is in thearea of pure water splitting. By adding appropriate cocatalysts such as carbonquantum dots (Figure 8.6) [99], or Co/Pt [100], H2 and O2 can be evolved in a2 : 1 ratio without the need of sacrificial reagents; however, the energy conversionefficiency of these systems requires improvement.

In general, research on C3N4-based junctions has concentrated on H2 evo-lution from suspension systems as the O2 evolution and photoanodic PECactivity are still low [101]; recent work in this field has attempted to improve thephotocurrent obtainable from C3N4 using junctions if suitable electrodes canbe synthesized. Examples include a C3N4 nanosheet/N-doped graphene/layeredMoS2 triple junction [102], where photoelectron transfer via highly conductivegraphene to MoS2 improves the photoresponse. Furthermore, a 3D branchedCoOx/C3N4/WO3 junction exhibited the highest anodic photocurrent for aC3N4-based electrode (about 1.5 mA cm−2 at 1.23 V vs RHE) [103]. This archi-tecture utilized WO3 as an electron acceptor and CoOx as a surface oxidationcatalyst, thus proving that a high anodic photocurrent from a C3N4 absorber isachievable.

8.2.5 Cu2O-Based Junctions

Cuprous oxide (Cu2O), a p-type semiconductor with a direct bandgap of ∼2 eV,exhibits a maximum theoretical photocurrent of ∼−15 mA cm−2 and 18% STHefficiency under AM1.5 light. Within the last 5–10 years, much research hasfocused on utilising Cu2O for photoelectrochemical hydrogen production. Oneof the main limiting factors in the use of Cu2O is its poor stability becausethe redox potentials for the reduction and oxidation of monovalent copperoxide lie within the bandgap. To address these issues, Grätzel et al. reporteda Cu2O/ZnO/Al2O3/TiO2/Pt electrode (Figure 8.7) capable of photocurrentsas high as −7.6 mA cm−2 at 0 V versus RHE with improved stability due to theprotective nature of TiO2 and high conductivity of ZnO/Al2O3 (AZO) [105]. Thiswork catalyzed further research into junctions that can not only protect Cu2O butalso enhance its activity. The same group reported a Cu2O/n-AZO/TiO2/MoS2+xheterojunction photocathode that exhibited improved stability in harsh acidicenvironments, returning a performance of −5.7 mA cm−2 at 0 V versus RHE atpH 1.0 [104]. Figure 8.8 shows the schematic representation of relative bandposition for the Cu2O/AZO/TiO2/MoS2+x photocathode after equilibration inthe dark, assuming band edge pinning at the interfaces and taking the built-inpotentials at the interfaces equal to the difference in Fermi levels. Electronscannot flow from TiO2 to Cu2O due to the high potential energy barrier at then-AZO/p-Cu2O interface. UV photons drive the photodeposition of MoS2−xHER onto the TiO2 surface, and photoelectrons can travel from the overlayersthrough to the Cu2O VB at higher applied potentials.

The coupling of a surface protected Cu2O with a MoS2 HER catalyst anda Ni–Mo catalyst layer resulted in the highest reported photocurrent for aCu2O-based photocathode, at −6.3 mA cm−2 at 0 V versus RHE in 1 M KOH


CB

CB CB

VB

VB VB

Cu2O

+0.00

Cu2O + H2O + 2e– <–––> 2Cu + 2OH–

2Cu2O + H2O + 2e– <–––> 2Cu2O + 2OH–

O2 + 4H+ + 4e– <–––> 2H2O

+0.47

+0.60

+1.23

E/V versus NHE

ZnO TiO2

2H+ + 2e– <–––> H2

Figure 8.7 Band alignment in a Cu2O photocathode using TiO2 and Al-doped ZnO protectionlayers. (Reproduced with permission from Ref. [104]. Copyright 2014, Nature Publishing Group.)

–2.0

–1.0

0.0

HER

70 nm

EVB

ECB

Ef

LiquidMoS2+xTiO2n-AZO

400 nm 20 nm 100 nm 50–100 nm

p-Cu2O

PotentialV versus RHE

1.0

2.0

3.0

Figure 8.8 Band energy positions for the Cu2O/AZO/TiO2/MoS2+x photocathode biased at 0 Vversus RHE in the dark, assuming pinning of the band edges of the semiconductor at theinterfaces. (Reproduced with permission from Ref. [105]. Copyright 2011, Nature PublishingGroup.)


electrolyte [106]. This appears to be the first report of the stability of MoS2in highly basic conditions, thus demonstrating significant potential to replaceplatinum as a cost-effective HER catalyst. Coupling of Cu2O to n-type WO3 is anefficient strategy to improve stability. For example, a Cu2O nanowire photocath-ode modified with a thin film of NiOx coupled to a WO3 nanosheet photoanodeexhibited a photocurrent density of −4.98 mA cm−2 at −0.33 V versus NHE andgood stability over 20 min illumination time [107], while a WO3/Cu2O NWalso exhibited improved photocurrent for water reduction [108]. Similarly, aCu/Cu2O/CuO composite electrode also exhibited improved activity [109],while Cu2O/TiO2 NW p–n junction exhibited a high photocurrent of 4 mA cm−2

at 1 V versus SCE in Na2SO4 electrolyte and was highly active under visiblelight for rhodamine B dye degradation [110] due to facile electron injectionfrom Cu2O into TiO2 [111]. Cu2O, protected by an ultrathin carbon sheath, wascoupled to TaON nanorods to yield a p-Cu2O/n-TaON junction photoanode,exhibiting an IPCE of 59% at 400 nm, a photocurrent of 3.06 mA cm−2 under1 sun illumination at 1.0 V versus RHE and retained about 87% of the initialactivity after 60 min irradiation [34]. The mechanism for the enhancement wasattributed to the fast transfer of photogenerated electrons from Cu2O to TaONcoupled with the high conductivity and protection from the electrolyte by thecarbon sheath. In a suspension system, Cu2O can actively photoreduce protonsto H2 under visible light; however, the stability proves to be a major limitingfactor over its efficiency. The coupling of Cu2O to RuOx was found to promoteCO2 photoreduction into CO and resulted in a remarkable enhancement instability [112]. Adding RuOx nanoparticles on Cu2O led to a twofold increasedyield of long-lived electrons, using transient absorption spectroscopy (TAS),indicating partially reduced electron–hole recombination losses, and correlateswith an approximately sixfold increase in the yield of CO2 reduction to CO [113].Likewise, coupling of Cu2O to RGO dramatically increased the activity for CO2photoreduction and the photocurrent of the junction was nearly double that ofthe bare Cu2O photocathode. The improved activity together with the enhancedstability of Cu2O was attributed to the efficient charge separation and transferto RGO as well as the protection function of RGO [114], while Tran et al. alsoobserved electron collection by RGO and enhanced stability when coupled toCu2O for photocatalytic hydrogen production [115]. Very recently an unusualmechanism was proposed for enhancement in activity and stability for CO2photoreduction by a Cu2O/carbon quantum dot heterojunction [116]. Photo-generated holes in Cu2O transfer to the surface of the CQDs for water oxidation;however, an additional photoexcitation mechanism in the CQDs followed byelectron transfer to Cu2O was put forward. Therefore, these examples representpromising solutions in addressing the problem of inherent poor stability inCu2O-based water-splitting photocatalysts; however, more work is required tounderstand the charge transfer mechanisms involved. The relationship betweenthe active components in CuO/Cu2O/TiO2 heterojunctions was realized byWang et al. [117]. Surprisingly, their CuO—TiO2 does not initially catalyze thereduction of water, but instead undergoes a continuous in situ restructuringprocess of reduction to Cu2O. A CuO/Cu2O/TiO2 triple junction was observedvia TEM, which evolves H2 faster than P25 TiO2 under solar irradiation. Hence,

8.3 Visible-Light-Driven Photocatalyst/OEC Junctions 207

their study reveals that Cu2O is the active component in these specific junctions,yet is more suited to an in situ restructuring process to inhibit postsynthesisoxidation.

8.3 Visible-Light-Driven Photocatalyst/OEC Junctions

8.3.1 BiVO4/OEC

The loading of OECs, such as cobalt phosphate (Co—Pi) [39, 118, 119] onBiVO4 helps to improve the kinetics for water oxidation and suppresses theaccumulation of holes at the photoanode/electrolyte interface, which can oftenresult in photocorrosion. It has been widely reported that BiVO4 suffers frompoor electron conductivity coupled with poor kinetics for water oxidation andsignificant bulk recombination [120]. Co—Pi OEC prepared by photodepositiononto BiVO4 has been shown to exhibit superior performance for water oxidationand exhibit favorable cathodic shifts in onset potential [51, 121]. Photodeposi-tion ensures the selective deposition of Co—Pi where photogenerated holes areavailable, resulting in a uniform coating of the catalyst. BiVO4 photoanodes withabout 30 nm Co—Pi OEC layer grown by Abdi and van de Krol [122] exhibitedphotocurrents of ∼1.7 mA cm−2 at 1.23 V versus RHE, more than double thatof bare BiVO4. Similarly, Pilli et al. showed that the photocatalytic activity ofMo-doped BiVO4 electrodes doubles with Co—Pi loading [51] and the activityof W-doped BiVO4 [123] and SiO2/BiVO4 [124] photoanodes have resultedin similar enhancements after addition of Co—Pi. A FeOOH OEC coupled toBiVO4 resulted in a near tenfold enhancement in photocurrent together witha 500 mV cathodic shift in onset potential. Furthermore, the FeOOH layersignificantly improved the stability of BiVO4 during prolonged illumination[53]. Coupling both FeOOH and NiOOH as dual-layer OECs onto porousBiVO4 electrodes has recently resulted in the report of photocurrents as high as2.73 mA cm−2 at a potential as low as 0.6 V versus RHE, clearly representing oneof the most encouraging results seen so far for BiVO4 [54]. This dual-layer OECreduces interface recombination at the BiVO4/OEC junction while creating amore favorable Helmholtz layer potential drop at the OEC/electrolyte junction.Nickel-borate (Ni—B) OEC has also been utilized with BiVO4 photoanodesto enhance the photocurrent generation by a factor of three to four times,cathodically shifts the onset potential and exhibits a near threefold improvementin IPCE [55]. Furthermore, we have recently demonstrated that this OEC hasfurther functionality as an inert, earth-abundant passivation layer for unstablephotoelectrodes, such as ZnO [125].

8.3.2 Fe2O3/OEC

The most successful junctions comprising α-Fe2O3 involve its coupling withsurface OECs that can trap the photohole, provide effective charge carrierseparation, and improve the kinetics for water oxidation. Zhong et al. [126] elec-trodeposited Co—Pi onto α-Fe2O3 electrodes and observed a 350 mV cathodic


shift in onset potential for water oxidation and a near twofold increase in IPCEat 450 nm to ∼20% at pH 13. The same group found that at pH 8, a more disperse,thinner 100 nm Co—Pi layer was able to significantly improve the photocur-rent density by nearly five times and cathodically shift the onset by ∼500 mVcompared to α-Fe2O3 at 1 V versus RHE [127], while further optimization ofthe Co—Pi deposition process yielded photocurrents approaching 3 mA cm−2 at1.23 V versus RHE [128]. Using TAS, it has been observed that Co—Pi suppressesthe kinetics for electron–hole recombination by over three orders of magnitudeand thus Co—Pi/α-Fe2O3 electrodes require smaller anodic potentials forphotocurrent generation [129], while under applied bias, slower electron–holerecombination is assigned primarily to enhanced electron depletion in α-Fe2O3[130]. NiO/α-Ni(OH)2–hematite electrodes also displayed improved activitytoward water oxidation, achieving photocurrents up to 16 mA cm−2 [131], whilea NiO/α-Fe2O3 p–n junction was used to promote charge separation throughthe use of NiO as an efficient hole acceptor, which reduced the overpotentialfor water oxidation [132]. Electrophoretic deposition of catalytic iridium oxide(IrO2) nanoparticles onto hematite photoanodes resulted in a dramatic shiftin the onset potential from +1.0 to +0.8 V versus RHE and an increase in theplateau photocurrent from 3.45 to 3.75 mA cm−2 under 1 sun illumination(Figure 8.9) [133]. However, during repeated scans the adherence of IrO2began to diminish and the shift in onset was decreased until more IrO2 wasloaded. The use of photodeposited Ni—B OEC on α-Fe2O3 nanorods resulted in∼200 mV cathodic shift of the onset potential to nearer its flat-band potential anda 9.5-fold enhancement in the photocurrent density at 0.86 V versus RHE [134].

8.3.3 WO3/OEC

Seabold and Choi [135] deposited Co—Pi OEC on WO3 photoanodes and foundthat not only did the onset potential shift cathodically toward the flat-band

4

3

2

1

0

0.8 1.0

Photocurrents

Dark currents

1.2

V/V vs RHE(a) (b)

Counter electrode

–+

H2O

O2

H2

IrO2

Fe2O3

FTO

Glass

J (

mA

cm

–2)

1.4 1.6

Figure 8.9 (a) Cross-section of the α-Fe2O3/IrO2 photoanode; (b) Performance of theunmodified α-Fe2O3 photoanode (solid black trace), and the same anode that wasfunctionalized with IrO2 nanoparticles (solid grey trace). The dashed trace is the photocurrentfor the former state-of-the-art hematite photoanode. (Reproduced with permission fromRef. [133]. Copyright 2010, John Wiley & Sons.)

8.4 Observation of Charge Carrier Kinetics in Heterojunction Structure 209

potential of WO3, the photocurrent-to-O2 conversion efficiency increased from∼61% for WO3 to ∼100% for Co—Pi/WO3 and stability is greatly improved.Strikingly, for bare WO3, it was found that 39% of photogenerated holes wereused to form peroxo species on the surface, which led to gradual decompositionof the electrode. Likewise, in order to improve the activity and stability of WO3toward water oxidation, Wang et al. deposited a Mn oxo-catalyst on ALD-grownWO3 films and found that although the activity enhancement over bare WO3was not significant, the stability of the films improved remarkably over a widepH range [136].

8.4 Observation of Charge Carrier Kineticsin Heterojunction Structure

8.4.1 Transient Absorption Spectroscopy

As stated in the Introduction, the moderate efficiencies of semiconductor watersplitting are, more often than not, due to nonradiative electron–hole recombina-tion, occurring prior to the surface reactions of electron and holes with water. Ascarrier lifetimes are on the order of femtoseconds to nanoseconds, to as muchas seconds after light absorption, a frequently used method to measure theirdecay kinetics is TAS. It is a technique most commonly used to measure kinet-ics of charge carriers in solar cell materials, but its use has been demonstratedfor PEC materials. The setup of the experiment has already been reviewed else-where [137], but in brief, it is comprised of a laser that can emit light pulses offemtosecond duration, together with a lock-in amplifier and a device to measurethe absorption spectra as a function of time, wavelength, or applied bias. By usingthe appropriate scavenger, either the electron or hole transient decay can be mea-sured. Berera et al. [137] describe the process by which fractions (0.1–10%) ofthe molecules are promoted to an electronically excited state by an excitation (orpump) pulse. A weak probe pulse (i.e., a pulse that has such a low intensity thatmultiphoton/multistep processes are avoided during probing) is sent through thesample with a delay 𝜏 with respect to the pump pulse. The absorption spectrumof the excited sample minus the absorption spectrum of the sample in the groundstate (ΔA) allows for calculation of the difference absorption spectrum. By vary-ing the time delay (𝜏) between the pump and the probe and recording a ΔAspectrum at each time delay, a ΔA profile as a function of 𝜏 and wavelength 𝜆,ΔA(𝜆,𝜏), is obtained. This is consequently very useful for researchers in the fieldof solar fuels because in the case of heterojunction systems, suppressing recom-bination and increasing carrier lifetimes are the main objectives.

For nanocrystalline TiO2 (nc-TiO2), it was found that the carrier lifetimestrongly depended on the pulse intensity and that water oxidation occurs onthe timescale of seconds, whereas recombination takes place on the orderof microseconds [5]. Looking at N-doped TiO2 under visible excitation,its lack of activity for water oxidation was assigned to the rapid decay ofvisible-light-generated photoholes, which occurs on a much faster timescalethan that required for water oxidation compared to nc-TiO2 [138]. Other


visible-driven photoanodes have been studied using this method by Pendleburyet al. who revealed that for α-Fe2O3 electrodes, the amplitude of the long-livedhole signal is only ∼10% of the initial hole signal, indicating that even underpositive applied bias the majority of photogenerated holes still undergo rapidelectron–hole recombination on the microsecond timescale [139]. This wasfollowed by the observation that recombination in α-Fe2O3, not surface kinetics,is the major limiting factor for water oxidation [66]. As expected, surfacemodification with Co—Pi resulted in the observation of a cathodic shift inphotocurrent and the appearance of long-lived hematite photoholes [129, 130],due to suppression of electron–hole recombination (Figure 8.10). The presenceof surface catalysts led to a decrease in the width of the space charge layer andFermi level pinning, thus enhancing the size of the electron depletion layer.

For Cu2O photocathodes, the introduction of RuOx particles at the surfaceresulted in a significant increase in the yield of long-lived (>100 μs) Cu2O elec-trons measured using TAS, attributed to a reduction in fast electron–hole recom-bination losses due to hole transfer from Cu2O to RuOx, thereby increasing thespatial separation of electrons and holes and facilitating the photooxidation reac-tion by holes [113].

In the case of BiVO4, Ma et al. [140] reported that the yield of long-lived(0.1–1 s) photogenerated holes is observed to correlate as a function of appliedelectrical bias, assigned to kinetic competition between water oxidation andrecombination of these surface-accumulated holes with bulk electrons acrossthe space charge layer. Two mechanisms were found to limit photocurrentgeneration in BiVO4 photoanodes: (i) rapid (≤μs) electron–hole recombinationand (ii) recombination of surface-accumulated holes with bulk BiVO4 electrons.In the case of WO3, rapid (<μs) electron–hole recombination dominates in theabsence of an electron scavenger and the production of long-lived holes with alifetime in the milliseconds to seconds timescale is required for water oxidationto occur [141]. Very recently, TAS was used to illustrate that the number and

200 100

50

0

–50

–100

–150

Δ A

bsorb

ance (

μOD

)

580 nm

900 nm

900 nm

580 nm

α-Fe2O3 α-Fe2O3/Co–Pi

150

100

50

0

10–5 10–3

Time (s)

10–1 10–5 10–3

Time (s)

10–1

(a) (b)

Figure 8.10 TAS spectra of Co—Pi/Fe2O3, indicating a long-lived hole (580 nm) population. (a)α-Fe2O3. (b) α-Fe2O3/Co—Pi (Reproduced with permission from Ref. [129]. Copyright 2011,American Chemical Society.)


lifetime of long-lived charges in a WO3/TiO2 junction increased remarkablyduring photocatalysis [82]. The use of TAS is becoming more popular withresearchers and could prove beneficial in providing experimental evidence ofcharge kinetics.

8.4.2 Electrochemical Impedance Spectroscopy

Other techniques that can probe charge transfer kinetics and recombinationinclude EIS, photoluminescence spectroscopy (PL), and SPS. As PL spectroscopydepends on radiative recombination, which is generally considered a minorityprocess, it will not be discussed in further detail here.

In photoelectrochemistry, there are three relevant effective parameters thatdetermine the performance of a PEC device: potential, photocurrent, and lightintensity. The basis of the EIS experiment is to apply a small amplitude sinusoidalac voltage, V (t), and then to measure the amplitude and phase angle (relativeto the applied voltage) of the resulting current, I(t) [142]. The impedance (Z)can then be calculated from Ohm’s law [Z =V (t)/I(t)]. Competition between theoverall rates of hole transfer and recombination determines the fraction of thehole current jh that is measured in the external circuit in EIS (Figure 8.11)

The rate constants kt and kr (s−1) in the equations for the EIS response are usedto express the rates of hole transfer and recombination (cm−2 s−1) in terms of thesurface concentration of trapped holes.

The radial frequency 𝜔max (HF) corresponding to the maximum imaginarycomponent of the high-frequency semicircle is given by

𝜔max =(

1Csc(Z2 − Rser)

)

Here Rser is the series resistance, Csc is the space charge capacitance, jh isthe current density corresponding to the flux of holes reaching the interface.The rate constants kt and kr are the first-order rate constants for interfacialtransfer and recombination, respectively. The equation for Z1 predicts the

–

+

+

–

hν

Semiconductor

(a) (b)Electrolyte

Im(Z

)

Real (Z)

Rser

Rser + Z2 Rser + Z1

kr

kt

jh

Figure 8.11 (a) Phenomenological kinetic scheme for PEIS analysis; (b) typical Nyquist plotshowing the origin of Z1 and Z2. (Reproduced with permission from Ref. [143]. Copyright 2011,The Royal Society of Chemistry.)


semicircles observed in Figure 8.11b. As the applied potential increases, therate of recombination decreases (due to band bending) and thus charge transferincreases. When the EIS is fitted using the appropriate equivalent circuit model,the resultant impedance semicircle (Nyquist plot) gives some indication as tothe efficiency of charge transfer between the semiconductor and electrolytewhen the value of the charge transfer resistance at the interface is computed.jh is likely to be lower than the product of qI0(I0 is the photon flux) becauseelectron–hole pairs generated outside the space charge region are mostly lostthrough recombination. Recently, this technique has been successfully appliedto bulk heterojunction solar cells based on the determination of chemicalcapacitance and recombination resistance [144]. Modern potentiostats areable to measure electrical impedance spectroscopy at a variety of frequenciesand will construct the Nyquist plot; the fitting of the equivalent circuit to anin-built model or to such developed by the user is a common function of theaccompanying software. There are many examples of its use in when applied tojunction architectures such as BiVO4/WO3 [46], CdS/WO3 [145], Cu2O/RGO[114], and g-C3N4/ZnO [146]to explain charge transfer.

Furthermore, in the equivalent Randle circuit, values of Rct may be compared(where Rs is the solution resistance, Q1 is the constant phase element (CPE)for the electrolyte–electrode interface, and Rct is the charge transfer resistanceacross the interface of electrode–electrolyte interface). Thus, a comparativelylower value of Rct is expected for a heterojunction, which exhibits favorablecharge transfer characteristics. Lee et al. [46] observed a significantly decreasedvalue for Rct in their optimized BiVO4/WO3 junction (780Ω) compared to bareBiVO4 (8803Ω), while for CdS-based junctions, values of Rct were found to be3856, 2038, and 1367Ω for TiO2, TiO2/CdS, and TiO2/CdS/Co—Pi electrodes,respectively (see Figure 8.12) [147]. Although one cannot obtain the rateconstants via this interpretation, it assumes pure charge transfer control and

4.0

3.2

2.4

1.6

0.8

0.00.0 0.8

CPE

Rs

Rct

1.6

Z″ (kΩ)

–Z

″ (k

Ω)

2.4 3.2

Pristine TiO2

TiO2/CdS

TiO2/CdS/Co–Pi

4.0

Figure 8.12 Nyquist plot and fitted equivalent circuit model of the CdS-based heterojunctionsproposed by Zhong et al. for improved PEC water splitting. (Reproduced with permission fromRef. [147]. Copyright 2015, John Wiley & Sons.)


no diffusion limitations or competing processes. Furthermore, the presenceof defects and multilayers is likely to yield different resistivities and requires amore complex model, so care must be taken when fitting the equivalent circuit.Nevertheless, this technique can provide information about the phenomeno-logical rate constants describing the competition between charge transfer andrecombination during water oxidation.

One can develop the EIS measurement further by probing the dynamicrelation between light irradiation and electrochemical response of the pho-toelectrode, using PEIS (photoelectrochemical impedance spectroscopy),IMPS (intensity-modulated photocurrent spectroscopy), and IMVS (intensity-modulated photovoltage spectroscopy). The use of these techniques to studyphotoelectrode performance is relatively new and so far limited to α-Fe2O3[148], and, therefore readers, are directed to the recent perspective by Klotzet al. for further details [149].

8.4.3 Surface Photovoltage Spectroscopy

SPS has been used for many years to probe charge carrier kinetics and recom-bination in dye sensitized solar cells, and this measurement technique hasonly recently been applied to the field of semiconductor photoelectrodesfor water splitting [150]. The surface photovoltage (SPV) is defined as theillumination-induced change in the surface potential; this nondestructive tech-nique measures change in band bending at the free semiconductor surface asa function of external illumination and can provide the researcher interestedin heterojunction photocatalysis a wealth of qualitative and quantitative infor-mation. This includes, but is not limited to, the relative locations of the bandpositions, the degree of band bending, defect states, surface dipole, diffusionlengths, recombination rates, and flat-band potentials.

The setup includes a light source, sample chamber containing a Kelvin probeenclosed within a Faraday cage, chopper, lock-in amplifier, and monochromatorand is described in more detail elsewhere [151]. The Kelvin probe can be usedto measure the contact potential difference (CPD), which is the differencebetween the work function of the metal tip of the probe and the semiconductorsurface. On illumination, the probe measures the change in CPD. The absorbedphotons induce the formation of free carriers by creating electron–hole pairsvia interband transitions and/or release captured carriers via trap-to-bandtransitions, resulting in a significant amount of charge transferring from thesurface to the bulk (or vice versa) and/or redistributed within the surface or thebulk. Since the electric potential and the charge distribution are interdependent,the potential drop across the surface space charge region, and surface potentialchange. It is important to note that the formation of a SPV occurs only if carriergeneration is followed by net charge redistribution; the SPV response is positivefor n-type materials and negative for p-type due to the different signs of theequilibrium surface potential. For a semiconductor with upward bent bands,irradiation will cause the band to flatten as the negative surface charge decreasesas a result of hole transport to the surface (Figure 8.13), whereas the oppositeis occurring in semiconductors with downward bent bands, causing downward


No illumination With illumination

VB

CB

Bulk Surface Bulk Surface

hv hv

SPVsat

SPV

Bulk Surface

Ef

SPV saturation

(a) (b) (c)

Figure 8.13 Schematic diagrams of the surface photovoltage effect. In panel (a), we observeupward band bending in a typical n-type semiconductor surface; in panel (b), the absorbedphotons produce free charge carriers resulting in a partial band flattening; and in panel (c) thelargest SPV saturation occurs to completely flatten bands. (Reproduced with permission fromRef. [152]. Copyright 2010, John Wiley & Sons.)

band bending to decrease. For increasing photon flux, the energy bands maybe completely flattened and the saturated SPV value is equivalent to the initialmagnitude of band bending of the semiconductor [19]. Readers are directedto the comprehensive overview by Kronik for details of the experiment [153].The photovoltage (in volts or similar units) can be measured as a steady stateexperiment (SS, function of light wavelength) or as a transient measurement(TS, as a function of time). There are many examples of SPV used to determinethe charge separation efficiency in heterojunction photoelectrodes. For example,Fu et al. [61] used TS-SPV and SS-SPS to probe charge carriers in BiVO4/TiO2composites. Coupling 5% TiO2 to BiVO4 resulted in a large increase in theSS-SPV trace, indicating better charge separation, and the TS-SPV signalrevealed a much longer carrier lifetime of ∼3 ms. This was attributed to theunusual spatial transfer of visible-light-excited high-energy electrons of BiVO4to TiO2. Osterloh et al. measured the internal photovoltages in single crystallineplatinum-/ruthenium-modified Rh-doped SrTiO3 nanocrystals [154]. Voltagesof −0.88 and −1.13 V were found between the absorber and the Ru and Ptcocatalysts, respectively, and a voltage of −1.48 V for a Rh:SrTiO3 film on anAu substrate (Figure 8.14). This showed that the Pt and Ru cocatalysts not onlyimprove the redox kinetics but also aid charge separation in the absorber.

Visible-light-driven hydrogen production from water has been observed onCuS/Zn0.8Cd0.2S composites, which have been shown to possess improvedcharge separation compared to CuS through SS-SPV. The TS-SPV measurementsuggested photogenerated electrons transfer from Zn0.8Cd0.2S to CuS [155].Zhao and Osterloh [156] reported the use of SPV spectroscopy to probe carrierdynamics in nanocrystal films of HCa2Nb3O10. For sufficiently thick films,the SPV spectrum was used to evaluate the efficiency of photochemical chargeseparation at the sample–support (Au) interface, which revealed that the

8.5 Conclusions 215

0.0

–0.5

–1.0

–1.5

–2.0

–2.5

1 2

Gold substrate

Gold substrate

Red

hv

h+

e–

Oxhv

h+

e–

3

–2.64 V

–2.12 V

–1.87 V

–1.48 V

–0.75 V

–0.24 V

MeOH

MV2+

Fe(CN)64–

Fe(CN)63–

blank

I–

E (eV)

ΔCP

D (

V)

4 5

Figure 8.14 SPV spectra of 0.5 mg Rh(3 mol%):SrTiO3 on 1 cm2 gold-coated glass in thepresence of redox reagents with schematic mechanism (inset). (Reproduced with permissionfrom Ref. [154]. Copyright 2015, The Royal Society of Chemistry.)

photovoltage increases linearly with film thickness and that in the presence ofan electron blocking PEDOT:PSS layer, the signal arises from electron transferacross the nanocrystal–gold interface. Similarly, BiVO4/Co3O4 composites weredemonstrated to have significantly improved activity for water oxidation undervisible-light irradiation (11 mmol h−1 g−1) [157]; SS-SPV measurement revealedp-type character for the Co3O4 particles and a junction at the Co3O4–BiVO4interface, resulting in improved electron–hole separation due to hole injectioninto Co3O4. SPV has also been used to show improved charge separation inother visible-driven heterojunction catalysts, such as Ag/Ag3PO4/graphene[158], NiS/CdS [159], Mn/ZnO [160], and V2O5/BiVO4 [161].

8.5 Conclusions

The development of a low-cost strategy to increase the STH conversion efficiencyis a fundamental part of our search to find renewable and sustainable energysources. For water-splitting photocatalysis, some key requirements need to besatisfied – appreciable absorption of sunlight, adequate band alignment for waterreduction/oxidation, fast reaction kinetics, and prolonged stability. One of thefundamental, dominating problems to address in semiconductor photocatalystsis the rapid recombination of charge carriers. The role of plasmonic layers,ultra-high energy electrons, and Z-scheme double excitation systems representjust a few of the very recent breakthroughs that contribute to our understandingof charge transfer and more work is needed to fully realize their potential.Carbon-based materials have great potential to be utilized for large-scale watersplitting either within a PEC cell or in a particle-based system, due to theirexcellent performance and low materials cost. In fact, suspension systems cannow reach energy conversion efficiencies of nearly 3%, and have been the subjectof numerous recent technological breakthroughs.


Overall, the concept of a heterojunction architecture is an emerging andclearly viable route to increasing efficiency and improving stability by facilitatingimproved charge separation and transfer, the evidence of which is discussed inthis chapter with our review of a wealth of recent reports of visible-light-drivenheterojunction photocatalysts. By far the most common methods to inves-tigate charge separation and transport on such timescales involve the use ofadvanced spectroscopies, especially time-resolved techniques, and these mayallow us to address the underlying reasons for low efficiencies in water-splittingreactions.

References


2 Li, Z., Luo, W., Zhang, M. et al. (2013) Photoelectrochemical cells for solarhydrogen production: current state of promising photoelectrodes, meth-ods to improve their properties, and outlook. Energy Environ. Sci., 6 (2),347–370.

3 Moniz, S.J.A., Quesada-Cabrera, R., Blackman, C.S. et al. (2014) A sim-ple, low-cost CVD route to thin films of BiFeO3 for efficient waterphoto-oxidation. J. Mater. Chem. A, 2, 2922–2927.

4 Bard, A. and Fox, M. (1995) Artificial photosynthesis: solar splitting of waterto hydrogen and oxygen. Acc. Chem. Res., 28, 141–145.

5 Tang, J., Durrant, J.R., and Klug, D.R. (2008) Mechanism of photocatalyticwater splitting in TiO2. Reaction of water with photoholes, importance ofcharge carrier dynamics, and evidence for four-hole chemistry. J. Am. Chem.Soc., 130 (42), 13885–13891.

6 Sivula, K., Zboril, R., Le Formal, F. et al. (2010) Photoelectrochemical watersplitting with mesoporous hematite prepared by a solution-based colloidalapproach. J. Am. Chem. Soc., 132 (21), 7436–7444.

7 Hodes, G., Cahen, D., and Manassen, J. (1976) Tungsten trioxide as a pho-toanode for a photoelectrochemical cell (PEC). Nature, 260, 312–313.

8 Bignozzi, C.A., Caramori, S., Cristino, V. et al. (2013) Nanostructured pho-toelectrodes based on WO3: applications to photooxidation of aqueouselectrolytes. Chem. Soc. Rev., 42 (6), 2228–2246.

9 Sayama, K., Nomura, A., Zou, Z. et al. (2003) Photoelectrochemical decom-position of water on nanocrystalline BiVO4 film electrodes under visiblelight. Chem. Commun., (23), 2908–2909.

10 Li, R., Zhang, F., Wang, D. et al. (2013) Spatial separation of photogeneratedelectrons and holes among {010} and {110} crystal facets of BiVO4. Nat.Commun., 4, 1–7.

11 Kudo, A. and Miseki, Y. (2009) Heterogeneous photocatalyst materials forwater splitting. Chem. Soc. Rev., 38 (1), 253–278.

12 Kudo, A., Ueda, K., Kato, H., and Mikami, I. (1998) Photocatalytic O2 evo-lution under visible light irradiation on BiVO4 in aqueous AgNO3 solution.Catal. Lett., 53, 229–230.

References 217

13 Martin, D.J., Umezawa, N., Chen, X. et al. (2013) Facet engineered Ag3PO4for efficient water photooxidation. Energy Environ. Sci., 6 (11), 3380–3386.

14 Martin, D.J., Qiu, K., Shevlin, S.A. et al. (2014) Highly efficient photocat-alytic H2 evolution from water using visible light and structure-controlledgraphitic carbon nitride. Angew. Chem. Int. Ed., 53, 9240–9245.

15 Rhee, K., Morris, E., Barber, J., and Kühlbrandt, W. (1998)Three-dimensional structure of the plant photosystem II reaction centreat 8 Å resolution. Nature, 396, 283–286.

16 Marschall, R. (2014) Semiconductor composites: strategies for enhancingcharge carrier separation to improve photocatalytic activity. Adv. Funct.Mater., 24 (17), 2421–2440.

17 Sivula, K. (2013) Metal oxide photoelectrodes for solar fuel production,surface traps, and catalysis. J. Phys. Chem. Lett., 4 (10), 1624–1633.

18 Walter, M.G., Warren, E.L., McKone, J.R. et al. (2010) Solar water splittingcells. Chem. Rev., 110 (11), 6446–6473.

19 Zhang, Z. and Yates, J.T. (2012) Band bending in semiconductors: chemicaland physical consequences at surfaces and interfaces. Chem. Rev., 112 (10),5520–5551.

20 Wang, H., Zhang, L., Chen, Z. et al. (2014) Semiconductor heterojunctionphotocatalysts: design, construction, and photocatalytic performances. Chem.Soc. Rev., 43 (15), 5234–5244.

21 Cowan, A.J. and Durrant, J.R. (2013) Long-lived charge separated states innanostructured semiconductor photoelectrodes for the production of solarfuels. Chem. Soc. Rev., 42 (6), 2281–2293.

22 Xiang, Q., Yu, J., and Jaroniec, M. (2012) Graphene-based semiconductorphotocatalysts. Chem. Soc. Rev., 41 (2), 782–796.

23 Huang, X., Qi, X., Boey, F., and Zhang, H. (2012) Graphene-based compos-ites. Chem. Soc. Rev., 41 (2), 666–686.

24 Tada, H., Mitsui, T., Kiyonaga, T. et al. (2006) All-solid-state Z-scheme inCdS-Au-TiO2 three-component nanojunction system. Nat. Mater., 5 (10),782–786.

25 Hisatomi, T., Kubota, J., and Domen, K. (2014) Recent advances insemiconductors for photocatalytic and photoelectrochemical water splitting.Chem. Soc. Rev., 43, 7520–7535.

26 Iwashina, K. and Kudo, A. (2011) Rh-doped SrTiO3 photocatalyst elec-trode showing cathodic photocurrent for water splitting under visible-lightirradiation. J. Am. Chem. Soc., 133, 13272–13275.

27 Kim, H.G., Hwang, D.W., and Lee, J.S. (2004) An undoped, single-phaseoxide photocatalyst working under visible light. J. Am. Chem. Soc., 126 (29),8912–8913.

28 Xu, X., Randorn, C., Efstathiou, P., and Irvine, J.T.S. (2012) A red metallicoxide photocatalyst. Nat. Mater., 11 (6), 1–4.

29 Maeda, K., Higashi, M., Siritanaratkul, B. et al. (2011) SrNbO2N as awater-splitting photoanode with a wide visible-light absorption band.J. Am. Chem. Soc., 133 (32), 12334–12337.

30 Kasahara, A. and Nukumizu, K. (2002) Photoreactions on LaTiO2N undervisible light irradiation. J. Phys. Chem. A, 106, 6750–6753.


31 Higashi, M., Domen, K., and Abe, R. (2012) Highly stable water splitting onoxynitride TaON photoanode system under visible light irradiation. J. Am.Chem. Soc., 134 (16), 6968–69671.

32 Kim, E., Nishimura, N., Magesh, G. et al. (2013) Fabrication ofCaFe2O4/TaON heterojunction photoanode for photoelectrochemical wateroxidation. J. Am. Chem. Soc., 135, 5375–5383.

33 Kim, H., Monllor-Satoca, D., Kim, W., and Choi, W. (2014) N-doped TiO2nanotubes coated with a thin TaOxNy layer for photoelectrochemical watersplitting: dual bulk and surface modification of photoanodes. Energy Environ.Sci. doi: 10.1039/c4ee02169j

34 Hou, J., Yang, C., Cheng, H. et al. (2014) High-performance p-Cu2O/n-TaONheterojunction nanorod photoanodes passivated with an ultrathin carbonsheath for photoelectrochemical water splitting. Energy Environ. Sci., 7 (11),3758–3768.

35 Deng, X. and Tüysüz, H. (2014) Cobalt-oxide-based materials as wateroxidation catalyst: recent progress and challenges. ACS Catal., 4, 3701–3714.

36 Bloor, L.G., Molina, P.I., Symes, M.D., and Cronin, L. (2014) Low pHelectrolytic water splitting using earth-abundant metastable catalysts thatself-assemble in situ. J. Am. Chem. Soc., 136, 3304–3311.

37 Gerken, J., McAlpin, J., Chen, J.Y.C. et al. (2011) Electrochemical water oxi-dation with cobalt-based electrocatalysts from pH 0–14: the thermodynamicbasis for catalyst structure, stability, and activity. J. Am. Chem. Soc., 133,14431–14442.

38 Bediako, D.K., Lassalle-Kaiser, B., Surendranath, Y. et al. (2012)Structure–activity correlations in a nickel–borate oxygen evolution catalyst.J. Am. Chem. Soc., 134 (15), 6801–6809.

39 Kanan, M.W. and Nocera, D.G. (2008) In situ formation of anoxygen-evolving catalyst in neutral water containing phosphate and Co2+.Science, 321 (5892), 1072–1075.

40 Kanan, M.W., Surendranath, Y., and Nocera, D.G. (2009) Cobalt-phosphateoxygen-evolving compound. Chem. Soc. Rev., 38 (1), 109–114.

41 Walsh, A., Yan, Y., and Huda, M. (2009) Band edge electronic structure ofBiVO4: elucidating the role of the Bi s and V d orbitals. Chem. Mater., 21(3), 547–551.

42 Abdi, F.F., Han, L., Smets, A.H.M. et al. (2013) Efficient solar water split-ting by enhanced charge separation in a bismuth vanadate-silicon tandemphotoelectrode. Nat. Commun., 4 (2195), 1–7.

43 Ye, K., Wang, Z., Gu, J. et al. (2017) Carbon quantum dots as a visible lightsensitizer to significantly boost the solar water splitting performance ofbismuth vanadate photoanodes. Energy Environ. Sci., 10, 772–779.

44 Park, Y., McDonald, K.J., and Choi, K.S. (2013) Progress in bismuth vana-date photoanodes for use in solar water oxidation. Chem. Soc. Rev., 42 (6),2321–2337.

45 Su, J., Guo, L., Bao, N., and Grimes, C.A. (2011) NanostructuredWO3/BiVO4 heterojunction films for efficient photoelectrochemical watersplitting. Nano Lett., 11 (5), 1928–1933.

References 219

46 Hong, S.J., Lee, S., Jang, J.S., and Lee, J.S. (2011) Heterojunction BiVO4/WO3electrodes for enhanced photoactivity of water oxidation. Energy Environ.Sci., 4 (5), 1781.

47 Saito, R., Miseki, Y., and Sayama, K. (2012) Highly efficient photoelectro-chemical water splitting using a thin film photoanode of BiVO4/SnO2/WO3multi-composite in a carbonate electrolyte. Chem. Commun., 48 (32),3833–3835.

48 Pihosh, Y., Turkevych, I., Mawatari, K. et al. (2014) NanostructuredWO3/BiVO4 photoanodes for efficient photoelectrochemical water splitting.Small, 10 (18), 3692–3699.

49 Shi, X., Choi, I.Y., Zhang, K. et al. (2014) Efficient photoelectrochemicalhydrogen production from bismuth vanadate-decorated tungsten trioxidehelix nanostructures. Nat. Commun., 5, 4775.

50 Moniz, S.J.A., Zhu, J., and Tang, J. (2014) 1D Co-Pi modified BiVO4/ZnOjunction cascade for efficient photoelectrochemical water cleavage. Adv.Energy Mater., 4 (10), 1301590.

51 Pilli, S.K., Furtak, T.E., Brown, L.D. et al. (2011) Cobalt-phosphate (Co-Pi)catalyst modified Mo-doped BiVO4 photoelectrodes for solar water oxida-tion. Energy Environ. Sci., 4 (12), 5028–5034.

52 Jeong, H.W., Jeon, T.H., Jang, J.S. et al. (2013) Strategic modification ofBiVO4 for improving photoelectrochemical water oxidation performance.J. Phys. Chem. C, 117, 9104–9112.

53 Seabold, J.A. and Choi, K.-S. (2012) Efficient and stable photo-oxidation ofwater by a bismuth vanadate photoanode coupled with an iron oxyhydroxideoxygen evolution catalyst. J. Am. Chem. Soc., 134 (4), 2186–2192.

54 Kim, T.W. and Choi, K.-S. (2014) Nanoporous BiVO4 photoanodes withdual-layer oxygen evolution catalysts for solar water splitting. Science, 343(6174), 990–994.

55 Choi, S.K., Choi, W., and Park, H. (2013) Solar water oxidation usingnickel-borate coupled BiVO4 photoelectrodes. Phys. Chem. Chem. Phys.,15 (17), 6499–6507.

56 Chatchai, P., Kishioka, S., Murakami, Y. et al. (2010) Enhanced photoelec-trocatalytic activity of FTO/WO3/BiVO4 electrode modified with goldnanoparticles for water oxidation under visible light irradiation. Electrochim.Acta, 55 (3), 592–596.

57 Pilli, S.K., Deutsch, T.G., Furtak, T.E. et al. (2013) BiVO4/CuWO4 hetero-junction photoanodes for efficient solar driven water oxidation. Phys. Chem.Chem. Phys., 15 (9), 3273–3278.

58 Ma, M., Kim, J., Zhang, K., and Shi, X. (2014) Double-deck inverse opalphotoanodes: efficient light absorption and charge separation in heterojunc-tion. Chem. Mater., 26, 5592–5597.

59 Fu, X., Xie, M., Luan, P., and Jing, L. (2014) Effective visible-excited chargeseparation in silicate-bridged ZnO/BiVO4 nanocomposite and its contri-bution to enhanced photocatalytic activity. ACS Appl. Mater. Interfaces, 6,18550–18557.


60 Ho-Kimura, S., Moniz, S.J.A., Handoko, A.D., and Tang, J. (2014) Enhancedphotoelectrochemical water splitting by nanostructured BiVO4–TiO2 com-posite electrodes. J. Mater. Chem. A, 2 (11), 3948–3953.

61 Xie, M., Fu, X., Jing, L. et al. (2013) Long-lived, visible-light-excited chargecarriers of TiO2/BiVO4 nanocomposites and their unexpected photoactivityfor water splitting. Adv. Energy Mater., 4, 1300995.

62 Resasco, J., Zhang, H., Kornienko, N. et al. (2016) TiO2/BiVO4 nanowireheterostructure photoanodes based on type II band alignment. ACS Cent.Sci., 2 (2), 80–88.

63 Ng, Y.H., Iwase, A., Kudo, A., and Amal, R. (2010) Reducing graphene oxideon a visible-light BiVO4 photocatalyst for an enhanced photoelectrochemicalwater splitting. J. Phys. Chem. Lett., 1 (17), 2607–2612.

64 Wan, X., Niu, F., Su, J., and Guo, L. (2016) Enhanced photoelectrochemicalwater oxidation of bismuth vanadate via a combined strategy of W dopingand surface RGO modification. Phys. Chem. Chem. Phys., 18, 31803–31810.

65 Sivula, K., Le Formal, F., and Grätzel, M. (2011) Solar water splitting:progress using hematite (α-Fe2O3) photoelectrodes. ChemSusChem, 4 (4),432–439.

66 Pendlebury, S.R., Cowan, A.J., Barroso, M. et al. (2012) Correlatinglong-lived photogenerated hole populations with photocurrent densi-ties in hematite water oxidation photoanodes. Energy Environ. Sci., 5 (4),6304–6312.

67 Barroso, M., Pendlebury, S.R., Cowan, A.J., and Durrant, J.R. (2013) Chargecarrier trapping, recombination and transfer in hematite (α-Fe2O3) watersplitting photoanodes. Chem. Sci., 4 (7), 2724–2734.

68 Sivula, K., Formal, F.L., and Grätzel, M. (2009) WO3−Fe2O3 photoanodesfor water splitting: a host scaffold, guest absorber approach. Chem. Mater.,21 (13), 2862–2867.

69 Han, H.S., Shin, S., Noh, J.H. et al. (2014) Heterojunction Fe2O3–SnO2nanostructured photoanode for efficient photoelectrochemical water split-ting. JOM, 66 (4), 664–669.

70 McDonald, K. and Choi, K. (2011) Synthesis and photoelectrochemicalproperties of Fe2O3/ZnFe2O4 composite photoanodes for use in solar wateroxidation. Chem. Mater., 23 (21), 4863–4869.

71 Moniz, S.J.A., Shevlin, S.A., An, X. et al. (2014) Fe2O3–TiO2 nanocompositesfor enhanced charge separation and photocatalytic activity. Chem. Eur. J., 20,15571–15579.

72 Rai, S., Ikram, A., Sahai, S. et al. (2014) Morphological, optical and photo-electrochemical properties of Fe2O3–GNP composite thin films. RSC Adv.,4 (34), 17671.

73 Mayer, M.T., Lin, Y., Yuan, G., and Wang, D. (2013) Forming heterojunc-tions at the nanoscale for improved photoelectrochemical water splitting bysemiconductor materials: case studies on hematite. Acc. Chem. Res., 46 (7),1558–1566.

74 Lin, Y., Zhou, S., Sheehan, S.W., and Wang, D. (2011) Nanonet-basedhematite heteronanostructures for efficient solar water splitting. J. Am.Chem. Soc., 133 (8), 2398–2401.

References 221

75 Lin, Y., Yuan, G., Sheehan, S. et al. (2011) Hematite-based solar watersplitting: challenges and opportunities. Energy Environ. Sci., 4 (12), 4862.

76 Lin, Y., Xu, Y., Mayer, M.T. et al. (2012) Growth of p-type hematite byatomic layer deposition and its utilization for improved solar water splitting.J. Am. Chem. Soc., 134 (12), 5508–5511.

77 Mayer, M.T., Du, C., and Wang, D. (2012) Hematite/Si nanowiredual-absorber system for photoelectrochemical water splitting at low appliedpotentials. J. Am. Chem. Soc., 134 (30), 12406–12409.

78 Liu, X., Wang, F., and Wang, Q. (2012) Nanostructure-based WO3 photoan-odes for photoelectrochemical water splitting. Phys. Chem. Chem. Phys.,14 (22), 7894–7911.

79 Kim, W., Tachikawa, T., Monllor-Satoca, D. et al. (2013) Promoting waterphotooxidation on transparent WO3 thin films using an alumina overlayer.Energy Environ. Sci., 6 (12), 3732–3739.

80 Lin, J., Hu, P., Zhang, Y. et al. (2013) Understanding the photoelectrochemi-cal properties of a reduced graphene oxide–WO3 heterojunction photoanodefor efficient solar-light-driven overall water splitting. RSC Adv., 3 (24), 9330.

81 Ma, S.S.K., Maeda, K., Abe, R., and Domen, K. (2012) Visible-light-drivennonsacrificial water oxidation over tungsten trioxide powder modified withtwo different cocatalysts. Energy Environ. Sci., 5 (8), 8390.

82 Sotelo-Vazquez, C., Quesada-Cabrera, R., Ling, M. et al. (2017) Evidenceand effect of photogenerated charge transfer for enhanced photocatalysis inWO3/TiO2 heterojunction films: a computational and experimental study.Adv. Funct. Mater., 27, 16051413

83 Wang, X., Maeda, K., Thomas, A. et al. (2009) A metal-free polymeric pho-tocatalyst for hydrogen production from water under visible light. Nat.Mater., 8 (1), 76–80.

84 Wang, Y., Wang, X., and Antonietti, M. (2012) Polymeric graphitic carbonnitride as a heterogeneous organocatalyst: from photochemistry to multi-purpose catalysis to sustainable chemistry. Angew. Chem. Int. Ed., 51 (1),68–89.

85 Kang, H.W., Lim, S.N., Song, D., and Park, S.B. (2012) Organic-inorganiccomposite of g-C3N4–SrTiO3:Rh photocatalyst for improved H2 evolutionunder visible light irradiation. Int. J. Hydrogen Energy, 37 (16), 11602–11610.

86 Ge, L. and Han, C. (2012) Synthesis of MWNTs/g-C3N4 composite pho-tocatalysts with efficient visible light photocatalytic hydrogen evolutionactivity. Appl. Catal., B, 117–118, 268–274.

87 Xiang, Q., Yu, J., and Jaroniec, M. (2011) Preparation and enhancedvisible-light photocatalytic H2-production activity of graphene/C3N4 com-posites. J. Phys. Chem. C, 115, 7355–7363.

88 Yuan, Y.-P., Cao, S.-W., Liao, Y.-S. et al. (2013) Red phosphor/g-C3N4 hetero-junction with enhanced photocatalytic activities for solar fuels production.Appl. Catal., B, 140–141, 164–168.

89 Yan, H. and Yang, H. (2011) TiO2–g-C3N4 composite materials for photocat-alytic H2 evolution under visible light irradiation. J. Alloys Compd., 509 (4),L26–L29.


90 Zang, Y., Zuo, Y., and Li, G. (2014) Hybridization of brookite TiO2 withg-C3N4: a visible-light-driven photocatalyst for As3+ oxidation, MO degra-dation and water splitting for hydrogen evolution. J. Mater. Chem. A.2, 15774–15780.

91 Xu, X., Liu, G., Randorn, C., and Irvine, J.T.S. (2011) g-C3N4 coated SrTiO3as an efficient photocatalyst for H2 production in aqueous solution undervisible light irradiation. Int. J. Hydrogen Energy, 36 (21), 13501–13507.

92 Wang, Y., Shi, R., Lin, J., and Zhu, Y. (2011) Enhancement of photocur-rent and photocatalytic activity of ZnO hybridized with graphite-like C3N4.Energy Environ. Sci., 4 (8), 2922–2929.

93 Hong, J., Wang, Y., Wang, Y. et al. (2013) Noble-metal-free NiS/C3N4 forefficient photocatalytic hydrogen evolution from water. ChemSusChem,6 (12), 2263–2268.

94 Chen, Z., Sun, P., Fan, B. et al. (2014) In situ template-free ion-exchangeprocess to prepare visible-light-active g-C3N4/NiS hybrid photocatalystswith enhanced hydrogen evolution activity. J. Phys. Chem. C, 118 (15),7801–7807.

95 Yin, L., Yuan, Y.-P., Cao, S.-W. et al. (2014) Enhanced visible-light-drivenphotocatalytic hydrogen generation over g-C3N4 through loading the noblemetal-free NiS2 cocatalyst. RSC Adv., 4 (12), 6127.

96 Fukasawa, Y., Shimojima, K.T.A., Antonietti, M. et al. (2011) Synthesis ofordered porous graphitic-C3N4 and regularly arranged Ta3N5 nanoparticlesby using self-assembled silica nanospheres as a primary template. Chem.Asian J., 6 (1), 103–109.

97 Ge, L., Zuo, F., Liu, J. et al. (2012) Synthesis and efficient visible light pho-tocatalytic hydrogen evolution of polymeric g-C3N4 coupled with CdSquantum dots. J. Phys. Chem. C, 116 (25), 13708–13714.

98 Zhang, J., Wang, Y., Jin, J. et al. (2013) Efficient visible-light photocatalytichydrogen evolution and enhanced photostability of core/shell CdS/g-C3N4nanowires. ACS Appl. Mater. Interfaces, 5 (20), 10317–10324.

99 Liu, J., Liu, Y., Liu, N. et al. (2015) Metal-free efficient photocatalyst forstable visible water splitting via a two-electron pathway. Science, 347 (6225),970–974.

100 Zhang, G., Lan, Z.-A., Lin, L. et al. (2016) Overall water splitting byPt/g-C3N4 photocatalysts without using sacrificial agents. Chem. Sci., 7 (5),3062–3066.

101 Jorge, A., Martin, D., Dhanoa, M.T.S. et al. (2013) H2 and O2 evolution fromwater half-splitting reactions by graphitic carbon nitride materials. J. Phys.Chem. C, 117, 7178–7185.

102 Hou, Y., Wen, Z., Cui, S. et al. (2013) Constructing 2D porous graphiticC3N4 nanosheets/nitrogen-doped graphene/layered MoS2 ternary nano-junction with enhanced photoelectrochemical activity. Adv. Mater., 25 (43),6291–6297.

103 Hou, Y., Zuo, F., Dagg, A.P. et al. (2014) Branched WO3 nanosheet arraywith layered C3N4 heterojunctions and CoOx nanoparticles as a flexiblephotoanode for efficient photoelectrochemical water oxidation. Adv. Mater.,26, 5043–5049.

References 223

104 Morales-Guio, C.G., Tilley, S.D., Vrubel, H. et al. (2014) Hydrogen evolutionfrom a copper(I) oxide photocathode coated with an amorphous molybde-num sulphide catalyst. Nat. Commun., 5 (I), 3059.

105 Paracchino, A., Laporte, V., Sivula, K. et al. (2011) Highly active oxide pho-tocathode for photoelectrochemical water reduction. Nat. Mater., 10 (6),456–461.

106 Morales-Guio, C.G., Liardet, L., Mayer, M.T. et al. (2014) Photoelectro-chemical hydrogen production in alkaline solutions using Cu2O coatedwith earth-abundant hydrogen evolution catalysts. Angew. Chem. Int. Ed.,54, 664–667.

107 Lin, C.-Y., Lai, Y.-H., Mersch, D., and Reisner, E. (2012) Cu2O|NiOxnanocomposite as an inexpensive photocathode in photoelectrochemicalwater splitting. Chem. Sci., 3 (12), 3482–3487.

108 Martínez-García, A., Vendra, V.K., Sunkara, S. et al. (2013) Tungsten oxide-coated copper oxide nanowire arrays for enhanced activity and durabilitywith photoelectrochemical water splitting. J. Mater. Chem. A, 1 (48), 15235.

109 Zhang, Z. and Wang, P. (2012) Highly stable copper oxide composite asan effective photocathode for water splitting via a facile electrochemicalsynthesis strategy. J. Mater. Chem., 22 (6), 2456–2464.

110 Wang, M., Sun, L., Lin, Z. et al. (2013) p–n heterojunction photoelectrodescomposed of Cu2O-loaded TiO2 nanotube arrays with enhanced photoelec-trochemical and photoelectrocatalytic activities. Energy Environ. Sci., 6 (4),1211.

111 Hou, Y., Li, X.Y., Zhao, Q.D. et al. (2009) Fabrication of Cu2O/TiO2 nan-otube heterojunction arrays and investigation of its photoelectrochemicalbehavior. Appl. Phys. Lett., 95 (9), 93108.

112 Handoko, A.D. and Tang, J. (2013) Controllable proton and CO2 photore-duction over Cu2O with various morphologies. Int. J. Hydrogen Energy, 38(29), 13017–13022.

113 Pastor, E., Pesci, F.M., Reynal, A. et al. (2014) Interfacial charge separa-tion in Cu2O/RuO(x) as a visible light driven CO2 reduction catalyst. Phys.Chem. Chem. Phys., 16 (13), 5922–5926.

114 An, X., Li, K., and Tang, J. (2014) Cu2O/reduced graphene oxide compositesfor the photocatalytic conversion of CO2. ChemSusChem, 7 (4), 1086–1093.

115 Tran, P.D., Batabyal, S.K., Pramana, S.S. et al. (2012) A cuprousoxide-reduced graphene oxide (Cu2O-rGO) composite photocatalyst forhydrogen generation: employing rGO as an electron acceptor to enhance thephotocatalytic activity and stability of Cu2O. Nanoscale, 4 (13), 3875–3878.

116 Li, H., Zhang, X., and MacFarlane, D.R. (2014) Carbon quantum dots/Cu2Oheterostructures for solar-light-driven conversion of CO2 to methanol. Adv.Energy Mater., 5, 1401077.

117 Wang, Z., Liu, Y., Martin, D.J. et al. (2013) CuOx-TiO2 junction: what isthe active component for photocatalytic H2 production? Phys. Chem. Chem.Phys., 15 (36), 14956–14960.

118 Jing, L., Zhou, J., Durrant, J.R. et al. (2012) Dynamics of photogeneratedcharges in the phosphate modified TiO2 and the enhanced activity forphotoelectrochemical water splitting. Energy Environ. Sci., 5 (4), 6552–6558.


119 Cao, Y., Jing, L., Shi, X. et al. (2012) Enhanced photocatalytic activity ofnc-TiO2 by promoting photogenerated electrons captured by the adsorbedoxygen. Phys. Chem. Chem. Phys., 14, 8530–8536.

120 Abdi, F.F., Savenije, T.J., May, M.M. et al. (2013) The origin of slow car-rier transport in BiVO4 thin film photoanodes. J. Phys. Chem. Lett., 4,2752–2757.

121 Jeon, T.H., Choi, W., and Park, H. (2011) Cobalt-phosphate complexes cat-alyze the photoelectrochemical water oxidation of BiVO4 electrodes. Phys.Chem. Chem. Phys., 13 (48), 21392–21401.

122 Abdi, F. and van de Krol, R. (2012) Nature and light dependence of bulkrecombination in Co-Pi-catalyzed BiVO4 photoanodes. J. Phys. Chem. C,116, 9398–9404.

123 Zhong, D.K., Choi, S., and Gamelin, D.R. (2011) Near-complete sup-pression of surface recombination in solar photoelectrolysis by “Co-Pi”catalyst-modified W:BiVO4. J. Am. Chem. Soc., 133 (45), 18370–18377.

124 Pilli, S.K., Deutsch, T.G., Furtak, T.E. et al. (2012) Light induced water oxi-dation on cobalt-phosphate (Co-Pi) catalyst modified semi-transparent,porous SiO2–BiVO4 electrodes. Phys. Chem. Chem. Phys., 14 (19),7032–7039.

125 Jiang, C., Moniz, S.J.A., Khraisheh, M., and Tang, J. (2014) Novel earth-abundant oxygen evolution catalysts coupled onto ZnO nanowire arraysfor efficient photoelectrochemical water cleavage. Chem. Eur. J., 20,12954–12961.

126 Zhong, D.K., Sun, J., Inumaru, H., and Gamelin, D.R. (2009) Solar wateroxidation by composite catalyst/α-Fe2O3 photoanodes. J. Am. Chem. Soc.,131 (17), 6086–6087.

127 Zhong, D.K. and Gamelin, D.R. (2010) Photoelectrochemical water oxidationby cobalt catalyst (“Co-Pi”)/alpha-Fe2O3 composite photoanodes: oxygenevolution and resolution of a kinetic bottleneck. J. Am. Chem. Soc., 132 (12),4202–4207.

128 Zhong, D.K., Cornuz, M., Sivula, K. et al. (2011) Photo-assisted electrode-position of cobalt–phosphate (Co-Pi) catalyst on hematite photoanodes forsolar water oxidation. Energy Environ. Sci., 4 (5), 1759–1764.

129 Barroso, M., Cowan, A.J., Pendlebury, S.R. et al. (2011) The role of cobaltphosphate in enhancing the photocatalytic activity of α-Fe2O3 toward wateroxidation. J. Am. Chem. Soc., 133 (38), 14868–14871.

130 Barroso, M., Mesa, C., Pendlebury, S.R. et al. (2012) Dynamics of photo-generated holes in surface modified α-Fe2O3 photoanodes for solar watersplitting. Proc. Natl. Acad. Sci. U.S.A., 109, 15640–15645.

131 Bora, D.K., Braun, A., Erni, R. et al. (2013) Hematite-NiO/α-Ni(OH)2 het-erostructure photoanodes with high electrocatalytic current density andcharge storage capacity. Phys. Chem. Chem. Phys., 15 (30), 12648–12659.

132 Li, J., Meng, F., Suri, S. et al. (2012) Photoelectrochemical performanceenhanced by a nickel oxide-hematite p-n junction photoanode. Chem. Com-mun., 48 (66), 8213–8215.

133 Tilley, S.D., Cornuz, M., Sivula, K., and Grätzel, M. (2010) Light-inducedwater splitting with hematite: improved nanostructure and iridium oxidecatalysis. Angew. Chem. Int. Ed., 49 (36), 6405–6408.

References 225

134 Hong, Y.-R., Liu, Z., Al-Bukhari, S.F.B. et al. (2011) Effect of oxygenevolution catalysts on hematite nanorods for solar water oxidation. Chem.Commun., 47 (38), 10653–10655.

135 Seabold, J.A. and Choi, K.S. (2011) Effect of a cobalt-based oxygen evolutioncatalyst on the stability and the selectivity of photo-oxidation reactions of aWO3 photoanode. Chem. Mater., 23 (5), 1105–1112.

136 Liu, R., Lin, Y., Chou, L.-Y. et al. (2011) Water splitting by tungsten oxideprepared by atomic layer deposition and decorated with an oxygen-evolvingcatalyst. Angew. Chem. Int. Ed., 50 (2), 499–502.

137 Berera, R., van Grondelle, R., and Kennis, J.T.M. (2009) Ultrafast tran-sient absorption spectroscopy: principles and application to photosyntheticsystems. Photosynth. Res., 101 (2–3), 105–118.

138 Tang, J., Cowan, A., Durrant, J.R., and Klug, D.R. (2011) Mechanism ofO2 production from water splitting: nature of charge carriers in nitrogendoped nanocrystalline TiO2 films and factors limiting O2 production. J. Phys.Chem. C, 115, 3143–3150.

139 Pendlebury, S.R., Barroso, M., Cowan, A.J. et al. (2011) Dynamics of pho-togenerated holes in nanocrystalline α-Fe2O3 electrodes for water oxidationprobed by transient absorption spectroscopy. Chem. Commun., 47 (2),716–718.

140 Ma, Y., Pendlebury, S., and Reynal, A. (2014) Dynamics of photogeneratedholes in undoped BiVO4 photoanodes for solar water oxidation. Chem. Sci.,5, 2964–2973.

141 Pesci, F. and Cowan, A. (2011) Charge carrier dynamics on mesoporousWO3 during water splitting. J. Phys. Chem. Lett., 2, 1900–1903.

142 Gelderman, K., Lee, L., and Donne, S. (2007) Flat-band potential of asemiconductor: using the Mott–Schottky equation. J. Chem. Educ., 84 (4),685–688.

143 Upul Wijayantha, K.G., Saremi-Yarahmadi, S., and Peter, L.M. (2011)Kinetics of oxygen evolution at α-Fe2O3 photoanodes: a study by photo-electrochemical impedance spectroscopy. Phys. Chem. Chem. Phys., 13 (12),5264–5270.

144 Garcia-Belmonte, G. (2013) Elucidating operating modes ofbulk-heterojunction solar cells from impedance spectroscopy analysis.J. Phys. Chem. Lett., 4, 877–886.

145 Li, H., Zhou, Y., Chen, L. et al. (2013) Rational and scalable fabrication ofhigh-quality WO3/CdS core/shell nanowire arrays for photoanodes towardenhanced charge separation and transport under visible light. Nanoscale,5 (23), 11933–11939.

146 Zhou, J., Zhang, M., and Zhu, Y. (2014) Preparation of visible light-driveng-C3N4@ZnO hybrid photocatalyst via mechanochemistry. Phys. Chem.Chem. Phys., 16, 17627–17633.

147 Ai, G., Li, H., Liu, S. et al. (2015) Solar water splitting by TiO2/CdS/Co-Pinanowire array photoanode enhanced with Co-Pi as hole transfer relay andCdS as light absorber. Adv. Funct. Mater., 25 (35), 5706–5713.

148 Thorne, J.E., Jang, J.-W., Liu, E.Y., and Wang, D. (2016) Understanding theorigin of photoelectrode performance enhancement by probing surfacekinetics. Chem. Sci., 7 (5), 3347–3354.


149 Klotz, D., Ellis, D., Dotan, H., and Rothschild, A. (2016) Empirical inoperando analysis of the charge carrier dynamics in hematite photoanodesby PEIS, IMPS and IMVS. Phys. Chem. Chem. Phys., 18, 23438–23457.

150 Donchev, V., Kirilov, K., Ivanov, T., and Germanova, K. (2006) Surface pho-tovoltage phase spectroscopy – a handy tool for characterisation of bulksemiconductors and nanostructures. Mater. Sci. Eng., B, 129 (1–3), 186–192.

151 Cavalcoli, D. and Cavallini, A. (2010) Surface photovoltage spectroscopy -method and applications. Phys. Status Solidi C, 7 (5), 1293–1300.

152 Sell, K., Barke, I., Polei, S. et al. (2010) Surface photovoltage of Ag nanopar-ticles and Au chains on Si(111). Phys. Status Solidi B, 247 (5), 1087–1094.

153 Kronik, L. and Shapira, Y. (1999) Surface photovoltage phenomena: theory,experiment, and applications. Surf. Sci. Rep., 37, 1–206.

154 Wang, J., Zhao, J., and Osterloh, F.E. (2015) Photochemical charge trans-fer observed in nanoscale hydrogen evolving photocatalysts using surfacephotovoltage spectroscopy. Energy Environ. Sci., 8, 2970–2976.

155 Zhang, L., Jiang, T., Li, S. et al. (2013) Enhancement of photocatalytic H2evolution on Zn0.8Cd0.2S loaded with CuS as cocatalyst and its photogener-ated charge transfer properties. Dalt. Trans., 42 (36), 12998–13003.

156 Zhao, J. and Osterloh, F.E. (2014) Photochemical charge separation innanocrystal photocatalyst films: insights from surface photovoltage spec-troscopy. J. Phys. Chem. Lett., 5 (5), 782–786.

157 Wang, J. and Osterloh, F.E. (2014) Limiting factors for photochemical chargeseparation in BiVO4/Co3O4, a highly active photocatalyst for water oxidationin sunlight. J. Mater. Chem. A, 2 (24), 9405–9411.

158 Jiang, B., Wang, Y., Wang, J.-Q. et al. (2013) In situ fabrication ofAg/Ag3PO4/graphene triple heterostructure visible-light photocatalystthrough graphene-assisted reduction strategy. ChemCatChem, 5 (6),1359–1367.

159 Meng, J., Li, F., Hu, Y. et al. (2013) Facile preparation of NiS/CdS-t com-posite photocatalyst for hydrogen evolution from aqueous solution ofsulphide/sulphite under visible light. Mater. Res. Bull., 48 (6), 2111–2116.

160 Lu, Y., Lin, Y., Xie, T. et al. (2012) Enhancement of visible-light-driven pho-toresponse of Mn/ZnO system: photogenerated charge transfer propertiesand photocatalytic activity. Nanoscale, 4 (20), 6393–6400.

161 Su, J., Zou, X., Li, G. et al. (2011) Macroporous V2O5–BiVO4 composites:effect of heterojunction on the behavior of photogenerated charges. J. Phys.Chem. C, 115, 8064–8071.

227

9

Conducting Polymers Nanostructures for Solar-LightHarvestingSrabanti Ghosh1, Hynd Remita2,3, and Rajendra N. Basu1

1CSIR – Central Glass and Ceramic Research Institute, Fuel Cell and Battery Division, 196, Raja S.C. MullickRoad, Kolkata 700032, India2Univ Paris-Sud, Université Paris-Saclay, Laboratoire de Chimie Physique, UMR 8000, UMR 8000 91405 Orsay,France3CNRS, Laboratoire de Chimie Physique, UMR 8000 91405 Orsay, France

9.1 Introduction

Recently, environmental pollution by hazardous chemicals has become seriousissue, and the development of sustainable technology using solar light is alreadyan alternative and established idea [1]. Upon solar-light irradiation of a semicon-ductor, excited electron–hole pairs are generated to initiate chemical reactionsfor the degradation and mineralization of pollutant compounds [2]. Oxide-basedsemiconductor, inorganic sulfides, and (oxy)nitrides have been extensively inves-tigated for photocatalytic applications [2, 3]. Particularly, titanium dioxide (TiO2)is the most widely used photocatalyst but limited utilization due to low quantumyield with the fast charge carriers (e−/h+) recombination and the necessity to useUV irradiation (3–4% of the solar light impinging on the Earth’s surface) [4–6].The main challenge in the field of photocatalytic degradation is the electronicstructure engineering of photoactive materials, which may be able to harvest solarradiation producing electron–hole couples and ensure efficient charge separa-tion for visible-light-driven degradation processes. The most important factorsdominating for efficient solar-light harvesting and converting solar energy intochemical energy are as follows:

a) Photo response of the material should optimally match the solar spectrum.b) Photoexcited charges must be efficiently separated to prevent recombination.c) Charges should have sufficient energy to carry out the desired chemical reac-

tions.d) Photocatalyst must be photostable, chemically inert, reusable, and of low cost.

There has been intense research activity for the development of visible-light-active semiconductor photocatalysts for solar energy conversion. Thedevelopment of innovative photocatalytic materials, based on the engineeringof electronic structure and heterojunctions produced by different techniques,and the modification of the catalysts surface by noble metals or cocatalysts to


228 9 Conducting Polymers Nanostructures for Solar-Light Harvesting

achieve increased charge separation have been widely explored [7–15]. However,the current photocatalytic efficiency under visible light is still moderate andsynthesis of stable visible-light-active photocatalysts remains a big challenge forlarge-scale applications. Detoxification of wastewater from dyes and pigmentsis a matter of great concern since the disposal of colored wastes such as dyesfrom industries (such as textiles, paper, plastics, tannery, and paints) into wateris highly toxic to aquatic life and creates serious environmental pollutions [16,17]. In this regard, solar-light-induced photocatalysis has been consideredas the economic way for the efficient removal of toxic dyes from wastewater[18]. In this chapter, we briefly overviewed the recent research activity onvisible-light-active conjugated polymer nanostructures for water depollution.Moreover, it has been found that the nanostructured polymer is the essentialcriteria for obtaining photoresponse in the visible spectral range. There has beenalso increasing interest in hybrid nanomaterials. The combination of conductingpolymers with inorganic compounds has proven to be attractive for a wide rangeof applications, such as optoelectronic device, energy storage, and solar cells.To efficiently harness the solar energy in the visible region, the combinationof organic conducting polymers and inorganic semiconductors is required tocontrol the composition and morphology at the nanoscale. The present chapterincludes synthesis and visible light driven photocatalytic activity of conductingpolymer nanostructures and experimental parameters such as nature of thephotocatalyst and evaluate their effectiveness in solar-light harvesting.

9.2 Conducting Polymers as Organic Semiconductor

In 1977, MacDiarmid, Shirakawa, and Heeger et al. discovered that the con-ductivity of polyacetylene (PA) after doping with chlorine, bromine, or iodinevapor increased up to ninefold and reached the order of 103 S cm−1 [19, 20].This was classified as conducting polymer (also known as conjugated polymers,CP), having intrinsic conducting properties, which opened the field of organicelectrical conductors [21, 22]. A significant progress has been accomplishedin the understanding of the structural and electronic requirements allowing toimprove the semiconducting characteristics for their potential use as electroac-tive materials in diverse organic electronic devices such as electroluminescence[23], light-emitting devices [24, 25], photovoltaics [26, 27], or chemical sensors[28]. Up to now, small conjugated oligomers have been synthesized and a varietyof CPs, for example, PA, polyaniline (PANI), polypyrrole (PPy), polythiophene(PT), poly(3,4-ethylenedioxythiophene) (PEDOT), and other PT derivativeshave been developed (Figure 9.1). These polymers are depicted as mainly p-typeorganic semiconductors with a bandgap lying in the range of 1.5–3 eV in theirfundamental neutral state, and the electronic structure of these polymershas been interpreted in a traditional inorganic semiconductor band picturewithout considering the electron–electron interactions [29, 30]. In contrast tothe inorganic semiconductors, coulomb and exchange interactions between

9.2 Conducting Polymers as Organic Semiconductor 229

Polyaniline (PANI)

Poly(acetylene) Poly(pyrrole)O

PEDOT

O

S

PA PPy

* * *

* **

Poly(3,4-ethylenedioxythiophene)

N N N Nn

nn

**

H

H

N

H

Figure 9.1 Molecular structure of some representative conducting polymers.

charges are significant parameters in π-conjugated polymers [31]. The polymerconsists of alternating single and double bonds known as conjugated doublebond. The conductivity of organic semiconductor is extrinsic and originatedfrom (i) oxidation with halogen (or p-doping) or (ii) reduction with alkali metal(called n-doping) and from the dissociation of photogenerated electron–holepairs that are bound by their mutual coulomb attraction. Unlike inorganicsemiconductors, doping in conjugated polymers is reversible in a way that upondedoping the original polymer can be retained with almost no degradation ofthe polymer backbone. The first example of conjugated polymer is PA [19].High electrical conductivity was observed when the polymer was “doped” withoxidizing or reducing agents. For example, oxidation of PA chain by iodine I2,which abstract an electron from the polymer chain, creates “soliton” defect(in which the change in bond alternation is extended over 5 to 9 repeatingunits) and forms I3

− counter anion (Scheme 9.1). Moreover, isolated solitonsare not stable in polymers and further charge exchange leads to the formationof S0–S+ (or S0–S−) pairs, which is strongly localized to form a polaron. Thecharge of the defect generally depends on the occupancy of the energy state as

doping

Polaron in polyacetylene Bipolaron in polyacetylene

Soliton+

Pure polyacetylene

+

+

Scheme 9.1 Schematic representation of the formation of different types of charge carriers inconducting polymers upon doping.


Neutral soliton

S0Positively charged

Soliton S+Negatively charged

Soliton S–

Conduction band

Valence band

EF

Band energy

CB

VB

CB

VB

CB

VB

Scheme 9.2 Schematic representation of the modifications in the band structure ofconducting polymers after the creation of the localized defects upon doping.

Conjugated polymerTraditional polymer Doped conjugated

polymer

Energy

Conduction

bandConduction

band

Conduction

band

Valence

band

Valence

band

Valence

band

EgEg > 5 eV

The bands

overlap

Electrons free

establish

conduction

Valence electrons

bound to the

atomic structure

SemiconductorInsulator

Conductor

Scheme 9.3 Difference between band structure of conventional polymer, undoped, anddoped conducting polymer.

shown in Scheme 9.2. When a further electron is removed, it is energeticallymore favorable to remove the second electron from the polaron than fromanother part of the polymer chain, which leads to the formation of one bipolaron[32]. The mobility of a polaron along the PA chain can be high and charge iscarried along the backbone. However, the counter anion I3

− is not very mobile,and a high concentration of counter anion is required, so that the polaroncan move close to the counter anion. The conjugated double bonds in thepolymer backbone allow the free movement of charges within the conjugatinglength, which makes them electrically conductive. Polaron as well as bipolaronstates are located probably in the middle of the bandgap (the Fermi energylevel) corresponds to the electrochemical potential of an electron as shown inScheme 9.3. Semiconducting organic polymers have many potential advantagesover traditional inorganic semiconductors including lightweight, flexible nature,and cost-effective manufacturing process. This chapter focuses on the synthesis,characterization of CPs, and current state of the art in the photocatalyticapplication of CPs under visible light.

9.4 Synthesis of Conducting Polymer Nanostructures 231

9.3 Conducting Polymer-Based NanostructuredMaterials

A dramatic change in various physicochemical properties of CPs in nanoscaledimensions has been realized compared to their bulk counterpart. CP-basednanostructured materials may be broadly classified into two categories:

a) CP nanostructures such as nanoparticles, nanowires, nanofibers, nanotubesof pure CP

b) CP nanocomposites, which are mixtures of metal or metal oxide nanoparticleswith CP at nanoscale.

CP nanostructures demonstrated high conductivity, stability, high surfacearea, unique optoelectronic properties, flexibility, and processability [33, 34].For example, CP nanostructures showed high electrical conductivity com-pared to their macrostructures [35]. Another example, electrical conductivityof the poly-(diphenylbutadiyne) (PDPB) nanofibers is 0.13 S cm−1, which isseveral orders of magnitude higher than that of the conductivity of bulk PA(10−11 S cm−1) [36]. In fact, the conductivity of PDPB nanofibers depends on thediameter; the conductivity increases with decreasing diameter of nanofibers.Notably, PANI nanoparticles (NPs) having very smaller size of 4 nm and highcrystallinity demonstrated high conductivity of 85 S cm−1 due to a highlycompact and ordered structure of PANI chains [37]. A correlation betweenmorphology and electric conductivity of polymer nanostructures has beenreported in several literatures [38, 39]. The electrical conductivity and the powerfactor follow this sequence: bulk PEDOT< globular nanoparticles<nanorod orellipsoidal nanoparticles<nanotubes<nanofibers. A significant effect of nano-scale dimensions on optical properties of CPs has been also observed for theutilization in light-emitting or photovoltaic devices [40, 41]. The size-dependentoptoelectronic properties of CP NPs having the continuous bathochromicabsorption and significantly enhanced emission drastically different from themacromolecular building blocks, which are useful for device fabrication [42].For inorganic semiconductor NPs, the particle size is comparable to or smallerthan the Bohr radius of the Wannier excitons. In contrast, the smaller Frenkelexcitons in organic semiconductor associated with the extended π-conjugationsystems, whose exciton size can be tuned by both the chemical alteration of theπ-conjugated molecular structures and their intermolecular interactions [43].Hence, compared with their bulk forms, nanostructured conductive polymersexhibit improved physicochemical properties and shortened pathways for chargeor mass transport which is ideal for applications in solar-energy harvesting.

9.4 Synthesis of Conducting Polymer Nanostructures

CPs have been usually synthesized via chemical, electrochemical, or photoin-duced oxidation (or reduction) of monomers, followed by the coupling of thecharged monomers to produce the polymer chains. There has been considerable


Synthesis

Hard template

Track-etch polycarbonate or

polyester and anodic aluminum

oxide (AAO) membranes

Self-organized template,

lyotropic liquid crystal (LC)

Layer-by-layer self-

assembly

Micro-/mini-emulsion

polymerization

Solid porous materials, zeolites,

metal oxides V2O5 fibers or MnO2

nanowires, polyoxometallates

Template free

Soft template

Scheme 9.4 Methods of synthesis of conducting polymer nanostructures.

improvement in the large-scale production of CP nanostructures (CPNs)with different approaches such as the conventional hard-template method,soft-template method, and template-free synthesis [44]. CP nanostructureshave been synthesized using different techniques, namely, micellar and reversemicellar polymerization, interfacial polymerization, rapid mixing polymeriza-tion, seeding polymerization, microemulsion polymerization, electrospinningand polymerization in the presence of hard and soft templates during thepolymerization process (Scheme 9.4).

9.4.1 Hard Templates

The hard templates such as track-etch polycarbonate (PC) or polyester (PE) mem-branes and anodic aluminum oxide (AAO) membranes have been employed forthe controlled synthesis of nanorods, nanofibers, nanotubes of CP such as PANI,PEDOT, PPy, and P3HT [45, 46]. In this method, a template membrane has beenusually used to grow nanostructures inside the pores or channels of the mem-branes to control the size and shape of the nanostructures. Moreover, zeolites,silica-based mesoporous molecular sieves, metal oxides, polyoxometalates, solidporous materials, and so on can be used as hard template for the synthesis ofnanostructures [45, 47, 48]. However, the majority of these approaches involvemultistep synthetic routes to premodify core templates, and not suitable for thelarge-scale production.

9.4.2 Soft Templates

Many molecular templates, surfactants, micelles, liquid crystalline phases,and structure-directing molecules, and so on, have been employed for thecontrolled synthesis of CP nanostructures based on self-assembly mechanismsusing hydrogen bonding, van der Waals forces, π–π stacking, electrostaticinteractions, and so on. The size and morphology of the CP are predominantlydetermined by the preassembled molecular templates [49–51]. For example,

9.5 Applications of Conducting Polymer 233

a soft-template, lyotropic liquid crystal (LC) with hexagonal mesophases hasbeen used for the synthesis of anisotropic CP nanostructures, which cannotbe achieved by using traditional bulk or solution polymerization methods[52, 53]. Remita and coworkers developed swollen hexagonal mesophasescomposed of oil-swollen tubes with tunable diameters. These soft templatescan be used to control the morphology and the size of the nanostructures [54].Ghosh et al. reported synthesis of the PEDOT nanostructures with spindle-likeor vesicle-like shapes in the hydrophobic domains of hexagonal mesophasesvia chemical oxidative polymerization of EDOT monomers using FeCl3 asoxidizing agent [55]. Ghosh et al. also reported the controlled synthesis ofpoly-(diphenylbutadiyne) nanofibers of 5–25 nm which is directly determinedby the diameter of the oil tubes of the hexagonal mesophases by photoinducedradical polymerization using a chemical initiator or by gamma irradiation[36]. Notably, this one-dimensional (1D) structure reflects the geometry of thehydrophobic domains of the hexagonal mesophases.

9.4.3 Template Free

The template-free method is considered a simple and straightforward techniquefor the synthesis of CPNs without the need of template and no posttreatment fortemplate removal [56]. Various CPNs such as nanotubes, nanofibers, and hollowspheres, and so on, have been successfully synthesized by the template-freemethod. Template-free synthetic strategies include interfacial polymerizationthrough self-assembly, electrospinning, and so on. Huang et al. employedaqueous/organic interfacial polymerization under ambient conditions [57]. Ininterfacial polymerization technique, the mass and charge diffusion through aliquid–liquid interface controls crystallinity, size, and shape of polymer [58–60].However, the mechanisms of self-assembly are complicated and not clearlyunderstood. Electrospinning is one of the most efficient techniques to generateCP nanofibers and composites under a high electric field [61]. Miao et al.reported the fabrication of hollow PANI nanofibers by electrospinning [62].The electrospinning technique produces continuous long nanofibers; however,non-CPs or support are usually added which lower the conductivity of the elec-trospun composite fibers. Other methods are also reported such as the directedelectrochemical nanowires assembly technique, which has been used to grow CPnanowires or soft lithography with the assistance of electrodeposition [63, 64].

9.5 Applications of Conducting Polymer

9.5.1 Conducting Polymer Nanostructures for Organic PollutantDegradation

Recently, the first experimental evidence of photocatalytic activity of conjugatedpolymers nanostructures has been demonstrated with PDPB nanofibers andPEDOT as active photocatalysts under visible light for water depollution [65].PDPB nanofibers with a diameter of ∼19 nm and a few micrometers long


Ar

Ar Ar

*

ArAr

Ar

Ar

*

Ar

n

1,4-diphenylbutadiyne

Ar = Phenyl groupPoly(diphenylbutadiyne)

(PDPB)

hν

Polymerization

Scheme 9.5 Schematic representation of polymerization of diphenylbutadiyne (DPB) by UVirradiation.

have been synthesized using hexagonal mesophases as soft templates and thehydrophobic domain of the mesophases can accommodate high concentrations(up to 20 wt%) of 1,4-diphenylbutadiyne (DPB) monomer, which can be directlypolymerized by photoirradiation in the presence of a free-radical initiator(benzoin methyl ether, BME, 1%) via 1,4-addition reaction to form alternatingene-yne polymer chains (Scheme 9.5). In contrast, micron-sized sphericalparticles (denoted as bulk PDPB) have been obtained by photopolymerizationof DPB (in the presence of BME) in cyclohexane in the absence of mesophases,clearly demonstrating the templating effect of the mesophase for the generationof the 1D nanostructure. The molar mass of polymer has been found to be1625 g mol−1 as determined by gel permeation chromatography (GPC), whichcorresponds to oligomers of degree of polymerization 8. The PDPB nanofibersare probably formed by π-stacking of the oligomers, and consequently withthe formation of relatively short polymer chains in the confined domains of oiltubes of the mesophases. The as-prepared PDPB nanostructures have a broadabsorption in the visible range. PDPB nanofibers exhibit a high photocatalyticactivity under both UV–visible (xenon lamp) and visible light (using a filter at𝜆> 450 nm) for the degradation of model pollutants (phenol and methyl orange,MO without using any sacrificial agent or cocatalysts. PDPB nanofibers showeda high photocatalytic activity, 75% photodegradation of MO under visible-lightirradiation much higher than that of plasmonic Ag-modified TiO2 (19%). Incontrast, bulk PDPB demonstrated a low photocatalytic activity under both UVand visible light. The difference in the photocatalytic activity between nano andbulk PDPB may be due to larger size and the presence of more defects in bulkPDPB favoring fast e−–h+ recombination. This dependence of the photocatalyticactivity on the size and morphology of the structure has been also reportedfor TiO2 [66, 67]. Moreover, cyclic voltammetry (CV) method can be used tomeasure electronic properties of CP and analysis of the CV profile of polymerstructures suggests onsets of oxidation and reduction processes occurring atlower potentials, having a much lower energy gap around 1.81 eV. Moreover,considering oligomeric PDPB structures comprising of various numbers of unitsfrom 1 to 8, the calculated value of the PDPB bandgap is ∼1.95 eV on the basisof density functional theory (DFT) calculation. Hence, PDPB nanostructuresacts as an organic semiconductor and when illuminated with photons of energyexceeding or equal to the bandgap (E ≥ 1.81 eV or 𝜆≤ 685 nm), excess electronsand holes are formed in the conjugated polymer chains. When electron–holeescape recombination, the electrons and holes can migrate to the semiconductor


surface and generate highly oxidative radicals (O2∙− superoxide radical, ∙OH

radicals, etc.), which causes degradation and mineralization of organic pollutants[6]. The details of oxidative radicals involved during the photocatalytic oxidationprocess have been described in Chapter 17.

Under irradiation, electrons are injected from the CP and react with oxygen toform the oxidizing O2

∙− superoxide radical (Eq. (9.1)):

O2 + e− → O2∙− (9.1)

According to the valence band energy level, OH− cannot react with the holesto yield oxidative ∙OH, however, a small amount of ∙OH radicals can be formedby the following reactions (Eqs. (9.2)–(9.5)):

O2∙− + H+ → HO2

∙ (9.2)2HO2

∙ → H2O2 + O2 (9.3)H2O2 + O2

∙− → ∙OH + O2 + OH− (9.4)H2O2 + h𝜈 → 2∙OH (9.5)

In parallel, the holes (h+) also diffuse to the surface and may also directly oxidizethe pollutant molecules during catalytic degradation reaction.

Another example, PEDOT nanostructures synthesized in soft templates viachemical oxidative polymerization demonstrate unprecedented photocatalyticactivity for organic pollutant degradation without the assistance of any sacrificialreagents or noble metal cocatalysts and better than TiO2 as benchmark catalyst[68]. Polymer nanostructures are synthesized via chemical oxidative polymer-ization of EDOT monomers using FeCl3 as an oxidizing agent as shown inScheme 9.6. The polymerization occurs by stepwise RC–RC (radical cation, RC)coupling of oxidized EDOT monomers or oligomers in the presence of an oxidant[69]. Depending on the mesophase composition, PEDOT spindle nanostructuresof 40 nm thick and several hundred nanometers long, while the PEDOT vesiclesof spherical hollow capsules of diameter around 1 μm with walls of thicknessaround 40 nm have been obtained after extraction from mesophases. Interest-ingly, PEDOT nanospindles showed an efficient photocatalytic activity underUV and visible light both for phenol and MO and the photocatalytic activity has

O

S

S S

S SO

O O O O

O O

n

O O

FeCl3

Polymerization

3,4-Ethylenedioxythiophene

(EDOT) Poly 3,4-ethylenedioxythiophene

(PEDOT)

Scheme 9.6 Schematic representation of chemical oxidative polymerization of3,4-ethylenedioxythiophene (EDOT) using FeCl3 as chemical oxidant.


0 50 100 150 200 250 300

Time of UV exposure (min)(a) (b)

(c) (d)

0 50 100 150 200 250

Time of visible exposure (min)

0 10 20 30 40 50

Time of UV exposure (min)0 50 100 150 200 250

Time of visible exposure (min)

C/C

0 (

Phenol)

0.0

0.2

0.4

0.6

0.8

1.0

C/C

0 (

MO

)

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0C

/C0 (

Phenol)

0.0

0.2

0.4

0.6

0.8

1.0

C/C

0 (

MO

)Phenol-UV

MO-UV

Phenol-Vis

PDPB NanoPEDOT SpindlePEDOT Vesicle

Ag-TiO2

P25-TiO2

PDPB Nano

PEDOT Spindle

PEDOT Vesicle

Ag-TiO2

P25-TiO2

MO-Vis

PDPB NanoPEDOT SpindlePEDOT Vesicle

Ag-TiO2

P25-TiO2

PDPB NanoPEDOT Spindle

PEDOT Vesicle

Ag-TiO2

P25-TiO2

Figure 9.2 Photocatalytic degradation of (a, b) phenol and (c, d) methyl orange (MO) in thepresence of commercial P25 TiO2 and Ag-TiO2, PDPB nanofibers and the synthesized PEDOTvesicles and PEDOT nanospindles under UV (a, c) and visible-light (>450 nm) (b, d) irradiation.(Reproduced with permission from Ref. [68]. Copyright 2015, Nature Publishers.)

been found to be much higher than that of commercial P25 TiO2 (Figure 9.2a–d).A complete degradation of phenol has been observed for TiO2, Ag-TiO2, andPEDOT nanospindles, after irradiation for 60, 15, and 10 min, respectively,whereas only 30% degradation has been achieved for PEDOT vesicles under UVlight and 100% of phenol is degraded with PEDOT nanospindles after 240 minirradiation under visible light (Figure 9.2a). Remarkably, 100% degradation ofMO was achieved by using PEDOT nanospindles after 15 min UV light irradi-ation as shown in Figure 9.2c. The PEDOT spindles displayed a significant MOphotodegradation achieving 100% degradation after 180 min under visible-lightirradiation (Figure 9.2d). It has to be noted that the photocatalytic activity ofPEDOT nanospindles has been found to be even higher than the one recentlyreported for PDPB nanofiber. In fact, 100% phenol degradation observed withPEDOT nanospindles under visible light, while only 64% of phenol is degradedwith PDPB. Notably, PEDOT vesicles remained totally inactive for both phenoland MO degradation under visible light. The stable and good recyclabilityof photocatalyst is highly desirable for industrial applications. The PEDOTnanospindles can be efficiently recycled and reused for repeated cycles withoutappreciable loss of activity. The PEDOT nanospindles are very stable photocat-alysts; no differences have been found in the morphology after photocatalyticdegradation as tested by microscopy characterizations. The photocatalytic


activity of PEDOT spindles was retained at over 98% and 95% of its originalactivity for phenol and MO, respectively, after six successive experimental runs.The total mineralization of the organic pollutants has been followed using acommon technique, the disappearance of the total organic carbon (TOC) forexpressing the detoxification level of water. Polymer-based photocatalysis wasable to fully oxidize the organic pollutant and dye, with an almost complete min-eralization of carbon into CO2 andH2O (TOC∼ 90%). CV study reveals the onsetof oxidation and reduction processes occurring at lower potentials having a muchlower energy gap around 1.69 eV than TiO2 (3.2 eV). When illuminated with pho-tons of energy exceeding (or equal to) the bandgap (E ≥ 1.69 eV or 𝜆≤ 733 nm),excess electrons and holes are formed in the PEDOT polymer chains. Theeffective separation of photogenerated electron–hole may promote the superiorphotocatalytic activity. Meanwhile, the photogenerated holes on PEDOT (ECB,+0.139 vs Ag corresponding to +0.667 eV vs SHE) cannot produce hydroxylradicals. Hence, photocatalysis mechanism involves O2

∙−, photoinduced h+, andHO∙ radicals mediated degradation of organic pollutant with effective chargeseparation in PEDOT nanospindles. Hence, the application of conjugated nanos-tructures in the field of photocatalysis can be generalized to other polymers.Poly(3-hexylthiophene) (P3HT) is one of the most used conjugated polymersin photovoltaics application. P3HT nanostructures have been synthesized inhexagonal mesophases, which showed high photocatalytic activity for degra-dation of phenol and rhodamine B under both UV and visible light [70]. TheseP3HT nanostructures photocatalysts can easily be deposited on flat supportssuch as quartz for photocatalytic applications avoiding a separation step by cen-trifugation. The photocatalytic activity of these supported P3HT nanostructuresis much enhanced with highly accelerated phenol degradation kinetics, whichreveal a new perspective in photocatalytic reactors and self-cleaning surfaces.

9.5.2 Conducting Polymer Nanostructures for Photocatalytic WaterSplitting

The hydrogen production by photocatalytic water splitting is one of the primechemical challenges in solar energy utilization for large-scale applications ofhydrogen [71, 72]. The first organic semiconductor such as poly(p-phenylene)and poly(azomethine) have been reported to allow a hydrogen evolution rateof 2.1 μmol h−1 from water in the presence of triethylamine or diethylamine assacrificial agents with lower performance in the UV region [73]. A linear poly-mer containing pyridyl units have been synthesized by Ni-catalyzed Yamamotocoupling of 2,5-dibromopyridine, which showed a reduced bandgap of 2.4 eVand due to enhanced photoinduced charge separation results in an almost10-fold enhancement for hydrogen evolution under visible-light irradiation(>400 nm) compared to poly-p-phenylene (PPP) [74]. The rate of photocatalysishas been improved further by a factor of 14 in the presence of colloidal Ru asa cocatalyst [75]. Moreover, Matsuoka et al. who compared a series of linearoligomeric p-phenylene chains OPP-n (n= 2–6) reported that three p-phenylene(OPP-3) units were necessary for photoinduced charge separation and forhydrogen evolution [76]. Hence, the hydrogen evolution efficiency increased


with increasing the chain length. Further, Cooper and coworkers also reported anincrease in photocatalytic hydrogen evolution with increasing oligomer size [77].It is important to note that polymeric carbon nitride (g-C3N4) with a bandgapof around 2.7 eV was also found to be an efficient metal-free photocatalyst thatproduces hydrogen from water under visible-light irradiation, but a sacrificialdonor is required. The presence of platinum nanoparticles as catalyst enhancedthe hydrogen production [78]. Loading the photoactive g-C3N4 material withRuO2 used as water oxidizing catalyst leads to water oxidation in the presenceof Ag+ ions as electron scavenger under visible-light irradiation. This discoveryhas spurred new possibilities to prepare photoactive catalytic polymers. Liuet al. have found that a metal-free new hybrid photocatalyst system made ofcarbon nanodot-carbon nitride nanocomposite (Cdots-C3N4) could split waterin two steps. The overall solar energy conversion efficiency was about 2.0%[79]. Incorporation of Cdots into the C3N4 matrix leads to an increase in theUV-visible absorption over the entire wavelength range and the proper positionof the reduction level for H2 and the oxidation level for H2O to H2O2 or O2in the bandgap of Cdots-C3N4 hybrid catalyst. Recently, Zhang et al. reporteddirect splitting of pure water by light-excited graphitic carbon nitride (g-C3N4)modified with Pt, PtOx, and CoOx as cocatalysts without using any sacrificialreagents. The as-prepared photocatalyst was stable for 510 h of reaction [80].For this modified g-C3N4, the apparent quantum yield (AQY) for the overallwater-splitting reaction was calculated to be 0.3% at 405 nm. Furthermore,Bhunia et al. [81] synthesized the triazine-based crystalline g-CN through thecombination of supramolecular aggregation and polycondensation by usingmelamine as a precursor and 2,4,6-triaminopyrimidine as a dopant, whichfacilitates the high crystallinity and a remarkably increased H2 evolution with aquantum yield of about 7% at 420 nm [81]. A detailed discussion about (g-C3N4)can be found in Chapter 12. Very limited organic semiconductors reported inliterature, which has enabled water to be reduced into hydrogen in the absenceof a sacrificial agent under visible-light irradiation. Further, Wang and coworkersdeveloped a novel organic semiconductor photocatalyst mimicking naturallight-harvesting antenna complexes in photosynthetic organisms, a disulfide(—S—S—) bridged C3N3S3 conjugated polymer for efficiently generating H2from pure water under visible-light irradiation without the need of a sacrificialelectron donor [82].

Recently, conjugated microporous polymers (CMPs) have been used asphotocatalysts for hydrogen evolution from water in the presence of a sacri-ficial electron donor. Cooper and coworkers synthesized CMP networks byan extended biphenyl analog using Pd(0)-catalyzed Suzuki–Miyaura poly-condensation [83]. All of the CMPs (having tunable optical gap from 1.94 to2.95 eV) demonstrated hydrogen production under visible light. A gradualincrease in the hydrogen evolution rate from CP-CMP1 to CP-CMP10 hasbeen obtained with decreasing the optical gap of the polymer. Further loweringof optical gaps from CP-CMP11 through CP-CMP15, a sudden decrease inthe photocatalytic performance was observed as shown in Figure 9.3. Thissuggests that with increasing conjugation across the polymer chain, a bridging


20H

2 e

vo

lutio

n r

ate

(μm

ol h

–1)

15

10

5

0

1.8 2.0 2.2 2.4

Optical gap (eV)

2.6 2.8 3.0

Figure 9.3 Rate of photocatalytic hydrogen production can be correlated with the optical gapin the polymers. Data shown for networks CPCMP1−15 (black squares) and analogous linearpolymers (discussed below), P16−18 (open squares); all measurements relate to 100 mgcatalyst in water containing 20 vol% diethylamine as an electron donor under filtered, visibleirradiation (𝜆> 420 nm, E< 2.95 eV). (Reproduced with permission from Ref. [82]. Copyright2011, The American Chemical Society.)

group between the phenyls in the linear oligomers happened which leads togreater charge delocalization. Moreover, longer charge carrier lifetimes inducedby the increased conjugation length of polymer may have beneficial effect inphotocatalysis. The structure–performance relationships of these polymersare well discussed in a recent review paper [84]. In another report, Cooperand coworkers also studied the effect of linker geometry, the comonomerlinker length, and the degree of planarization with respect to the photocatalytichydrogen evolution rate [85]. They established a strong correlation between pho-tocatalytic performance and their light absorption profiles, and 1,3,5-linkagesare strongly detrimental to the photocatalytic activity because the absorptiononsets are significantly blue-shifted and photoluminescence lifetime is partiallycorrelated with the photocatalytic activity. In general, efficient light capture bythe photocatalyst is a primary step in the photocatalytic process. Hence, organicchromophores may provide a virtually unlimited variety of optoelectronic andphotophysical properties, high absorption cross section in the visible range andtunable optical bandgaps, high charge carrier mobilities and long-lived excitedstates that can be translated into the polymer. In another example, Li et al.reported synthesis of a new series of n-type porous conjugated polymers (PCPs)based on perylenediimide (PDI) with bandgap energies ranging from 1.54 to2.25 eV [86]. Interestingly, the incorporation of PDI and bipyridyl moietiesinto the network leads to the highest hydrogen evolution rate (∼7.2 μmol h−1),which is attributed to the improved charge transport properties (coplanarstructure) and better wettability properties. This indicates that catalytic effi-ciency strongly depends on monomer composition. Sprick et al. also showedthat addition of planarized fluorene, carbazole, dibenzo[b,d]thiophene ordibenzo-[b,d]thiophene sulfone units into linear poly(p-phenylene)s enhancesphotocatalytic activity for hydrogen generation under visible light [87].


For example, the dibenzo-[b,d]thiophene sulfone copolymer has an AQY of 2.3%at 420 nm, as compared to 0.1% for platinized commercial pristine carbon nitride.Very recently, Li et al. prepared PCP photocatalysts with conjugated donorchromophore 4,8-di(thiophen-2-yl)benzo[1,2-b:4,5-b′]-dithiophene (DBD,Figure 9.4a) and bipyridyl (bpy) unit by Suzuki Coupling [88]. Figure 9.4billustrates the photocatalytic hydrogen production of these polymers inwater/triethylamine mixture. Low photocatalytic activity of hydrogen evolution(1.9−10.1 μmol h−1) has been obtained and the hydrogen production increasesas the chain length of oligo-(phenylene) linker for donor–donor-based PCPs(PCP1−3), due to the lack of internal polarization for effective charge separationprocess and hydrophobicity of the PCPs. Further PCP4−8 based on a set ofpyridine ligand demonstrates structure-dependent photocatalytic activities.High hydrogen generation rate (59.8 μmol h−1) has been observed at para-substituted pyridine (M6) in comparison to meta-substituted pyridines (PCP5,PCP7, and PCP8) in the range of 18.2−34.9 μmol h−1. Moreover, PCP9−11containing stronger acceptor diazines as the building units exhibited highcatalytic activity, in which PCP10 and PCP11 (103.6 and 106.9 μmol h−1)significantly enhanced activities compared to PCP9 with moderate hydrogengeneration rate ∼30.4 μmol h−1 (Figure 9.4c). Figure 9.4c shows loading of2 wt% Pt cocatalyst, the AQYs at 400 nm can be enhanced from 1.05% to 1.93%,confirming the key role of metal cocatalyst in H2 formation. Hence PCP madeof fully conjugated donor chromophore DBD and bpy unit showed the bestphotocatalytic performance of ∼106 μmol h−1. Yu and coworkers also developedanother series of heterogeneous photocatalysts for hydrogen evolution based onin-chain cobalt-chelating conjugated polymers [89]. Two conjugated polymers,electron-donating block benzodithiophene (BDT) and the electron-acceptingblock PDI were copolymerized with the ligand block bpy by palladium catalyzedC—C coupling as shown in Figure 9.5a. To combine functions of the conjugatedbackbone as a light-harvesting antenna and electron-transfer with the in-chainbipyridyl-chelated transition metal centers as catalytic active sites, these twoconjugated polymers chelated with cobalt ions and used to generate photo-catalytically hydrogen over a period of 27 h (Figure 9.5b) from diethylamine(DEA)/water mixture under visible light. The rate of hydrogen generationdepend on the Co(II) concentration for both PBDT-bpy and PPDI-bpy with anoptimal hydrogen evolution rate of 0.71 μmol h−1 at significantly higher Co(II)loadings (maximum at ∼60%, Figure 9.5c). The catalytic activity of PCP forhydrogen generation has been found to be wavelength-dependent AQYs, wherePPDI-bpy shows an AQY profile that closely tracks the UV–vis absorptionspectrum activity under visible-light excitation (400–600 nm) (Figure 9.5d).On the other hand, the donor-BDT-based PBDT-bpy demonstrates increasein activity for wavelengths <400 nm. Due to the amalgamation of photoactiveπ–electron backbone and microporous properties, CMPs have recently emergedas an efficient and stable catalysts not only for hydrogen evolution but also forother chemical transformations promoted by visible light such as selective oxi-dation of organic sulfides, C—C bond formation, molecular oxygen activation,reductive dehalogenation reaction, oxidative hydroxylation of arylboronic acidsand cationic polymerization [90].

Br

Br

B

B

S

S

OO

DBD

(a)

(b) (c)

Br

Br

Br

Br

Br

Br Br BrN N

M0

M1

M2

M7 M8

M10 M11

M9

N N

N N

NN

N–N

NM4

NM5

NM6

Br

S

S

BO

OO

B

O

O

O

Br

Br Br

Br

Br Br

Br

M3

Br

120 2.5

PCP10 with 2% Pt

PCP10 with 0% Pt2.0

1.5

1.0

0.5

0.0350 400 450 500

Wavelength (nm)

550 600 650

H2 (μm

ol h

−1)

AQ

Y (

%)

100

80

60

40

20

0

PCP0

PCP1

PCP2

PCP3

PCP4

PCP5

PCP6

PCP7

PCP8

PCP9

PCP10

PCP11

Br Br Br Br Br

Figure 9.4 (a) Structures of comonomers (M0−M11) used for the preparation of PCP photocatalysts PCP0−PCP11, where the number refers to the number incomonomer used by Suzuki Coupling. (b) Photocatalytic hydrogen production rates of PCP0−11 under full-arc irradiation for 2 h. (c) Wavelength-dependentapparent quantum yields (AQY) for PCP10 with and without loading 2 wt% Pt cocatalyst. (Reproduced with permission from Ref. [88]. Copyright 2016, TheAmerican Chemical Society.)


0.8

PBDT-bpy PPDI-bpy(a)

(c)

(b)

(d)

0.0003

Time (h)

2.5

2.0

1.5

1.0

N2

bubbling

N2

bubbling

0.5

0.00.0 3.0

O

O

O

OS

S N N nO

O

O

O

O ON

O ON

NN

n

6.0 9.0 12.0 15.0 18.0 21.0 24.0 27.0

0.0002

0.0001A

QY

0.0000

0.6

0.4

H2 (μm

ol h

–1)

H2 (μm

ol)

0.2

0.00.0 0.2 0.4

[CoCl2]/[bpy]

0.6 0.8 1.0 300 350 400 450

PBDT-bpy

PDI-bpy

PBDT-bpy

PPDI-bpy

Wavelength (nm)

500 550 600 650

Figure 9.5 (a) Chemical structures of PBDT-bpy and PPDI-bpy conjugated polymers used forhydrogen generation. (b) Time course of H2 production from water for polymer PBDT-bpy with[CoCl2]/[bpy]= 0.1. (c) [CoCl2]/[bpy] dependence of the rate of photocatalytic hydrogenproduction from water. (d) Wavelength-dependent AQYs of water splitting for polymersPBDT-bpy and PPDI-bpy. (Reproduced with permission from Ref. [89]. Copyright 2016, TheAmerican Chemical Society.)

9.5.3 Conducting Polymer-Based Heterostructures

In order to enhance the catalytic activity, hybrid materials consisting ofπ-conjugated polymers (or oligomers) and semiconductor nanocrystals caninduce higher visible-light absorption and limit the charge carrier recombi-nation. These synergistic optical and electronic properties can lead to higherphotocatalytic activity and also allow immobilization of the nanocrystals forphotocatalytic applications [15, 91]. The electronic energy levels of the organicand inorganic components of the hybrid can be tuned, and this can directlyinfluence the charge carriers generation and separation. Therefore a large varietyof different approaches has been developed for assembling conjugated polymersand semiconductor nanocrystals. A series of semiconductor nanocrystals andbulk conjugated polymer-based heterojunctions such as TiO2-PANI, TiO2-P3HT,ZnO-PANI has shown an improved efficiency under solar light [10, 91, 92].However, the reported photoconversion efficiency is limited for semiconductorand CP heterojunctions [93]. Moreover, very few direct experimental evidencesor spectroscopic observations have been explored to establish the photoinducedcharge transfer mechanism at the heterostructure interface [91, 94]. Thus,investigations of the role of crystal defects, trapping sites and exciton-relaxationon the interfacial processes through spectroscopic studies toward the rationaldesign of light-harvesting nanoheterojunction (LHNH) are essential. The


(a) (b)

100 nm 5 nm

0.26 nm

(002)

Figure 9.6 Transmission electron microscopic images of (a) PDPB nanofibers and(b) PDPB-ZnO light-harvesting nanoheterojunction (LHNH). (Reproduced with permissionfrom Ref. [95]. Copyright 2015, Nature Publishers.)

synthesis of heterostructures with the appropriate morphological orientationsand band positions of the semiconductor and the polymer unit is challenging.Recently, Ghosh and coworkers developed an efficient light-harvesting het-erostructures based on poly(diphenylbutadiyne) (PDPB) nanofibers and ZnONPs via a solution phase synthetic route [95]. Figure 9.6a shows transmissionelectron microscopy (TEM) image of PDPB nanofibers of uniform diameterof ∼19 nm and a few micrometers long. The ZnO NPs (∼20 nm) were loadedonto the PDPB nanofibers as evident from a high-resolution TEM (HRTEM)image (Figure 9.6b). The ZnO NPs possess a good degree of cystallinity and theinterplanar distances of ∼0.26 nm corresponding to the spacing between two(002) planes of ZnO.

As ZnO NPs with an approximate size ∼20 nm, which does not have intrinsicdefect state emission, ZnO NPs (∼5 nm) were synthesized in situ on PDPBnanofibers to investigate the role of defect states in the photoinduced chargetransfer processes. The photoinduced electron dynamics from PDPB nanofibersto ZnO NPs has been studied by steady state and picosecond-resolved photo-luminescence studies. The cosensitization for multiple photon harvesting (withdifferent energies) at the heterojunction has been achieved via a systematicextension of conjugation from monomeric to polymeric diphenyl butadiynemoiety in the proximity of the ZnO NPs (Scheme 9.7a). On the other hand, theenergy transfer from the surface defects of ZnO NPs (∼5 nm) to PDPB nanofibersthrough Förster resonance energy transfer (FRET) confirms the close proximitywith molecular resolution (Scheme 9.7b). From FRET calculations, the distancebetween the donor ZnO NPs and acceptor PDPB nanofibers are determined tobe 3.4 and 3.1 nm for defect state 1 and defect state 2, respectively. The FRETdistance is consistent with the size of the ZnO NPs (radius= 2.5 nm). Thisobservation confirms the closer proximity between the PDPB and ZnO as well asUV light harvesting in the light-harvesting nanohybrids via energy transfer fromZnO to PDPB nanofibers. Scheme 9.7a illustrates the interfacial electron/energytransfer pathways for ZnO/PDPB nanoheterojunction. The demonstration of


efficient charge separation has been recognized with approximately fivefoldincrease in photocatalytic degradation of MO taken as a model pollutant inwater in comparison to polymer nanofibers counterpart under visible-lightirradiation. The activity of the polymer nanostructures is much lower ∼17% incomparison to PDPB-ZnO LHNH (80%) under similar illumination conditions.These observations suggest that due to efficient charge separation at the inter-face between PDPB and ZnO, the PDPB-ZnO LHNH is suitable for solar-lightharvesting. For PDPB-ZnO, the photocurrent response was 7.2 μA cm−2, about2.5 times higher than pure ZnO NPs which also reflects higher separation andtransfer efficiency of photo excited electrons from conjugated polymer to theconduction band of ZnO due to the formation of the heterostructure. The PDPB

Po

ten

tia

l vs v

acu

um

(e

V)

–2 e

e

e

h

e e

–

FRETO2

O2

H2O

O2

Poly(diphenyl butadiyne) (PDPB)

h h

h

h h

CB

VB

ZnO NP

ZnO NP

e e e

e e e e–3

–4

–5

–6

–7

–8*

*n

(a)

Visible

light

UV

light

Before

After

3.1 nm

PD

PB

nanofiber

(b)

e– transfer

Scheme 9.7 Schematic presentation of (a) the cosensitization of different PDPB oligomers toZnO NPs and molecular structure of PDPB polymer (b) the interfacial carrier dynamics at theheterojunction showing the photocatalytic degradation of MO in aqueous solution.(Reproduced with permission from Ref. [95]. Copyright 2015, Nature Publishers.)

9.6 Conclusion 245

is a p-type, organic semiconductor; ZnO is an inorganic, n-type semiconductorand forming a donor–acceptor junction (heterojunction). When the PDPB-ZnOLHNH is illuminated under visible light, electrons are excited from the highestoccupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital(LUMO) of PDPB, leaving holes behind in the HOMO of PDPB (Scheme 9.7).The excited state electrons are readily injected into the conduction band of ZnO.Thus enhanced photocatalytic activity and photoresponse has been achieveddue to formation of the nanoheterojunction. Very recently, Hou et al. preparedconjugated microporous poly(benzothiadiazole, BBT)/TiO2 heterojunction forboth photocatalytic H2 production and pollutant degradation under visible-light(𝜆> 420 nm) irradiation [96]. BBT/TiO2 heterojunction demonstrated signifi-cantly enhanced photocatalytic activities ∼18.0 and 20.4 times for H2 evolutionand ciprofloxacin degradation, respectively, as compared to polymer aloneunder visible light. Another example, PT/Bi2MoO6 nanocomposites showedhigh photocatalytic activity for photodegradation of rhodamine B than bareBi2MoO6 under visible light [97]. The superior performance of the PT/Bi2MoO6composite could be attributed to the high charge separation via the synergiceffect between polythiophene and Bi2MoO6, as well as the high charge transferrate due to the hole transporting ability of polythiophene. Various attempts havebeen carried out to improve the performance, using a series of other CP withsemiconductor nanocrystals [15].

9.6 Conclusion

In conclusion, semiconducting polymer nanostructures having narrowerbandgap illustrate superior photocatalytic activity compared to TiO2 for thedegradation of organic pollutants under UV and visible light. The photocatalyticactivity of CP emerges only at the nanoscale dimension and catalytic activitydepends also on the size and shape of the polymer nanostructures. As CPnanostructures possess large surface area, short path length for ion transport,and a huge absorption enhancement in the visible region make them excellentcandidates for photocatalytic applications. These photocatalysts are in generalvery stable with cycling. The detailed understanding of the mechanism as well ascocatalyst free polymeric catalysts could be useful for large-scale applications innear future. The concept of using polymer nanostructures as visible-light-activephotocatalysts could be extended to other CPs. The application of conjugatedpolymer nanostructures in the field of photocatalysis can be generalized to longerchain polymers. Recently, conjugated microporous polymers have been exploredfor photocatalytic hydrogen evolution under solar light. With the tunable opto-electronic properties of conjugated polymer including absorption in the visibleregion, extended conjugation for exciton/polaron percolation, bicontinuousdonor–acceptor architecture and facile charge transfer from the catalyst to thesubstrate or cocatalyst make them suitable for heterogeneous photocatalysis.The tunability of polymer structures associated with the synthetic flexibilitywhich allows the incorporation of functional groups, while providing a rigid and


lightweight backbone. However, research in conjugated polymeric materials forsolar-light-induced water splitting is still in its infancy, but the recent resultsreported in literature are very promising. Moreover, CPs have the uniquefeature of hybridization with semiconductors nanoparticles to generate novelhybrid nanostructures with appropriate bandgap alignment in combination withefficient charge separation at the interface leads to enhanced photocatalyticactivity and photoresponse. Hence, CP nanostructures offer the perspectivefor the development of a new generation visible-light-driven photocatalysts forenvironmental protection, hydrogen generation and also find applications inself-cleaning surfaces, and in the field of photovoltaic and solar cells.

References

1 Chu, S. and Majumdar, A. (2012) Opportunities and challenges for a sustain-able energy future. Nature, 488, 294–303.

2 Serpone, N. and Emeline, A.V. (2012) Semiconductor photocatalysis – past,present, and future outlook. J. Phys. Chem. Lett., 3, 673–677.




6 Ohtani, B. (2013) Titania photocatalysis beyond recombination: a criticalreview. Catalysts, 3, 942–953.

7 Asahi, R., Morikawa, T., Ohwaki, T., Aoki, K., and Taga, Y. (2001) Visible-lightphotocatalysis in nitrogen-doped titanium oxides. Science, 293, 269–271.

8 Baker, D.R. and Kamat, P.V. (2009) Photosensitization of TiO2 nanostructureswith CdS quantum dots: particulate versus tubular support architectures. Adv.Funct. Mater., 19, 805–811.

9 Wang, Y., Wang, Q., Zhan, X., Wang, F., Safdara, M., and He, J. (2013) Vis-ible light driven type II heterostructures and their enhanced photocatalysisproperties: a review. Nanoscale, 5, 8326–8339.

10 Luo, Q., Bao, L., Wang, D., Li, X., and An, J. (2012) Preparation and stronglyenhanced visible light photocatalytic activity of TiO2 nanoparticles mod-ified by conjugated derivatives of polyisoprene. J. Phys. Chem. C, 116,25806–25815.

11 Kubacka, A., Garcia, M.F., and Colon, G. (2012) Advanced nanoarchitecturesfor solar photocatalytic applications. Chem. Rev., 112, 1555–1614.

12 Hai, Z., Kolli, N.E., Uribe, D.B., Beaunier, P., Yacaman, M.J., Vigneron, J.,Etcheberry, A., Sorgues, S., Justin, C.C., Chen, J., and Remita, H. (2013)Modification of TiO2 by bimetallic Au–Cu nanoparticles for waste watertreatment. J. Mater. Chem. A, 1, 10829–10835.

13 Grabowska, E., Zaleska, A., Sorgues, S., Kunst, M., Etcheberry, A.,Justin, C.C., and Remita, H. (2013) Modification of titanium(IV) dioxide

References 247

with small silver nanoparticles: application in photocatalysis. J. Phys. Chem. C,117, 1955–1962.

14 Clavero, C. (2014) Plasmon-induced hot-electron generation atnanoparticle/metal-oxide interfaces for photovoltaic and photocatalyticdevices. Nat. Photonics, 8, 95–103.

15 Su, Y.W., Lin, W.H., Hsu, Y.J., and Wei, K.H. (2014) Conjugated poly-mer/nanocrystal nanocomposites for renewable energy applications inphotovoltaics and photocatalysis. Small, 10, 4427–4442.

16 Hunger, K. (2003) Industrial Dyes: Chemistry, Properties, Applications,Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

17 Christie, R.M. (2007) Environmental Aspects of Textile Dyeing, WoodheadPublishing, Great Abington, Cambridge.

18 Legrini, O., Oliveros, E., and Braun, A.M. (1993) Photochemical processes forwater treatment. Chem. Rev., 93, 671–698.

19 Shirakawa, H., Louis, E.J., MacDiarmid, A.G., Chiang, C.K., and Heeger,A.J. (1977) Synthesis of electrically conducting organic polymers: halogenderivatives of polyacetylene, (CH)x. J. Chem. Soc. Chem. Commun., 579–580.

20 Chiang, C.K., Fischer, C.R., Park, Y.W., Heeger, A.J., Shirakawa, H., Louis,E.J., Gau, S.C., and MacDiarmid, A.G. (1977) Electrical conductivity in dopedpolyacetylene. Phys. Rev. Lett., 39, 1098–1101.

21 Hadziioannou, G. and Malliaras, G.G. (eds) (2007) Semiconducting Polymers,2nd edn, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

22 Nalwa, H.S. (ed.) (1997) Handbook of Organic Conductive Molecules and Poly-mers, John Wiley & Sons, Inc, New York.

23 Burroughes, J.H., Bradley, D.D.C., Brown, A.R., Marks, R.N., Mackay, K.,Friend, R.H., Burns, P.L., and Holmes, A.B. (1990) Light-emitting diodesbased on conjugated polymers. Nature, 347, 539–541.

24 Grimsdale, A.C., Chan, K.L., Martin, R.E., Jokisz, P.G., and Holmes, A.B.(2009) Synthesis of light-emitting conjugated polymers for applications inelectroluminescent devices. Chem. Rev., 109, 897–1091.

25 Müllen, K. and Scherf, U. (2005) Organic Light-Emitting Devices, Wiley-VCHVerlag GmbH & Co. KGaA, Weinheim, Germany.

26 Cheng, Y.J., Yang, S.H., and Hsu, C.S. (2009) Synthesis of conjugated polymersfor organic solar cell applications. Chem. Rev., 109, 5868–5923.

27 Brabec, C., Dyakonov, V., and Scherf, U. (eds) (2008) Organic Photovoltaics,Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

28 Thomas, S.W. III,, Joly, G.D., and Swager, T.M. (2007) Chemical sensors basedon amplifying fluorescent conjugated polymers. Chem. Rev., 107, 1339–1386.

29 Su, W.P., Schrieffer, J.R., and Heeger, A.J. (1979) Solitons in polyacetylene.Phys. Rev. Lett., 42, 1698–1701.

30 Su, W.P., Schrieffer, J.R., and Heeger, A.J. (1980) Soliton excitations in poly-acetylene. Phys. Rev. B, 22, 2099–2111.

31 Sariciftci, E.N.S. (1997) Primary Photoexcitations in Conjugated Polymers:Molecular Exciton Versus Semiconductor Band Model, World Scientific,Singapore. ISBN: 978–9810228804

32 Bredas, J.L. and Street, G.B. (1985) Polarons, bipolarons, and solitons inconducting polymers. Acc. Chem. Res., 18, 309–315.


33 Shi, Y., Peng, L., Ding, Y., Zhao, Y., and Yu, G. (2015) Nanostructured conduc-tive polymers for advanced energy storage. Chem. Soc. Rev., 44, 6684–6696.

34 Pecher, J. and Mecking, S. (2010) Nanoparticles of conjugated polymers.Chem. Rev., 110, 6260–6279.

35 Ramanathan, K., Bangar, M.A., Yun, M.H., Chen, W., Mulchandani, A., andMyung, N.V. (2004) Individually addressable conducting polymer nanowiresarray. Nano Lett., 4, 1237–1239.

36 Ghosh, S., Ramos, L., Dazzi, A., Besseau, A.D., Remita, S., Beaunier, P., andRemita, H. (2015) Conducting polymer nanofibers with controlled diameterssynthesized in hexagonal mesophases. New J. Chem., 39, 8311–8320.

37 Jang, J., Ha, J., and Kim, S. (2007) Fabrication of polyaniline nanoparticlesusing microemulsion polymerization. Macromol. Res., 15, 154–159.

38 Łuzny, W. and Banka, E. (2000) Relations between the structure and electricconductivity of polyaniline protonated with camphorsulfonic acid. Macro-molecules, 33, 425–429.

39 Li, W., Zhang, Q., Zheng, G., Seh, Z.W., Yao, H., and Cui, Y. (2013) Under-standing the role of different conductive polymers in improving the nanos-tructured sulfur cathode performance. Nano Lett., 13, 5534–5540.

40 Berson, S., Bettignies, R.D., Bailly, S., and Guillerez, S. (2007)Poly(3-hexylthiophene) fibers for photovoltaic applications. Adv. Funct. Mater.,17, 1377–1384.

41 Bavel, S.S.V., Sourty, E., With, G.D., and Loos, J. (2009) Three-dimensionalnanoscale organization of bulk heterojunction polymer solar cells. Nano Lett.,9, 507–513.

42 Wang, F., Han, M.Y., Mya, K.Y., Wang, Y., and Lai, Y.H. (2005) Aggregation-driven growth of size-tunable organic nanoparticles using electronicallyaltered conjugated polymers. J. Am. Chem. Soc., 127, 10350–10355.

43 Xiao, D., Yang, W., Yao, J., Xi, L., Yang, X., and Shuai, Z. (2004)Size-dependent exciton chirality in (R)-(+)-1,1′-bi-2-naphthol dimethyl ethernanoparticles. J. Am. Chem. Soc., 126, 15439–15444.

44 Ghosh, S., Thandavarayan, M., and Basu, R.N. (2016) Nanostructured con-ducting polymers for energy applications: towards a sustainable platform.Nanoscale, 8, 6921–6947.

45 Lee, W. and Park, S.J. (2014) Porous anodic aluminum oxide: anodizationand templated synthesis of functional nanostructures. Chem. Rev., 114,7487–7556.

46 Fei, J., Cui, Y., Yan, X., Yang, Y., Wang, K., and Li, J. (2009) Controlledfabrication of polyaniline spherical and cubic shells with hierarchical nanos-tructures. ACS Nano, 3, 3714–3718.

47 Treuba, M., Montero, M.A.L., and Rieumont, J. (2004) Pyrrole nanoscaledelectro polymerization: effect of the proton. J. Electrochim. Acta, 49,4341–4349.

48 Cho, M.S., Choi, H.J., and Ahn, W.S. (2004) Enhanced electrorheologyof conducting polyaniline confined in MCM-41 channels. Langmuir, 20,202–207.

49 Jang, J. and Yoon, H. (2005) Formation mechanism of conducting polypyrrolenanotubes in reverse micelle systems. Langmuir, 21, 11484–11489.

References 249

50 Ahn, K.J., Lee, Y., Choi, H., Kim, M.S., Im, K., Noh, S., and Yoon, H. (2015)Surfactant-templated synthesis of polypyrrole nanocages as redox mediatorsfor efficient energy storage. Sci. Rep., 5, 14097–14107.

51 Xia, L., Wei, Z., and Wan, M. (2010) Conducting polymer nanostructures andtheir application in biosensors. J. Colloid Interface Sci., 341, 1–11.

52 Hulvat, J.F. and Stupp, S.I. (2003) Liquid-crystal templating of conductingpolymers. Angew. Chem. Int. Ed., 42, 778–781.

53 Hulvat, J.F. and Stupp, S.I. (2004) Anisotropic properties of conducting poly-mers prepared by liquid crystal templating. Adv. Mater., 16, 589–592.

54 Surendran, G., Tokumoto, S.M., Santos, E.P.D., Remita, H., Ramos, L.,Kooyman, P.J., Santilli, C.V., Bourgaux, C., Dieudonné, P., and Prouzet, E.(2005) Highly swollen liquid crystals as new reactors for the synthesis ofnanomaterials. Chem. Mater., 17, 1505–1514.

55 Ghosh, S., Remita, H., Ramos, L., Dazzi, A., Besseau, A.D., Beaunier, P.,Goubard, F., Aubert, P.H., Brisset, F., and Remita, S. (2014) PEDOT nanos-tructures synthesized in hexagonal mesophases. New J. Chem., 38, 1106–1115.

56 Wan, M. (2008) A template-free method towards conducting polymer nanos-tructures. Adv. Mater., 20, 2926–2932.

57 Huang, J., Virji, S., Weiller, B.H., and Kaner, R.B. (2003) Polyanilinenanofibers: facile synthesis and chemical sensors. J. Am. Chem. Soc., 125,314–315.

58 Dallasa, P. and Georgakilas, V. (2015) Interfacial polymerization of conduc-tive polymers: generation of polymeric nanostructures in a 2-D space. Adv.Colloid Interface Sci., 224, 46–61.

59 Yuan, C., Zhang, L., Hou, L., Lin, J., and Pang, G. (2014) Green interfacialsynthesis of two-dimensional poly(2,5-dimethoxyaniline) nanosheets as apromising electrode for high performance electrochemical capacitors. RSCAdv., 4, 24773–24776.

60 Ma, H., Luo, Y., Yang, S., Li, Y., Cao, F., and Gong, J. (2011) Synthesis ofaligned polyaniline belts by interfacial control approach. J. Phys. Chem. C,115, 12048–12053.

61 Huang, Z.M., Zhang, Y.Z., Kotaki, M., and Ramakrishna, S. (2003) A reviewon polymer nanofibers by electrospinning and their applications in nanocom-posites. Compos. Sci. Technol., 63, 2223–2253.

62 Miao, Y.E., Fan, W., Chen, D., and Liu, T.X. (2013) High-performancesupercapacitors based on hollow polyaniline nanofibers. ACS Appl. Mater.Interfaces, 5, 4423–4428.

63 Kannan, B., Williams, D.E., Laslau, C., and Sejdic, J.T. (2012) The elec-trochemical growth of highly conductive single PEDOT (conductingpolymer):BMIPF6 (ionic liquid) nanowires. J. Mater. Chem., 22, 18132–18135.

64 Nguyen, V.Q., Schaming, D., Martin, P., and Lacroix, J.C. (2015) Highlyresolved nanostructured PEDOT on large areas by nanosphere lithographyand electrodeposition. ACS Appl. Mater. Interfaces, 7, 21673–21681.

65 Ghosh, S., Natalie, K.A., Ramos, L., Remita, S., Dazzi, A., Besseau, A.D.,Goubard, F., Aubert, P.H., and Remita, H. (2015) Conducting polymernanostructures for photocatalysis under visible light. Nat. Mater., 14,505–511.


66 Mahaney, O.O.P., Murakami, N., Abe, R., and Ohtani, B. (2009) Correlationbetween photocatalytic activities and structural and physical properties oftitanium(IV) oxide powders. Chem. Lett., 38, 238–239.

67 Ohtani, B., Mahaney, O.O.P., Amano, F., Murakami, N., and Abe, R. (2010)What are titania photocatalysts – an exploratory correlation of photocatalyticactivity with structural and physical properties. J. Adv. Oxid. Technol., 13,247–261.

68 Ghosh, S., Kouame, N.A., Remita, S., Ramos, L., Goubard, F., Aubert, P.H.,Dazzi, A., Besseau, A.D., and Remita, H. (2015) Visible-light active conductingpolymer nanostructures with superior photocatalytic activity. Sci. Rep., 5,18002–18011.

69 Leclerc, M., Diaz, F.M., and Wegner, G. (1989) Structural analysis ofpoly(3-alkylthiophene)s. Makromol. Chem., 190, 3105–3116.

70 Floresyona, D., Goubard, F., Aubert, P.H., Lampr, I., Mathurin, J., Dazzi, A.,Ghosh, S., Beaunier, P., Brisset, F., Remita, S., Ramos, L., and Remita, H.(2017) Highly active poly(3-hexylthiophene) nanostructures for photocatalysisunder solar light. Appl. Catal., B, 209, 23–32.

71 Navarro, R.M., Sánchez, M.C.S., Galvan, M.C.A., Valle, F., and Fierro, J.L.G.(2009) Hydrogen production from renewable sources: biomass and photocat-alytic opportunities. Energy Environ. Sci., 2, 35–54.

72 Turner, J.A. (2004) Sustainable hydrogen production. Science, 305, 972–974.73 Yanagida, S., Kabumoto, A., Mizumoto, K., Pac, C., and Yoshino, K. (1985)

Poly(p-phenylene)-catalysed photoreduction of water to hydrogen. J. Chem.Soc. Chem. Commun., 474–475.

74 Matsuoka, S., Kohzuki, T., Nakamura, A., Pac, C., and Yanagida, S. (1991)Efficient visible-light-driven photocatalysis. Poly(pyridine-2,5-diyl)-catalysedhydrogen photoevolution and photoreduction of carbonyl compounds.J. Chem. Soc. Chem. Commun., 580–581.

75 Matsuoka, S., Kohzuki, T., Kuwana, Y., Nakamura, A., and Yanagida, S. (1992)Visible-light-induced photocatalysis of poly(pyridine-2,5-diyl). Photoreductionof water, carbonyl-compounds and alkenes with triethylamine. J. Chem. Soc.Perkin Trans., 2, 679–685.

76 Matsuoka, S., Fujii, H., Yamada, T., Pac, C., Ishida, A., Takamuku, S.,Kusaba, M., Nakashima, N., and Yanagida, S. (1991) Photocatalysis ofoligo(para-phenylenes) – photoreductive production of hydrogen and ethanolin aqueous triethylamine. J. Phys. Chem., 95, 5802–5808.

77 Sprick, R.S., Bonillo, B., Clowes, R., Guiglion, P., Brownbill, N.J., Slater, B.J.,Blanc, F., Zwijnenburg, M.A., Adams, D.J., and Cooper, A.I. (2016)Visible-light-driven hydrogen evolution using planarized conjugated polymerphotocatalysts. Angew. Chem. Int. Ed., 55, 1792–1796.

78 Wang, X.C., Maeda, K., Thomas, A., Takanabe, K., Xin, G., Domen, K., andAntonietti, M. (2009) A metal-free polymeric photocatalyst for hydrogenproduction from water under visible light. Nat. Mater., 8, 76–80.

79 Liu, J., Liu, Y., Liu, N., Han, Y., Zhang, X., Huang, H., Lifshitz, Y., Lee, S.-T.,Zhong, J., and Kang, Z. (2015) Water splitting. Metal-free efficient photocata-lyst for stable visible water splitting via a two-electron pathway. Science, 347,970–974.

References 251

80 Zhang, G., Lan, Z.-A., Lin, L., Lin, S., and Xinchen, W. (2016) Overall watersplitting by Pt/g-C3N4 photocatalysts without using sacrificial agents. Chem.Sci., 7, 3062–3066.

81 Bhunia, M.K., Yamauchi, K., and Takanabe, K. (2014) Harvesting solar lightwith crystalline carbon nitrides for efficient photocatalytic hydrogen evolu-tion. Angew. Chem. Int. Ed., 53, 11001–11005.

82 Zhang, Z., Long, J., Yang, L., Chen, W., Dai, W., Fu, X., and Wang, X. (2011)Organic semiconductor for artificial photosynthesis: water splitting intohydrogen by a bioinspired C3N3S3 polymer under visible light irradiation.Chem. Sci., 2, 1826–1830.

83 Sprick, R.S., Jiang, J.X., Bonillo, B., Ren, S.J., Ratvijitvech, T., Guiglion, P.,Zwijnenburg, M.A., Adams, D.J., and Cooper, A.I. (2015) Tunable organicphotocatalysts for visible-light-driven hydrogen evolution. J. Am. Chem. Soc.,137, 3265–3270.

84 Vyas, V.S., Lau, V.W., and Lotsch, B.V. (2016) Soft photocatalysis: organicpolymers for solar fuel production. Chem. Mater., 28, 5191–5204.

85 Sprick, R.S., Bonillo, B., Sachs, M., Clowes, R., Durrant, J.R., Adams, D.J., andCooper, A.I. (2016) Extended conjugated microporous polymers for photocat-alytic hydrogen evolution from water. Chem. Commun., 52, 10008–10011.

86 Li, L. and Cai, Z. (2016) Structure control and photocatalytic performanceof porous conjugated polymers based on perylenediimide. Polym. Chem., 7,4937–4943.

87 Sprick, R.S., Bonillo, B., Clowes, R., Guiglion, P., Brownbill, N.J., Slater, B.J.,Blanc, F., Zwijnenburg, M.A., Adams, D.J., and Cooper, A.I. (2016)Visible-light-driven hydrogen evolution using planarized conjugated polymerphotocatalysts. Angew. Chem. Int. Ed., 128, 1824–1828.

88 Li, L., Lo, W.-Y., Cai, Z., Zhang, N., and Yu, L. (2016) Donor–acceptor porousconjugated polymers for photocatalytic hydrogen production: the importanceof acceptor comonomer. Macromolecules, 49, 6903–6909.

89 Li, L., Hadt, R.G., Yao, S., Lo, W.-Y., Cai, Z., Wu, Q., Pandit, B., Chen, L.X.,and Yu, L. (2016) Photocatalysts based on cobalt-chelating conjugated poly-mers for hydrogen evolution from water. Chem. Mater., 28, 5394–5399.

90 Lang, X., Chen, X., and Zhao, J. (2014) Heterogeneous visible light photo-catalysis for selective organic transformations. Chem. Soc. Rev., 43, 473–486.

91 Reiss, P., Couderc, E., Girolamo, J., and Pron, A. (2011) Conjugated poly-mers/semiconductor nanocrystals hybrid materials – preparation, electricaltransport properties and applications. Nanoscale, 3, 446–489.

92 Liao, G., Chen, S., Quan, X., Chen, H., and Zhang, Y. (2010) Photonic crystalcoupled TiO2/polymer hybrid for efficient photocatalysis under visible lightirradiation. Environ. Sci. Technol., 44, 3481–3485.

93 Pei, Z., Ding, L., Lu, M., Fan, Z., Weng, S., Hu, J., and Liu, P. (2014) Synergis-tic effect in polyaniline-hybrid defective ZnO with enhanced photocatalyticactivity and stability. J. Phys. Chem. C, 118, 9570–9577.

94 Liu, R. (2014) Hybrid organic/inorganic nanocomposites for photovoltaic cells.Materials, 7, 2747–2771.

95 Sardar, S., Kar, P., Remita, H., Liu, B., Lemmens, P., Pal, S.K., andGhosh, S. (2015) Enhanced charge separation and fret at heterojunctions


between semiconductor nanoparticles and conducting polymer nanofibers forefficient solar light harvesting. Sci. Rep., 5, 17313–17327.

96 Hou, H.-J., Zhang, X.-H., Huang, D.-K., Ding, X., Wang, S.-Y., Yang, X.-L.,Li, S.-Q., Xiang, Y.-G., and Chen, H. (2017) Conjugated microporouspoly(benzothiadiazole)/TiO2 heterojunction for visible-light-driven H2 produc-tion and pollutant removal. Appl. Catal. B Environ., 203, 563–571.

97 Zhijie, Z., Zheng, T., Xu, J., and Zeng, H. (2016) Polythiophene/Bi2MoO6: anovel conjugated polymer/nanocrystal hybrid composite for photocatalysis.J. Mater. Sci., 51, 3846–3853.

253

Part III

Visible Light Active Photocatalysis for Solar Energy Conversionand Environmental Protection

255

10

Sensitization of TiO2 by Dyes: A Way to Extend the Rangeof Photocatalytic Activity of TiO2 to the Visible RegionMarta I. Litter1, 2, Enrique San Román3, the late María A. Grela4, 5,Jorge M. Meichtry1, and Hernán B. Rodríguez6

1Gerencia Química, Comisión Nacional de Energía Atómica, Consejo Nacional de lnvestigaciones Cientificas yTécnicas, Buenos Aires, Argentina2Instituto de Investigación e Ingeniería Ambiental, Universidad Nacional de General San Martín, Prov. deBuenos Aires, Argentina3Universidad de Buenos Aires. Consejo Nacional de lnvestigaciones Cientificas y Técnicas. Instituto de QuímicaFísica de los Materiales, Medio Ambiente y Energía (INQUIMAE). Facultad de Ciencias Exactas y Naturales,Buenos Aires, Argentina4Departamento de Química, Facultad de Ciencias Exactas y Naturales (FCEyN), Universidad Nacional de Mardel Plata (UNMdP), Mar del Plata, Argentina5CONICET–Universidad Nacional de Mar del Plata, Instituto de Física de Mar del Plata (IFIMAR), Mar del Plata,Argentina6INIFTA (UNLP-CONICET), Facultad de Ciencias Exactas, Universidad Nacional de La Plata, Diag. 113 y Calle 64,La Plata, Argentina

10.1 Introduction

TiO2-heterogeneous photocatalysis has been widely investigated for oxidationand reduction of pollutants in water and air. However, a serious limitation forTiO2 technological applications is the requirement of UV light to promote chem-ical processes due to its wide bandgap (about 3 eV). Thus, to extend the usefulrange of irradiation of TiO2, several strategies have been studied [1–12]:

1) Coupling of TiO2 with a smaller bandgap semiconductor (SC): after irradia-tion with visible light, charge carriers (electrons and holes) are generated inthe smaller bandgap SC, active in the visible range; electrons are then injectedinto the TiO2 conduction band, leading to further reactions [1]. Alternatively,model photosynthetic systems involving a composite of two SCs with a p–nheterojunction and a staggered (Type II) band alignment can be designed tomimic the Z-scheme of photosynthesis [13, 14]. Especially, coupling with CdShas been attempted, but it is not recommended due to CdS toxicity and car-cinogenicity.

2) Coupling of TiO2 with graphene and other carbon compounds: these materialsshow an enhanced activity toward heterogeneous photocatalysis under visiblelight [15].


256 10 A Way to Extend the Range of Photocatalytic Activity of TiO2 to the Visible Region

3) Deposition, doping, or modification of TiO2 with metals, where generallythe metal acts as a sink of electrons, due to dopant energy levels within thebandgap of TiO2, with a shift of the band edge absorption threshold to thered [1].

4) Deposition of noble-metal nanoparticles over TiO2, which can induce photo-catalytic processes under visible light by the plasmonic effect [16, 17].

5) Modification of TiO2 with nonmetals such as N (the most used element), C, F,P, S, and B, with substitution of oxygen sites in the TiO2 lattice [4, 10].

6) Ion implantation by bombarding with high-energy ions, usually by means ofan ion accelerator [5].

7) Reaction with dyes acting as sensitizers (S) after light excitation, and electroninjection into the conduction band (CB) of the SC [11, 18, 19]. Hole injec-tion to metal oxide SCs would require dyes with unreasonably high standardpotentials, but it is possible in the case of SCs with less positive valence bandpotentials like p-type gallium phosphide (p-GaP) [20]; however, this case willbe not included in the discussion.

In this chapter, the literature associated with the sensitization of TiO2 by mod-ification with dyes (point 7) will be reviewed, including reactions leading to thetransformation of dyes themselves, of colorless pollutants, and hydrogen pro-duction, with emphasis on the involved mechanisms. Additionally, the field ofdye-sensitized solar cells (DSSC) will be briefly addressed.

10.2 Mechanisms Involved in the Use of Dye-ModifiedTiO2 Materials for Transformation of Pollutants andHydrogen Production under Visible Irradiation

Mechanistically, the role of the SC in systems for chemical transformation is dif-ferent from that proposed for photocatalytic processes under UV irradiation.Taking into account that the time period between the absorption of two suc-cessive photons under typical conditions (i.e., solar and low intensity laboratoryillumination, either UV or visible) is much higher than those corresponding toback electron transfer (ET) and recombination pathways, multielectron transferprocess on a single SC particle would be highly improbable [21–23]. Consideringthe earlier reports on the excited state redox properties of several dyes adsorbedon the TiO2 surface [24], the following mechanism has been suggested:

TiO2-(dye) + h𝜈 → TiO2-(dye∗) (10.1)TiO2-(dye∗) → TiO2-(dye) + h𝜈′ (10.1′)TiO2-(dye∗) → TiO2-(dye∙+) + eCB

− (10.2)TiO2-(dye∙+) + eCB

− → TiO2-(dye) (10.2′)TiO2-(dye∙+) + P → TiO2-(dye) + P∙+ (10.3)eCB

− + O2 → O2∙− (10.4)

O2∙− + H+ ⇆ HO2

∙ (10.5)HO2

∙ + HO2∙ → H2O2 + O2 (10.6)

H2O2 + O2∙− → HO∙ + OH− + O2 (10.7)

10.2 Dye-Modified TiO2 Materials for Transformation of Pollutants and Hydrogen Production 257

H2O2 + TiO2(eCB−) → HO∙ + OH− (10.8)

O2∙−, HO2

∙, H2O2, HO∙ + P → oxidized products of P (10.9)

eCB− + Q → Q∙− (10.10)

When the dye adsorbed or chemically linked to a SC, TiO2 in this case, isirradiated under visible light, it is promoted to an excited state (dye*, Eq. (10.1)).Generally, the singlet state of the dye participates in electron injection process,although triplet states can also be formed after excitation, but with a substantiallyslower electron injection [25–27]. Eventually, the back reaction (Eq. (10.1′)) canreturn the dye to its ground state, either radiatively or nonradiatively. Althoughsensitization can take place also by energy transfer reactions (e.g., production ofsinglet oxygen by energy transfer from the excited dye to molecular oxygen), onlyET reactions will be considered in the present work [11, 18, 28, 29]. Injectionof electrons from the excited state leads to the formation of a radical cation(dye∙+, Eq. (10.2), where eCB

− are electrons in the TiO2 conduction band); thetrapping can occur also in shallow traps within the bandgap. This reaction isthermodynamically possible [30], as the excited state of the dye has generally amore negative reduction potential than the corresponding to the ground state,E(dye*/dye∙+) ranging around from−1.0 to−1.6 V versus SHE [12], which is lowerthan that of the TiO2 CB at pH 0 (ECB ≈−0.5 V vs SHE [1]). Electron injection(Eq. (10.2)) is generally a very fast process (in the subpicosecond–femtosecondtime scale), much faster than the intrinsic deactivation of the excited state ofthe sensitizer (Eq. (10.1′)), though, in some cases, multiexponential decays withresidual components exceeding the nanosecond range are observed [31]. Onthe other hand, charge recombination, in which the electron is back-transferredfrom the CB to the dye radical (Eq. (10.2′)), occurs at a rate several orders ofmagnitude slower than the forward charge injection (Eq. (10.2)) [1]. Note thatreaction (10.2′) is not exactly the reverse of reaction (10.2) because the dyereturns to its ground state. Such a sluggish unwanted recombination is advanta-geous for the charge separation as this offers more chance for CB transport of theinjected electrons to surface reaction sites and for the oxidized dyes to react [32].Then, dye∙+ can oxidize a donor species P, the dye returning to its reduced state(Eq. (10.3)). During the oxidative processes, holes on the valence band (VB) ofTiO2, with stronger oxidizing power (E0 =+2.7 V vs SHE for anatase) than thatof dye∙+, are not involved. The described process has been typically applied alsoto photoelectrochemical DSSCs [1, 6], aspect that will be treated in Section 10.3.After trapping of the injected electrons in surface sites, these electrons can bescavenged by surface-adsorbed dissolved molecular oxygen to produce super-oxide/hydroperoxide (O2

∙−/HO2∙) radicals (Eqs. (10.4) and (10.5)). Following

reactions (10.6)–(10.8) contribute to the formation of hydrogen peroxide andeven hydroxyl radicals (HO∙), all of them constituting the reactive oxygen species(ROS), which may also contribute to the degradation of P (Eq. (10.9)). Reductionof chemical organic and inorganic electron acceptor species (Q, Eq. (10.10)) isalso possible by eCB

− produced in reaction (10.2), normally in the absence ofmolecular oxygen. Electron scavenging is generally several orders of magnitudefaster than back ET to dye∙+ (Eq. (10.2′)) [1]. If P is absent or reaction (10.3) isslow, the process will lead to the self-sensitized dye degradation; this aspect willbe discussed in Section 10.4. In other cases, the dye-modified TiO2 can be used


H+

CB

O2•−

VB

dye•+

O2

TiO2Dye

dye*

dye

1.8–2.7 eV

(690–450 nm)

HO2•

H2O2

OH•

dye

CB

VB

Q

H+

CB

O2•−

VB

HO2•

H2O2

OH•

H2O

(b)(a) TiO2

3.2 eV

(390 nm)

CB

VB

e−

hν(UV light)

hν(Vis light)

dye*

P

P•+

eCB−

P

P•+

O2Q•− Q•−Q

eCB−

hνhν

hVB+

–H+

Figure 10.1 Comparison of the photocatalytic mechanisms with TiO2 particles under (a) UVirradiation and (b) by the self-photosensitized pathway under visible light irradiation.

for visible-light-assisted degradation of colorless pollutants, either by oxidativeor reductive processes, as will be discussed in Section 10.5.

Figure 10.1 shows the differences between the photocatalytic mechanismsunder UV and visible-light irradiation, adapted from Ref. [33].

To act as sensitizers, molecules have to fulfill the following conditions [12]:1) The absorption spectrum should cover the wide visible region.2) The dye should be photostable (unless the self-sensitized degradation is

desired).3) Dye aggregation on the TiO2 surface should be avoided (see Section 10.3).4) Dyes should have suitable anchoring groups (—SO3H, —COOH, —PO3H2,

etc.) to facilitate the binding onto the TiO2 surface.5) The lowest unoccupied molecular orbital (LUMO) of the dye should be higher

in energy than the edge of the TiO2 CB to allow ET between the excited dyeand TiO2.

6) For an efficient sensitization, injection of electrons into CB (Eq. (10.2))must be much faster than the decay of the excited state to the ground state(Eq. (10.1′)) [34].

The ET from the dye to the SC can take place either to the CB (Eq. (10.2)) ordirectly from the dye to shallow trap sites; in aerated systems, the electron canbe easily transferred to adsorbed O2. In contrast, in homogeneous solution, thephotoinduced ET from the dye to O2 is not as efficient. This is the reason whymost of the dyes are rather stable in homogeneous solution under visible-lightirradiation in air [32, 35]. The CB of the SC provides sufficient unoccupiedelectron-acceptor states with a wide and continuous energy distribution(Figure 10.2) [36], allowing rapid and efficient electron injection, thus avoiding

10.2 Dye-Modified TiO2 Materials for Transformation of Pollutants and Hydrogen Production 259

Figure 10.2 Electron injection from theexcited state of a dye (dye*) into the CB ofthe SC.

E (V)

Dye*

Dye

CB

VB

e−

TiO2

hν

useless (radiative or nonradioactive) loss of the dye excited states. The electroninjection is generally ultrafast and can be explained by either adiabatic (strongchromophore coupling to one or a few TiO2 states) or nonadiabatic transfermechanisms (weak dye–SC coupling) (see Section 10.3) [37].

The efficiency for the photooxidation of the substrate P can be lowered some-what by electron capture by the radical cation [1, 6]. This will cause a decreaseon the photoefficiency if the reaction between eCB

− and O2 or Q (Eqs. (10.4) and(10.10)) is slower than the back ET (Eq. (10.2′)), as observed in the degradationof perchlorinated compounds over TiO2 modified with a Ru complex [38]. In theabsence of O2 or other reducible substrate, this process would dominate, inhibit-ing Eq. (10.3), that is, the degradation of P.

The most common used sensitizers have been organic dyes such as thiazinesand xanthenes (rose bengal (RB), erythrosin B, thionine), metal free or metalporphyrins and phthalocyanines, substituted and unsubstituted bipyridines, andtransition-metal complexes such as polypyridine complexes (e.g., [Ru(bpy)2(4,4-(PO3H2)2bpy)]2+) [11, 12, 39–45]. Phthalocyanines (Pc) are suitable sensitizersbecause they absorb in the red (around 670 nm) and show high chemical andthermal stability [46].

The modification of TiO2 with dyes has been carried out by different techniquesas, for example, covalent bonding, ion-pair association, physisorption, entrap-ment in cavities or pores, and hydrophobic interactions leading to self-assemblyof monolayers [1, 6, 47, 48].

The identification of the ET product in the charge injection process (Eq. (10.1))was first attempted by Moser and Grätzel in the aqueous eosin (EO)/colloidalTiO2 system [49] by following the spectrum obtained immediately after laserexcitation (532 nm) of a deaerated EO solution in water containing colloidal TiO2.The transient spectrum showed a prominent peak at 475 nm, which was unam-biguously attributed to the semioxidized eosin, EO∙+. Dye radical cations havebeen also in situ monitored by electron spin resonance and time-resolved laserflash photolysis techniques [1 and Refs. therein, 50–53].


10.3 Use of Dye-Modified TiO2 Materials for EnergyConversion in Dye-Sensitized Solar Cells

Energy conversion in DSSC is reviewed here only with the aim of gatheringevidence that may help in understanding the role of dye–SC interactions for thedesign of efficient dye-sensitized photocatalysts and for inferring the involvedmechanisms. The following examples refer to n-type SCs, mainly TiO2, andelectron injecting dyes, but the involved concepts may be extrapolated to othersystems. Only a few representative examples are selected from the huge amountof publications in this area; various reports on this topic are found in theliterature [1, 6, 54].

The first steps in the mechanism involved in DSSC processes are the same asdescribed in Section 10.2 (Eqs. (10.1)–(10.2′)). The efficiency of DSSCs depends,among other design properties like charge transport in the photoanode and coun-terelectrode and redox electrolyte, on the light harvesting yield of the dye andthe efficiency of electron injection into the CB, and it is limited by losses due torecombination of charge carriers [55]. Furthermore, the overpotentials for elec-tron injection and regeneration of the oxidized dye are in general quite large [56].

Engineering the dye properties to achieve high light harvesting yields is centralto increase the efficiency [57]. Light absorption depends on the molar absorptioncoefficient and spectral bandwidths of the dye, the dye concentration (packingdensity) at the SC surface and the available surface area. Extended conjugation inorganic dyes increases the absorption coefficients by several times, as comparedwith the rather low absorption of Ru dyes. However, extended π systems are proneto interact among themselves, particularly at high dye-surface coverage, yield-ing dimers and higher aggregates, which decay rapidly to the ground state [58],and unstable radical cations. On the other side, Ru complexes have larger band-width, show several stable oxidation states and are resistant to ambient condi-tions. Various strategies have been developed to avoid aggregation and to increasethe absorption bandwidth of organic dyes. Bulky substituents and 3D structures[59–61], and the use of coadsorbates [62, 63] proved to reduce dye aggregation onthe SC surface, improving the sensitization efficiency. On the other hand, cosen-sitization (panchromatic engineering), that is, the use of multiple sensitizing dyeswith complementary absorption spectra, has been proposed as an alternative toincrease the absorption spectral bandwidth of single dyes [57]. Cosensitizationstrategies involve from the homogeneous mixture deposition of dyes on the SCsurface to tandem incorporation of sensitizers via layer-by-layer deposition; thislast option includes, as an alternative, the incorporation of dyes in different SClayers to avoid detrimental cross-talking between the different sensitizers, thatis, the occurrence of interdye reactions leading to unwanted processes. Further-more, in the absence of electronic interactions between the different sensitizingdyes, coadsorption may result also in the inhibition of molecular aggregation [64].

The electronic coupling between the dye and the SC governs the electron injec-tion dynamics. Depending on its magnitude, two models are currently applied:the Sakata–Hashimoto–Hiramoto model of photoinduced electron transfer(PET) [65] and the Creutz–Brunschwig–Sutin model of optical electron transfer

10.3 Use of Dye-Modified TiO2 Materials for Energy Conversion in Dye-Sensitized Solar Cells 261

(OET) [66]. In the first case, the coupling is weak and, on irradiation, an electronis promoted to the LUMO of the dye (a π* state for aromatic photosensitizers).If this state is higher in energy than the bottom of the CB, ET takes placeto the continuum levels of the CB in resonance with the LUMO energy (seeFigure 10.2). In this case, ET rates increase monotonically with the energy gap(no inverted region is observed). In the weak coupling limit, the excited dyemay decay by fluorescence emission or nonradiative processes (Eq. (10.1′)),competing with ET. In the second case, the interaction is so strong that interfacialcharge transfer absorption takes place. This absorption is spectrally observablewhen transition energies occur at wavelengths lower than the bandgap. TheLUMO is in this case localized in the CB of the SC. Therefore, excitation andET take place simultaneously [67]. Both models represent extreme cases andintermediate situations are possible. While in the weak-coupling limit ET occursnonadiabatically from a localized chromophore state into delocalized SC states,in the strong-coupling limit, ET proceeds adiabatically into localized surfacestates during the optical excitation. [68]. Dye-SC electronic coupling can beenhanced by the appropriate selection of anchoring groups in the dye structure[69], while dye redox potentials are relevant to the feasibility of electron injection,with threshold overpotential values in the order of 200–300 mV for an efficientelectron injection from the excited dyes to the SC [70].

Different photosensitizers, from metal complexes to porphyrins, phthalocya-nines and metal-free organic dyes have been tested for DSSCs [57, 71]. Thewidely used and versatile Ru dyes show drawbacks related to the high cost of themetal and, generally, low absorption coefficients. Organic dyes, on the contrary,have received great attention due to their low cost and high absorption coeffi-cients, together with their tunable properties. In particular, various strategieshave been developed to favor charge carrier separation after dye excitation andcharge injection into the CB of the SC. Push–pull chromophores, that is, thosehaving donor–π bridge–acceptor (D–π–A) structures, induce intramolecularcharge transfer from D to A through the π bridge on excitation, favoring chargeseparation, hole localization in D and electron injection from the anchored A tothe SC, reducing the chance of charge recombination [72–74]. The expansionof the π-conjugated spacer can induce highest occupied molecular orbital(HOMO) and LUMO shifts, thus resulting in tunable photophysical properties,while extending the absorption spectral bandwidth [75]. Moreover, it wasdemonstrated that the incorporation of an additional electron-withdrawingacceptor between D and the π bridge, in the so-called D–A–π–A motif, notonly favors the modulation of energy levels, absorption bandwidth and DSSCperformance, but also enhance dye photostability [76]. Different donors, such asporphyrins, indolines, arylamines, phenothiazines, coumarines, among others,and anchoring and auxiliary acceptors like cyanoacrylic acid, benzothiadiazole,benzothriazole, quinoxaline, phthalimide, and diketo-pirrolopirrole, have beenused to design organic sensitizers [76, 77]. Regarding the π-spacer, oligoenesare generally used. However, simple oligoene bridges are chemically and pho-tochemically unstable because they are prone to isomerize geometrically andare photooxidizable in the presence of O2. For this reason, conjugated aromaticrings, especially thiophene derivatives [78], are currently considered [79]. The


development of new sensitizers with tunable properties is often a trial-and-errorapproach, but advances in the comprehension of the relationship between dyestructure and photophysical and photochemical properties, with the help oftheoretical calculations [80], can allow the rational design of sensitizers.

Most DSSCs use synthetic dyes. However, limitations related to costs, photo-stability and toxicity of the materials have opened the use of natural sensitiz-ers as potential alternatives. Plant pigments, such as chlorophylls, carotenoids,flavonoids, and anthocyanines, have been tested as TiO2 sensitizers for DSSCs[81]. Even though these solar cells have generally low conversion efficiencies,research in natural sensitizers constitutes a promising alternative for the devel-opment of low-cost and environmentally friendly DSSCs and dye-sensitized pho-tocatalysts.

On the other hand, narrow-bandgap SC quantum dots (QDs) have been pro-posed as photosensitizers instead of organic dyes because of their size-dependenttunable bandgap and energy levels, high absorption coefficients and photostabil-ity [82, 83].

Advances in the development of DSSCs, including the above-mentioned andalternative strategies, are inspiring subjects for the design and improvement ofdye-sensitized photocatalysts active under visible light irradiation. Some of thesestrategies, such as the use of sensitizers with D–π–A and D–A–π–A structures[74, 84] and cosensitization [85], mainly for photocatalytic hydrogen evolution,have been previously tested for improved photocatalytic applications.

10.4 Self-Sensitized Degradation of Dye Pollutants

Nonregenerative dye sensitization of TiO2, that is, when the organic dye actsboth as a sensitizer of the SC and a substrate to be degraded, has been proposedas a promising alternative for the visible-light driven photocatalytic degradationof organic dye pollutants in wastewater effluents. This aspect has been recentlyreviewed (e.g., [12, 30, 32, 86–89]), and Table 8.2 in Ref. [12] includes the earliestreferences.

In 1977, Watanabe et al. first reported that the ET from adsorbed rhodamine B(RhB) in its singlet excited state to the CB of CdS powders led to an efficient pho-tochemical N-deethylation of the dye [90]. In the 1990s, Kamat and coworkersshowed that visible-light irradiation could induce the bleaching of preadsorbedazo dyes on dry TiO2 powders in the presence of O2 [18, 27, 29]. Since then, a lotof efforts have been invested in getting insights into the involved self-sensitizedphotodegradation mechanisms and the improvement of the process [6, 30, 53,89, 91–93], and various dyes have been successfully degraded in oxygenated TiO2aqueous suspensions under visible light irradiation (e.g., [33, 94–101]).

In 1998, Wu et al. [33] proposed a mechanism for the self-sensitized oxidativetransformation of RhB through reactions (10.1) and (10.2), generation of ROS inthe presence of oxygen (Eqs. (10.4)–(10.8)), followed by reaction (10.11):

dye or dye∙+ + ROS∕O2 → peroxy∕hydroxylated intermediates→ degraded∕mineralized prods. (10.11)

10.4 Self-Sensitized Degradation of Dye Pollutants 263

The photodegradation of the dye induced by visible light would be useful, forexample, in the treatment of textile wastewaters containing the colored pollutant,a very much wanted process in view of the huge amount of dyes produced per yearand their extensive use worldwide, considering that an important percentage ofthese products enter without treatment into the environment. The process mightbe more advantageous than UV-TiO2 conventional photocatalysis because of thepossibility of use of solar light, with a significant economic advantage.

Although the above-mentioned initial steps described by Eqs. (10.1) and (10.2)are not questioned, the subsequent details of the degradation pathway remainstill rather unclear, especially the role of O2 and ROS and the fate of the dyeradical cation during the degradation process. For example, two competitive dyedegradation mechanisms were recognized in the sulforhodamine-B (SRB)/TiO2system: sequential N-dealkylation and destruction of the conjugated structureof the chromophore, the former leading generally to absorption spectral changeswhile the later leading particularly to discoloration [12, 30, 97]. The degradationof squarylium cyanine, in contrast, goes through the chromophoric cleavage[30, 98].

The characterization of the degradation process involves the evaluation of thediscoloration kinetics, the degree of transformation of the dye and the identifi-cation of intermediates. Discoloration does not guarantee the complete miner-alization of the organic dye, but only reflects the destruction of the conjugationstructure of the chromophore. While the complete mineralization of the dye ispreferable, the reaction produces generally a set of intermediates, its identifica-tion being necessary in order to evaluate the potential production of toxic orrecalcitrant species [12, 30]. An effective adsorption of the dye on the SC surfaceenhances the photoinjection of electrons after excitation with visible light. Whilerecombination of photoinduced charge carriers has to be avoided, enhanced ROSgeneration is a key parameter to produce an efficient photodegradation of thedye. For that reason, strategies to improve self-sensitized photodegradation onTiO2 involve: (i) the preadsorption of the dyes on the SC surface; (ii) the increasein charge transfer and charge separation processes, avoiding recombination; and(iii) the enhancement of ROS generation.

Dye adsorption on the TiO2 surface may involve nonspecific electrostaticinteractions between charged dyes and the surface, or more specific interactionsthrough anchoring chemical groups present in the dyes via simple or multiden-tate complex formation. In the first case, the charge surface density of the SC is akey parameter. The point of zero charge of Degussa P-25 TiO2 is 6.8, the SC beingpositively charged in acid medium and negatively charged in alkaline medium[102]. For that reason, the pH dependence of the 𝜁-potential of TiO2 affects dyephotodegradation. For example, self-sensitized degradation of eosin (a nega-tively charged dye) in the presence of TiO2 is enhanced in acid medium due toadsorption via electrostatic interaction compared with neutral or alkaline media[91]. On the contrary, a slow photodegradation rate was observed for cationicRhB in acid medium, while incorporation of the anionic surfactant dodecyl-benzenesulfonate, which strongly adsorbs on the TiO2 surface in acid mediuminverting the sign of the 𝜁-potential, accelerates RhB degradation [93]. TiO2surface modifications, such as the use of coadsorbates and thermal or chemical


treatments, may have critical effects on dye adsorption and photodegradationefficiencies [103, 104]. Anchoring groups present in the dye structure, such ascarboxylates, sulfonates, phosphonates, and hydroxyl groups, may facilitate itsbinding to the TiO2 surface via complex formation, not only favoring dye adsorp-tion but also enhancing the electronic coupling between the dye and the SCand the photoinduced electron injection into the CB [105]. The photoinjectiondynamics is also governed by the adsorption state of the dye [12, 30].

Other factors such as the presence of inorganic ions, dissolved organic matter,or humic substances, generally present in wastewaters, may compete for theactive sites in the TiO2 surface and interfere in the photocatalytic processes;in a mixture of pollutants, one of them can interfere the degradation of theother [106].

After electron injection, the charge separation efficiency between the dyeradical cation and the injected electron in the TiO2 CB is one of the main factorsdetermining the rate of dye photodegradation. Two competitive processes areinvolved: depletion of charge carriers via charge transfer processes or chemicalreactions, and recombination of charge carriers. Favoring separation of carriersand increasing rates of depletion of electrons are common strategies to enhanceoxidative dye degradation. Any surface modification of the SC that facilitatethe electron scavenging by O2, such as doping with Pt(IV) species [107] or withnoble metals like Pt [108], promote the photooxidation and mineralization ofdye pollutants. For example, the incorporation of small amounts of hexachloro-platinate(IV) chemisorbed on the TiO2 surface enhanced self-photodegradationof ethyl orange under visible-light irradiation. The phenomenon was ascribedto an efficient charge separation via ET of the eCB

− to the chemisorbedPt(IV) species, which in turns assists the ET to O2, favoring ROS generation(Eqs. (10.4)–(10.8)) [107]. In general, the incorporation of metallic dopants toSCs is a recognized strategy to improve the photocatalytic activity via favoringcharge separation and electron scavenging by O2 [89]. Platinized TiO2 has shownenhanced self-degradation rates of sulforhodamine B, the Pt dopant acting asan electron sink, promoting electron scavenging by O2 and initiating ROSgeneration processes [108]. Other metallic dopants, such as Ag [109], Au [110],Pd [111], among others, associated with wide bandgap SCs, were proposed asphotocatalysts for self-sensitized degradation of dyes. More recently, core shellstructures (Me@SC) between a noble metal, Me=Au, Ag, and a wide bandgapSC have also shown an enhancement in the dye-sensitized degradation process incomparison with pure SCs. It has been demonstrated that the electrons injectedinto the SC layer are quickly transferred to the metal core, lowering the rate of theback ET (Eq. (10.2′)) [112, 113]. In another example, TiO2 nanoparticles dopedwith Zn enhanced the photocatalytic degradation of RhB under 𝜆> 400 nm.The improved activity by Zn doping was attributed to the appropriate energeticposition between ZnO and the excited state of dye, which enhances the electroninjection into the TiO2 CB and promotes ROS formation [114]. On the otherhand, the presence of transition metal ions, generally encountered in wastewatereffluents, might have detrimental effects on the self-sensitized degradation ofdye pollutants. For example, the presence of Cu2+ and Fe3+, which have suitableredox potentials to compete with O2, alters the ET processes by reducing the

10.5 Use of Dye-Modified TiO2 for Visible-Light-Assisted Degradation of Colorless Pollutants 265

electron scavenging by O2, and depresses the self-sensitized degradation of dyes[96]. These facts also evidence the main role of O2 reduction, which starts ROSgeneration processes leading to dye degradation through Eq. (10.11). Actually,it has been demonstrated that O2

∙−/HO2∙ are the main species responsible

for the photooxidation and mineralization of the dye pollutants [107, 115] inmany cases.

Regarding the dye, diverse reactivities have been found among dye familieshaving different chemical structures. In the case of azo dyes, the first step is thecleavage of the azo double bond [18]. Triphenylmethane dyes are found to reacteasier than anthraquinone dyes, and food dyes are easier to bleach than others[116]. Electron-withdrawing groups retard the photosensitized oxidation rate.

10.5 Use of Dye-Modified TiO2 for Visible-Light-AssistedDegradation of Colorless Pollutants

Organic radical cations or ROS formed in the dye-TiO2 photocatalytic systemcan degrade other coexisting pollutants to drive their decontamination undervisible-light irradiation through Eqs. (10.1)–(10.9). It has been proposed that theoxidation of colorless compounds is caused by the radical cation of the dyes [91,117] and the ROS, mainly through HO∙ formed in reaction (10.8) [118].

In the dye-modified TiO2 system, the mineralization extent is greatly depen-dent on the redox potential of the radical cation derived from the dye, which islower than that of the hole originated in the TiO2 VB after UV irradiation, or thatof the derived HO∙. Therefore, not all pollutants can be oxidized by this mecha-nism, as it occurs in the case of conventional UV-TiO2 photocatalysis; moreover,complete mineralization under visible light irradiation is generally not possible[117]. In contrast, eCB

− generated in reaction (10.2) have a similar role to thoseproduced under UV light, and they can be scavenged by O2 or by other electronacceptors Q (see Figure 10.1, Eqs. (10.4) and (10.10)), for example halogenatedpollutants or Cr(VI), as it will be discussed below [32, 119, 120].

An early review on degradation of pollutants by dye-modified TiO2 undervisible light has been published by Chatterjee and Dasgupta more than10 years ago [47], which includes Chatterjee’s group works on the pho-todegradation of various organic pollutants, such as phenols, chlorophenols,halocarbons (trichloroethylene, 1,2-dichloroethane and 1,4-dichlorobenzene),surfactants and pesticides, using TiO2 modified with different dyes, for example,thionine, eosinY, RhB, methylene blue, nile blue A, and safranine O (e.g., Refs.[121–124]). Works of other groups can also be consulted in Ref. [47]. Later,a plethora of works appeared, including oxidative and reductive processes,increasing the number of examples (e.g., [40, 42, 117]). As early as 1994,Ross et al. [91] used RB as TiO2 sensitizer for the photocatalytic oxidation ofterbutylazine under visible light. Our group [117] reported the degradation ofphenol, thiophenol, 4-chlorophenol (4-CP) and hydroquinone under 𝜆> 665 nmby sensitization with hydroxoaluminumtricarboxy-monoamide phthalocyanine(AlTCPc) adsorbed on the TiO2 surface. A mechanism has been proposed, with


the Pc radical cation produced by electron injection of the dye into the CB as thespecies responsible for the oxidation of the substrates (Eq. (10.3)). As indicatedabove, it was observed that not all compounds could be degraded using thismodified catalyst under visible light: electron donating compounds such asethylenediaminetetraacetic acid (EDTA), oxalic acid, and benzoquinone did notshow any reaction, in agreement with the more positive one-electron redoxpotential of these species in relation with the redox potential of the dye∙+/dyecouple. This was also proved later for citric acid [125]. It was pointed out thatHO∙ would be formed by reactions (10.4)–(10.8) only if the dye radical cationwere scavenged by reaction (10.3). Otherwise, carrier recombination would takeplace through reaction (10.2′).

Iliev [40] reported the oxidative degradation of phenolic compounds usingPc complexes on TiO2 at pH 9 under irradiation with 𝜆≥ 450 nm. Phenolscould be effectively degraded into compounds such as fumaric, maleic, andformic acids along with CO2 production, indicating incomplete degradation.p-Benzoquinone was also formed, but it could be easily oxidized further inalkaline solutions. A similar work was made for the photooxidation of sulfideand thiosulfate ions [126].

TiO2 powders impregnated with metal-free or Cu [5,10,15,20-tetra(4-tert-butylphenyl)]porphyrin were used in the degradation of 4-nitrophenol in aque-ous suspension. Time-resolved microwave conductivity (TRMC), electronicparamagnetic resonance (EPR), and X-ray photoelectron (XPS) techniques wereused for elucidation of the mechanism. TRMC measurements indicated that thenumber and lifetime of the photoinduced excess of charge carriers increasedin the presence of the macrocycles. A cooperative mechanism involving thephotoactivation of both TiO2 and sensitizer had been then proposed [42].

Dye radical cations having appropriate reduction potentials can be used inselective oxidation of alcohols. For example, a composite with alizarin red (AR)anchored on TiO2 was combined with the nitrosyl radical (2,2,6,6-tetramethyl-piperidin-1-yl)oxyl or (2,2,6,6-tetramethylpiperidin-1-yl)oxidanyl (TEMPO),producing an efficient photocatalytic system for selective oxidation of alcoholsunder visible-light irradiation, with a continuous oxidation and regeneration ofTEMPO by the dye radical cation, and the oxidation of various alcohols to thecorresponding aldehydes [91].

Zhao et al. [118] achieved complete mineralization of 4-CP using Pt(dcbpy)Cl2/TiO2 at 𝜆> 420 nm in the presence of O2. Other examples were the use of acidred 44 for the dye-sensitized photocatalysis under visible light for phenoldecomposition [127], sensitization of TiO2 with Zn(II) and Co(II) tetracar-boxyphthalocyanine (TCPcM, linked through ester bond) also for phenoldegradation [46], the use of a coumarin-343/TiO2 photocatalyst for 4-CPdegradation using a simple high intensity LED-based photoreactor [128],an anthocyanin/TiO2 system effectively catalyzing the photodegradationof methyl orange [129], a set of Co, Cu, Zn and metal-free phenylpor-phyrins for the degradation of luminol and photooxidation of terephthalicacid [130], and so on; the list continues to be very extensive. Recently,Boyer et al. [131] studied cis-dichlorobis(2,2′-bipyridyl-4,4′-dicarboxylicacid)ruthenium (II) (Ru(dcbpyH2)2Cl2) as a visible photosensitizer bound to

10.5 Use of Dye-Modified TiO2 for Visible-Light-Assisted Degradation of Colorless Pollutants 267

the surface of TiO2 electrospun fibers for the degradation of phenazopyridine(2,6-pyridinediamine,3-(phenylazo) monohydrochloride, PAP), as a modelbiopharmaceutical waste. Analogously, two Ru(II) polyaza complexes, N1-(2-aminobenzyliden)-N2,N2-bis(2-(2-aminobenzyliden)aminoethyl)ethane-1,2-diaminoruthenium(II) and N1,N2-bis(2-aminobenzyliden)ethane-1,2-diaminoruthenium(II), were incorporated to TiO2 via metal-ligand direct reaction, andthe photocatalytic activity under visible light irradiation was tested for ibuprofendegradation [132]. A3B-type nonsymmetrically tetrasubstituted zinc(II) andcobalt(II) phthalocyanines bearing one carboxy group (4-mercaptobenzoicacid/4-hydroxybenzoic acid) and three 4-tert-butylphenoxy substituents wereincorporated into TiO2 as reported in Ref. [117], and their photocatalytic activitywas tested on 4-CP decomposition under visible light, the modified TiO2 beingable of reuse with low dye decomposition [133].

Effect of pH was also investigated on the catalytic activity of TiO2. For example,TiO2 nanoparticles sensitized with a metal-free organic dye ((E)-3-(5-(5-(4-(bis(4-((2-(2 methoxyethoxy)ethoxy)methyl)phenyl)amino)phenyl)-thiophen-2-yl)thiophen-2-yl)-2-cyanoacrylic acid), exhibited higher or comparablevisible-light activities for conversion of pollutants (4-CP, As(III), Cr(VI)), in awider pH range in comparison with TiO2 sensitized with the RuL3 complex [134].

The effect of the presence of noble metals on the semiconductor was also inves-tigated. Photodeposition of Pt on the above cited Zn(II) and Co(II) TCPcM/TiO2systems enhanced the reaction photoefficiency for phenol degradation in com-parison with the system without Pt [46], attributed to the formation of a Schottkybarrier at the metal SC contact: Eq. (10.12) competes with the back ET, enhancingO2

∙− formation (Eq. (10.13)).

eCB− + Pt → Pt(e−) (10.12)

Pt(e−) + O2 → Pt + O2∙− (10.13)

Noble metal deposits also enhanced the reductive degradation of pollutantslike perchlorinated compounds, as it will be exemplified later [38] (Eq. (10.14))(Figure 10.3).

Pt(e−) + Q → Pt + Q∙− (10.14)

As indicated above, examples on photocatalytic oxidations performed by sensi-tized TiO2 under visible light increased in the last times. In contrast, few reportsexist on the reduction of compounds by this procedure, as most papers deal withhydrogen production (see Section 10.6). Reductive reactions take place in theabsence of O2 when suitable electron acceptors are able to capture eCB

− orig-inated in reaction (10.2) (Eq. (10.10)). The first example on these processes hasbeen N,N,N′,N′-tetraethyloxonine reduction by charge injection from the excitedanthracene-9-carboxylic acid into the TiO2 CB [27]. Later, reduction of halo-gen containing compounds and toxic high-valence transition metal ions suchas Cr(VI) has been attempted. For example, Cho et al. found that TiO2 mod-ified with tris-(4,4-dicarboxy-2,2-bipyridyl)ruthenium(II) complexes (Ru(II)L3)could reduce CCl4 through eCB

− injection from the excited dye in the absenceof dissolved O2 [119]. To sustain the reduction process, sacrificial electron and


hydrogen donors (e.g., 2-propanol) were used to regenerate the dye and to pro-vide a hydrogen source for the dehalogenation reaction. As also said, the inclusionof noble metal deposits such as Pt, Ag, Au and Pd improves the reaction, withPt showing the best activity [38, 135]. In the case of Pt, the back ET to the oxi-dized dye proceeds in the microsecond timescale, while the electron trappingprocess by Pt is in the order of few picoseconds [136] (i.e., rate10.2′ ≪ rate10.13 inFigure 10.3). Additionally, the metal could act as a catalyst for the C—Cl bondcleavage by stabilizing intermediate chlorinated carbon radicals.

Kyung et al. [120] found simultaneous and synergistic conversion in TiO2/dye/metal ion systems (ternary components) under visible light (𝜆> 420 nm) forremoval of acid orange 7 and metal ions such as Cr(VI) or Ag(I). In those cases,complexes between metal ions and dyes were suggested to induce intracomplexET upon visible-light absorption. In contrast, the Cr(VI)/RhB system exhibitedinsignificant visible-light reactivity, which was ascribed to the low adsorption ofthe dye on TiO2.

Our group [137] proved that AR chelated to TiO2 promoted Cr(VI) reductionunder visible light (𝜆∼ 470 nm); the involvement of monoelectronic steps in theCr(VI) transformation was proved by EPR detection of Cr(V). In the same paper,spectroscopic evidences were presented for the first time that Cr(VI) forms acharge transfer complex with TiO2 nanoparticles, indicating a strong interactionbetween Cr(VI) and the SC. In this line, we adsorbed AlTCPc at different load-ings on Degussa P-25 and tested the system for Cr(VI) photocatalytic reductionunder visible light irradiation in the presence of 4-CP as sacrificial donor [125].A rapid reaction took place in spite of the presumable aggregation of the dye onthe TiO2 surface. In this case, it was proposed that the complex between Cr(VI)and TiO2 was responsible for the fast capture of eCB

− by Cr(VI), inhibiting theformation of ROS in the reductive pathway. Under UV irradiation, AlTCPc-TiO2was more efficient than bare TiO2, and no bleaching of the dye was observedas long as 4-CP was present in the system. In a subsequent paper [58], photocur-rent and absorption spectra of AlTCPc/TiO2 films were studied together with theabsorption and fluorescence of the dye in solution as a function of the dye con-centration. Results identified the actual photoactive species as the monomericdye electronically coupled to the SC.

CB O2 / O2•−

VB

eCB− + dye•+

(10.1)hν

dye*

Q / Q•−

(10.2)

(10.2′)

e−

e–

Schottky barrier

e–

Noble metalnanoparticle

(10.14)

(10.15)

(10.13)

dye

Figure 10.3 Electron transfer mechanism in TiO2 nanoparticles modified with dyes andnoble-metal nanoparticles under visible-light irradiation.

10.6 Water Splitting and Hydrogen Production using Dye-Modified TiO2 Photocatalysts 269

10.6 Water Splitting and Hydrogen Production usingDye-Modified TiO2 Photocatalysts under Visible Light

Water splitting is an endoergic process of paramount interest for storing solarenergy as chemical energy.

2H2O → 2H2 + O2 (ΔG0 = 237.1 kJ mol−1) (10.15)In 1972, the liminal work by Fujishima and Honda demonstrated the feasibility

of the photoelectrochemical water splitting under UV irradiation using a rutilesingle-crystal TiO2 photoanode and a Pt cathode with an electrochemical (powersupply) or chemical (pH difference) bias [138]. Since then, overall water splittingfor the production of H2 using electrode or particulate photocatalyst systemshas received a lot of attention, and very interesting reviews have been publishedin recent years (e.g., [139–141]). In a photocatalytic device, light is absorbed(directly or through a photosensitizer) and the electrons and holes reduce waterto hydrogen and oxidize water to oxygen, respectively. One critical drawback isthat in photocatalytic systems, hydrogen and oxygen are produced in the sameenvironment and easily recombine before they can be separated. For this reason,the investigation is generally restricted to only one half-reaction, that is, eitherreduction or oxidation; the common approach is to limit the process to reductionto hydrogen in the presence of sacrificial electron donors, for example, alcohols,amines, EDTA, formic acid, or a redox system, such as I3

−/I−, added to the solu-tion to sustain the reaction cycle, which can prevent recombination processes.

As previously indicated, multielectron transfer processes under typical condi-tions are highly improbable [21–23], and the first step of the generation of H2 willbe the conversion of protons to atomic hydrogen on the photocatalyst surface.The process can be described as follows [23]:

Hsurf+ + e− → Hsurf

•

Hsurf• + Hsurf

• → H2(g)

Hsurf+ + e− → Hsurf

•

(10.16)

The one-electron redox potential for this couple is reported to be very nega-tive, at least in homogeneous solutions (E0 =−2.3 eV vs SHE) [142], and the useof metals (e.g., Pt, Pd, Au, Rh, Ni, Cu, and Ag) deposited on the TiO2 surfaceis necessary to make hydrogen evolution rate measurable. The metal not onlycatalyzes the reduction of protons but contributes also to hydrogen evolutionthrough electron trapping across the Schottky barrier [38].

Hydrogen generation is also possible under visible light by dye photosensitiza-tion to overcome the lack of response of TiO2 in the visible. Here, the dye acts asa light antenna, TiO2 as a chemically stable charge transporter and Pt as a chargereservoir [84, 143] (Eq. (10.17)):

Pt∕TiO2∕dyeH+ + P + h𝜈 → · · · → · · · → 1/2H2 + P∙+ (10.17)

where P is a sacrificial electron donor.


The following mechanism, adapted from Ref. [143], has been proposed for H2generation, initiated by excitation of the dye under visible light:

Pt∕TiO2∕dye + h𝜈 → · · · → Pt(e−)∕TiO2∕dye∙+ (10.18)Pt(e−)∕TiO2∕dye∙+ + P → Pt(e−)∕TiO2∕dye + P∙+ (10.19)Pt(e−)∕TiO2∕dye + H+ → Pt∕TiO2∕dye + Hsurf

∙ (10.20)Hsurf

∙ + Hsurf∙ → H2(g) (10.21)

The simplified scheme (10.18)–(10.21) is actually a complex process involvingseveral competitive pathways: electron injection in competition with relaxationof the excited state of the dye, charge recombination between the radical cationand eCB

− as a crucial energy-wasting process stimulated by the slow H2 genera-tion and/or dye regeneration, reduction of the dye radical cation by the donorsto leave long-lived electrons, accumulation of electrons on Pt, and reduction ofprotons (or water) to H2 [84].

A reasonable H2 production rate can be obtained by efficient absorption of vis-ible light and ET from the excited dyes to the TiO2 CB. The back ET is mostly inthe order of nanoseconds to microseconds or even milliseconds, while the elec-tron injection times are in the order of femtoseconds [71, 140]. The fast electroninjection and slow backward reaction make dye-sensitized semiconductors fea-sible for energy conversion. Fast dye regeneration from the sacrificial electrondonor must take place through reaction (10.19), resulting in recovery of the dyestarting redox state [84].

Numerous studies have been carried out for visible-light-driven H2 generationby water reduction with aqueous dispersions of Pt-loaded TiO2 particles in thepresence of a dye, typically a Ru(II) complex, together with a sacrificial electrondonor. Dyes used for this purpose are thiazines, phenazines, xanthenes, andtriphenylmethane derivatives. Several papers report results on this application,using safranines, acridines, proflavine [144], xanthene dyes [145], rutheniumbipyridyl complexes (RuL3) [135, 146–150], porphyrins [150, 151], phthalo-cyanines [85], eosin [152], organic dyes containing quinoxaline and pyrido-[3,4-b]pyrazine in the D–A–π–A configuration [74], (diphenylaminophenyl)dithiopheneacrylic acid [153], phenothiazine-based organic dyes [154–156], tinporphyrin [157], and so on. The valuable reviews by Cecconi et al. [84] and byZhang et al. [141] list many examples of this application.

10.7 Conclusions

Over the past decades, considerable efforts have been put on the feasibility ofvisible-light-activated TiO2 to extend the usable solar energy spectrum. Fromthe different strategies for this purpose, photosensitization by organic dyes ormetal complexes has proven to be one of the most effective for enabling TiO2to be used upon visible-light illumination. Present research focuses mainly onthe design of TiO2-based systems for environmental (dye-sensitized photocatal-ysis) and energy (hydrogen evolution, solar cells) applications. Different dyes andvisible absorbing metal complexes can be used as sensitizers.

The mechanisms involved in these processes are different from those proposedfor semiconductors under UV irradiation. Briefly, after light absorption, a very

References 271

fast ET from the excited dye to the SC conduction band takes place, leaving adye radical cation of mild oxidant power and an eCB

−, whose transfer can occurto a reducible substrate, eventually O2 (in photocatalytical oxidations), an oxi-dized metal or other electron acceptors (in photocatalytical reductions), or H+

(for hydrogen generation); electricity may be generated as well as in DSSCs. Theproduced dye radical cation must be further reduced to regenerate the dye, unlessthe degradation of the dye itself is the objective.

One of the most important issues, when the dye itself is not the target to bedegraded, is the need to improve the catalyst stability. Most of the dye-modifiedTiO2 photocatalysts rapidly lose their activity with repeated usage or are grad-ually deactivated even in the dark. In addition, most sensitizers are not stable ifthe irradiation source delivers also UV. Stability in aqueous solution is a criticalrequirement as most applications of visible-light photocatalysts use water assolvent.

The feasibility of ETs is strongly dependent on the interaction between dyeand SC. Strong coupling ensures ultrafast charge injection, minimizing losses byradiative or nonradiative deactivation, while chelation or chemical binding pre-vents dye leakage. Physical adsorption is needed if the objective is the degradationof the dye.

Dye-sensitized photocatalysis does not allow oxidation of pollutants with veryhigh redox potentials because the oxidant is the dye radical cation. ROS, partic-ularly the HO∙ radical produced in the reductive pathway, though being strongeroxidants, would not be formed if the dye radical cation is not reduced by the targetpollutant. In turn, lowering the oxidizing power in dye-sensitized photocatalysiscan be relevant if the objective is not mineralization but selective oxidation.

Summarizing, recent advances and strategies to improve dye-sensitized photo-catalytic activity of TiO2 under visible-light irradiation were described. The goalis to obtain stable and reusable materials with an extended photoactivity overthe visible range of the solar spectrum. The clear understanding of the photo-catalytic processes and mechanisms involved in this materials is crucial for thedevelopment of TiO2-based reusable heterogeneous photocatalysts for low-costapplications.

Acknowledgement

M.I.L., E.S.R., J.M.M., and H.B.R. wish to remember M.A.G. after her recent pass-ing away for the engagement, discipline and friendship with which she faced ourcommon activities.

References

1 Kamat, P.V. (1993) Photochemistry on nonreactive and reactive (semiconduc-tor) surfaces. Chem. Rev., 93, 267–300.

2 Linsebigler, A.L., Lu, G., and Yates, J.T. Jr. (1995) Photocatalysis on TiO2surfaces: principles, mechanisms, and selected results. Chem. Rev., 95,735–758.


3 Serpone, N., Lawless, D., Disdier, J., and Herrmann, J.-M. (1994) Spectro-scopic, photoconductivity, and photocatalytic studies of TiO2 colloids: nakedand with the lattice doped with Cr3+, Fe3+, and V5+ cations. Langmuir, 10,643–652.

4 Banerjee, S., Pillai, S.C., Falaras, P., O’Shea, K.E., Byrne, J.A., and Dionysiou,D.D. (2014) New insights into the mechanism of visible light photocatalysis.J. Phys. Chem. Lett., 5, 2543–2554.

5 Wade, J. (2005) An Investigation of TiO2-ZnFe2O4 Nanocomposites forVisible-Light Photocatalysis, Master of Science in Electrical EngineeringThesis, Department of Electrical Engineering, College of Engineering,University of South Florida.

6 Hagfeldt, A. and Grätzel, M. (1995) Light-induced redox reactions innanocrystalline systems. Chem. Rev., 95, 49–68.

7 Emeline, A.V., Kuzmin, G.N., and Serpone, N. (2008) Wavelength-dependentphotostimulated adsorption of molecular O2 and H2 on second generationtitania photocatalysts: the case of the visible-light-active N-doped TiO2system. Chem. Phys. Lett., 454, 279–283.

8 Kalyanasundaram, K., Grätzel, M., and Pelizzetti, E. (1986) Interfacialelectron transfer in colloidal metal and semiconductor dispersions andphotodecomposition of water. Coord. Chem. Rev., 69, 57–125.

9 Rehman, S., Ullah, R., Butt, A.M., and Gohar, N.D. (2009) Strategiesof making TiO2 and ZnO visible light active. J. Hazard. Mater., 170,560–569.

10 Pelaez, M., Nolan, N.T., Pillai, S.C., Seery, M.K., Falaras, P., Kontos, A.G.,Dunlop, P.S.M., Hamilton, J.W.J., Byrne, J.A., O’Shea, K., Entezari, M.H., andDionysiou, D.D. (2012) A review on the visible light active titanium dioxidephotocatalysts for environmental applications. Appl. Catal. B Environ., 125,331–349.

11 Carp, O., Huisman, C.L., and Reller, A. (2004) Photoinduced reactivity oftitanium dioxide. Prog. Solid State Chem., 32, 33–177.

12 Wang, Z., Chen, C., Ma, W., and Zhao, J. (2013) Chapter 8: Sensitization oftitania semiconductor: a promising strategy to utilize visible light, in Photo-catalysis and Water Purification (ed. P. Pichat), Wiley-VCH Verlag GmbH,Weinheim, Germany, pp. 199–240.

13 Wang, J.-C., Zhang, L., Fang, W.-X., Ren, J., Li, Y.-Y., Yao, H.-C., Wang, J.-S.,and Li, Z.-J. (2015) Enhanced photoreduction CO2 activity over directZ-scheme α-Fe2O3/Cu2O heterostructures under visible light irradiation.ACS Appl. Mater. Interfaces, 7, 8631–8639.

14 Aguirre, M.E., Zhou, R., Eugene, A.J., Guzman, M.I., and Grela, M.A.(2017) Cu2O/TiO2 heterostructures for CO2 reduction through a directZ-scheme: Protecting Cu2O from photocorrosion. Appl. Catal. B Environ.,217, 485–493.

15 Wang, P., Wang, J., Ming, T., Wang, X., Yu, H., Yu, J., Wang, Y., and Lei, M.(2013) Dye-sensitization-induced visible-light reduction of graphene oxidefor the enhanced TiO2 photocatalytic performance. ACS Appl. Mater. Inter-faces, 5, 2924–2929.

16 Aguirre, M.E., Armanelli, A., Perelstein, G., Feldhoff, A., Tolley, A.J., andGrela, M.A. (2015) Modulation of electron transfer process in Au–ZnOnanostructures. Nanoscale, 7, 6667–6674.

References 273

17 Aguirre, M.E., Custo, G., Goes, M.S., Bueno, P.R., Zampieri, G., andGrela, M.A. (2014) Critical water effect on the plasmon band and visiblelight activity of Au/ZnO nanocomposites. J. Phys. Chem. C, 118, 2018–2027.

18 Vinodgopal, K., Wynkoop, D.E., and Kamat, P.V. (1996) Environmentalphotochemistry on semiconductor surfaces: photosensitized degradation of atextile azo dye, Acid Orange 7, on TiO2 particles using visible light. Environ.Sci. Technol., 30, 1660–1666.

19 Litter, M.I., Quici, N., Meichtry, J.M., and Montesinos, V.N. (2017) Pho-tocatalytic treatment of inorganic materials with TiO2 nanoparticles, inEncyclopedia of Nanoscience and Nanotechnology (ed. H.S. Nalwa), AmericanScientific Publishers, Valencia, California.

20 Price, M. and Maldonado, S. (2011) Dye-sensitized hole injection at p-typeIII–V electrode interfaces. ECS Trans., 35, 153–163.

21 Levy, I.K., Brusa, M.A., Aguirre, M.E., Custo, G., San Román, E., Litter, M.I.,and Grela, M.A. (2013) Exploiting electron storage in TiO2 nanoparticles fordark reduction of As(V) by accumulated electrons. Phys. Chem. Chem. Phys.,15, 10335–10338.

22 Kamat, P.V. (2012) Manipulation of charge transfer across semiconductorinterface. A criterion that cannot be ignored in photocatalyst design. J. Phys.Chem. Lett., 3, 663–672.

23 Serpone, N., Emeline, A.V., Ryabchuk, V.K., Kuznetsov, V.N., Artem’ev, Y.M.,and Horikoshi, S. (2016) Why do hydrogen and oxygen yields fromsemiconductor-based photocatalyzed water splitting remain disappoint-ingly low? Intrinsic and extrinsic factors impacting surface redox reactions.ACS Energy Lett., 1, 931–948.

24 Kamat, P.V. (1985) Photoelectrochemistry in colloidal systems. Part 2. – Aphotogalvanic cell based on TiO2 semiconductor colloid. J. Chem. Soc.Faraday Trans. 1, 81, 509–518.

25 Asbury, J.B., Anderson, N.A., Hao, E., Ai, X., and Lian, T. (2003) Parametersaffecting electron injection dynamics from ruthenium dyes to titanium diox-ide nanocrystalline thin film. J. Phys. Chem. B, 107, 7376–7386.

26 Koops, S., O’Regan, C., Barnes, P.R.F., and Durrant, J.R. (2009) Parametersinfluencing the efficiency of electron injection in dye-sensitized solar cells.J. Am. Chem. Soc., 131, 4808–4818.

27 Kamat, P.V. (1989) Photoelectrochemistry in particulate systems.9. Photosensitized reduction in a colloidal titania system usinganthracene-9-carboxylate as the sensitizer. J. Phys. Chem., 93, 859–864.

28 Gerischer, H. and Willig, F. (1976) Reaction of excited dye molecules at elec-trodes. Top. Curr. Chem., 61, 31–84.

29 Nasr, C., Vinodgopal, K., Fisher, L., Hotchandani, S., Chattopadhyay, A.K.,and Kamat, P.V. (1996) Environmental photochemistry on semiconductorsurfaces. Visible light induced degradation of a textile diazo dye, NaphtholBlue Black, on TiO2 nanoparticles. J. Phys. Chem., 100, 8436–8442.

30 Zhao, J., Chen, C., and Ma, W. (2005) Photocatalytic degradation of organicpollutants under visible light irradiation. Top. Catal., 35, 269–278.

31 Bell, T.D.M., Pagba, C., Myahkostupov, M., Hofkens, J., and Piotrowiak, P.(2006) Inhomogeneity of electron injection rates in dye-sensitized TiO2:comparison of the mesoporous film and single nanoparticle behavior. J. Phys.Chem. B, 110, 25314–25321.


32 Chen, C., Ma, W., and Zhao, J. (2010) Semiconductor-mediated photodegra-dation of pollutants under visible-light irradiation. Chem. Soc. Rev., 39,4206–4219.

33 Wu, T., Liu, G., Zhao, J., Hidaka, H., and Serpone, N. (1998) Photoassisteddegradation of dye pollutants. V. Self-photosensitized oxidative transfor-mation of rhodamine B under visible light irradiation in aqueous TiO2dispersions. J. Phys. Chem. B, 102, 5845–5851.

34 Balzani, V., Juris, A., Venturi, M., Campagna, S., and Serroni, S. (1996)Luminescent and redox-active polynuclear transition metal complexes.Chem. Rev., 96, 759–833.

35 Bandara, J. and Kiwi, J. (1999) Fast kinetic spectroscopy, decoloration andproduction of H2O2 induced by visible light in oxygenated solutions of theazo dye Orange II. New J. Chem., 23, 717–724.

36 Takeshita, K., Sasaki, Y., Kobashi, M., Tanaka, Y., Maeda, S., Yamakata, A.,Ishibashi, T.-A., and Onishi, H. (2004) Effect of annealing temperature onback electron transfer and distribution of deep trap sites in dye-sensitizedTiO2, studied by time-resolved infrared spectroscopy. J. Phys. Chem. B, 108,2963–2969.

37 Prezhdo, O.V., Duncan, W.R., and Prezhdo, V.V. (2008) Dynamics of the pho-toexcited electron at the chromophore–semiconductor interface. Acc. Chem.Res., 41, 339–348.

38 Bae, E. and Choi, W. (2003) Highly enhanced photoreductive degradation ofperchlorinated compounds on dye-sensitized metal/TiO2 under visible light.Environ. Sci. Technol., 37, 147–152.

39 Fan, F.-R.F. and Bard, A.J. (1979) Spectral sensitization of the heteroge-neous photocatalytic oxidation of hydroquinone in aqueous solutions atphthalocyanine-coated titanium dioxide powders. J. Am. Chem. Soc., 101,6139–6140.

40 Iliev, V. (2002) Phthalocyanine-modified titania-catalyst for photooxidationof phenols by irradiation with visible light. J. Photochem. Photobiol. A, 151,195–199.

41 Yu, J.C., Xie, Y., Tang, H.Y., Zhang, L., Chan, H.C., and Zhao, J. (2003)Visible light-assisted bactericidal effect of metalphthalocyanine-sensitizedtitanium dioxide films. J. Photochem. Photobiol. A, 156, 235–241.

42 Mele, G., Del Sole, R., Vasapollo, G., García López, E., Palmisano, L., andSchiavello, M. (2003) Photocatalytic degradation of 4-nitrophenol in aqueoussuspension by using polycrystalline TiO2 impregnated with functionalizedCu(II)–porphyrin or Cu(II)–phthalocyanine. J. Catal., 217, 334–342.

43 Herrera-Melián, J.A., Doña Rodríguez, J.M., Viera Suárez, A., Tello Rendón,E., Valdes do Campo, C., Araña, J., and Pérez Peña, J. (2000) The photocat-alytic disinfection of urban waste waters. Chemosphere, 41, 323–327.

44 Ranjit, K.T., Willner, I., Bossmann, S., and Braun, A. (1998) Iron(III)phthalocyanine-modified titanium dioxide: a novel photocatalyst for theenhanced photodegradation of organic pollutants. J. Phys. Chem. B, 102,9397–9403.

References 275

45 Héquet, V., Le Cloirec, P., Gonzalez, C., and Meunier, B. (2000) Photocat-alytic degradation of atrazine by porphyrin and phthalocyanine complexes.Chemosphere, 41, 379–386.

46 Granados, O.G., Paez M, C.A., Martínez, O.F., and Páez-Mozo, E.A. (2005)Photocatalytic degradation of phenol on TiO2 and TiO2/Pt sensitized withmetallophthalocyanines. Catal. Today, 107/108, 589–594.

47 Chatterjee, D. and Dasgupta, S. (2005) Visible light induced photocatalyticdegradation of organic pollutants. J. Photochem. Photobiol., C, 6, 186–205.

48 Beranek, R. and Kisch, H. (2008) Tuning the optical and photoelectrochemi-cal properties of surface-modified TiO2. Photochem. Photobiol. Sci., 7, 40–48.

49 Moser, J. and Grätzel, M. (1984) Photosensitized electron injection in col-loidal semiconductors. J. Am. Chem. Soc., 106, 6557–6564.

50 He, J., Zhao, J., Hidaka, H., and Serpone, N. (1998) EPR characteristics ofa dye/colloidal TiO2 system under visible light irradiation. J. Chem. Soc.Faraday Trans., 94, 2375–2378.

51 He, J., Chen, F., Zhao, J., and Hidaka, H. (1998) Adsorption model of ethylester of fluorescein on colloidal TiO2 and the mechanism of the interfacialelectron transfer. Colloids Surf. A, 142, 49–57.

52 Wang, Z., Ma, W., Chen, C., Ji, H., and Zhao, J. (2011) Probing paramag-netic species in titania-based heterogeneous photocatalysis by electron spinresonance (ESR) spectroscopy – a mini review. Chem. Eng. J., 170, 353–362.

53 Qu, P., Zhao, J., Shen, T., and Hidaka, H. (1998) TiO2-assisted photodegra-dation of dyes: a study of two competitive primary processes in the degra-dation of RB in an aqueous TiO2 colloidal solution. J. Mol. Catal. A, 129,257–268.

54 Regan, B.O. and Grätzel, M. (1991) A low-cost, high-efficiency solar cellbased on dye-sensitized colloidal TiO2 films. Nature, 353, 737–740.

55 Sugathan, V., John, E., and Sudhakar, K. (2015) Recent improvements in dyesensitized solar cells: a review. Renew. Sustain. Energy Rev., 52, 54–64.

56 Hardin, B.E., Snaith, H.J., and McGehee, M.D. (2012) The renaissance ofdye-sensitized solar cells. Nature Photon., 6, 162–169.

57 Lee, C.-P., Lin, R.Y.-Y., Lin, L.-Y., Li, C.-T., Chu, T.-C., Sun, S.-S.,Lin, J.T., and Ho, K.-C. (2015) Recent progress in organic sensitizers fordye-sensitized solar cells. RSC Adv., 5, 23810–23825.

58 Rodríguez, H.B., Di Iorio, Y., Meichtry, J.M., Grela, M.A., Litter, M.I., andSan Román, E. (2013) Evidence on dye clustering in the sensitization of TiO2by aluminum phthalocyanine. Photochem. Photobiol. Sci., 12, 1984–1990.

59 Reddy, P.Y., Giribabu, L., Lyness, C., Snaith, H.J., Vijaykumar,C., Chandrasekharam, M., Lakshmikantam, M., Yum, J.-H.,Kalyanasundaram, K., Grätzel, M., and Nazeeruddin, M.K. (2007) Effi-cient sensitization of nanocrystalline TiO2 films by a near-IR absorbingunsymmetrical zinc phthalocyanine. Angew. Chem. Int. Ed., 46, 373–376.

60 Mori, S., Nagata, M., Nakahata, Y., Yasuta, K., Goto, R., Kimura, M., andTaya, M. (2010) Enhancement of incident photon-to-current conversionefficiency for phthalocyanine-sensitized solar cells by 3D molecular struc-turalization. J. Am. Chem. Soc., 132, 4054–4055.


61 Hua, Y., Chang, S., Huang, D., Zhou, X., Zhu, X., Zhao, J., Chen, T.,Wong, W.-Y., and Wong, W.-K. (2013) Significant improvement ofdye-sensitized solar cell performance using simple phenothiazine-baseddyes. Chem. Mater., 25, 2146–2153.

62 Amao, Y. and Komori, T. (2003) Dye-sensitized solar cell using a TiO2nanocrystalline film electrode modified by an aluminum phthalocyanine andmyristic acid coadsorption layer. Langmuir, 19, 8872–8875.

63 Nazeeruddin, M.K., Humphry-Baker, R., Grätzel, M., Wöhrle, D.,Schnurpfeil, G., Schneider, G., Hirth, A., and Trombach, N. (1999) Efficientnear-IR sensitization of nanocrystalline TiO2 films by zinc and aluminumphthalocyanines. J. Porphyrins Phthalocyanines, 3, 230–237.

64 Xie, Y., Tang, Y., Wu, W., Wang, Y., Liu, J., Li, X., Tian, H., and Zhu, W.-H.(2015) Porphyrin cosensitization for a photovoltaic efficiency of 11.5%: arecord for non-ruthenium solar cells based on iodine electrolyte. J. Am.Chem. Soc., 137, 14055–14058.

65 Sakata, T., Hashimoto, K., and Hiramoto, M. (1990) New aspects of electrontransfer on semiconductor surface: dye-sensitization system. J. Phys. Chem.,94, 3040–3045.

66 Creutz, C., Brunschwig, B.S., and Sutin, N. (2006) Interfacial charge-transferabsorption: 3. Application to semiconductor-molecule assemblies. J. Phys.Chem. B, 110, 25181–25190.

67 Pilarczyk, K., Kwolek, P., Podborska, A., Gaweda, S., Oszajca, M., andSzaciłowski, K. (2017) Unconventional computing realized with hybridmaterials exhibiting the photoelectrochemical photocurrent switching(PEPS) effect, in Advances in Unconventional Computing, Volume 2: Pro-totypes, Models and Algorithms (ed. A. Adamatzky), Springer, Switzerland,pp. 429–467.

68 Duncan, W.R. and Prezhdo, O.V. (2007) Theoretical studies of photoin-duced electron transfer in dye-sensitized TiO2. Annu. Rev. Phys. Chem., 58,143–184.

69 Zhang, L. and Cole, J.M. (2015) Anchoring groups for dye-sensitized solarcells. ACS Appl. Mater. Interfaces, 7, 3427–3455.

70 Fujisawa, J., Osawa, A., and Hanaya, M. (2016) A strategy to minimize theenergy offset in carrier injection from excited dyes to inorganic semiconduc-tors for efficient dye-sensitized solar energy conversion. Phys. Chem. Chem.Phys., 18, 22244–22253.

71 Hagfeldt, A., Boschloo, G., Sun, L., Kloo, L., and Pettersson, H. (2010)Dye-sensitized solar cells. Chem. Rev., 110, 6595–6663.

72 Yella, A., Mai, C.-L., Zakeeruddin, S.M., Chang, S.-N., Hsieh, C.-H., Yeh,C.-Y., and Grätzel, M. (2014) Molecular engineering of push–pull por-phyrin dyes for highly efficient dye-sensitized solar cells: the role of benzenespacers. Angew. Chem. Int. Ed., 126, 3017–3021.

73 Mathew, S., Yella, A., Gao, P., Humphry-Baker, R., Curchod, B.F.E.,Ashari-Astani, N., Tavernelli, I., Rothlisberger, U., Nazeeruddin, M.K., andGrätzel, M. (2014) Dye-sensitized solar cells with 13% efficiency achievedthrough the molecular engineering of porphyrin sensitizers. Nat. Chem.,6, 242–247.

74 Li, X., Cui, S., Wang, D., Zhou, Y., Zhou, H., Hu, Y., Liu, J.-G., Long, Y., Wu,W., Hua, J., and Tian, H. (2014) New organic donor–acceptor–π–acceptor

References 277

sensitizers for efficient dye-sensitized solar cells and photocatalytic hydrogenevolution under visible-light irradiation. ChemSusChem, 7, 2879–2888.

75 Liang, M. and Chen, J. (2013) Arylamine organic dyes for dye-sensitizedsolar cells. Chem. Soc. Rev., 42, 3453–3488.

76 Wu, Y., Zhu, W.-H., Zakeeruddin, S.M., and Grätzel, M. (2015) Insight intoD–A–π–A structured sensitizers: a promising route to highly efficient andstable dye-sensitized solar cells. ACS Appl. Mater. Interfaces, 7, 9307–9318.

77 Mishra, A., Fischer, M.K.R., and Bäuerle, P. (2009) Metal-free organic dyesfor dye-sensitized solar cells: From structure: property relationships todesign rules. Angew. Chem. Int. Ed., 48, 2474–2499.

78 Katoh, R., Furube, A., Mori, S., Miyashita, M., Sunahara, K., Koumura, N.,and Hara, K. (2009) Highly stable sensitizer dyes for dye-sensitized solarcells: role of the oligothiophene moiety. Energy Environ. Sci., 2, 542–546.

79 Bureš, F. (2014) Fundamental aspects of property tuning in push–pullmolecules. RSC Adv., 4, 58826–58851.

80 Ding, W.-L., Wang, D.-M., Geng, Z.-Y., Zhao, X.-L., and Xu, W.-B. (2013)Density functional theory characterization and verification of high-performance indoline dyes with D–A–π–A architecture for dye-sensitizedsolar cells. Dyes Pigm., 98, 125–135.

81 Shalini, S., Balasundara prabhu, R., Prasanna, S., Mallick, T.K., andSenthilarasu, S. (2015) Review on natural dye sensitized solar cells: Oper-ation, materials and methods. Renew. Sustain. Energy Rev., 51, 1306–1325.

82 Tian, J. and Cao, G. (2013) Semiconductor quantum dot-sensitized solarcells. Nano Rev., 4, 22578–22586.

83 Duan, J., Zhang, H., Tang, Q., He, B., and Yu, L. (2015) Recent advances incritical materials for quantum dot-sensitized solar cells: a review. J. Mater.Chem. A, 3, 17497–17510.

84 Cecconi, B., Manfredi, N., Montini, T., Fornasiero, P., and Abbotto, A. (2016)Dye-sensitized solar hydrogen production: the emerging role of metal-freeorganic sensitizers. Eur. J. Org. Chem., 2016, 5194–5215.

85 Zhang, X., Peng, B., Peng, T., Yu, L., Li, R., and Zhang, J. (2015) A newroute for visible/near-infrared-light-driven H2 production over titania:co-sensitization of surface charge transfer complex and zinc phthalocyanine.J. Power Sources, 298, 30–37.

86 Rauf, M.A. and Ashraf, S.S. (2009) Fundamental principles and applicationof heterogeneous photocatalytic degradation of dyes in solution. Chem. Eng.J., 151, 10–18.

87 Khataee, A.R. and Kasiri, M.B. (2010) Photocatalytic degradation of organicdyes in the presence of nanostructured titanium dioxide: influence of thechemical structure of dyes. J. Mol. Catal. A Chem., 328, 8–26.

88 Teh, C.M. and Mohamed, A.R. (2011) Roles of titanium dioxide andion-doped titanium dioxide on photocatalytic degradation of organic pol-lutants (phenolic compounds and dyes) in aqueous solutions: a review.J. Alloys Compd., 509, 1648–1660.

89 Han, F., Kambala, V.S.R., Srinivasan, M., Rajarathnam, D., and Naidu, R.(2009) Tailored titanium dioxide photocatalysts for the degradation oforganic dyes in wastewater treatment: a review. Appl. Catal. A, 359, 25–40.

90 Watanabe, T., Takizawa, T., and Honda, K. (1977) Photocatalysis throughexcitation of adsorbates. 1. Highly efficient N-deethylation of rhodamine Badsorbed to cadmium sulfide. J. Phys. Chem., 81, 1845–1851.


91 Ross, H., Bendig, J., and Hecht, S. (1994) Sensitized photocatalytical oxida-tion of terbutylazine. Sol. Energy Mater. Sol. Cells, 33, 475–481.

92 Zhang, M., Chen, C., Ma, W., and Zhao, J. (2008) Visible-light-induced aero-bic oxidation of alcohols in a coupled photocatalytic system of dye-sensitizedTiO2 and TEMPO. Angew. Chem. Int. Ed., 120, 9876–9879.

93 Zhao, J., Wu, T., Wu, K., Oikawa, K., Hidaka, H., and Serpone, N. (1998)Photoassisted degradation of dye pollutants. 3. Degradation of the cationicdye rhodamine B in aqueous anionic surfactant/TiO2 dispersions under vis-ible light irradiation: evidence for the need of substrate adsorption on TiO2particles. Environ. Sci. Technol., 32, 2394–2400.

94 Liu, G., Wu, T., Zhao, J., Hidaka, H., and Serpone, N. (1999) Photoassisteddegradation of dye pollutants. 8. Irreversible degradation of alizarin redunder visible light radiation in air-equilibrated aqueous TiO2 dispersions.Environ. Sci. Technol., 33, 2081–2087.

95 Liu, G., Li, X., Zhao, J., Horikoshi, S., and Hidaka, H. (2000) Photooxidationmechanism of dye alizarin red in TiO2 dispersions under visible illumina-tion: an experimental and theoretical examination. J. Mol. Catal. A, 153,221–229.

96 Chen, C., Li, X., Ma, W., Zhao, J., Hidaka, H., and Serpone, N. (2002) Effectof transition metal ions on the TiO2-assisted photodegradation of dyesunder visible irradiation: a probe for the interfacial electron transfer processand reaction mechanism. J. Phys. Chem. B, 106, 318–324.

97 Liu, G., Li, X., Zhao, J., Hidaka, H., and Serpone, N. (2000) Photooxidationpathway of sulforhodamine-B. Dependence on the adsorption mode on TiO2exposed to visible light radiation. Environ. Sci. Technol., 34, 3982–3990.

98 Wu, T., Lin, T., Zhao, J., Hidaka, H., and Serpone, N. (1999) TiO2-assistedphotodegradation of dyes. 9. Photooxidation of a squarylium cyanine dye inaqueous dispersions under visible light irradiation. Environ. Sci. Technol., 33,1379–1387.

99 Liu, G. and Zhao, J. (2000) Photocatalytic degradation of dye sulforhodamineB: a comparative study of photocatalysis with photosensitization. New J.Chem., 24, 411–417.

100 Chen, C., Zhao, W., Li, J., Zhao, J., Hidaka, H., and Serpone, N. (2002)Formation and identification of intermediates in the visible-light-assistedphotodegradation of sulforhodamine-B dye in aqueous TiO2 dispersion.Environ. Sci. Technol., 36, 3604–3611.

101 Zhang, T., Oyama, T., Horikoshi, S., Hidaka, H., Zhao, J., and Serpone, N.(2002) Photocatalyzed N-demethylation and degradation of methylene bluein titania dispersions exposed to concentrated sunlight. Sol. Energy Mater.Sol. Cells, 73, 287–303.

102 Zhao, J., Hidaka, H., Takamura, A., Pelizzetti, E., and Serpone, N. (1993)Photodegradation of surfactants. 11. 𝜁-potential measurements in the photo-catalytic oxidation of surfactants in aqueous TiO2 dispersions. Langmuir, 9,1646–1650.

103 Pan, L., Zou, J.-J., Zhang, X., and Wang, L. (2011) Water-mediated promo-tion of dye sensitization of TiO2 under visible light. J. Am. Chem. Soc., 133,10000–10002.

References 279

104 Pan, L., Zou, J.-J., Liu, X.-Y., Liu, X.-J., Wang, S., Zhang, X., and Wang, L.(2012) Visible–light–induced photodegradation of rhodamine B over hierar-chical TiO2: effects of storage period and water-mediated adsorption switch.Ind. Eng. Chem. Res., 51, 12782–12786.

105 Wiberg, J., Marinado, T., Hagberg, D.P., Sun, L., Hagfeldt, A., andAlbinsson, B. (2009) Effect of anchoring group on electron injection andrecombination dynamics in organic dye-sensitized solar cells. J. Phys. Chem.C, 113, 3881–3886.

106 Konstantinou, I.K. and Albanis, T.A. (2004) TiO2-assisted photocatalyticdegradation of azo dyes in aqueous solution: kinetic and mechanistic investi-gations: a review. Appl. Catal. B, 49, 1–14.

107 Zhao, W., Chen, C., Ma, W., Zhao, J., Wang, D., Hidaka, H., andSerpone, N. (2003) Efficient photoinduced conversion of an azo dye onhexachloroplatinate(IV)-modified TiO2 surfaces under visible light irradia-tion – a photosensitization pathway. Chem. Eur. J., 9, 3292–3299.

108 Zhao, W., Chen, C., Li, X., Zhao, J., Hidaka, H., and Serpone, N. (2002)Photodegradation of sulforhodamine-B dye in platinized titania disper-sions under visible light irradiation: influence of platinum as a functionalco-catalyst. J. Phys. Chem. B, 106, 5022–5028.

109 Wu, Z., Xue, Y., Wang, H., Wu, Y., and Yu, H. (2014) ZnO nanorods/Pt andZnO nanorods/Ag heteronanostructure arrays with enhanced photocatalyticdegradation of dyes. RSC Adv., 4, 59009–59016.

110 Cho, S., Jang, J.-W., Hwang, S., Lee, J.S., and Kim, S. (2012) Self-assembledgold nanoparticle–mixed metal oxide nanocomposites for self-sensitized dyedegradation under visible light irradiation. Langmuir, 28, 17530–17536.

111 Liu, Y., Ohko, Y., Zhang, R., Yang, Y., and Zhang, Z. (2010) Degradationof malachite green on Pd/WO3 photocatalysts under simulated solar light.J. Hazard. Mater., 184, 386–391.

112 Sudeep, P.K., Takechi, K., and Kamat, P.V. (2007) Harvesting photons inthe infrared. Electron injection from excited tricarbocyanine dye (IR-125)into TiO2 and Ag@TiO2 core–shell nanoparticles. J. Phys. Chem. C, 111,488–494.

113 Aguirre, M.E., Rodríguez, H.B., San Román, E., Feldhoff, A., and Grela, M.A.(2011) Ag@ZnO core–shell nanoparticles formed by the timely reduc-tion of Ag+ ions and zinc acetate hydrolysis in N,N-dimethylformamide:mechanism of growth and photocatalytic properties. J. Phys. Chem. C, 115,24967–24974.

114 Zhao, Y., Li, C., Liu, X., Gu, F., Du, H.L., and Shi, L. (2008) Zn-dopedTiO2 nanoparticles with high photocatalytic activity synthesized byhydrogen–oxygen diffusion flame. Appl. Catal. B, 79, 208–215.

115 Stylidi, M., Kondarides, D.I., and Verykios, X.E. (2004) Visible light-inducedphotocatalytic degradation of acid Orange 7 in aqueous TiO2 suspensions.Appl. Catal. B, 47, 189–201.

116 Epling, G.A. and Lin, C. (2002) Photoassisted bleaching of dyes utilizingTiO2 and visible light. Chemosphere, 46, 561–570.

117 Hodak, J., Quinteros, C., Litter, M.I., and San Román, E. (1996) Sensi-tization of TiO2 with phthalocyanines. Part 1 – photo-oxidations using


hydroxoaluminium tricarboxymonoamidephthalocyanine adsorbed on TiO2.J. Chem. Soc. Faraday Trans., 92, 5081–5088.

118 Zhao, W., Sun, Y., and Castellano, F.N. (2008) Visible-light induced waterdetoxification catalyzed by PtII dye sensitized titania. J. Am. Chem. Soc., 130,12566–12567.

119 Cho, Y., Choi, W., Lee, C.-H., Hyeon, T., and Lee, H.-I. (2001) Visiblelight-induced degradation of carbon tetrachloride on dye-sensitized TiO2.Environ. Sci. Technol., 35, 966–970.

120 Kyung, H., Lee, J., and Choi, W. (2005) Simultaneous and synergistic con-version of dyes and heavy metal ions in aqueous TiO2 suspensions undervisible-light illumination. Environ. Sci. Technol., 39, 2376–2382.

121 Chatterjee, D. and Mahata, A. (2001) Demineralization of organic pollutantson the dye modified TiO2 semiconductor particulate system using visiblelight. Appl. Catal. B, 33, 119–125.

122 Chatterjee, D. and Mahata, A. (2002) Visible light induced photodegradationof organic pollutants on dye adsorbed TiO2 surface. J. Photochem. Photobiol.A, 153, 199–204.

123 Chatterjee, D. and Mahata, A. (2004) Evidence of superoxide radical forma-tion in the photodegradation of pesticide on the dye modified TiO2 surfaceusing visible light. J. Photochem. Photobiol. A, 165, 19–23.

124 Chatterjee, D., Dasgupta, S., and Rao, N.N. (2006) Visible light assisted pho-todegradation of halocarbons on the dye modified TiO2 surface using visiblelight. Sol. Energy Mater. Sol. Cells, 90, 1013–1020.

125 Meichtry, J.M., Rivera, V., Di Iorio, Y., Rodríguez, H.B., San Román, E.,Grela, M.A., and Litter, M.I. (2009) Photoreduction of Cr(VI) using hydrox-oaluminiumtricarboxymonoamide phthalocyanine adsorbed on TiO2.Photochem. Photobiol. Sci., 8, 604–612.

126 Iliev, V., Tomova, D., Bilyarska, L., Prahov, L., and Petrov, L. (2003) Phthalo-cyanine modified TiO2 or WO3-catalysts for photooxidation of sulfide andthiosulfate ions upon irradiation with visible light. J. Photochem. Photobiol.A, 159, 281–287.

127 Moon, J.Y., Yun, C.Y., Chung, K.W., Kang, M.-S., and Yi, J. (2003) Photocat-alytic activation of TiO2 under visible light using Acid Red 44. Catal. Today,87, 77–86.

128 Ghosh, J.P., Langford, C.H., and Achari, G. (2008) Characterizationof an LED based photoreactor to degrade 4-chlorophenol in an aque-ous medium using coumarin (C-343) sensitized TiO2. J. Phys. Chem. A,112, 10310–10314.

129 Zyoud, A., Zaatar, N., Saadeddin, I., Helal, M.H., Campet, G., Hakim, M.,Park, D., and Hilal, H.S. (2011) Alternative natural dyes in water purification:anthocyanin as TiO2-sensitizer in methyl orange photo-degradation. SolidState Sci., 13, 1268–1275.

130 Granados-Oliveros, G., Martínez Ortega, F., Páez-Mozo, E., Ferronato, C.,and Chovelon, J.-M. (2010) Photoactivity of metal-phenylporphyrinsadsorbed on TiO2 under visible light radiation: influence of central metal.Open Mater. Sci. J., 4, 15–22.

References 281

131 Boyer, S.M., Liu, J., Zhang, S., Ehrlich, M.I., McCarthy, D.L., Tong, L.,DeCoste, J.B., Bernier, W.E., and Jones, W.E. Jr. (2016) The role of ruthe-nium photosensitizers in the degradation of phenazopyridine with TiO2electrospun fibers. J. Photochem. Photobiol. A, 329, 46–53.

132 Góngora, J.F., Elizondo, P., and Hernández-Ramírez, A. (2017) Photocatalyticdegradation of ibuprofen using TiO2 sensitized by Ru(II) polyaza complexes.Photochem. Photobiol. Sci., 16, 31–37.

133 Sevim, A.M. (2017) Synthesis and characterization of Zn and Comonocarboxy-phthalocyanines and investigation of their photocatalyticefficiency as TiO2 composites. J. Organomet. Chem., 832, 18–26.

134 Park, Y., Lee, S.-H., Kang, S.O., and Choi, W. (2010) Organic dye-sensitizedTiO2 for the redox conversion of water pollutants under visible light. Chem.Commun., 46, 2477–2479.

135 Bae, E., Choi, W., Park, J., Shin, H.S., Kim, S.B., and Lee, J.S. (2004)Effects of surface anchoring groups (carboxylate vs phosphonate) inruthenium-complex-sensitized TiO2 on visible light reactivity in aqueoussuspensions. J. Phys. Chem. B, 108, 14093–14101.

136 Iwata, K., Takaya, T., Hamaguchi, H., Yamakata, A., Ishibashi, T., Onishi, H.,and Kuroda, H. (2004) Carrier dynamics in TiO2 and Pt/TiO2 powdersobserved by femtosecond time-resolved near-infrared spectroscopy at aspectral region of 0.9–1.5 μm with the direct absorption method. J. Phys.Chem. B, 108, 20233–20239.

137 Di Iorio, Y., San Román, E., Litter, M.I., and Grela, M.A. (2008) Photoin-duced reactivity of strongly coupled TiO2 ligands under visible irradiation:an examination of an alizarin red@TiO2 nanoparticulate system. J. Phys.Chem. C, 112, 16532–16538.




141 Zhang, X., Peng, T., and Song, S. (2016) Recent advances in dye-sensitizedsemiconductor systems for photocatalytic hydrogen production. J. Mater.Chem. A, 4, 2365–2402.

142 Breitenkamp, M., Henglein, A., and Lilie, J. (1976) Mechanism of thereduction of lead ions in aqueous solution (a pulse radiolysis study).Ber. Bunsen-Ges. Phys. Chem., 80, 973–979.

143 Lee, S.-H., Park, Y., Wee, K.-R., Son, H.-J., Cho, D.W., Pac, C., Choi, W., andKang, S.O. (2010) Significance of hydrophilic characters of organic dyes invisible-light hydrogen generation based on TiO2. Org. Lett., 12, 460–463.

144 Bi, Z.-C. and Ti Tien, H. (1984) Photoproduction of hydrogen bydye-sensitized systems. Int. J. Hydrogen Energy, 9, 717–722.

145 Shimidzu, T., Iyoda, T., and Koide, Y. (1985) An advancedvisible-light-induced water reduction with dye-sensitized semiconductorpowder catalyst. J. Am. Chem. Soc., 107, 35–41.


146 Bae, E. and Choi, W. (2006) Effect of the anchoring group (carboxylate vsphosphonate) in Ru-complex-sensitized TiO2 on hydrogen production undervisible light. J. Phys. Chem. B, 110, 14792–14799.

147 Park, J. and Choi, W. (2009) TiO2−Nafion photoelectrode hybridized withcarbon nanotubes for sensitized photochemical activity. J. Phys. Chem. C,113, 20974–20979.

148 Dhanalakshmi, K.B., Latha, S., Anandan, S., and Maruthamuthu, P. (2001)Dye sensitized hydrogen evolution from water. Int. J. Hydrogen Energy,26, 669–674.

149 Furlong, D.N., Wells, D., and Sasse, W.H.F. (1986) Colloidal semiconduc-tors in systems for the sacrificial photolysis of water: sensitization oftitanium dioxide by adsorption of ruthenium complexes. J. Phys. Chem.,90, 1107–1115.

150 Hirano, K., Suzuki, E., Ishikawa, A., Moroi, T., Shiroishi, H., andKaneko, M.J. (2000) Sensitization of TiO2 particles by dyes to achieve H2evolution by visible light. Photochem. Photobiol. A, 136, 157–161.

151 Malinka, E.A., Kamalov, G.L., Vodzinskii, S.V., Melnik, V.I., and Zhilina, Z.I.(1995) Hydrogen production from water by visible light using zincporphyrin-sensitized platinized titanium dioxide. J. Photochem. Photobiol.A, 90, 153–158.

152 Abe, R., Hara, K., Sayama, K., Domen, K., and Arakawa, H. (2000) Steadyhydrogen evolution from water on eosin Y-fixed TiO2 photocatalyst using asilane-coupling reagent under visible light irradiation. J. Photochem. Photo-biol. A, 137, 63–69.

153 Han, W.-S., Wee, K.-R., Kim, H.-Y., Pac, C., Nabetani, Y., Yamamoto, D.,Shimada, T., Inoue, H., Choi, H., Cho, K., and Kang, S.O. (2012)Hydrophilicity control of visible-light hydrogen evolution and dynamicsof the charge-separated state in dye/TiO2/Pt hybrid systems. Chem. Eur. J.,18, 15368–15381.

154 Lee, J., Kwak, J., Ko, K.C., Park, J.H., Ko, J.H., Park, N., Kim, E., Ryu, D.H.,Ahn, T.K., Lee, J.Y., and Son, S.U. (2012) Phenothiazine-based organic dyeswith two anchoring groups on TiO2 for highly efficient visible light-inducedwater splitting. Chem. Commun., 48, 11431–11433.

155 Watanabe, M., Hagiwara, H., Iribe, A., Ogata, Y., Shiomi, K., Staykov, A.,Ida, S., Tanaka, K., and Ishihara, T. (2014) Spacer effects in metal-freeorganic dyes for visible-light-driven dye-sensitized photocatalytic hydrogenproduction. J. Mater. Chem. A, 2, 12952–12961.

156 Manfredi, N., Cecconi, B., Calabrese, V., Minotti, A., Peri, F., Ruffo, R.,Monai, M., Romero-Ocaña, I., Montini, T., Fornasiero, P., and Abbotto,A. (2016) Dye-sensitized photocatalytic hydrogen production: distinctactivity in a glucose derivative of a phenothiazine dye. Chem. Commun.,52, 6977–6980.

157 Kim, W., Tachikawa, T., Majima, T., Li, C., Kim, H.-J., and Choi, W. (2010)Tin-porphyrin sensitized TiO2 for the production of H2 under visible light.Energy Environ. Sci., 3, 1789–1795.

283

11

Advances in the Development of Novel Photocatalystsfor DetoxificationCiara Byrne1,2, Michael Nolan3, Swagata Banerjee4, Honey John5,Sheethu Jose6, Pradeepan Periyat6,7, and Suresh C. Pillai1,2

1Institute of Technology Sligo, Department of Environmental Science, Nanotechnology and Bio-EngineeringResearch Group, Ash Lane, Sligo, Ireland2Institute of Technology, Department of Environmental Science, Centre for Precision Engineering, Materialsand Manufacturing Research (PEM), Ash Lane, Sligo, Ireland3University College Cork, Tyndall Theory Modelling & Design Centre, Tyndall National Institute, Lee Maltings,Cork, Ireland4University College of Science; Osmania University, Department of Biochemistry, Amberpet, Hyderabad500007, India5Cochin University of Science and Technology, Department of Polymer Science and Rubber Technology, AlfiyaNagar, Kochi 682022, Kerala, India6Central University of Kerala, Department of Chemistry, Tejaswini Hills, Periye Post 671314, Kerala, India7University of Calicut, Department of Chemistry, Trishur-Calicut Road, Thenhipalam 673635, Kerala, India

11.1 Introduction

Since the discovery of photocatalysis technology by Fujishima and Honda, therehas been a significant increase in research into this methodology for variousfunctional applications [1–3]. The reaction is usually initiated by a photocatalystsuch as TiO2 or ZnO being bombarded with photons from natural or artificiallight of appropriate wavelength depending on the bandgap of the semiconductor(Figure 11.1) [3, 5, 6]. The irradiation causes the electrons to move from thevalence band (e−CB ) to the conduction band by creating a hole in the valenceband (h+

VB) (Eq. (11.1) and Figure 11.1) [6, 7].TiO2 + hv → h+

VB + e−CB (11.1)The excited electrons (e−CB) can react with atmospheric oxygen (O2) to form

superoxide radicals (∙O2−) or hydroperoxide radicals (∙HO2) Eq. (11.2) [4, 7].

These reactive oxygen species will take part in the degradation of organicpollutants into water (H2O) and carbon dioxide (CO2) as given in Figure 11.1.In a similar manner, the ∙OH reacts with pollutants to form H2O and CO2 orsimilar smaller molecules as the end products.

e−CB + O2→∙O2

− (11.2)h+

VB + H2O→∙OH + H+ (11.3)Over the past two decades, titanium dioxide (TiO2) has been widely applied in

heterogeneous photocatalysis including detoxification of air and water, hydrogen


284 11 Advances in the Development of Novel Photocatalysts for Detoxification

TiO2 + hv → h+VB + e–

CB

•O2– + Pollutant → → → H2O + CO2

Valance band

Conduction bande–

CB

h+VB

•OH + Pollutant → → → H2O + CO2

O2 + e–CB → •O2

–

H2O + h+VB → •OH + H+

Ba

nd

ga

p E

ε

Reduction

Oxidation

Figure 11.1 The mechanism of photocatalysis. (Reproduced with permission from Ref. [4].Copyright 2015, Elsevier.)

production owing to its high stability, nontoxicity, and relative abundance [8].The photocatalytic detoxification of organic pollutants using TiO2 is the mostextensively studied method of overcoming the problem of water contaminationby organic pollutants. The doping of metal ions in TiO2 can significantly enhanceits photocatalysis efficiency for detoxification. The introduction of dopants intothe TiO2 matrix can be able to overcome major limitations of pure TiO2, suchas wide bandgap (3.2 eV for crystalline anatase phase) and high photogeneratedelectron–hole pair recombination rate thereby to improve the efficiency of pho-tocatalytic detoxification of TiO2 [7].

Most semiconductors with suitable band structures for water splitting are widebandgap materials that absorb in the UV part of the solar spectrum, which isunfortunately<5% of the solar spectrum. For large-scale renewable energy appli-cations, it is desirable to absorb in the visible-light region as this makes up approx-imately 42% of the solar spectrum. At the same time, for indoor applications inhospital settings, UV photocatalysis can be effective in bacterial removal andit would also be of benefit to not sacrifice the UV activity of a photocatalyst inachieving visible-light absorption.

Photocatalytic detoxification has been examined and studied as a methodfor water treatment since the late 1970s[9, 10]. In the decades since, photocat-alytic detoxification has also been used for industrial applications [9, 11]. Thischapter will discuss in extensive detail the theoretical studies of photocatalysis(Section 11.2), metal-doped photocatalysts for detoxification (Section 11.3),graphene–TiO2 composites for detoxification (Section 11.4), and commercialapplications of photocatalysis in environmental detoxification (Section 11.5).

11.2 Theoretical Studies of Photocatalysis 285

11.2 Theoretical Studies of Photocatalysis

While TiO2 is the most studied example of a photocatalytic material, with otherexamples including Fe2O3, ZnO, BiVO4, GaN, and GaP, its bandgap of 3.2 eVmeans that it will only absorb in the UV, which means that the band gap require-ment cannot be met [12]. Thus, there has been a significant level of activity indeveloping approaches to narrow the bandgap and allow visible-light absorption[6, 12–19]. We review this topic of bandgap modification in Section 11.2.1. Inaddition, when we consider the requirements related to the oxidation process, thefocus has been on aligning the valence and conduction band edges (CBEs), pri-marily to water oxidation, but also to the hydrogen reduction potentials, Section11.2.2 [20–23]. This means that the fate of the photogenerated electrons andholes, which is clearly a key contributor to efficiency [24–27], has rarely beenstudied using simulations and we review this in Section 11.2.3. Finally, whilewater adsorption, activation, and oxidation on TiO2 surfaces are also important[28–44], these are not the focus of this chapter and are reviewed in other chaptersof this volume.

11.2.1 Doping and Surface Modification of TiO2 for BandgapEngineering

The solutions proposed to achieve the optimal bandgap for visible-light absorp-tion can fall into one of the following:

A) Doping with different metals on the Ti siteB) Doping with nonmetals on the O-siteC) Metal/nonmetal codoping on Ti and O sitesD) Quantum dot modificationE) Composites and surface modified TiO2.

Metal doping is a well-studied and, in principle, useful strategy to activateTiO2 by inducing visible-light absorption. Since 2001, the substitutional dopingof nonmetal species on the oxygen site in TiO2 has been studied, with C, N, andS being dominating [6, 12–19]. The classic work on this topic is the Science paperfrom Asahi et al. [45] in which N-doped TiO2 was shown to be visible-lightactive. The change to visible-light absorption in TiO2 arises because the intro-duction of the metal or nonmetal dopants onto Ti sites in TiO2 results in theformation of new electronic states in the previous TiO2 valence-to-conductionband energy gap. In the undoped bulk, this is the fundamental bandgap, butin a doped material it is more accurate to speak of an energy gap, as the TiO2bandgap persists (see e.g., [22]).

Figure 11.2 highlights the effect of doping of TiO2 with metallic and nonmetal-lic dopants from a selection of examples in the extensive literature on doping ofTiO2. In Figure 11.2a, we can see how doping of TiO2 with Fe (which enters asFe3+, left panel) can lead to formation of a new empty state below the CBE ofTiO2 [46]. This would have the effect of reducing the energy gap by lowering theenergy of the empty CB states. Also shown in Figure 11.2a is how the presence ofnonmetal B and N dopants on the anion site produces dopant levels that lie

CB

e

h hh

h h h h h h h h h

eV

0

+1

CB

(A) (B)

(C) (D)

VB

Dopant

level

Ebg = 3.20 eV

2.05 eV2.05 eV

2EF+•

F+2.55 eV

2.90: 2.55 ev2T2

2.90 eV

+2

+3

hVB VB VB

h h h h h h h hh h

Shallow trapFe

(III)/Fe

(II)Fe

(II)/Fe

(I)e e

e e e e e e e e eCB

(a)

Visible light

Fe-TiO2 B or N-TiO2 (B,N)-TiO2(Fe,N)-TiO2 (Fe,N,B)-TiO2

Dopant le

vels

(b) (c)

(a) (b) (c) (d) (e)

Visible light Visible light

CB CB

Substitutional

B C N FO

CB

VB

3.9

2.18 1.39 0.13

0.82

1.19

Ti3+

F 2p6

3d1

N 2p5C 2p

4B 2p3

VB

1.7

eV

0.0

6 e

V

EF

Fe3+

Oxygen vacancies

Fe/TiO2


Figure 11.2 Examples of results on formation of new electronic states in metal- andnonmetal-doped TiO2 from DFT simulations. (a) Formation of in-gap electronic states in Fe-and N-doped TiO2. (Reproduced with permission from Ref. [46]. Copyright 2010, The RoyalSociety of Chemistry.) (b) Formation of in-gap electronic states in nonmetal-doped anataseTiO2 and location of excess spins arising from anion oxidation states. (Reproduced withpermission from Ref. [47]. Copyright 2013, Elsevier.) (c) Schematic of positions of Fe3+ and Ti3+

electronic states in Fe-doped TiO2. (Reproduced with permission from Ref. [48]. Copyright2011, The Royal Society of Chemistry.) (d) Schematic of the positions of in-gap electronicstates in modified TiO2. (Reproduced with permission from Ref. [49]. Copyright 2009, The RoyalSociety of Chemistry.)

above the VB edge of TiO2 and which then reduce the energy gap between thehighest occupied electronic states and the lowest CB states.

Figure 11.2b summarizes findings from the work of Di Valentin et al. onnonmetal-doped TiO2 [36], showing the position of the dopant and the local-ization of spin on the dopant sites as a result of the imbalance in the dopantand oxygen oxidation states, O2− versus B3−, N3−, and C4−. The position of thedopant-derived levels inside the TiO2 bandgap depends on the identity of thedopant and the dopant levels shift closer to the VB edge. As we go across thisrow from B to N to C to O to F; F doping introduces excess electrons into thesystem, which reduces a Ti atom to Ti3+. Formation of Ti3+ then results in theappearance of a defect state in the energy gap. In principle, this permits lowerenergy electronic transitions from the occupied state to the conduction band orfrom the valence band to the defect state. Figures 11.2c and d present schematicfigures of the positions of dopant and defect derived electronic states within theTiO2 bandgap [49, 50].

While formation of Ti3+ has traditionally been studied using density functionaltheory (DFT) modeling from the perspective of defect chemistry, more recentwork has shown that the production of these species may be beneficial forphotocatalysis. Therefore, there has been a significant amount of researchfocused on the production of Ti3+ states [50–58] which can be introduced bythermal treatments without or with hydrogen, particle bombardment or undertypical reaction conditions. The extended absorption range of defective TiO2arises from formation of Ti3+ states that lie below the CBE into which valenceband electrons can be excited with visible light. The high concentration of Ti3+

results in the formation of a continuum of electronic states below the conductionband, rather than localized states which are detrimental to the activity of TiO2.

An interesting material is TiO2 with a high degree of hydrogenation [54, 55],which is also known as black TiO2. The introduction of H2 creates the Ti3+ statesthrough incorporation of H as a proton and transfer of an electron to the TiO2.Figure 11.3 shows some results from the literature on these interesting systems.Figure 11.3a indicates that introduction of hydrogen modifies the TiO2 and theelectron paramagnetic resonance (EPR) data shows the types of defects that arepresent [54]. Figure 11.3b compares light absorption in unmodified TiO2 (indi-cated as TiO2) and after hydrogen incorporation (indicated as TiO2−xHx) with thesolar spectrum. It is clear from this result that the hydrogenated TiO2 is able toabsorb over the entire solar spectrum, whereas the unmodified TiO2 only absorbsin the short wavelength region [55]. The insets show the color change on going

0.04

1.0

0.8

0.6

0.4

0.2

0.0400

4.2 eV

1.74 eV

1.06 eV

TiO2

HP-TiO2

Solar spectrum

800 1200

TiO2–xHx

TiO2–xHx

TiO2

Wavelength (nm)

Ultravioletlight

Holes oxidation

O2 reductionCB

VB

Vo-Ti3+

e–

e–

h+

h+ h+

e–

e–

Ab

so

rba

nce

(a

.u.)

1600 2000

0.03

0.02

0.01

0

0

(a) (b)

(c) (d) (e)

(i)

(iii) (iv)

(ii)

DO

S

DO

S o

f vib

ratio

na

l m

od

es

–5 0

Model A

Dynamics of H atoms

in model B

Model B

5

Energy (eV) Frequency (cm–1)

10 0 1000

LUMOs

3d-Ti

HOMOs

2p-O

Ti3+ levels

2000 3000 4000

5

Hydrogenation time (h)

Ph

oto

activity (

min

–1)

10

Ti3+ Ti3+

Ti3+

O–

O–

O–

15


Figure 11.3 Examples of results on TiO2 with a high degree of hydrogen incorporation.(a) Schematic of the effect of hydrogenation on the optical and EPR characteristics of TiO2.(Reproduced with permission from Ref. [54]. Copyright 2013, The American Chemical Society.)(b) Absorption spectrum of TiO2 and hydrogenated TiO2 overlain on the solar spectrum,together with images of TiO2 and hydrogenation TiO2 to highlight the color of each sample.(Reproduced with permission from Ref. [55]. Copyright 2014, The Royal Society of Chemistry.)(c) Atomic structure of lightly and heavily hydrogenated TiO2 particle together with the densityof states. (Reproduced with permission from Ref. [56]. Copyright 2013, Nature PublishingGroup.) (d) Schematic of the positions of in-gap electronic states with a large number of Ti3+

species present. (Reproduced with permission from Ref. [57]. Copyright 2012, The AmericanChemical Society.) (e) Schematic of the energies of Ti3+ species in the TiO2 bandgap.(Reproduced with permission from Ref. [58]. Copyright 2015, The Royal Society of Chemistry.)

from TiO2 to TiO2−xHx, showing the black color of the latter. DFT studies of thissystem are more scarce, but Figure 11.3c presents the results of the DFT levelsimulation of a TiO2 nanocluster, which is perfect and which is saturated withhydrogen atoms (the small white spheres) [56]. The important results are the(i) hydrogen disrupts the TiO2 structure and (ii) the electronic density of statesshows clear changes to both the valence and conduction band regions with a nar-rowing of the energy gap compared to unmodified TiO2. Finally, Figures 11.3dand e show schematic cartoons of the effect of the introduction of Ti3+ sites onthe energy bands – a spread of Ti3+ induced energy states is present betweenthe valence and conduction bands which facilitates lower energy electronic exci-tations compared to unmodified (oxidized) TiO2 [57, 58]. Finally, codoping ofTiO2 with cations and anions is a more recent approach and is inspired by over-coming the problem that if a metal or nonmetal species has a different oxidationstate to the substituted species then this results in formation of charge trappingstates [6, 13, 20, 49, 59–64]; Figure 11.4a shows a schematic of different codop-ing arrangements, namely cation–cation, cation–anion, and anion–anion [20].If, for example, the pair of dopants is introduced such that the oxidation statesof the cations or cation–anion are chosen so that they are balanced, relative toTi4+, then the valence or CBE can be shifted, but there are no charge trappingstates present, thus shifting the bandgap while reducing charge recombination[60]. Considering the Ti and O oxidation states of +4 and −2, then Mo (+6) andC (−4) or Ta/Nb (+5) and N (−3) are examples of cation–anion pairs that are cor-rectly charge balanced and will not produce charge trapping states [60–65]. It hasbeen demonstrated that codoping of TiO2 with P/N, N/W, N/H, Cr/Sb [64–67]results in better visible-light photocatalytic activities when compared to singledopants. Modeling studies allow detailed studies of codoping of TiO2.

Figure 11.4b shows the computed density of states projected onto electronicstates of the cations (V, Nb, Sc, Mo, Ga) and anions (N, C, F) for differentcation–anion (V/N and Nb/N on Ti/O sites), cation–cation (V/Sc and V/Ga onTi/Ti sites), and anion–anion (F/N on O/O sites) from DFT calculations [20].The presence of the dopants can leads to new states above the valence band edgeand in the original TiO2 bandgap, for example, Sc/C or Mo/C. Alternatively, theCBE can be lowered relative to undoped TiO2 in cation–cation doping scenarios.

Figure 11.4c shows a similar set of DFT level results for charge balanced codop-ing combinations, where of note are the Nb/N and Mo/C combinations, which


Cation–cation pair

(i)(a)

(b)

(c)

100

DO

S (

a.u

.)

50

0

50

50

50

50

50

50

50(i)

(ii)

(iii)

(iv) (Mo+C)

(Cr+C)

(Nb+N)

(V+N)

25

25

25

25

50

50

50

0–6 –4 –2 0

Energy (eV)

DO

S

2 4 6

0

0

0

0

0

0–6 –4 –2

Energy (eV)

VTi

No (×5)

NbTi (×5)

No (×5)

Co (×3)

Co (×3)

ScTi

MoTi (×1.5)

VTi

GaTi

Fo (×3)

No

VTi

SeTi

20

(ii)

(i)

(ii)

(iii)

(iv)

(v)

(vi)

(vii)

(iii) (iv) (v) (vi) (vii)

Cation–anion pair Anion–anion pair

Figure 11.4 (a) Schematic of different cation, anion codoping scenarios in bulk anatase TiO2.(Reproduced with permission from Ref. [20]. Copyright 2017, Elsevier.) (b) Computedelectronic density of states for different cation and anion codoping combinations in bulkanatase TiO2. (Reproduced with permission from Ref. [20]. Copyright 2017, Elsevier.)(c) Computed electronic density of states for V/N, Nb/N, Cr/C and Mo/C codoped TiO2.(Reproduced with permission from Ref. [59]. Copyright 2009, The American Physical Society.)


push the valence band edge to higher energy but do not result in any in-gap defectstates [59]. This would then redshift light absorption compared to undoped TiO2.Indeed, experimental results for Mo/C [62, 63] and Nb/N [68] codoping of TiO2published subsequently confirm the predictions from simulations.

In conclusion, doping of TiO2 with metal cations can be a useful way to improvethe photocatalytic activity [69–87]. A large number of metal ions, in particularthe main group metals, transition metal, and lanthanide metal ions are studiedfor enhanced the photocatalytic activity of anatase TiO2. The high crystallinity ofanatase phase, suitable surface properties such as high surface area, mesoporos-ity and pore volume are the important factors that contribute to photocatalyticactivity of anatase TiO2.

11.2.2 Alignment of Valence and Conduction Band Edges with WaterOxidation and Reduction Potentials

While understanding the effect of modifications on TiO2 on its light absorptionproperties and on the nature of the valance and CBEs is obviously important andhas dominated DFT studies of modified TiO2 to date, there are other factors thatneed to be explored. It may be the case that one can propose a materials composi-tion that will have visible-light absorption, but the alignments of the valence andconduction bands relative to the water oxidation and reduction potentials maynot be favorable. Relative to the standard hydrogen electrode (SHE), Figure 11.5,the energy of the valence band states must be more positive than the water oxi-dation potential (or in terms of electronic energies, lie at a more negative energy)so that holes have sufficient oxidizing power. The energy of the conduction bandstates must lie more negative than the reduction potential, be it for O2 or water(Figure 11.5a,b) [20, 21].

In Figure 11.5b–d, we summarize the results from a selection of studies ofdoped TiO2 in which the valence and CBEs of doped TiO2 are aligned withthe oxidation and reduction potentials of water [20–23]. In Figure 11.5b this isshown for the examples of different anion–anion and cation–anion codopingsystems [21]. In all cases, the doping shifts the valence band edge to less positivepotentials (relative to SHE) or higher energy. With the exception of N/P codop-ing, these energies still lie below the oxidation potential of water so that theoxidation reaction would still be possible. All doping scenarios position the CBEat more negative potentials than the reduction potential so that this reaction isalso possible. Thus, one can conclude that C/S, C/Se, and Mo/C codoping wouldinduce visible-light absorption and permit the water oxidation reaction, but N/Pcodoping would not.

In Figure 11.5c, a similar energy band alignment diagram is shown for a rangeof single dopants and codopants in an anatase TiO2 nanowire system [22]. TheEi values are the intrinsic anatase nanowire bandgap, which is little modified bydoping. We can see that simple doping does not always give electronic statesaligned suitably with the water oxidation and reduction potentials. However,with codoping, C/Cr and C/V can shift light absorption toward the visible, whilehaving suitable band alignments, making these reasonable candidates for wateroxidation under visible-light absorption.


(–0.52)

(+2.53)

0.24 eV

–1

0

Energy relative to SHE

0.0

NHE (eV)Pure

1.62

Eg =1.58

1.14

2.27

eV

eV3.22

eV

eV

eV

C,S C,SeN,P Mo,C

H2/H2O

H2O/O2

H+/H2

H2O/O2

0.5

1.0

1.5

2.0

2.5

3.0

–1

0

E (

eV

) vs N

HE

(V

)

1

2

3

4

3.07

TiO2 F@TiO2Rh@TiO2

(Rh+F)@TiO2

–0.30.05

–0.28–0.37

3.16

2.44

3.05

3.37 3.11 2.05 2.31

0.4 eV

0.67

(–0.28)

+1

O2/O

2

•–

(+2.27)+OH/H

2O

ANW

(a)

(c)

(b)

(d)

En

erg

y (

eV

)

–4–3.81 –3.70 –3.76 –3.76 –3.71 –3.75 –3.79

–4.29

–5.07

–5.54

–6.34–6.62

–7.06

Eg (eV)

Ei (eV)

3.72

3.72

1.27

3.72 3.73 3.81 3.78 3.71 3.61 3.65 3.67

0.87 1.14 2.61 2.49 2.85 2.49 2.00

–6.59–6.94

–4.94

–3.96 –3.92 H+/H2

–4.5 eV

–5.73 eV

O2/H2O

–7.53 –7.42 –7.49 –7.57 –7.49 –7.46 –7.40–7.61 –7.59

–6

–8

C N V Cr C&Cr N&V C&V N&Cr

+2

+3

0.26 eV

TiO2 2.73 eV

eCB–

hVB+


Figure 11.5 (a) Schematic of alignment of TiO2 energy levels with water oxidation andreduction potentials. (Reproduced with permission from Ref. [20]. Copyright 2017, Elsevier.)(b) Alignments of different doped TiO2 energy levels with the water oxidation and reductionpotentials. (Reproduced with permission from Ref. [21]. Copyright 2010, The American PhysicalSociety.) (c) Alignments of different doped TiO2 energy levels with the water oxidation andreduction potentials. (Reproduced with permission from Ref. [22]. Copyright 2013, The RoyalSociety of Chemistry.) (d) Energy levels (valence and conduction bands and the dopantderived states) in F- and Rh-doped and codoped TiO2 aligned with the water oxidation andreduction potentials. (Reproduced with permission from Ref. [23]. Copyright 2014, TheAmerican Chemical Society.)

Finally, Figure 11.5d shows band alignments for F and Rh single-doped TiO2as well as the Rh/F codoped system [23]. F-doping shifts the CBE to below thereduction potential so this is not favorable and Rh doping introduces a new statein the energy gap of TiO2. This state could act as a recombination center, thuskilling the efficiency of the system. The combination of Rh/F doping produces anew state lying above the TiO2 VB edge, shifting the energy gap to 2.31 eV (inthe visible region) and maintaining the required band alignments with the wateroxidation and reduction potentials.

There are obviously many studies of doping of TiO2, which have investigatedthe electronic properties and focus on the modifications to the valence andconduction bands, introduction of new electronic states and the alignmentswith the necessary potentials for different reactions. From these studies, it ispossible to predict which dopants in TiO2 (and other materials such as Fe2O3or WO3) would give the required properties. The actual realization of these newcompositions is beyond the scope of DFT simulations and requires experimentalwork to prepare and characterize so that the DFT results may be confronted.This gives good guidance as to the validity of the DFT approach in predictingnew photocatalyst compositions.

11.2.3 Electron and Hole Localization

Compared to the volume of the literature on doping of TiO2, the investigationof the fate of the electron and hole after photoexcitation, which is a further cru-cial aspect in photocatalysis, is much less prominent. It is of course difficult tomodel the excited states of solids and surface routinely, especially when comparedto molecules, and, therefore, some different approaches are used. A well-knownapproach to introduce an electron into the system, which then populates an orig-inally empty conduction band state, is to increase the charge by one electron in abulk system; similarly, removing one electron creates a hole in the valence band.In a surface, which is clearly of more relevance for the mechanism of oxidation orreduction, this is not possible within standard periodic boundary conditions andto model addition of an electron, a H atom can be added (producing an —OHspecies), which then transfers its electron to TiO2. To model a hole, one can havea surface with water present and remove a neutral hydrogen atom (which is anelectron and a proton), leaving behind a hole. A final approach to model pho-toexcited systems is to impose a triplet state on the system – this has the effect ofpromoting an electron to the previously empty conduction band states (obeyingthe Pauli exclusion principle) and leaving behind a valence band hole. Figure 11.6


3.5

e–tr

e–

(a)

(b)

(c)

h+

e–

h+

Ti5c3+

h+ pairTi3+ O–tr

O2c–

3

2.5

2

1.5

0.5

0

0 0.5 1

i

(i) (ii) (iii) (iv)

H

Ti Ob

Oa

[100] [100]

[010]

[001]

[010]

0.10 Å

0.10 Å 0.05 Å

5

6

4

3

2

1

0.10 Å

0.10 Å0.05 Å

[001]

ii iii iv

1.5 2 2.5

Time (ps)

Spin

popula

tion

Dis

tance (

Å)

3 3.5 4 4.5

dOa–ObdOa–Ti

dOa–H

OaOb

5

1

1


Figure 11.6 (a) Computed spin density and electronic density of states for (top panel) excesselectron (middle panel) excess hole and (bottom panel) electron–hole pair in anatase (101);the yellow isosurfaces show the location of the spins. (Reproduced with permission from Ref.[24]. Copyright 2011, The American Chemical Society.) (b) Location of hole state in rutile (110)immersed in water from AIMD simulation at 330 K. The green isosurfaces show the location ofthe hole on oxygen. The graph shows the spin population on different oxygen atoms duringthe first 5 ps of the simulation. (Reproduced with permission from Ref. [25]. Copyright 2014,The American Chemical Society.) (c) Electron localization on different Ti sites (blue spinisosurfaces) of a rutile (110) slab. The left image shows localization in the center of the slab,while the right image shows localization on a surface Ti atom (indicated as Ti 6). The insetshows the local geometry around the electron. (Reproduced with permission from Ref. [26].Copyright 2015, The American Chemical Society.)

shows some representative examples of calculations using these approaches tomodel electrons and holes in different TiO2 systems [24–26].

Di Valentin and Selloni used the triplet electronic state in modeling electronand hole localization in anatase (101) and within the CRYSTAL code they couldalso add or remove an electron [24]. Figure 11.6a (from the top to bottom panel)shows the location of electrons and holes for the following studies: (i) additionof an electron, (ii) addition of a hole, and (iii) an excited electron and hole. In allcases, using hybrid DFT, the electron and hole localize, preferentially on one Tior oxygen site, respectively. This gives Ti3+ and O− species in anatase and theelectronic density of states shows localized electron states consistent with theformation of these species, which are important in photocatalysis. Subsequentwork from Di Valentin showed that this model can capture the key processes inoxidation of molecules such as methanol on TiO2 surfaces [88]. We have also usedthis simple model in characterizing electron and hole localization in TiO2 rutileand anatase surfaces modified with nanoscale metal oxide nanoclusters [27].

Figure 11.6b shows some results from an ab initio molecular dynamics(AIMD, within hybrid DFT) simulation of a water layer on rutile (110), whichfocused on the dynamics of the hole (formed by removing a proton andan electron) at a temperature of 330 K [25]. This shows the atomic struc-ture of the water–TiO2 system (top left panel) and the spin on particularoxygen atoms (cyan, purple curves) as the simulations proceeds over a 5 pstimeframe. These results show that the hole localizes on one of two oxygenatoms in the first 3 ps. These are the terminal hydroxyl (Oa) and a nearbysurface oxygen atom (Ob), the fluctuating hole localization correlates withchanges in geometry around the oxygen atoms involved. After 3ps, there is abreaking of the OH bond and transfer of the proton to the surface. This leavesbehind an O− species on the surface. These results suggest this process will be fast.

From the perspective of understanding electron trapping in TiO2 nanocrystals,which can expose different TiO2 surfaces, Wallace and McKenna [26] studiedelectron trapping at low-index rutile surfaces, those being (110), (100), (101),(001), and (111) using DFT+U calculations. Figure 11.6c shows an example ofelectron trapping at the rutile (100) surface; the Ti localization sites are numbered1–6, with site 1 in the center of the slab and number 6 at the surface. The electronlocalization results in structural distortions around the localization site, whichare characterized by elongations of the Ti—O distances around the reduced Ti3+


site. When the electron trapping energies are evaluated, the most stable site in thissurface is in the bulk (site 1 in Figure 11.6c), with the surface site (site 6) being theleast stable. When the other surfaces are considered, these authors find that thesurfaces with the most favorable nonbulk trapping sites for electrons are (110)and (001), in which (110) traps an electron in a subsurface site. Finally, crystalmorphologies exposing predominantly (110) and (001) were proposed to enhanceelectron trapping at nonbulk sites.

11.3 Metal-Doped Photocatalysts for Detoxification

11.3.1 High-Temperature Stable Anatase TiO2 Photocatalyst

Among two common types of TiO2 (anatase and rutile), anatase is meta-stableand more photoactive than rutile [89, 90]. Control of the phase structureof anatase TiO2 at high temperature is very important because the perfor-mance of this material for definite application depends on the phase structure.The higher adsorption affinity toward organic compounds, along with thelower electron–hole recombination rate of anatase phase makes it a superiorphotocatalyst [91, 92]. Photocatalytically active stable TiO2 coatings for detox-ification, self-cleaning and hygienic applications (e.g., bathroom tiles, sanitarywares, and self-cleaning glass) for the control of organic contaminants requirehigh-temperature stability of anatase TiO2 [93–96] .

11.3.2 Main Group Metal Ions on Anatase Stability and PhotocatalyticActivity

Various main group metal ions were studied as dopants to increase the anatasephase stability up to a temperature of 1100 ∘C [97–105]. Among the various maingroup metal ion dopants, silica and alumina are the best candidates to increasethe anatase phase stability and photocatalytic performance of TiO2. For example,Anderson et al. [102] reported that silica-doped TiO2 improved the anatasestability up to 900 ∘C, and this doped TiO2 is a more efficient photocatalystfor decomposition of Rhodamine 6G, compared to pure TiO2. The enhancedphotoactivity of TiO2 in the presence of alumina or silica was explained basedon their high-temperature stable anatase structure, adsorbent property, whichincreases the concentration of the degradant near the anatase TiO2 sites relativeto the solution concentration. Warrier et al. have studied the independent dopingof silica or alumina to increase the anatase stability up to 1100 ∘C [99, 106, 107].

11.3.3 Effect of Transition Metals on Anatase Stabilityand Photocatalytic Activity

A large number of transition metal ions (Cr, Cu, Ta, Nb, Zn, Pt, Ni, Fe, Ag) [69,74, 76, 82, 108–113] are also been studied and reports showed that transitionmetal ion doping increases the anatase to rutile transformation temperatureand photocatalytic activity of TiO2 in the visible region [79–85]. The transition

11.3 Metal-Doped Photocatalysts for Detoxification 297

Undoped TiO2 (anatase)

UV3.2 eV

h+

h+

e–e–

Doping ofmetal ions

CB CB

VB

Visible

Figure 11.7 Representation of narrowing the bandgap of anatase TiO2 by doping ofmetal ions.

metal ion doping in the TiO2 matrix narrow down the bandgap of TiO2 and,consequently, a redshift in the absorption edge from UV to visible region due tothe charge transfer transition from transition metal d electrons to CB/VB of TiO2[77] as shown in Figure 11.7. Hidalgo et al. [78] reported the photodepositionof gold on TiO2 samples at a lower light intensity of 120 min with 0.5–1% of goldis enough for the phenol photodegradation. The similar result was obtained forAu—TiO2 samples synthesized by photodeposition at 60 min for catalyst with a2% nominal content of gold.

Maicu et al. [79] carried out a comparative study of the photo deposition of Pt,Au, and Pd under the same experimental conditions onto presulfated and non-sulfated TiO2 and found that the photoactivity of the catalysts was in the orderof Pt>Pd>Au. Sulfation and metallization of samples were created to producea synergistic enhancement in photoactivity for the degradation of phenol.

11.3.4 Effect of Rare Earth Metal Ions on Anatase Stabilityand Photocatalytic Activity

Lanthanides are another class of metal ions, which are extensively used as adopant to increase the anatase TiO2 stability. Among this Gd3+ ion doping inTiO2 may be the important one because gadolinium in the +3 valence state hasno free electron in its outer shell. This gives an extra stability due to the half-filledf-orbital. Hishita et al.[80] and Zhao et al.[81] reported the effect of Gd3+ ionin anatase-to-rutile phase transformation. The synthesis of Gd3+-doped TiO2using titanium alkoxide method improved anatase stability up to 800 ∘C. Zhanget al. [82] also explained the higher anatase phase stability and photocatalyticactivity of Gd3+-doped TiO2 using titanium alkoxide and gadolinium nitrate asprecursor. Baiju et al. [83] reported an aqueous sol–gel method to synthesizea high-temperature stable mesoporous Gd3+- and La3+-doped anatase TiO2,which is stable above 800 ∘C without using any surfactant. This systematic studyusing different mol% of Gd3+ ion doping in the TiO2 matrix explains the effectof doping on the anatase-to-rutile phase transformation, along with texturalproperties and higher photocatalytic activity of Gd3+-doped anatase TiO2.Gadolinium (Gd)-doped TiO2 prepared by El-Bahy et al. [84], a doped TiO2with high surface area, large pore volume, small particle size, and small bandgap,exhibited the highest photocatalytic activity. Also, TiO2 doped with cerium (Ce)


and holmium (Ho) were found to be able to retard grain growth of TiO2 as wellas decrease its crystallite size while increasing its specific surface area [85, 86].The redox pair of Ce (Ce3+/Ce4+) could act as an electron scavenger that trapsthe bulk electrons in TiO2. Ce also extends the photoresponse of TiO2 into thevisible region by reducing the bandgap of the original material. All these factorsguide to a boost in the photocatalytic performance of TiO2 [86, 94]. Burn et al.[87] reported the doping of Nd3+ on TiO2 enhance its photocatalytic activity dueto the higher temperature anatase stabilization. The different metal-ion-dopedTiO2 as photocatalyst is listed in Table 11.1.

Metal-doped TiO2 has also been used for degradation of pollutants such astoxic organic materials, pesticides, and drugs. For example, Umar et al. [114]reported a Mn-doped TiO2 for destruction of a pesticide namely glyphosateusing a visible-light halogen lamp with constant stirring and bubbling ofatmospheric oxygen. In this study, 80% destruction of pesticide glyphosatetook place after 300 min of irradiation in the presence of Mn-doped TiO2,whereas in the absence of photocatalyst, no observable decrease in the pesticideconcentration find. Devi et al. [115] reported Mo6+-ion-doped TiO2 was usedto remove an organochlorine pesticide “Tebuconazole” from aqueous solutionunder visible-light irradiation. TiO2 and B-doped TiO2 photocatalysts were

Table 11.1 The different metal-ion-doped TiO2 as photocatalyst for detoxification of water.

Doped metal ion Pollutant References

Gold Phenol [74]Gold, palladium, platinum Phenol [75]Silver Rhodamine 6G, methylene blue, methyl orange [110]Iron Methylene blue, malachite green dye, phenol [111]Copper Methyl orange [112]Vanadium Methylene blue and 2,4-dichlorophenol [109]Tungsten Methylene blue [113]Gadolinium Direct blue [80]Cerium Phenol [82]Neodymium Methyl orange [83]Cerium Chlorophenol [80]Silica Rhodamine 6G [106]Manganese Glyphosate (pesticide) [114]Molybdenum Tebuconazole [115]Boron Diuron, o-phenylphenol, terbuthylazine, and

2-methyl-4-chlorophenoxyacetic acid (MCPA)[116]

Silver Endosulfan (organochlorine pesticide) [117]Copper Bisphenol A [118]Nickel and boron Trichlorophenol (TCP), 2,4-dichlorophenol

(2,4-DCP), and sodium benzoate[119]

Zinc and iron Phenol [120]

11.4 Graphene-TiO2 Composites for Detoxification 299

used for the degradation of four recalcitrant pesticides (diuron, o-phenylphenol,2-methyl-4-chlorophenoxyacetic acid (MCPA), and terbuthylazine) along withozonation under simulated solar irradiation [116]. Thomas et al. [117] preparedsilver-doped anatase TiO2 photocatalyst via low-temperature hydrothermalroute for the degradation of the organochlorine pesticide, Endosulfan. TheB-doped TiO2 catalysts, with 0.5–0.8 wt% of interstitial boron, were more activethan bare TiO2 for the removal and mineralization of the target pesticides.This combination of ozonation and photocatalysis led to faster mineralizationrates than the individual methods and allowed the complete removal of thepesticides below the regulatory standards [116]. Cu—TiO2 nanorods were alsoused as an effective photocatalyst for Bisphenol-A (BPA) under the irradiation ofUV and visible light [118]. The different metal-ion-doped TiO2 as photocatalystused for degradation of various types of pollutants are listed in Table 11.1.

11.4 Graphene-TiO2 Composites for Detoxification

The practical use of anatase phase TiO2 as photocatalyst is limited due to therapid recombination rate of photogenerated electron–hole pair in TiO2 [121].The bandgap and electron–hole pair recombination of TiO2 can be modifiedby the combination of carbon-based materials and TiO2 [122]. TiO2–carbon(Ti—C) is now considered as potential photocatalyst in purification of water.The TiO2—C composites can be generally categorized into three kinds:TiO2-mounted activated carbon, carbon-doped TiO2, and carbon-coatedTiO2, and each of them exhibits good photocatalytic activity. However, severalproblems still hinder further promotion of efficiency of the present TiO2—Ccomposites, such as the marked decrease in the adsorptivity during photodegra-dation, the weakening of the light intensity arriving at the catalyst’s surface,and the lack of reproducibility due to the preparation and treatment variation.Several research groups have studied the incorporation of organic monolayerssuch as porphyrins, MWCNTs, and polyaniline to hybridize with semiconduc-tors to slow down the recombination of photoinduced electron−hole pairs,enhance charge transfer rate, and increase surface adsorption of molecules to bedecomposed [123–125]. The excellent physical, chemical, optical, and electricalproperties of graphene have motivated many researchers to tune their researchwork related to the design of high-performance catalysts. The combination ofTiO2 and graphene is promising to simultaneously possess excellent adsorptivity,transparency, conductivity, and controllability, which could facilitate effectivephotodegradation of pollutants [126].

The presence of many oxygen-bearing functional groups in graphite oxide[GO] and reduced graphene oxide [rGO] helps TiO2 to anchor on graphenesheets [127, 128]. Studies reported that TiO2 nanocrystals can grow directlyon the surface of GO sheets via two steps: firstly, coat TiO2 on GO sheets byhydrolysis, and secondly, it is subjected to hydrothermal treatment [127]. Manyother synthesis strategies also have been applied to prepare TiO2–graphenecomposites. A two-step solvo-/hydrothermal process was used to preparegraphene-wrapped TiO2 nanoflower composites (G–TiO2) consisting of


nanosheets and nanoparticles by Lui et al. [128]. When TiO2 is wrapped withgraphene, a planar conjugated surface is developed for dye adsorption andthereby reducing recombination through accepting electrons from TiO2, whichin turn improves the photo catalytic performance. In this study, the authorsreported that highest photocatalytic performance is obtained for the compositewith graphene loading of 5 wt% when methylene blue is used as the model dye,which outperforms commercial P25 by a factor of 3.4. The proposed mechanismof the photocatalysis is shown in Figure 11.8.

Graphite oxide–TiO2 composite can also be synthesized through liquid-phasedeposition method. Pastrana-Martínez et al. synthesized GO–TiO2 compos-ite through liquid-phase disposition followed by thermal reduction in N2atmosphere [129]. During this reduction process, partial reduction in GOand simultaneous deposition of TiO2 on partially reduced GO took place.The photodegradation studies on diphenhydramine under near-UV/vis andvisible-light irradiation showed that the composite is having varying photocat-alytic activity depending on the surface area and morphology of the composite.Mohamed synthesized TiO2–rGO composite for the detoxification of Sarin,a nerve agent. TiO2–rGO composite was synthesized through UV-assistedphotocatalytic reduction in GO in the presence of TiO2 nanoparticles inethanol. The composite having 3.0 wt% rGO showed the highest Sarin removalrate of about 99.5%. The enhanced photocatalytic performance is due to the

•O2–

O2

H2O

(a)

(c)

(b)

TiO2

VB

hν

hν

•OH

CB(2)

MB*

MB2 μm

2 μm

En

erg

y le

vel

Graphene

h+

e–

(1)

Figure 11.8 (a) Proposed diagram of the photocatalytic mechanism for graphene- wrappedTiO2 nanoflowers. The main photodegradation pathways include (1) the reduction andoxidation of adsorbed water species by a photogenerated electron–hole pair and (2) oxidationof MB by donating an electron to graphene (or the photocatalyst); (b) SEM images of theas-prepared TiO2 nanoflowers; and (c) G–TiO2 composite (TiO2 is highlighted in red, andgraphene is highlighted in blue). (Reproduced with permission from Ref. [128]. Copyright2010, The American Chemical Society.)

11.4 Graphene-TiO2 Composites for Detoxification 301

increased light absorption intensity and the reduction in photoelectron–holepair recombination in TiO2 in the presence of rGO [130]. Verma et al. synthe-sized graphene–TiO2 composite through simple ex situ hydrothermal method[131]. They assessed the photocatalytic activity using methylene blue as themodel pollutant and antibacterial activity using Bacillus subtilis and Escherichiacoli . They reported that the photodegradation capacity of TiO2–rGO compositewas four times higher than that of TiO2 alone. The photodegradation results andproposed mechanism are shown in Figure 11.9.

Similarly, the antibacterial property and photodetoxification activity ofTiO2–rGO composite is much higher than that of TiO2 alone and GO aloneunder visible light [132]. The active species such as O2− and OH− will attackbacterial cell membrane and ultimately leading to the cell death. The authorsclaimed that the colonies grown on the agar plates are reduced drastically whenexposed to visible light for 60 min in the presence of this composite catalyst.A biphasic TiO2–rGO composite can also be prepared by one-step hydrothermalmethod to study the photodegradation of organic pollutants using Rhodamineas model dye [133]. This composite has the capability of degrading colorlessdyes such as benzoic acid even better than the model catalyst P25 under visiblelight.

1.0

–3.0

–2.5

–2.0

–1.5

–1.0

–0.5

0.0

0 15 30 45 60 75 90

0.8

0.6

0.4

0.2

0.0

0 15 30 45

(a) (b)

(c)

Time (min)

Graphene sheet

CB

3.2 eV3.0 eV

Rutile

TiO2

hυ

OH–/H2O+

+ + + + +

h+

OH•

VB

0.2 μm

Anatase

Impurity level

O2 + e–

e– e– e–e– e–

e–O2–

Time (min)

C/C

0

In(C

/C0)

60

MB control

MB control

RGO-TiO2 (M)

TiO2 (M)

RGO-TiO2 (M)

TiO2 (M)

75 90

Figure 11.9 (a) Photodegradation of MB with time under visible-light irradiation, (b) lnC/Coversus time plot for determination of rate of constant. (c) Schematic representation of thepossible mechanism of photocatalytic activity for degradation of MB under visible light.(Reproduced with permission from Ref. [131]. Copyright 2017, Elsevier.)

TiO2

(a) Non-selectiveadsorption

(A)

(B) (C)

PhNH3+

PhNH3+

PhNH3+

PhNH

PhNH

(b) Selectiveadsorption

(c) Spatially separatedselective adsorption

TiO2TiO2

TiO2OR

TiO2 TiO2 TiO2TiO2

PhNH2

160 °C 8 h

GO

rGO

b a

(1)

+

Amidation reaction

Rela

tive

inte

nsity (

a.u

.)

398

a

c

b

d

AdventitiousN element

+H3N-C

C-N

400

Binding energy (eV)

402

Epoxide

ring-opening reaction

(2)

c d

r.t. 5 h

PhNH2

NH2

NH2

Ti Ti

Ti Ti

O: O:

H–OH HO• COO–

O

–O

O–O–

HN

OC

NH

O

O+H3N+H3N

NH3++

+

+

O=C O–

r.t. 5 h

Figure 11.10 (A) Schematic diagram illustrating the adsorption behaviors of cationic and anionic dyes on TiO2 surface with controllable microstructures.(B) Schematic diagram illustrating the controllable preparation of various photocatalysts and (C) their corresponding XPS N 1s spectra: (a) TiO2, (b)PhNH2/TiO2, (c) rGO-TiO2, and (d) PhNH2/rGO-TiO2. (Reproduced with permission from Ref. [134]. Copyright 2016, The American Chemical Society.)

11.5 Commercial Applications of Photocatalysis in Environmental Detoxification 303

The photocatalytic selectivity is a big challenge for the complete mineralizationof targeted organics in a complex mixture. In this aspect, Yu et al. have demon-strated a quite interesting strategy of preparing TiO2 photocatalysts withexcellently preferential adsorption for both typical cationic and anionic dyesto realize their controllable photocatalytic selectivity [134]. In their study, thenegative aniline-functionalized-reduced graphene oxide (rGO) nanosheets andpositive phenylamine (PhNH2) molecules were successfully loaded on the TiO2surface with spatially separated loading sites, and the resultant PhNH2- modifiedrGO-TiO2 (PhNH2/rGO-TiO2) photocatalysts exhibit tunable photocatalyticselectivity. In this case, the negative rGO and positive PhNH2 molecules on theTiO2 surface work as the preferentially adsorption-active sites for cationic andanionic dyes, respectively. Also, the resultant PhNH2/rGO-TiO2 photocatalystnot only realizes tunable photocatalytic selectivity but also can completelydecompose the oppositely cationic and anionic dyes. The absorption behavior ofthe oppositely charged dyes on TiO2 surface is demonstrated in Figure 11.10.

Recently, apart from binary hybrids, tertiary hybrids are also being usedfor effective detoxification applications [135, 136]. MoS2–graphene–TiO2composite is an effective ternary composition for photocatalytic applications.Gao et al. [137] reported a simple one-pot solvothermal approach to fabricate aphotocatalyst with MoS2 quantum dot–graphene–TiO2, which shows significantimprovement in photocatalytic properties. The main advantage of such ternarycomposites is the increased charge separation, visible-light absorbance, andspecific surface area and reaction sites upon the introduction of MoS2 QDs.

11.5 Commercial Applications of Photocatalysisin Environmental Detoxification

Over the last few decades, semiconductor photocatalysis has emerged as a pow-erful alternative in the field of environmental remediation over the conventionalpurification and disinfection techniques [4, 138, 139]. Solar photocatalysis iswidely applied in developing functional materials for wastewater treatments,air purification, and self-cleaning applications. Photocatalytic products wereinitially commercialized in Japan in the mid-1990 and later in Europe andAmerica. The global market for photocatalyst-based products is estimated toexpand at a compound annual growth rate of 12.6% in the coming five years,and expected to reach nearly $2.9 billion by 2020 [140], the majority of whichaccounts for use in construction materials.

11.5.1 Self-Cleaning Materials

The combination of photocatalysis and photoinduced hydrophilicity exhibitedby TiO2 and other semiconductors has been applied to develop self-cleaningsurfaces [4, 141]. These photocatalytic materials have gained significant pop-ularity over the last few decades due to their applications in constructingself-cleaning, antibacterial, antifogging coatings for various applications [142].The photocatalytic self-cleaning surfaces are useful against environmentalpollution and involve lower maintenance cost as the surfaces can be easily


Hydrophilic surface

θ θ

Hydrophobic surface

Superhydrophilic surface

(a)

Water goesbeneath contaminant

Water picks upcontaminant

WaterContaminant

(b)

Superhydrophobic surface

Figure 11.11 (a) Schematic representation of a liquid drop on a hydrophilic and a hydrophobicsurface. (Reproduced with permission from Ref. [4]. Copyright 2015, Elsevier.). (b) Schematicrepresentation of self-cleaning action of a superhydrophilic and a superhydrophobic surface.(Reproduced with permission from Ref. [141]. Copyright 2013, The Royal Society of Chemistry.)

cleaned by a stream of water. The self-cleaning surfaces can be broadly dividedinto two classes: (i) hydrophilic and (ii) hydrophobic surfaces. For a hydrophilicsurface, the contact angle (𝜃) of a liquid drop over a solid surface remains low(𝜃 < 90∘), while for a hydrophobic surface the contact angle assumes a highervalue (𝜃 > 90∘) (Figure 11.11a). For superhydrophilic surfaces (𝜃 < 10∘), theliquid drop spreads evenly on the surface and efficiently washes away any dust orother contaminants present on the surface with the water flow (Figure 11.11b).Several studies showed that surface roughness and porosity are crucial factors indetermining the contact angle and wettability of a surface [143–145].

Various mechanisms have been proposed to understand the phenomenonof photoinduced hydrophilicity exhibited by TiO2. The widely accepted modelexplains the phenomenon through formation of TiO2 surface defects uponillumination by UV light [146]. It was postulated that irradiation of TiO2 surfaceby UV light results in the formation of “oxygen vacancies,” which in turnconverts Ti4+ ions into Ti3+ and increases the affinity for water molecules. Sakaiet al. proposed that hydroxyl groups at the TiO2 surface undergo reconstructionas a consequence of UV-light illumination and extent of this change is directlyrelated to the density of surface hydroxyl groups [147]. Additionally, the positiveholes generated upon illumination of TiO2 by UV irradiation can also diffuseto the surface, where they can be trapped at sites of lattice oxygen. This can in


turn weaken the Ti—O bond, increase the susceptibility to water attack, andthus facilitate formation of new O—H bonds. Takeuchi et al. proposed thatthermal energy generated by illumination of TiO2 by UV light results in thedesorption of surface bound water molecules and consequently decreases theH-bonded network on the surface and reduces the surface tension of the watercluster, which is essential for surface wetting [148]. Despite various mechanismsproposed, no consensus has been arrived till date and a combination of differentmodels is often necessary to explain the phenomenon.

Transparent thin films of photocatalytic TiO2 have been used to fabricateself-cleaning tiles, window glasses, and other building materials, where the pho-tocatalytic material utilizes sun light and rain water to remove dust and otherorganic pollutants (Figure 11.12). In 2001, the first commercial self-cleaning win-dows (Pilkington ActivTM) were developed by Pilkington Glass. The self-cleaningwindows utilize sun light and rain water to remove dust and other organicpollutants [149]. The environmentally friendly self-cleaning surfaces minimizethe energy expenditure, exposure to chemicals and reduce maintenance cost.This was followed by other major glass companies delivering products on similarline like Saint Gobain’s Bioclean, Impact Safety Glass, and PPG’s SuncleanTM.HydrotechTM, introduced by the Japanese company TOTO Ltd., uses the mecha-nism of photoinduced superhydrophilicity of TiO2 to eliminate pollutants fromthe surface and has been widely applied in indoor and outdoor applications [150].A cricket stadium in Dubai (Figure 11.13a) represents the first example outsideJapan, where the roof is built with a self-cleaning PTFE (Polytetrafluoroethylene)membrane coated with photocatalytic TiO2 [151].

Dirt

Coating

Glass

UV light Rain

Figure 11.12 Schematic representation of the working principle of self-cleaning glassesshowing (from the left to right) the accumulation of dust/pollutants on glass, activation of thephotocatalytic material by sunlight, photocatalytic decomposition of the pollutants, andfinally cleaning of the degraded materials by rain water. (Reproduced with permission fromRef. [142]. Copyright 2013, Elsevier.)


(a) (b)

(c) (d)

Figure 11.13 Self-cleaning cement coated on (a) Roof of Dubai Sports City’s Cricket Stadium.(Reproduced with permission from Ref. [151]. Copyright 2012, Royal Society of Chemistry.)(b) Dives in Misericordia Church in Rome. (c) Cité de la Musique et des Beaux-Arts inChambéry, France. (d) European Photoreactor (EUPHORE). (Reproduced with permission fromRef. [139]. Copyright 2015, Elsevier.)

An Italian cement company, named Italcementi, has marketed TiO2-basedcements namely TX AriaTM, TX ActiveTM, and TX MillenniumTM and used inmany building constructions (Figure 11.13b,c). These TiO2-based cementswere also used in highway constructions and have been tested for degradinggases such as NOx, SOx from automobile exhaust [152]. Heidelberg cementtechnology has manufactured TioCem®, which showed significant degradationof NOx and other air pollutants [153]. Mitsubishi Materials Corp, Japan, hasdesigned NOxerTM for removal of NOx from air [154]. These materials candegrade NOx emitted from automobile exhaust and improve air quality. Devel-opment of European Photoreactor (EUPHORE) in Valencia, Spain, representsa remarkable advancement in the field of removal of air pollutants using solarirradiation [155]. The EUPHORE photoreactor (Figure 11.13d) is composed oftwo outdoor hemispherical Teflon chambers, which allow transmittance of over80% of the sunlight in the wavelength range 280–640 nm. The chambers areequipped with sensitive analytical devices for sensing of even trace amount ofpollutants such as VOCs, O3, NO, NO2, PAN, organic nitrates, hydroperoxidesand organic acids. In order to improve the visible-light-induced photocatalysisby TiO2, effect of doping with metal, nonmetal, formation of heterojunction of


TiO2 with other semiconductors have been extensively studied in recent years[6, 7]. Visible-light-active TiO2 thin films for self-cleaning building materialswere developed by doping with Ni2+, Fe3+ ions into TiO2 lattice [131], where thevisible-light activity arises due to various intrinsic defects (oxygen vacancies orTi interstitial) created by the dopant. Tetraethyl orthosilicate (TEOS) modifiedDegussa P25–TiO2 nanoparticles have been used to fabricate glazed ceramic tilesfor construction applications [156]. The resulting tiles showed high photocat-alytic activity and photoinduced hydrophilicity upon visible-light illumination,which have been assigned to larger surface area and increased surface roughnessof the coatings. Fabrication of self-cleaning materials for construction purposesinvolves processing at very high temperatures. Therefore, increased thermal sta-bility of photocatalytically active anatase TiO2 is desirable for these applications.Pillai and coworkers reported that nonmetal-doped anatase TiO2 exhibits highthermal stability and thus suitable for self-cleaning applications [157]. Moreover,the same group also reported development of visible-light-active, oxygen-richTiO2, where anatase phase shows stability up to 900 ∘C and can be useful forconstruction of self-cleaning building materials [158].

11.5.2 Bactericidal

Photoinduced bactericidal activity of TiO2 has gained significant attention inrecent years [1]. Photocatalytic antimicrobial effects of TiO2 are extremelyimportant in various fields such as medical applications, construction of ster-ilized coatings for hospitals, indoor applications, food industry to overcomemicrobial contamination [159]. The UV-light induced photocatalytic antibacte-rial action of TiO2 thin films was initially demonstrated by Kikuchi et al. [160].The photocatalytic bactericidal effect of illuminated TiO2 has been proposedto arise from the production of OH∙, O2

∙—, and H2O2 (Figure 11.14). Sun et al.demonstrated that TiO2 microspheres with reactive (111) facets exposed on theexternal surface caused a higher extent of bactericidal effect compared to EvonikDegussa P25 upon UV irradiation [162]. The improved activity results from theproduction of higher levels of OH∙ due to reduced electron–hole recombination

CB

VB

UV

TiO2 nanotubes

K+ leakage

Membrane ruptureDNA/RNA damage

e–

h+

O2

•O2–

1O2

H2O

•O2–

•OH

Figure 11.14 Schematic diagram showing the mechanism of antibacterial action of TiO2nanotubes. (Reproduced with permission from Ref. [161]. Copyright 2015, Elsevier.)


in the faceted TiO2 microspheres [162]. Rengifo-Herrera et al. reported thatin the case of N,S codoped TiO2, photocatalytic bacterial inactivation underUV-light irradiation is caused by highly oxidizing OH∙ [163, 164]. However, theholes generated by visible-light irradiation do not possess sufficient reductionpotential to oxidize H2O to produce OH∙ [163, 164]. The bactericidal effect ofN,S codoped TiO2 under visible-light illumination is thought to arise predomi-nantly from the production of less oxidative O2

∙— and 1O2 [163, 164]. In recenttime, TiO2 modified with metal ions such as Ni2+ [165], Cu2+ [158], Ag+[166],nonmetals including N [167], S [168], C [169], and more recently graphenenanosheet–TiO2 composites [170, 171] have been developed and examined fortheir bactericidal effects on various bacterial strains.

Recently, bactericidal effect of undoped titania nanotubes synthesized byelectrochemical anodization method against E. coli (97.5%) and Staphylococcusaureus (99.9%) under UV illumination has been reported [161]. Such highantibacterial activity of these nanotubes is thought to be governed by theirsurface morphology and physicochemical properties. The antifungal andantibacterial activity of rutile–TiO2 nanorod arrays synthesized by hydrothermalprocess and sintering has been recently demonstrated [172]. The nanorodarrays displayed high activity against Candida albicans, Aggregatibacter actino-mycetemcomitans, and Porphyromonas gingivalis, under UV irradiation, whichis believed to be due to production of various ROS.

Contamination of drinking water by cyanotoxins, particularly microcystins, isa major concern in recent time. Visible-light-active TiO2 doped with various non-metals including N [173], S [174], N—F [175], C—N [176], P—F [177] have beenevaluated for their ability to degrade cyanotoxins. The effect of dopant mate-rials on the crystallite size, anatase–rutile phase transition, surface roughnessand porosity, modification of electronic energy levels, and formation of OH∙,O2∙—,1O2 responsible for the degradation of toxin have been discussed in variousreports [175–178].

In 1992, the first bactericidal photocatalytic coating was developed by theJapanese Arc-Flash Company, using TiO2 nanoparticles [179]. The Arc-Flash®photocatalytic coating displays antibacterial activity and can be used to effi-ciently sanitize environments including hospitals, schools, and households.In addition to be used as antimicrobial agents, TiO2 has the potential to beused to increase product shelf-life and prevent spoilage of perishable goodsby treating the air in vegetable, fruit and flower storage zones. Products like“ABSOGER” from ABSOGER Sas Co. (Les Barthes, FR), “Bio-KES” from KESScience & Technology Inc. (GA, USA), are front runners in this market [180].Also, similar products like “AiroCide®” , “FRESH+TM ” have also entered intothe market. This technology exploits the photocatalytic properties of TiO2 toprevent premature maturation of fruit by destroying gas-phase ethylene, whichin higher concentration (>1 ppm) triggers ripening of vegetables and fruits [180].

11.5.3 Wastewater Detoxification

A major application of semiconductor photocatalysis has emerged in thetreatment of wastewater through removal/mineralization of organic especially


halogenated compounds in the presence of dissolved oxygen. Photoreactors aredevices that collect and utilize solar energy for the treatment of toxic water inthe presence of photocatalyst. The initially designed photoreactors for solarphotocatalytic applications were based on parabolic trough collector (PTC)[181, 182]. Parabolic troughs are usually made of reflective surfaces such asalumina, and concentrate the solar radiation on a transparent reactor tubecontaining the reactant fluid placed along the parabolic focal line (Figure 11.15).Due to increased intensity of the incident radiation, PTC allows use of lesserphotocatalytic load.

The first outdoor engineering-scale PTC was designed and installed at theNational Solar Thermal Test Facility, Sandia laboratories in Albuquerque, NewMexico (USA) [183]. The facility consists of 6 aligned PTCs with single-axissolar tracking for a total of 465 m2 aperture area and was developed for treat-ment of wastewater containing heavy metals and chlorinated solvents. PTCsSOLARIS and PROPHIS were constructed for industrial reactions at the GermanAerospace Centre (DLR), Cologne, Germany, and at the Plataforma Solar deAlmeria (PSA), Spain, respectively (Figure 11.16). These collectors are used forsolar-light-driven organic photochemical synthesis, with a capacity of 35–120 lreaction mixture. Each of these collectors consists a crossbeam, a turret, pipingand electrical equipment, four troughs, and a two-axis solar tracker for aligningthe collector with the position of Sun. The rectangular collector aperture isstruck by direct solar radiation followed by their reflection through parabolicmirrors into the transparent receiver–reactor tubes as they are positioned inthe four focal lines. The receiver–reactor tubes are connected in series and; thereaction mixture is pumped through these until the desired product is produced.

Heatexchanger

Stirrer

Oxidantaddition

Recirculatingpump

Effluent

Recirculatingtank

Figure 11.15 Schematic representation of a parabolic trough collector. (Reproduced withpermission from Ref. [182]. Copyright 2016, MDPI.)


(a) (b)

Figure 11.16 Schematic representations of (a) SOLARIS reactor and (b) PROPHIS reactor.(Reproduced with permission from Ref. [139]. Copyright 2015, Elsevier.)

The excess heat is removed by an external heat exchanger and an air cooledprocess cooler is placed to reject that excess heat to the ambient [184].

The main disadvantage of PTCs arises from the intrinsic geometry of thetroughs, which allow using only the direct radiation beam, thus making themunsuitable for use on cloudy days and at places with high or intermediate solarzenith angles [185]. Additionally, use of tracking system increases the overallcost of the system.

Nonconcentrating solar collectors (NCCs) represent an attractive cost–effective alternative to PTCs [186]. The fluid is pumped onto the surface of flat orcorrugated inclined planes supporting the photocatalyst. The back plate of NCCscan be made of glass, metal, or stone. These collectors do not have any solartracking devices and thus have low manufacturing and maintenance costs. Thereactor design allows capturing both direct beam of radiation and diffusing light.A solar catalytic pilot plant based on NCC reactor using TiO2 has been built atthe site of a textile factory, in Tunisia, for removal of recalcitrant compoundsand color from wastewater. The plant has a total illuminated area of 50 m2 and isdesigned to treat 1 m3 h−1 wastewater. However, due to fixed orientation of thenonconcentrating static solar collectors, these devices are less energy efficientand better suited for small-scale applications [187]. Due to low installationand maintenance cost, these devices are useful in less-developed places, whereinstallation of photoreactors for wastewater treatment is not feasible.

Compound parabolic collectors (CPCs) represent a combination of PTC andNCC photoreactors, displaying excellent energy efficiency, and are promisingcandidates for solar photochemical applications [185]. CPCs consist of station-ary collectors with parabolic reflective surfaces arranged around a cylindricalreactor tube (Figure 11.17). The parabolic collector surfaces are designed in away such that the focal points of the two halves remain close. In this design,axes of the two parabolic halves are inclined to each other in a manner suchthat rays incident within the angle between the axes undergo single or multi-ple internal reflections in the region between the two focal points and thus getconcentrated in that section. The reflector geometry allows capturing of bothdirect and diffuse sunlight, thus enabling its use even on cloudy days [188]. Asthe total amount of sunlight absorbed is significantly higher in CPC, the reac-tor can be smaller sized. CPCs with collector areas varying from 3 m2 to 150 m2


Stirrer

Oxidantaddition

Recirculatingpump

Effluent

Recirculatingtank

Figure 11.17 Schematic representation of a compound parabolic concentrator. (Reproducedwith permission from Ref. [36]. Copyright 2013, Elsevier.)

have been successfully employed for removal of pollutants, pathogenic bacteria,organic dyes, chlorinated compounds, pesticides from water, treatment of urbanwastewater [10, 189–192].

The European industrial consortium called SOLARDETOX has designedwastewater treatment plants along with several other plants in Spain basedon solar heterogeneous (TiO2) and homogeneous photocatalysis [193]. Thesephotocatalytic CPC reactor module-based demonstration plants are capable oftreating large volume of contaminated water within a short time span havingphoton collection area as large as 150 m2. Another Spanish pharmaceutical com-pany named DSM DIRETIL has also been operating since 2007 for pretreatmentof saline industrial wastewater containing biorefractory pharmaceuticals usinga 100 m2 large homogeneous solar photocatalytic CPC plant. The removal ofcontaminant by this system is partially done as organic carbon and rest of it byaerobic biological treatment [194].

Another CPC facility, named SOLFIN (SOLar synthesis of FINe chemicals),uses a photoreactor of 25 l volume capacity circulated by a centrifugal pump. Theapparatus of 1 m length and 20 cm width employs sunlight at a low concentrationfactor (CF= 2–3 suns) [195]. The efficient cooling system associated with thisunit keeps the heat of the reaction mixture below 20 ∘C by discharging thegenerated heat. The SOLFIN apparatus has been used for large-scale production


of a number of precursors and intermediates for industrial polymers andpharmaceuticals [196].

One of the interesting efforts in this line of research is to overcome the environ-mental problems caused by the development of a green house agriculture sectorin the Mediterranean Basin using CPC-type facility. These green house facilitiescovering over 200 000 hectares are proliferating throughout the EU countries,largely in Almeria, Spain, causing concerns due to their uncontrolled dumpingof plastic containers containing reagents. A huge population of flora and faunaon land and sea alike is getting polluted by those wastes’ groundwater filtration,leaching and finally polluting the ocean after being borne by river and/or rainwater. The CPC reactors are actively being employed to recycle those productsthus making commercially viable products in addition to the minimization ofenvironmental hazards [197].

Although different solar photoreactors are largely used for the removal ofpathogenic microorganisms and other toxic materials from water, scaling upthe photocatalytic reactor is a huge challenge in this field, especially whencompared to scaling up of conventional chemical reactors. Design of the reactorand chemical as well as environmental parameters play crucial roles in processoptimization. Hence, controlling the temperature, mass transfer, reagent andcatalyst contact, mixing of the reagents, and flow patterns should be optimizedin each reaction scale such that catalysts receive sufficient solar radiation. Toachieve uniform sunlight distribution and maximize surface area exposure,radial and axial scale-up are followed, respectively. Operating costs can beminimized with significant reduction in the reactor dimensions by achieving ahigher illuminated surface-to-volume ratio [139].

In addition to the aforementioned efforts, a lot of large- and small-scalecompanies are using TiO2-based photocatalyst system for water purification,and some of them are successful with important product development. TheJapanese company Ishihara Sangyo Kaisha (ISK) is a large TiO2 manufacturer,having a patented technology on water purification named Tipaque®, usingphotocatalysis and a novel, porous catalyst [198]. In comparison, Hyosung Ebara,a South Korean company specialized in water purification, has developed waterpurification systems based on semiconductor photocatalyst [199]. However,no obvious commercial system has been reported. On the other hand, a smalluniversity-based company like Clear Water Industries, located in Florida, hasbeen reported that they can treat more than 2200 l water per minute usingflatbed semiconductor photocatalyst [200]. Another similar academia industryconglomerate named Photox Bradford Limited, originating from BradfordUniversity, UK, uses semiconductor photocatalytic reactor with a capacity ofabout 170 l min−1 using TiO2 as photocatalytic slurry. Established companieslike Lynntech Inc. had also started as a small company that sells semiconductorphotocatalyst reactor on a fixed bed [201]. One of the most successful effortshas come from Purifics® Environmental Technologies Inc., Canada, who haddeveloped several US patents on water and air purifying systems. Based on theirautomated treatment system for water and air named Photo-Cat® technology,they have emerged as one of the largest suppliers of industrial semiconductorphotocatalyst treatment systems [202] .

References 313

11.6 Conclusions

The detoxification of pollutants using photocatalysis is one of the mostextensively investigated methods of overcoming water contamination by organicpollutants. While TiO2 is the most researched photocatalyst, other semiconduc-tors such as Fe2O3, ZnS, g-C3N4, ZnO, BiVO4, GaN, and GaP are also studiedby a number of researchers. These semiconductors are wide bandgap materialsthat absorb in the UV part of the solar spectrum, which is only <5% of the solarspectrum. For large-scale and commercial applications, it is desirable to developa catalyst which absorb in the visible-light region as this makes up approximately42% of the solar spectrum. A number of different methodologies such as oxygendefect creation, anion doping, and cation doping have been previously triedto improve the efficiency of the photocatalysts in the visible region. Thoughresearch in the area of solar photocatalysis has made major progress in recentyears, the commercialization of this technology is still at a very early stage.Several critical issues such as use of visible-light-active photocatalysis, reducedcharge-carrier recombination, increased photocatalytic efficiency, and suitablescaling-up options need to be addressed for successful implementation of thetechnology.

References

1 Fujishima, A. (1972) Electrochemical photolysis of water at a semiconductorelectrode. Nature, 238, 37–38.

2 Linsebigler, A.L., Lu, G., and Yates, J.T. (1995) Photocatalysis on TiO2 sur-faces: principles, mechanisms, and selected results. Chem. Rev., 95, 735–758.

3 Scanlon, D.O., Dunnill, C.W., Buckeridge, J., Shevlin, S.A., Logsdail, A.J.,Woodley, S.M., Catlow, C.R.A., Powell, M.J., Palgrave, R.G., Parkin, I.P.,and Watson, G.W. (2013) Band alignment of rutile and anatase TiO2. Nat.Mater., 12 (9), 798–801.

4 Banerjee, S., Dionysiou, D.D., and Pillai, S.C. (2015) Self-cleaning applica-tions of TiO2 by photo-induced hydrophilicity and photocatalysis. Appl.Catal., B, 176, 396–428.

5 Hanaor, D.A. and Sorrell, C.C. (2011) Review of the anatase to rutile phasetransformation. J. Mater. Sci., 46 (4), 855–874.

6 Pelaez, M., Nolan, N.T., Pillai, S.C., Seery, M.K., Falaras, P., Kontos, A.G.,Dunlop, P.S., Hamilton, J.W., Byrne, J.A., O’shea, K., and Entezari, M.H.(2012) A review on the visible light active titanium dioxide photocatalystsfor environmental applications. Appl. Catal., B, 125, 331–349.

7 Banerjee, S., Pillai, S.C., Falaras, P., O’Shea, K.E., Byrne, J.A., and Dionysiou,D.D. (2014) New insights into the mechanism of visible light photocatalysis.J. Phys. Chem. Lett., 5, 2543–2554.

8 Su, R., Tiruvalam, R., He, Q., Dimitratos, N., Kesavan, L., Hammond, C.,Lopez-Sanchez, J.A., Bechstein, R., Kiely, C.J., Hutchings, G.J., andBesenbacher, F. (2012) Promotion of phenol photodecomposition over TiO2using Au, Pd, and Au—Pd nanoparticles. ACS Nano, 6, 6284–6292.


9 Malato, S. (2007) Waste Water Treatment by Advanced Oxidation Processes(Solar Photocatalysis in Degradation of Industrial Contaminants), Innova-tive Processes and Practices for Wastewater Treatment and Re-use AnkaraUniversity, Turkey.

10 Kositzi, M., Poulios, I., Malato, S., Caceres, J., and Campos, A. (2004) Solarphotocatalytic treatment of synthetic municipal wastewater. Water Res., 38,1147–1154.

11 da Hora Machado, A.E., Xavier, T.P., de Souza, D.R., de Miranda, J.A.,Duarte, E.T.F.M., Ruggiero, R., de Oliveira, L., and Sattler, C. (2004) Solarphoto-Fenton treatment of chip board production waste water. Sol. Energy,77 (5), 583–589.

12 Kumar, S.G. and Devi, L.G. (2011) Review on modified TiO2 photocatalysisunder UV/visible light: selected results and related mechanisms on interfa-cial charge carrier transfer dynamics. J. Phys. Chem. A, 115, 13211–13241.

13 Etacheri, V., Di Valentin, C., Schneider, J., Bahnemann, D., and Pillai, S.C.(2015) Visible-light activation of TiO2 photocatalysts: advances in theory andexperiments. J. Photochem. Photobiol., C, 25, 1–138.

14 Serpone, N. (2006) Is the band gap of pristine TiO2 narrowed by anion-and cation-doping of titanium dioxide in second-generation photocatalysts?J. Phys. Chem. B, 110, 24287–24293.

15 Fujishima, A., Zhang, X., and Tryk, D.A. (2008) TiO2 photocatalysis andrelated surface phenomena. Surf. Sci. Rep., 63, 515–582.

16 Fujishima, A., Rao, T.N., and Tryk, D.A. (2000) Titanium dioxide photocatal-ysis. J. Photochem. Photobiol., C, 1, 1–35.

17 Yates, J.T. (2009) Photochemistry on TiO2: mechanisms behind the surfacechemistry. Surf. Sci., 603, 1605–1612.

18 Kapilashrami, M., Zhang, Y., Liu, Y.-S., Hagfeldt, A., and Guo, J. (2014) Prob-ing the optical property and electronic structure of TiO2 nanomaterials forrenewable energy applications. Chem. Rev., 114, 9662–9707.

19 Nie, X.L., Zhou, S.P., Maeng, G., and Sohlberg, K. (2009) Doping of TiO2polymorphs for altered optical and photocatalytic properties. Int. J. Photoen-ergy, article 294042.

20 Na Phattalung, S., Limpijumnong, S., and Yu, J. (2017) Passivated co-dopingapproach to bandgap narrowing of titanium dioxide with enhanced photo-catalytic activity. Appl. Catal., B, 200, 1–11.

21 Wang, P., Liu, Z., Lin, F., Zhou, G., Wu, J., Duan, W., Gu, B.-L., andZhang, S.B. (2010) Optimizing photoelectrochemical properties of TiO2by chemical codoping. Phys. Rev. B, 82 (19), 193103.

22 Zhang, D. and Yang, M. (2013) Band structure engineering of TiO2nanowires by n–p codoping for enhanced visible-light photoelectrochemicalwater-splitting. Phys. Chem. Chem. Phys., 15, 18523–18529.

23 Wang, J., Sun, H., Huang, J., Li, Q., and Yang, J. (2014) Band structure tuningof TiO2 for enhanced photoelectrochemical water splitting. J. Phys. Chem. C,118, 7451–7457.

24 Di Valentin, C. and Selloni, A. (2011) Bulk and surface polarons in photoex-cited anatase TiO2. J. Phys. Chem. Lett., 2, 2223–2228.

References 315

25 Cheng, J., Vandevondele, J., and Sprik, M. (2014) Identifying trappedelectronic holes at the aqueous TiO2 interface. J. Phys. Chem. C, 118,5437–5444.

26 Wallace, S.K. and McKenna, K.P. (2015) Facet-dependent electron trappingin TiO2 nanocrystals. J. Phys. Chem. C, 119, 1913–1920.

27 Nolan, M., Iwaszuk, A., Lucid, A.K., Carey, J.J., and Fronzi, M. (2016) Designof novel visible light active photocatalyst materials: surface modified TiO2.Adv. Mater., 28, 5425–5446.

28 Fronzi, M., Iwaszuk, A., Lucid, A.K., and Nolan, M. (2016) Metal oxidenanocluster-modified TiO2 as solar activated photocatalyst materials. J. Phys.Condens. Matter, 28, 074006/1–074006/23.

29 Galynska, M. and Persson, P. (2014) Energetic materials, advances in quan-tum chemistry, vol. 69, Elsevier, Amsterdam.

30 Hahn, K.R., Tricoli, A., Santarossa, G., Vargas, A., and Baiker, A. (2012) Firstprinciples analysis of H2O adsorption on the (110) surfaces of SnO2, TiO2and their solid solutions. Langmuir, 28, 1646–1656.

31 Hussain, A., Gracia, J., Nieuwenhuys, B.E., and Niemantsverdriet, J.W.H.(2013) Explicit roles of Au and TiO2 in a bifunctional Au/TiO2 cata-lyst for the water–gas shift reaction: a DFT study. ChemCatChem, 5,2479–2488.

32 Koocher, N.Z., Martirez, J.M.P., and Rappe, A.M. (2014) Theoretical modelof oxidative adsorption of water on a highly reduced reconstructed oxidesurface. J. Phys. Chem. Lett., 5, 3408–3417.

33 Patel, M., Sanches, F.F., Mallia, G., and Harrison, N.M. (2014) A quantummechanical study of water adsorption on the (110) surfaces of rutile SnO2and TiO2: investigating the effects of intermolecular interactions usinghybrid-exchange density functional theory. Phys. Chem. Chem. Phys., 16,21002–21015.

34 Patel, M., Mallia, G., Liborio, L., and Harrison, N.M. (2012) Water adsorp-tion on rutile TiO2 (110) for applications in solar hydrogen production:a systematic hybrid-exchange density functional study. Phys. Rev. B, 86,045302/1–045302/15.

35 Sahoo, S.K., Nigam, S., Sarkar, P., and Majumder, C. (2013) DFT study ofH2O adsorption on TiO2 (110) and SnO2 (110) surfaces. AIP Conf. Proc.,1512, 292–293.

36 Di Valentin, C. and Pacchioni, G. (2013) Trends in non-metal doping ofanatase TiO2: B, C, N and F. Catal. Today, 206, 12–18.

37 Sun, C., Liu, L.-M., Selloni, A., Lu, G.Q., and Smith, S.C. (2010)Titania–water interactions: a review of theoretical studies. J. Mater. Chem.,20, 10319–10334.

38 Kristoffersen, H.H., Hansen, J.O., Martinez, U., Wei, Y., Matthiesen, R.S.,Bechstein, R., Laegsgaard, E., Besenbacher, F., Hammer, B., and Wendt, S.(2013) Role of steps in the dissociative adsorption of water on rutile TiO2(110). Phys. Rev. Lett., 110, 146101/1–146101/4.

39 Tilocca, A. and Selloni, A. (2012) DFT-GGA and DFT+U simulations of thinwater layers on reduced TiO2 anatase. J. Phys. Chem. C, 116, 9114–9121.


40 Valdés, Á., Qu, Z.-W., Kroes, G.-J., Rossmeisl, J., and Nørskov, J.K. (2008)Oxidation and photo-oxidation of water on TiO2 surface. J. Phys. Chem. C,112, 9872–9879.

41 Muuronen, M., Parker, S.M., Berardo, E., Le, A., Zwijnenburg, M.A., andFurche, F. (2017) Mechanism of photocatalytic water oxidation on smallTiO2 nanoparticles. Chem. Sci., 8, 2179–2183.

42 Li, Y.-F., Liu, Z.-P., Liu, L., and Gao, W. (2010) Mechanism and activity ofphotocatalytic oxygen evolution on titania anatase in aqueous surroundings.J. Am. Chem. Soc., 132, 13008–13015.

43 Liao, P., Keith, J.A., and Carter, E.A. (2012) Water oxidation on pure anddoped hematite (0001) surfaces: prediction of Co and Ni as effective dopantsfor electrocatalysis. J. Am. Chem. Soc., 134, 13296–13309.

44 Nguyen, M.-T., Piccinin, S., Seriani, N., and Gebauer, R. (2015)Photo-oxidation of water on defective hematite (0001). ACS Catal., 5,715–721.

45 Asahi, R., Morikawa, T., Ohwaki, K., Aoki, K., and Taga, Y. (2001)Visible-light photocatalysis in nitrogen-doped titanium oxides. Science,293, 269–271.

46 Xing, M., Wu, Y., Zhang, J., and Chen, F. (2010) Effect of synergy on thevisible light activity of B, N and Fe co-doped TiO2 for the degradation ofMO. Nanoscale, 2, 1233–1239.

47 Di Valentin, C. and Pacchioni, G. (2013) Trends in non-metal doping ofanatase TiO2: B, C, N and F. Catal. Today, 206, 12–18.

48 Zhang, J., Chen, X., Shen, Y., Li, Y., Hu, Z., and Chu, J. (2011) Synthe-sis, surface morphology, and photoluminescence properties of anataseiron-doped titanium dioxide nano-crystalline films. Phys. Chem. Chem.Phys., 13, 13096–13105.

49 Zhang, H., Chen, G., and Bahnemann, D.W. (2009) Photoelectrocatalyticmaterials for environmental applications. J. Mater. Chem., 19, 5089–5121.

50 Wang, G., Ling, Y., and Li, Y. (2012) Oxygen-deficient metal oxide nanos-tructures for photoelectrochemical water oxidation and other applications.Nanoscale, 4, 6682–6691.

51 Muhlich, C.L., Zhou, Y., Holder, A.M., Weimer, A.W., and Musgrave, C.B.(2012) Effect of surface deposited Pt on the photoactivity of TiO2. J. Phys.Chem. C, 116, 10138–10149.

52 Wang, G.M., Wang, H.Y., Ling, Y.C., Tang, Y.C., Yang, X.Y., Fitzmorris, R.C.,Wang, C.C., Zhang, Z.J., and Li, Y. (2011) Hydrogen-treated TiO2 nanowirearrays for photoelectrochemical water splitting. Nano Lett., 11, 3026–3033.

53 Wang, H.Y., Wang, G.M., Ling, Y.C., Lepert, M., Wang, C.C., Zhang, Z.J.,and Li, Y. (2012) Oxide nanowires for solar cell applications. Nanoscale, 4,1436–1445.

54 Yu, X., Kim, B., and Kim, Y.K. (2013) Highly enhanced photoactivity ofanatase TiO2 nanocrystals by controlled hydrogenation-induced surfacedefects. ACS Catal., 3, 2479–2486.

55 Wang, Z., Liu, Y., Huang, B., Dai, Y., Lou, Z., Wang, G., Zhang, X., andQin, X. (2014) Progress on extending the light absorption spectra of photo-catalysts. Phys. Chem. Chem. Phys., 16, 2758–2784.

References 317

56 Chen, X., Liu, L., Liu, Z., Marcus, M.A., Wang, W.-C., Oyler, N.A.,Grass, M.E., Mao, B., Glans, P.-A., Yu, P.Y., Guo, J., and Mao, S.S. (2013)Properties of disorder-engineered black titanium dioxide nanoparticlesthrough hydrogenation. Sci. Rep., 3, p. 1510.

57 Andreev, A.S., Kuznetsov, V.N., and Chizhov, Y.V. (2012) Atomic hydro-gen activated TiO2 nanocluster: DFT calculations. J. Phys. Chem. C, 116,18139–18145.

58 Lium, Y., Wang, J., Yang, P., and Matras-Postolek, K. (2015) Self-modificationof TiO2 one-dimensional nano-materials by Ti3+ and oxygen vacancy usingTi2O3 as precursor. RSC Adv., 5, 61657–61663.

59 Gai, Y., Li, J., Li, S.-S., Xai, J.-B., and Wei, S.-H. (2009) Design of narrow-gapTiO2: a passivated codoping approach for enhanced photoelectrochemicalactivity. Phys. Rev. Lett., 102, 036402/1–036402/4.

60 Liu, G., Wang, L., Yang, H.G., Cheng, H.M., and Lu, G.Q. (2010)Titania-based photocatalysts — crystal growth, doping and heterostructuring.J. Mater. Chem., 20, 831–843.

61 Long, R. and English, N.J. (2012) Band gap engineering of (N, Si)-codopedTiO2 from hybrid density functional theory calculations. New J. Phys., 14,053007/1–053007/11.

62 Yang, F., Yang, H., Tian, B., Zhang, J., and He, D. (2013) Preparation of anMo and C co-doped TiO2 catalyst by a calcination–hydrothermal method,and degradation of rhodamine B in visible light. Res. Chem. Intermed., 39,1685–1699.

63 Zhang, J., Pan, C., Fang, P., Wei, J., and Xiong, R. (2010) Mo+C Co-dopedTiO2 using thermal oxidation for enhancing photocatalytic activity. ACSAppl. Mater. Interfaces, 2, 1173–1182.

64 Long, R. and English, N.J. (2011) New insights into the band-gap narrowingof (N, P)-Co-doped TiO2 from hybrid density functional theory calculations.ChemPhysChem, 12, 2604–2608.

65 Mi, L., Xu, P., and Wang, P. (2007) First-principles calculation of N:Hcodoping effect on energy gap narrowing of TiO2. Appl. Phys. Lett., 90,171909/1–171909/3.

66 Long, R. and English, N.J. (2009) First-principles calculation of nitrogen–tungsten codoping effects on the band structure of anatase–titania. Appl.Phys. Lett., 94, 132102/1–132102/3.

67 Di Valentin, C., Pacchioni, G., Onishi, H., and Kudo, A. (2009) Cr/Sbco-doped TiO2 from first principles calculations. Chem. Phys. Lett., 469,166–171.

68 Chadwick, N.P., Glover, E.N.K., Sathasivam, S., Basahel, S.N., Althabaiti, S.A.,Alyoubi, A.O., Parkin, I.P., and Carmalt, C.J. (2016) Photo-activity and lowresistivity in N/Nb Co-doped TiO2 thin films by combinatorial AACVD.J. Mater. Chem. A, 4, 407–415.

69 (a) Sun, L., Li, J., Wang, C.L., Li, S.F., Chen, H.B., and Lin, C.J. (2009) Anelectrochemical strategy of doping Fe3+ into TiO2 nanotube array films forenhancement in photocatalytic activity. Sol. Energy Mater. Sol. Cells, 93,1875–1880; (b) Asilturk, M., SayIlkan, F., and Arpa, E. (2009) Effect of Fe3+

ion doping to TiO2 on the photocatalytic degradation of malachite green


dye under UV and vis-irradiation. J. Photochem. Photobiol., A, 203, 64–71;(c) Ambrus, Z., Balazs, N., Alapi, T., Wittmann, G., Sipos, P., Dombi, A., andMogyorosi, K. (2008) Synthesis, structure and photocatalytic properties ofFe(III)-doped TiO2 prepared from TiCl3. Appl. Catal., B, 81, 27–37.

70 Ranjit, K.T., Willner, I., Bossmann, S.H., and Braun, A.M. (2001) Lanthanideoxide-doped titanium dioxide photocatalysts: novel photocatalysts for theenhanced degradation of p-chlorophenoxyacetic acid. Environ. Sci. Technol.,35, 1544–1549.

71 Periyat, P.K., Baiju, V., Mukundan, P., Pillai, P.K., and Warrier, K.G.K. (2007)Aqueous colloidal sol–gel route to synthesize nanosized ceria-doped titaniahaving high surface area and increased anatase phase stability. J. Sol–Gel Sci.Technol., 43, 299–304.

72 LeDuc, C.A., Campbell, J.M., and Rosssin, J.A. (1996) Effect of lanthana asa stabilizing agent in titanium dioxide support. Ind. Eng. Chem. Res., 35,2473–2476.

73 Sibu, C.P., Kumar, S.R., Mukundan, P., and Warrier, K.G.K. (2002) Structuralmodifications and associated properties of lanthanum oxide doped sol−gelnanosized titanium oxide. Chem. Mater., 14, 2876–2881.

74 Peng, T., Zhao, D., Sing, H., and Yan, C. (2005) Preparation oflanthana-doped titania nanoparticles with anatase mesoporous walls andhigh photocatalytic activity. J. Mol. Catal. A: Chem., 238, 119–126.

75 Xie, Y. and Yuan, C. (2004) Photocatalysis of neodymium ion modified TiO2sol under visible light irradiation. Appl. Surf. Sci., 221, 17–24.

76 Lin, J. and Yu, J.C. (1998) An investigation on photocatalytic activities ofmixed TiO2-rare earth oxides for the oxidation of acetone in air. J. Pho-tochem. Photobiol., A, 116, 63–67.

77 Teh, C.M. and Mohamed, A.R. (2011) Roles of titanium dioxide andion-doped titanium dioxide on photocatalytic degradation of organic pol-lutants (phenolic compounds and dyes) in aqueous solutions: a review.J. Alloys Compd., 509, 1648–1660.

78 Hidalgo, M.C., Murcia, J.J., Navío, J.A., and Colón, G. (2011) Photodeposi-tion of gold on titanium dioxide for photocatalytic phenol oxidation. Appl.Catal., A, 397, 112–120.

79 Maicu, M., Hidalgo, M.C., Colón, G., and Navío, J.A. (2011) Comparativestudy of the photodeposition of Pt, Au and Pd on pre-sulphated TiO2 forthe photocatalytic decomposition of phenol. J. Photochem. Photobiol., A, 217,275–283.

80 Hishita, S., Mutoh, I., Koumoto, K., and Yanagida, H. (1983) Inhibitionmechanism of the anatase-rutile phase transformation by rare earth oxides.Ceram. Int., 9, 61–67.

81 Zhao, D., Peng, M., Liu, L., Lu, P., and Cai (2008) Fabrication, characteri-zation and photocatalytic activity of Gd3+-doped titania nanoparticles withmesostructure. Microporous Mesoporous Mater., 114, 166–174.

82 Zhang, Y., Zhang, H., Xu, Y., and Wang, Y. (2004) Significant effect of lan-thanide doping on the texture and properties of nanocrystalline mesoporousTiO2. J. Solid State Chem., 177, 3490–3498.

References 319

83 Baiju, K.V., Periyat, P., Shajesh, P., Wunderlich, W., Manjumol, K.A.,Smitha, V.S., Jaimy, K.B., and Warrier, K.G.K. (2010) Mesoporous gadolin-ium doped titania photocatalyst through an aqueous sol–gel method. J.Alloys Compd., 505, 194–200.

84 El-Bahy, Z.M., Ismail, A.A., and Mohamed, R.M. (2009) Enhancement oftitania by doping rare earth for photodegradation of organic dye (directblue). J. Hazard. Mater., 166, 138–143.

85 Fan, C., Xue, P., and Sun, Y. (2006) Preparation of nano-TiO2 doped withcerium and its photocatalytic activity. J. Rare Earths, 24, 309–313.

86 Shi, J.W., Zheng, J.T., and Wu, P. (2009) Preparation, characterization andphotocatalytic activities of holmium-doped titanium dioxide nanoparticles.J. Hazard. Mater., 161, 416–422.

87 Burns, A., Hayes, G., Li, W., Hirvonen, J., Demaree, J.D., and Shah, S.I.(2004) Neodymium ion dopant effects on the phase transformation insol–gel derived titania nanostructures. Mater. Sci. Eng., B, 111, 150–155.

88 Di Valentin, C. and Fittipaldi, D. (2013) Hole scavenging by organic adsor-bates on the TiO2 surface: a DFT model study. J. Phys. Chem. Lett., 4,1901–1906.

89 Rao, C.N.R. and Rao, K.J. (1978) Phase Transition in Solids, McGraw-Hill,New York.

90 Zhang, H. and Banfield, J.F. (1998) Thermodynamic analysis of phase stabil-ity of nanocrystalline titania. J. Mater. Chem., 8, 2073–2076.

91 Stafford, U., Gray, K.A., Kamat, P.V., and Varma, A. (1993) An in situdiffuse reflectance FTIR investigation of photocatalytic degradation of4-chlorophenol on a TiO2 powder surface. Chem. Phys. Lett., 205, 55–61.

92 Riegel, G. and Bolton, J.R. (1995) Photocatalytic efficiency variability in TiO2particles. J. Phys. Chem., 99, 4215–4224.

93 Machida, M., Norimoto, W.K., and Kimura, T. (2005) Antibacterial activityof photocatalytic titanium dioxide thin films with photodeposited silver onthe surface of sanitary ware. J. Am. Ceram. Soc., 88, 95–100.

94 Hashimoto, K., Irie, H., and Fujishima, A. (2005) TiO2 photocatalysis:a historical overview and future prospects. Jpn. J. Appl. Phys., Part 1, 44,8269–8285.

95 Zhang, X.T., Sato, O., Taguchi, M., Einaga, Y., Murakami, T., andFujishima, A. (2005) Self-cleaning particle coating with antireflection proper-ties. Chem. Mater., 17, 696–700.

96 Kamat, P.V. (1993) Photochemistry on nonreactive and reactive (semiconduc-tor) surfaces. Chem. Rev., 93, 267–300.

97 Kominami, H., Kohno, M., Matsunaga, Y., and Kera, Y. (2001) Thermaldecomposition of titanium alkoxide and silicate ester in organic solvent:a new method for synthesizing large-surface-area, silica-modified tita-nium(IV) oxide of high thermal stability. J. Am. Ceram. Soc., 84, 1178–1180.

98 Okada, K., Yamamoto, N., Kameshima, Y., and Yasumori, Y. (2001) Effect ofsilica additive on the anatase-to-rutile phase transition. J. Am. Ceram. Soc.,84, 1591–1596.


99 Kumar, S.R., Suresh, C., Vasudevan, A.K., Suja, N.R., Mukundan, P., andWarrier, K.G.K. (1999) Phase transformation in sol–gel titania containingsilica. Mater. Lett., 38, 161–166.

100 Yang, Y. and Ferreira, J.M.F. (1998) Inhibitory effect of the Al2O3—SiO2mixed additives on the anatase–rutile phase transformation. Mater. Lett., 36,320–324.

101 Jung, K.Y. and Pask, S.B. (2000) Enhanced photoactivity of silica-embeddedtitania particles prepared by sol–gel process for the decomposition oftrichloroethylene. Appl. Catal., B, 25, 249–256.

102 Anderson, C. and Bard, A.J. (1995) An improved photocatalyst of TiO2/SiO2prepared by a sol–gel synthesis. J. Phys. Chem., 99, 9882–9885.

103 Anderson, C. and Bard, A.J. (1997) Improved photocatalytic activity andcharacterization of mixed TiO2/SiO2 and TiO2/Al2O3 materials. J. Phys.Chem. B, 101, 2611–2616.

104 Mabakazu, A., Kawamura, T., and Kodama, S. (1998) Photocatalysis ontitanium–aluminum binary metal oxides: enhancement of the photocatalyticactivity of titania species. J. Phys. Chem., 92, 438–440.

105 Fu, X., Clark, L.A., Yang, Q., and Anderson, M.A. (1996) Enhanced photo-catalytic performance of titania-based binary metal oxides: TiO2/SiO2 andTiO2/ZrO2. Environ. Sci. Technol., 30, 647–653.

106 Sivakumar, S., Sibu, C., Mukundan, P., Krishna Pillai, P., and Warrier, K.G.K.(2004) Nanoporous titania–alumina mixed oxides – an alkoxide free sol–gelsynthesis. Mater. Lett., 58, 2664–2669.

107 Attia, A., Zukalova, M., Rathousky, J., Zukal, A., and Kavan, L. (2005) Meso-porous electrode material from alumina-stabilized anatase TiO2 for lithiumion batteries. J. Solid State Electrochem., 9, 138–145.

108 (a) Francisco, M.S.P. and Mastelaro, V.R. (2002) Inhibition of the anatase-rutile phase transformation with addition of CeO2 to CuO—TiO2 system:Raman spectroscopy, X-ray diffraction, and textural studies. Chem. Mater.,14, 2514–2518; (b) Bokhimi, X., Morales, A., and Novaro, O. (1997) Effectof copper precursor on the stabilization of titania phases, and the opticalproperties of Cu/TiO2 prepared with the sol−gel technique. Chem. Mater., 9,2616–2620.

109 (a) Tian, B., Li, C., Gu, F., Jiang, H., Hu, Y., and Zhang, J. (2009) Flamesprayed V-doped TiO2 nanoparticles with enhanced photocatalytic activityunder visible light irradiation. Chem. Eng. J., 151, 220–227; (b) Doong, R.A.,Chang, P.Y., and Huang, C.H. (2009) Microstructural and photocatalyticproperties of sol–gel-derived vanadium-doped mesoporous titanium dioxidenanoparticles. J. Non-Cryst. Solids, 355, 2302–2308.

110 Nair, J., Nair, P., Mizukami, F., Oosawa, Y., and Okubo, T. (1999) Microstruc-ture and phase transformation behavior of doped nanostructured titania.Mater. Res. Bull., 34, 1275–1290.

111 (a) Seery, M.K., George, R., Floris, P., and Pillai, S.C. (2007) Silverdoped titanium dioxide nanomaterials for enhanced visible light pho-tocatalysis. J. Photochem. Photobiol., A, 189, 258–263; (b) Tomas, S.A.,Luna-Resendis, A., Cortes-Cuautli, L.C., and Jacinto, D. (2009) Optical and

References 321

morphological characterization of photocatalytic TiO2 thin films doped withsilver. Thin Solid Films, 518, 1337–1340.

112 Chen, R.F., Zhang, C.X., Deng, J., and Song, G.Q. (2009) Preparation andphotocatalytic activity of Cu2+-doped TiO2/SiO2. Int. J. Miner. Metall.Mater., 16, 220–225.

113 Yang, Y., Wang, H., Li, X., and Wang, C. (2009) Electrospun mesoporousW6+-doped TiO2 thin films for efficient visible-light photocatalysis. Mater.Lett., 63, 331–333.

114 Umar, K., Aris, A., Ahmad, H., Parveen, T., Jaafar, J., Majid, Z.A.,Reddy, A.V.B., and Talib, J. (2016) Synthesis of visible light active dopedTiO2 for the degradation of organic pollutants-methylene blue andglyphosate. J. Anal. Sci. Technol., 7, 29–32.

115 Devi, L.G. and Murthy, N. (2008) Characterization of Mo doped TiO2 andits enhanced photo catalytic activity under visible light. Catal. Lett., 125,320–330.

116 Quinonesa, D.H., Reya, A., Alvareza, P.M., Beltrana, F.J., and Pumab, G.L.(2015) Boron doped TiO2 catalysts for photocatalytic ozonation of aque-ous mixtures of common pesticides: diuron, o-phenylphenol, MCPA andterbuthylazine. Appl. Catal., B, 178, 74–81.

117 Thomas, J., Kumar, K.P., and Chitra, K.R. (2011) Synthesis of Ag doped nanoTiO2 as efficient solar photocatalyst for the degradation of endosulfan. Adv.Sci. Lett., 4, 108–114.

118 Chiang, L.F. and Doong, R. (2014) Cu–TiO2 nanorods with enhancedultraviolet- and visible-light photoactivity for bisphenol a degradation.J. Hazard. Mater., 277, 84–92.

119 Zhao, W., Ma, W., Chen, C., Zhao, J., and Shuai, Z. (2004) Efficient degra-dation of toxic organic pollutants with Ni2O3/TiO2−xBx under visibleirradiation. J. Am. Chem. Soc., 126, 4782–4783.

120 Yuan, Z.-H., Jia, J.-H., and Zhang, L.-H. (2002) Influence of co-doping ofZn(II)+ Fe(III) on the photocatalytic activity of TiO2 for phenol degradation.Mater. Chem. Phys., 73, 323–332.

121 Fujishima, A., Rao, T.N., and Tryk, D.A. (2000) Titanium dioxide photocatal-ysis. J. Photochem. Photobiol., C, 1, 1–21.

122 Zhang, Y., Zhang, N., Tang, Z.R., and Xu, Y. (2012) Improving the photo-catalytic performance of graphene–TiO2 nanocomposites via a combinedstrategy of decreasing defects of graphene and increasing interfacial contact.Phys. Chem. Chem. Phys., 14, 9167–9175.

123 Soja, G.R. and Watson, D.F. (2009) TiO2-catalyzed photodegradation of por-phyrins: mechanistic studies and application in monolayer photolithography.Langmuir, 25 (9), 5398–5403.

124 Kamil, A.M., Hussein, F.H., Halbus, A.F., and Bahnemann, D.W. (2014)Preparation, characterization, and photocatalytic applications ofMWCNTs/TiO2 composite. Int. J. Photoenergy, 2014, 1–8.

125 Kumar, R., Omaish Ansari, M., Parveen, N., Oves, M., Barakat, M.A.,Alshahri, A., Yusuf Khan, M., and Hwan Cho, M. (2016) Facile routeto a conducting ternary polyaniline@TiO2/GN nanocomposite for


environmentally benign applications: photocatalytic degradation of pollutantsand biological activity. RSC Adv., 6, 111308–111317.

126 Marcano, D.C., Kosynkin, D.V., Berlin, J.M., Sinitskii, A., Sun, Z.Z.,Slesarev, A., Alemany, L.B., Lu, W., and Tour, J.M. (2010) Improved syn-thesis of graphene oxide. ACS Nano, 4, 4806–4814.

127 Liang, Y.Y., Wang, H.L., Casalongue, H.N.S.C., Chen, Z., and Dai, H.J. (2010)TiO2 nanocrystals grown on graphene as advanced photocatalytic hybridmaterials. Nano Res., 3, 701–705.

128 Lui, G., Liao, J.Y., Duan, A., Zhang, Z., and Yu, A. (2013) Graphene-wrappedhierarchical TiO2 nanoflower composites with enhanced photocatalyticperformance. J. Mater. Chem. A, 1, 12255–12262.

129 Pastrana-Martínez, L.M., Morales-Torres, S., Likodimos, V., Figueiredo, J.L.,Faria, J.L., Falaras, P., and Silva, A.M.T. (2012) Advanced nanostructuredphotocatalysts based on reduced grapheneoxide–TiO2 composites for degra-dation of diphenhydramine pharmaceutical and methyl orange dye. Appl.Catal., B, 123–124, 241–256.

130 Mohamed, R.M. (2012) UV-assisted photocatalytic synthesis ofTiO2-reduced graphene oxide with enhanced photocatalytic activity indecomposition of sarin in gas phase. Desalin. Water Treat., 50 (1–3),147–156.

131 Verma, R., Samdarshi, S.K., Kalpana, S., and Konwar, B.K. (2017) Nanos-tructured bi-phasic TiO2 nanoparticles grown on reduced graphene oxidewith high visible light photocatalytic detoxification. Mater Chem. Phys., 186,202–211.

132 Verma, R., Samdarshi, S.K., Sagar, K., and Konwar, B.K. (2017) Nanostruc-tured bi-phasic TiO2 nanoparticles grown on reduced graphene oxide withhigh visible light photocatalytic detoxification. Mater. Chem. Phys., 186,202–211.

133 Selim Arif Sher Shah, M., Reum Park, A., Zhang, K., Park, J.H., and Yoo, P.J.(2012) Green synthesis of biphasic TiO2-reduced graphene oxide nanocom-posites with highly enhanced photocatalytic activity. ACS Appl. Mater.Interfaces, 4 (8), 3893–3901.

134 Yu, H., Xiao, P., Tian, J., Wang, F., and Yu, J. (2016)Phenylamine-functionalized rGO/TiO2 photocatalysts: spatially separatedadsorption sites and tunable photocatalytic selectivity. ACS Appl. Mater.Interfaces, 8, 29470.

135 Wang, M., Han, J., Xiong, H., Guo, R., and Yin, Y. (2015) Nanostructuredhybrid shells of r-GO/AuNP/m-TiO2 as highly active photocatalysts. ACSAppl. Mater. Interfaces, 7, 6909.

136 Leong, K.H., Sim, L.C., Bahnemann, D., Jang, M., Ibrahim, S., andSaravanan, P. (2015) Reduced graphene oxide and Ag wrapped TiO2 pho-tocatalyst for enhanced visible light photocatalysis. APL Mater., 3, 104503.

137 Gao, W., Wang, M., Ran, C., and Li, L. (2015) Facile one-pot synthesisof MoS2 quantum dots–graphene–TiO2 composites for highly enhancedphotocatalytic properties. Chem. Commun., 51, 1709–1712.

138 Fagan, R., McCormack, D.E., Dionysiou, D.D., and Pillai, S.C. (2016) Areview of solar and visible light active TiO2 photocatalysis for treatingbacteria, cyanotoxins and contaminants of emerging concern. Mater. Sci.Semicond. Process., 42, 2–14.

References 323

139 Spasiano, D., Marotta, R., Malato, S., Fernandez-Ibañez, P., and Di Somma, I.(2015) Solar photocatalysis: materials, reactors, some commercial, andpre-industrialized applications: a comprehensive approach. Appl. Catal., B,170-171, 90–123.

140 BCC Research (2015) Market Research Report: Photocatalysts: Technolo-gies and Global Markets, http://www.bccresearch.com/market-research/advanced-materials/photocatalysts-technologies-markets-report-avm069b.html (accessed 19 March 2017).

141 Nishimoto, S. and Bhushan, B. (2013) Bioinspired self-cleaning surfaces withsuperhydrophobicity, superoleophobicity, and superhydrophilicity. RSC Adv.,3, 671–690.

142 Midtdal, K. and Jelle, B.P. (2013) Self-cleaning glazing products: astate-of-the-art review and future research pathways. Sol. Energy Mater.Sol. Cells, 109, 126–141.

143 Cebeci, F.Ç., Wu, Z., Zhai, L., Cohen, R.E., and Rubner, M.F. (2006)Nanoporosity-driven superhydrophilicity: a means to create multifunctionalantifogging coatings. Langmuir, 22, 2856–2862.

144 Wang, J.-J., Wang, D.-S., Wang, J., Zhao, W.-L., and Wang, C.-W. (2011)High transmittance and superhydrophilicity of porous TiO2/SiO2 bi-layerfilms without UV irradiation. Surf. Coat. Technol., 205, 3596–3599.

145 Bayati, M.R., Molaei, R., and Golestani-Fard, F. (2011) Enhancing photoin-duced hydrophilicity of micro arc oxidized TiO2 nanostructured porouslayers by V-doping. Colloids Surf., A, 373, 51–60.

146 Wang, R., Hashimoto, K., Fujishima, A., Chikuni, M., Kojima, E.,Kitamura, A., Shimohigoshi, M., and Watanabe, T. (1997) Light-inducedamphiphilic surfaces. Nature, 388, 431–432.

147 Sakai, N., Fujishima, A., Watanabe, T., and Hashimoto, K. (2003) Quanti-tative evaluation of the photoinduced hydrophilic conversion properties ofTiO2 thin film surfaces by the reciprocal of contact angle. J. Phys. Chem. B,107, 1028–1035.

148 Takeuchi, M., Sakamoto, K., Martra, G., Coluccia, S., and Anpo, M. (2005)Mechanism of photoinduced superhydrophilicity on the TiO2 photocatalystsurface. J. Phys. Chem. B, 109, 15422–15428.

149 Pilkington http://www.pilkington.com/products/bp/bybenefit/selfcleaning/activ/home.htm (accessed 19 March 2017).

150 TOTO LTD Super-Hydrophilic Photocatalyst And Its Applications, http://www.toto.co.jp/hydro_e/ (accessed 19 March 2017).

151 Zhang, L., Dillert, R., Bahnemann, D., and Vormoor, M. (2012)Photo-induced hydrophilicity and self-cleaning: models and reality. EnergyEnviron. Sci., 5, 7491–7507.

152 Italcementi (2009) Research and Innovation TX Active, http://www.italcementigroup.com/ENG/Research+and+Innovation/Innovative+Products/TX+Active/ (accessed 19 March 2017).

153 HEIDELBERGCEMENT (2010) TioCem® for the Reduction of Air Pollu-tants, http://www.heidelbergcement.de/de/tiocem-en (accessed 19 March2017).

154 Mitsubishi Materials http://www.mmc.co.jp/corporate/en/csr/environment/products.html (accessed 19 March 2017).


155 CEAM Euphore Home Page, http://www.ceam.es/WWWEUPHORE/home.htm (accessed 19 March 2017).

156 Zhang, P., Tian, J., Xu, R., and Ma, G. (2013) Hydrophilicity, photocatalyticactivity and stability of tetraethyl orthosilicate modified TiO2 film on glazedceramic surface. Appl. Surf. Sci., 266, 141–147.

157 Periyat, P., Pillai, S.C., McCormack, D.E., Colreavy, J., and Hinder, S.J. (2008)Improved high-temperature stability and Sun-light-driven photocatalyticactivity of sulfur-doped anatase TiO2. J. Phys. Chem. C, 112, 7644–7652.

158 Etacheri, V., Seery, M.K., Hinder, S.J., and Pillai, S.C. (2011) Oxygen richtitania: a dopant free, high temperature stable, and visible-light activeanatase photocatalyst. Adv. Funct. Mater., 21, 3744–3752.

159 Chen, J. and Poon, C.-S. (2009) Photocatalytic construction and build-ing materials: from fundamentals to applications. Build. Environ., 44,1899–1906.

160 Kikuchi, Y., Sunada, K., Iyoda, T., Hashimoto, K., and Fujishima, A. (1997)Photocatalytic bactericidal effect of TiO2 thin films: dynamic view of theactive oxygen species responsible for the effect. J. Photochem. Photobiol., A,106, 51–56.

161 Podporska-Carroll, J., Panaitescu, E., Quilty, B., Wang, L., Menon, L., andPillai, S.C. (2015) Antimicrobial properties of highly efficient photocatalyticTiO2 nanotubes. Appl. Catal., B, 176-177, 70–75.

162 Sun, L., Qin, Y., Cao, Q., Hu, B., Huang, Z., Ye, L., and Tang, X. (2011)Novel photocatalytic antibacterial activity of TiO2 microspheres exposing100% reactive {111} facets. Chem. Commun., 47, 12628–12630.

163 Rengifo-Herrera, J.A., Pierzchała, K., Sienkiewicz, A., Forró, L., Kiwi, J.,and Pulgarin, C. (2009) Abatement of organics and Escherichia coli by N, SCo-doped TiO2 under UV and visible light. Implications of the formation ofsinglet oxygen (1O2) under visible light. Appl. Catal., B, 88, 398–406.

164 Rengifo-Herrera, J.A. and Pulgarin, C. (2010) Photocatalytic activity of N, SCo-doped and N-doped commercial anatase TiO2 powders towards phenoloxidation and E. Coli inactivation under simulated solar light irradiation. Sol.Energy, 84, 37–43.

165 Ananpattarachai, J., Boonto, Y., and Kajitvichyanukul, P. (2016) Visible lightphotocatalytic antibacterial activity of Ni-doped and N-doped TiO2 onStaphylococcus aureus and Escherichia coli bacteria. Environ. Sci. Pollut. Res.Int., 23, 4111–4119.

166 Li, H., Cui, Q., Feng, B., Wang, J., Lu, X., and Weng, J. (2013) Antibacterialactivity of TiO2 nanotubes: influence of crystal phase, morphology and Agdeposition. Appl. Surf. Sci., 284, 179–183.

167 Dunnill, C.W., Ansari, Z., Kafizas, A., Perni, S., Morgan, D.J., Wilson, M.,and Parkin, I.P. (2011) Visible light photocatalysts-N-doped TiO2 by sol–gel,enhanced with surface bound silver nanoparticle islands. J. Mater. Chem., 21,11854–11861.

168 Yu, J.C., Ho, W., Yu, J., Yip, H., Wong, P.K., and Zhao, J. (2005) Efficientvisible-light-induced photocatalytic disinfection on sulfur-doped nanocrys-talline titania. Environ. Sci. Technol., 39, 1175–1179.

References 325

169 Etacheri, V., Michlits, G., Seery, M.K., Hinder, S.J., and Pillai, S.C. (2013)A highly efficient TiO2−xCx nano-heterojunction photocatalyst for visiblelight induced antibacterial applications. ACS Appl. Mater. Interfaces, 5,1663–1672.

170 Chang, Y.-N., Ou, X.-M., Zeng, G.-M., Gong, J.-L., Deng, C.-H., Jiang, Y.,Liang, J., Yuan, G.-Q., Liu, H.-Y., and He, X. (2015) Synthesis of magneticgraphene oxide–TiO2 and their antibacterial properties under solar irradia-tion. Appl. Surf. Sci., 343, 1–10.

171 Cao, B., Cao, S., Dong, P., Gao, J., and Wang, J. (2013) High antibacterialactivity of ultrafine TiO2/graphene sheets nanocomposites under visible lightirradiation. Mater. Lett., 93, 349–352.

172 Jia, L., Qiu, J., Du, L., Li, Z., Liu, H., and Ge, S. (2017) TiO2 nanorod arraysas a photocatalytic coating enhanced antifungal and antibacterial efficiencyof Ti substrates. Nanomedicine, 12, 761–776.

173 Choi, H., Antoniou, M.G., Pelaez, M., de la Cruz, A.A., Shoemaker, J.A., andDionysiou, D.D. (2007) Mesoporous nitrogen-doped TiO2 for the photocat-alytic destruction of the cyanobacterial toxin microcystin-LR under visiblelight irradiation. Environ. Sci. Technol., 41, 7530–7535.

174 Zhao, C., Pelaez, M., Dionysiou, D.D., Pillai, S.C., Byrne, J.A., andO’Shea, K.E. (2014) UV and visible light activated TiO2 photocatalysisof 6-hydroxymethyl uracil, a model compound for the potent cyanotoxincylindrospermopsin. Catal. Today, 224, 70–76.

175 Pelaez, M., Falaras, P., Likodimos, V., Kontos, A. G., de la Cruz, A. A.,O’Shea, K. and Dionysiou, D. D. (2010) Synthesis, structural charac-terization and evaluation of sol–gel-based NF-TiO2 films with visiblelight-photoactivation for the removal of microcystin-LR. Appl. Catal., B,99, 378–387.

176 Liu, G., Han, C., Pelaez, M., Zhu, D., Liao, S., Likodimos, V., Kontos, A.G.,Falaras, P., and Dionysiou, D.D. (2013) Enhanced visible light photocatalyticactivity of CN-codoped TiO2 films for the degradation of microcystin-LR. J.Mol. Catal. A: Chem., 372, 58–65.

177 Han, C., Pelaez, M., Likodimos, V., Kontos, A.G., Falaras, P., O’Shea, K., andDionysiou, D.D. (2011) Innovative visible light-activated sulfur doped TiO2films for water treatment. Appl. Catal., B, 107, 77–87.

178 Pelaez, M., Falaras, P., Kontos, A.G., de la Cruz, A.A., O’Shea, K.,Dunlop, P.S.M., Byrne, J.A., and Dionysiou, D.D. (2012) A comparative studyon the removal of cylindrospermopsin and microcystins from water withNF-TiO2-P25 composite films with visible and UV-Vis light photocatalyticactivity. Appl. Catal., B, 121, 30–39.

179 Zhang, J., Li, S., Chen, L., Pan, Y., and Yang, S. (2012) The progress of TiO2photocatalyst coating. IOSR J. Eng., 2, 50–53.

180 Keller, N., Ducamp, M.-N., Robert, D., and Keller, V. (2013) Ethyleneremoval and fresh product storage: a challenge at the frontiers of chem-istry: toward an approach by photocatalytic oxidation. Chem. Rev., 113,5029–5070.


181 Fernández-García, A., Zarza, E., Valenzuela, L., and Pérez, M. (2010)Parabolic-trough solar collectors and their applications. J. Renewable Sus-tainable Energy Rev., 14, 1695–1721.

182 Abdel-Maksoud, Y., Imam, E., and Ramadan, A. (2016) TiO2 solar pho-tocatalytic reactor systems: selection of reactor design for scale-up andcommercialization-analytical review. Catalysts, 6, 138.

183 Alrousan, D.M.A., Polo-López, M.I., Dunlop, P.S.M., Fernández-Ibáñez, P.,and Byrne, J.A. (2012) Solar photocatalytic disinfection of water with immo-bilised titanium dioxide in re-circulating flow CPC reactors. Appl. Catal., B,128, 126–134.

184 Oelgemoller, M., Jung, C., Ortner, J., Mattay, J., and Zimmermann, E. (2005)Green photochemistry: solar photooxygenations with medium concentratedsunlight. Green Chem., 7, 35–38.

185 Malato, S., Blanco, J., Vidal, A., and Richter, C. (2002) Photocatalysis withsolar energy at a pilot-plant scale: an overview. Appl. Catal., B, 37, 1–15.

186 Dillert, R., Cassano, A.E., Goslich, R., and Bahnemann, D. (1999) Largescale studies in solar catalytic wastewater treatment. Catal. Today, 54,267–282.

187 Guillard, C., Disdier, J., Monnet, C., Dussaud, J., Malato, S., Blanco, J.,Maldonado, M.I., and Herrmann, J.-M. (2003) Solar efficiency of a newdeposited titania photocatalyst: chlorophenol, pesticide and dye removalapplications. Appl. Catal., B, 46, 319–332.

188 Malato, S., Blanco, J., Richter, C., Curcó, D., and Gimenez, J. (1997)Low-concentrating CPC collectors for photocatalytic water detoxifica-tion: comparison with a medium concentrating solar collector. Water Sci.Technol., 35, 157–164.

189 García-Montaño, J., Pérez-Estrada, L., Oller, I., Maldonado, M.I.,Torrades, F., and Peral, J. (2008) Pilot plant scale reactive dyes degrada-tion by solar photo-Fenton and biological processes. J. Photochem. Photobiol.,A, 195, 205–214.

190 Oller, I., Gernjak, W., Maldonado, M.I., Pérez-Estrada, L.A.,Sánchez-Pérez, J.A., and Malato, S. (2006) Solar photocatalytic degradationof some hazardous water-soluble pesticides at pilot-plant scale. J. Hazard.Mater., 138, 507–517.

191 Sarria, V., Kenfack, S., Guillod, O., and Pulgarin, C. (2003) An innovativecoupled solar-biological system at field pilot scale for the treatment ofbiorecalcitrant pollutants. J. Photochem. Photobiol., A, 159, 89–99.

192 Gernjak, W., Maldonado, M.I., Malato, S., Cáceres, J., Krutzler, T., Glaser, A.,and Bauer, R. (2004) Pilot-plant treatment of olive mill wastewater (OMW)by solar TiO2 photocatalysis and solar photo-Fenton. Sol. Energy, 77,567–572.

193 Zapata, A., Oller, I., Sirtori, C., Rodríguez, A., Sánchez-Pérez, J.A., López, A.,Mezcua, M., and Malato, S. (2010) Decontamination of industrial wastew-ater containing pesticides by combining large-scale homogeneous solarphotocatalysis and biological treatment. Chem. Eng. J., 160, 447–456.

194 Oller, I., Malato, S., Sánchez-Pérez, J.A., Maldonado, M.I., Gernjak, W.,Pérez-Estrada, L.A., Muñoz, J.A., Ramos, C., and Pulgarín, C. (2007)

References 327

Pre-industrial-scale combined solar photo-Fenton and immobilized biomassactivated-sludge biotreatment. Ind. Eng. Chem. Res., 46, 7467–7475.

195 Blanco, J., Malato, S., Fernández, P., Vidal, A., Morales, A., Trincado, P.,Oliveira, J.C., Minero, C., Musci, M., Casalle, C., Brunotte, M., Tratzky, S.,Dischinger, N., Funken, K.H., Sattler, C., Vincent, M., Collares-Pereira, M.,Mendes, J.F., and Rangel, C.M. (1999) Compound parabolic concentratortechnology development to commercial solar detoxification applications. Sol.Energy, 67, 317–330.

196 Dondi, D., Protti, S., Albini, A., Carpio, S.M., and Fagnoni, M. (2009) Syn-thesis of 𝛾-lactols, 𝛾-lactones and 1,4-monoprotected succinaldehydes undermoderately concentrated sunlight. Green Chem., 11, 1653–1659.

197 Blanco-Galvez, J., Fernández-Ibáñez, P., and Malato-Rodríguez, S. (2006)Solar photocatalytic detoxification and disinfection of water: recentoverview. J. Sol. Energy Eng., 129, 4–15.

198 Ishihara Sangyo Kaisha, Ltd Contact, http://www.iskweb.co.jp/eng/contact/index.html (accessed 19 March 2017).

199 Hyosung Ebara Co., Ltd http://hyosungeba.tradekorea.com/, (accessed 19March 2017).

200 CLEARWATER Industries, Inc http://www.clearwaterind.com/ (accessed 19March 2017).

201 Lynntech http://lynntech.com/ (accessed 19 March 2017).202 Purifics Complete Water Purification Using Ceramic Membrane Technolo-

gies, http://www.purifics.com/photo-cat-photocatalytic-membrane-system(accessed 19 March 2017).

329

12

Metal-Free Organic Semiconductors for Visible-Light-ActivePhotocatalytic Water SplittingS. T. Nishanthi, Battula Venugopala Rao, andKamalakannan Kailasam

Habitat Centre, Institute of Nano Science & Technology, Sector 64, Phase X, 160062 Mohali, Punjab, India

12.1 Introduction

With the rapid growth of the population and the fast development of economy,driving the global energy system into a sustainable path is progressively becom-ing a major concern and worldwide policy objective [1, 2]. It is crucial to developalternative “clean and renewable” energy sources with urgency in the comingyears. Hydrogen is considered to be one of the most promising alternative energycarriers. One of the most economical approaches to address these problems is touse photocatalytic water-splitting process, utilizing solar energy to split waterinto clean hydrogen and oxygen, which is an ideal way that does not emit anyharmful products [3, 4]. The key to achieving solar hydrogen production is todevelop stable, efficient, and inexpensive photocatalysts which should be activein visible light especially should split water in natural sunlight. But still there aremany material-related issues that hinder its widespread use to develop a suit-able photocatalyst [5]. According to the thermodynamics, there are three mainfeatures involved in water-splitting reaction. One, it is a highly endothermic pro-cess. Second, the change in Gibb’s free energy is positive (ΔG∘ = 238 kJ mol−1).Third, it is a four electron process that requires 1.23 V potential [1, 2]. The fun-damental equation that is showing the required energy to produce hydrogen andoxygen from water is as follows:

2H2O −−−−→ 4H+ + O2 +1.23 V versus NHE4H+ −−−−→ 2H2 0 V versus NHE

2H2O −−−−→ 2H2 + O2

Therefore, according to these equations, the basic requirements for a semicon-ductor to split water are suitable bandgap (more than 1.23 V) and the conductionband minima should be more negative than the reduction potential of H+/H2(0 V vs NHE), the valence band maxima should be more positive than reductionpotential of O2/H2O (1.23 V vs NHE).The solar-light-driven water splitting usingg-CN is shown in Scheme 12.1. It is particularly difficult to obtain a simple,


330 12 Metal-Free Organic Semiconductors for Visible-Light-Active Photocatalytic Water Splitting

CB

H2

H+

NH NH2

NH2

NN

NN

N N

N

N

N

N NH

N N N N

N N

NN N N N N N N

N

N

N N

N N

N

N N

N

N

N

N

N

N

O2

H2O

e–

h+ h+ h+

e– e–

VB

Scheme 12.1 Solar-light-driven water splitting using g-CN.

cost–effective, and highly active semiconductor material that satisfies all thecrucial requirements:

1) The ideal bandgap should be 1.5–3.2 eV for effective utilization of the solarspectrum.

2) The conduction band minimum (CBM) should be more negative than thewater reduction (H+/H2) level and the valence band maximum (VBM) shouldbe more positive than the water oxidation (O2/H2O) level with respect to thepotential of NHE.

3) Stability in both acidic and basic medium.4) Photocorrosion against resistance.5) The surface area, porosity, or reactive facets should be larger for higher active

sites [6, 7].

In addition to all these requirements, a photocatalyst has to suppress therecombination of photoinduced charge carriers generated during the reactionand enhance the photocatalytic process. In 1972, Fujishima and Honda [8] ignitedthe overall photocatalytic water splitting using TiO2 electrodes under ultraviolet(UV) light which has been considered as the landmark event in the photo-catalytic field. However, the main drawback in TiO2 photocatalyst possessingwider bandgap (3.2 eV) and faster recombination rates of photoinduced chargecarriers. It should be noted that still more efforts are being made to achievewater splitting in the visible light using modified TiO2 as photocatalyst. Thedesign of effective photocatalyst is important to achieve the higher efficiencyfor hydrogen generation and to transfer the process to large-scale hydrogengeneration realization.

In this chapter, we briefly overviewed the recent research in visible-light-drivenphotocatalytic water splitting mainly through organic polymeric semiconductormaterials. First, the necessity of a photocatalyst for achieving water splitting with

12.2 Organic Semiconductors for Photocatalytic Water Splitting 331

drawbacks from metal-based systems is mentioned followed by a brief descrip-tion on the organic semiconductor for photocatalytic water splitting. Second,the designing concept of graphitic carbon nitrides for visible-light-driven watersplitting is reviewed. Finally, recent advances in visible-light-driven organic semi-conductors, including covalent organic frameworks (COFs), other new arrivals,and perspectives are discussed in detail. And the readers who are most interestedin photocatalytic water splitting using inorganic metal-based photocatalysts canrefer lot of book chapters and reviews especially dedicated to the same [9–12].

12.2 Organic Semiconductors for Photocatalytic WaterSplitting and Emergence of Graphitic Carbon Nitrides

During the past three decades, various UV and visible-light-driven photo-catalysts have been explored including metal oxides, sulfides, nitrides, phos-phides, metal (oxy) nitrides, and graphene-based materials [13–23]. Notablymetal oxides with d0and d10 electronic configurations such as SrTiO3, Ta2O5,Zn2GeO4, Bi2WO6, K4Nb6O17, Sr0.25H1.5Ta2O6, and Sr0.4H1.2Nb2O6⋅H2O mate-rials have been employed as catalysts for photocatalytic water splitting in thevisible light [24–28]. Although most of these photocatalysts are limited inpractical applications due to its poor visible-light absorbance, photocorrosionin sulfide compounds, and fast recombination of charge carriers. Out of this,recently, oxy nitrides and oxy sulphides emerged as visible light active photo-catalyst utilize visible light with promising performance with high quantumyields for overall water splitting [29–36]. The above-mentioned materials arebasically inorganic semiconductors, whereas they have some disadvantages suchas their earth abundance, crystallinity, toxicity of heavy metals, and cumbersomesynthesis process. Therefore, it is utmost urgency to develop cost–effective andearth-abundant materials as new visible-light-driven photocatalysts for watersplitting to increase the efficiency of the process. Metal-free photocatalysts aremore advantageous because of their earth abundance, lightweight, cost–effective,easy fabrication, and good mechanical flexibility, which make them promisingcatalysts in photocatalytic water splitting. The first organic semiconductor usedfor photoreduction of water to hydrogen is poly(p-phenylene) where it is onlyactive in UV light (1–1.7 μmol H2 h−1 at 𝜆> 290 nm) [37]. Conjugated linearpoly(phenylene) scan catalyze hydrogen evolution in conjunction with methylviologen (1 μmol H2 h−1 in 𝜆> 420 nm), but they are only modestly active underUV irradiation and their performance under visible light is very poor [38–40].In addition, poly(azomethine), a conjugated polymer system, generates around7 μmol H2 h−1 in 𝜆> 300 nm [41].

Very recently, polymeric graphitic carbon nitride, g-CN1 (commonly knownas g-C3N4), has been discovered as a polymeric semiconductor, metal-free

1 g-C3N4 is most commonly used to denote polymeric graphitic carbon nitrides, which is incorrectas always there are 1–3% of hydrogen left in the carbon nitride structure depend on the temperatureused during the thermal condensation process, for example, at 550 ∘C, it is just polymeric melon. Itshould be noted that the pure g-CN, a crystalline material, is prepared by ionothermal conditionsand refer [42] for more details. But for reader’s understanding, polymeric graphitic carbon nitrideswill be denoted as g-CN in the whole chapter.


photocatalyst, which fulfills the basic requirements for a water-splitting catalystand revolutionizing the field after its first phenomenal report by Wang et al. [37].Generally, g-CN has been considered the most stable allotrope among variouscarbon nitrides under mild conditions. The main advantages of using thesepolymers are abundance, stability, and visible-light response in the presence of asacrificial donor or recent reports on overall water splitting in the presence ofsuitably located cocatalysts [37, 43–46]. However, the photocatalytic efficiency ofbulk g-CN (0.1 %) is still low due to its nonporous nature, faster recombination,and lower electrical conductivity [47]. The photocatalytic activity of g-CN isgreatly influenced by modifications, for example, through nanostructurationand improving crystallinity, which will be elaborated in the dedicated sectionsbelow. Promising results have been achieved in recent years, in particular forphotocatalytic and photoelectrochemical water splitting through adopting fewstrategies, which will be discussed in the further sections. Photoelectrochemicalwater splitting is not part of this chapter, and lot of book chapters and reviewsespecially are dedicated for the same [48–52].

12.3 Graphitic Carbon Nitrides for Photocatalytic WaterSplitting

Binary carbon nitride (CN) materials have attracted extensive attention as oneof the most promising candidates to complement carbon in energy applications.In theory, there are various hypothetical phases of covalent carbon nitrides, forexample, 𝛼, 𝛽, cubic, pseudocubic, and graphitic [53]. Among these, g-CN isconsidered as the most stable one at ambient conditions and has the bandgapof 2.7 eV [54–58]. g-CN is a prototypical two-dimensional (2D) polymer thatis composed of conjugated planes packed with N-bridged triazine (melam) ortri-s-triazine2 (melem) repeating units via van der Waals interactions [37, 59,60]. To date, the condensation process of cyanamide (CA) to dicyandiamide(DCDA) and later to melamine (MA) has been adopted to generate g-CNpolymers (Figure 12.1) [42, 62, 63]. However, g-CN solids obtained from thermalcondensation of monomers are not completely condensed and have a C/Nmolar ratio of about 0.72 and a tiny amount of H (about 1–2 wt%), close to thestructure of “melon” polymer [51]. g-CN polymer, as mentioned earlier, exhibitsphotocatalytic activity for water splitting in the presence of a proper sacrificialelectron donor or acceptor, even in the absence of noble metal catalysts [37].Density functional theory (DFT) calculations suggest that the visible-lightresponse of the photocatalyst originates from an electron transition from thevalence band formed by N2p orbitals to the conduction band populated by C2porbitals (Figure 12.2) [37].

g-CN features some beneficial prerequisites that are required for a heteroge-neous water-splitting photocatalysis. First of all, it has an appropriate electronic

2 g-CN is made up of either triazine or heptazine based basic building blocks and mostly heptazinebased carbon nitrides are utilized/reported for photocatalytic water-splitting applications.

NH2

N

NH2

Melam

Melon

Polymeric

N-brigded

tri-s-triazine

Melam

(Tri-s-triazine, heptazine)Melamine

Melamine-melem composite

(a)

(b)

N N

NH2N

NH2

N

HN

N N

NH2N

NH2

NH2

N N

NH2N

NH2

N

N NH2

NH2

N

N

N

NH2

N

N N

N

H2N NH2

N

N

N

NH2

N

N N

N

H2N NH2

N

N

N

NH2

N

N N

N

NH

NH

NH

N

N

N

NH2

N

N N

N

Figure 12.1 Polymerization and pathways (a and b) for graphitic carbon nitride synthesis. (Reproduced with permission from Ref. [61]. Copyright 2014, TheAmerican Chemical Society.)


–2.0

V versus NHEE

–1.5

–1.0CB

2.7 eV

VB 1.6 V

g-C3N4

H+/H2 (0.0 V)

OH–/O2 (+1.23 V)

–1.1 V

–0.5b

a 0.0

0.5

1.0

1.5

2.0

2.5(a) Tri-s-triazine (melem) unit

Perfect graphitic carbon nitride sheet

(b)

N

N

N

N

N

N N

N

N N

Figure 12.2 (a) Chemical structure of graphitic carbon nitride sheets, and (b) bandgapstructure (pH= 7) comparison of g-CN with titanium dioxide (TiO2), a reference photocatalyst.(Reproduced with permission from Ref. [64]. Copyright 2012, The Royal Society of Chemistry.)

structure with a bandgap of 2.7 eV, corresponding to an optical absorptionedge of 460 nm [64]. This bandgap is large enough to overcome the endother-mic character of water-splitting reaction. Furthermore, the highest occupiedmolecular orbital and lowest unoccupied molecular orbital (highest occupiedmolecular orbital–lowest unoccupied molecular orbital (HOMO–LUMO)) bandpositions of g-CN cover the redox potentials of water so that the photogeneratedhole has enough oxidation strength to oxidize water to O2 and the photogener-ated electron is reductive enough to reduce water to H2 [65]. In comparison withother conducting polymer semiconductors, g-CN is chemically and thermallystable and does not undergo photocorrosion during the whole water-splittingprocess [65]. Due to the high nitrogen content of g-CN in the form of bothgraphitic and triazine nitrogen moieties, which can serve as active sites, g-CNcan be potentially developed to a metal-free electrocatalyst after improvementof its conductivity. Last but not least, g-CN has an appropriate microstructure,with surface termination as defects and nitrogen atoms for electron localizationor anchoring inorganic/organic functional motifs as the active sites [65]. This isa successful new example of a robust, metal-free semiconductor photocatalystthat offers new opportunities in the field of artificial photosynthesis and energyconversion.

12.3.1 Precursor-Derived g-CN

The main advantage of preparing g-CN is the simple synthesis route through thethermal condensation of any of the low-cost nitrogen-rich precursors such ascyanamide [54], dicyandiamide [66], melamine [67], thiourea [68], urea [69], ormixtures thereof [70]. The crystalline phase of carbon nitrides is usually deter-mined using X-ray powder diffraction (XRD) pattern. The XRD patterns of g-CN

12.3 Graphitic Carbon Nitrides for Photocatalytic Water Splitting 335

10

13.1

(a) (b) (c)

27.5 C1s C1s

284.6

2

286.2

6

284.6

2

286.2

6

288.01 288.01

20 30

Inte

nsity (

a.u

.)

Inte

nsity (

a.u

.)

Inte

nsity (

a.u

.)

40

2 θ Binding energy (eV)

50 60 70 282 284 286 288 290 292

Binding energy (eV)

282 284 286 288 290 292

Figure 12.3 (a) XRD pattern of g-CN sample and XPS spectra of the g-CN prepared bypyrolysis of urea at 550 ∘C (b) C1s spectra. (c) N1s spectra. (Reproduced with permission fromRef. [71]. Copyright 2011, The Royal Society of Chemistry.)

feature two diffraction peaks at around 27∘ and 13∘ (Figure 12.3a). For graphitictype materials, the former can be indexed as the 002 peak corresponds to theinterlayer stacking of aromatic sheets and the latter can be indexed as the 100peak that corresponds to the interplanar separation within a sheet. X-ray pho-toelectron spectroscopy (XPS) measurements are used to investigate the natureof carbon (Figure 12.3b) and nitrogen elements ( Figure 12.3c) in g-CN, includ-ing sp2-bonded carbon of C—C (about 284.6 eV) and N—C=N (about 288.1 eV),the sp2-bonded nitrogen of C—N=C (about 398.7 eV), the nitrogen of tertiaryN—(C) 3 groups (about 400.3 eV) and the presence of amino groups (C—N—H,about 401.4 eV) caused by incomplete polymerization. Consequently, elementalanalysis is employed to determine the C and N elemental content and the C:Nratio of g-CN materials. UV–Vis diffuse reflectance spectra are commonly usedto calculate the bandgaps (Eg) of g-CN samples by employing the Kubelka–Munkfunction. Roughly, Eg can be determined using the simple equation: Eg = 1240/𝜆,in which 𝜆 [nanometers] is the absorption band edge of a given sample. Generally,the physicochemical properties of g-CN are related to the type of the precursorsand reaction parameters used.

Yan et al. [67] heated melamine in a semiclosed system at different tempera-tures and found that the C:N ratio of the product increased from 0.721 to 0.742,and the bandgaps decreased from 2.8 to 2.75 eV as the heating temperatureincreased from 500 to 580 ∘C. The C:N molar ratio than that of an ideal g-CN(0.75) is not reached so far, as 1∘ and 2∘ amino groups can be present due toincomplete condensation. It is noteworthy that the fabrication of an ideal g-CNwith a C:N stoichiometric ratio of 0.75 is rather difficult and can be carried outin ultravacuum conditions. The specific surface area (SSA) of g-CN dependson the precursors and synthesis conditions used; for instance, a surface area (ofabout 8 m2 g−1) was reported for melamine-derived g-CN [67]. Notably, Zhanget al. [68] reported the fabrication of g-CN by heating thiourea at differenttemperatures. An increase in temperature from 450 to 600 ∘C favored polycon-densation of g-CN and subsequently improved the structural interconnectivityand increased electron delocalization in aromatic sheets. However, a highertemperature than 650 ∘C could cause the decomposition of g-CN and thus


reduce the particle size. As a result, the bandgap first decreased and thenincreased to 2.71, 2.58, and 2.76 eV for the g-CN prepared at 450, 550, and650 ∘C, respectively. The specific surface area of thiourea-derived g-CN wasenhanced to 52 m2 g−1. Recently, urea was found to be a superior precursor forpreparing g-CN with high SSA because it produces sheet-like g-CN of muchsmaller thickness [71–75]. The surface area of g-CN has been improved byfollowing various approaches in synthesis conditions and by using different pre-cursors [76–78]. A sustainable approach of one-step self-supporting solid-statepyrolysis was developed for the low-cost and large-scale production of orderedcarbon nitrides host material, which contains tens of micron 2D rippled sheets,micro-mesopores, and oxygen heteroatoms. A higher SSA of ∼605 m2 g−1 wasachieved from the urea-derived ordered carbon nitrides [79]. Nanostructuredg-CN photocatalysts can be synthesized by the supramolecular preorganizationof hydrogen-bonded molecular assemblies, such as cyanuric acid–melamine[80], melamine–cyanuric acid (MCA) [81–83], or melamine–trithiocyanuricacid mixtures [76]. These findings suggest that the selection of different pre-cursors, combined with suitable control over the reaction parameters, suchas the time and temperature of the thermal treatment, is an effective strategyfor optimizing the electronic structure as well as the SSA of g-CN. Amongvarious precursors used for the synthesis of g-CN, urea is found to be aneffective source to prepare thin-layer g-CN with high specific surface area.Martin et al. [84] synthesized g-CN from different precursors (urea, DCDA,and thiourea) under identical conditions for comparison and employed it forphotocatalytic activity. The urea-derived g-CN exhibited superior hydrogenevolution (20 000 μmol g−1) in comparison to either the widely used DCDA(1350 μmol g−1) or thiourea-derived g-CN (2470 μmol g−1) under both full-arcand visible-light irradiation for 10 h. However, to simplify the synthesis ofg-CN materials and further improve their properties, various precursors andexperimental conditions should be explored, for instance, like recently reportedexfoliation methods which provide more surface area could greatly enhance thephotocatalytic performance due to more active edges [85–92].

12.3.2 Nanoporous g-CN by Templating Methods

Porous polymers are an emerging class of materials with pores in the mesoporousregime, that is, with pore diameters >2 nm, were mainly prepared using tem-plating methods [93–95], analogous to their inorganic counterparts. Generally,mesophases formed from surfactants or amphiphilic block copolymers have beenused as templates for the generation of a series of mesoporous inorganic materi-als, including mainly silica and metal oxides [96, 97]. However, for organic mate-rials, it seemed that such organic templates are applicable on a case-to-case basis,especially the compatibility between the template and the polymer precursor hasto be carefully adjusted so that mesophase formation of the template can occur.Also, the removal of the organic template from the organic replica can causesevere problems either by simple extraction or by calcination. Therefore, mostlyhard templating procedures using inorganic templates have been often appliedfor the preparation of mesoporous organic or carbon materials [98].


12.3.2.1 Hard TemplatingIt is well known that a higher SSA of catalysts can provide more active siteson the surface for enhanced photocatalytic activity. For pristine g-CN, it isunfortunate that the surface area is usually lower than 10 m2 g−1 as describedbefore in Section 12.2. Modification on g-CN began with the creation of a porousstructure and the increase of SSAs. Table 12.1 lists the templating methods andthe resulting textural properties of various g-CN catalysts. Wang et al. [108]synthesized the silica-templated mesoporous g-CN (mpg-CN) by generatingnanoporous structures into the polymeric matrix to improve the structural andelectronic properties for photocatalytic water-splitting process. The surfacearea (68–373 m2 g−1) and porosity can be tailored by optimized synthesizedconditions. The hydrogen production activity of mpg-CN was nearly eight timeshigher than that of g-CN; however, the activity was improved by a factor of ∼750after addition of 3 wt% Pt as cocatalyst promoting the charge transfer and creat-ing hydrogen desorption sites. Li et al. [100] synthesized the mesoporous g-CNusing different silica materials as hard templates and compared the hydrogenactivity with bulk g-CN. The SBA-15/g-CN showed the surface area of 10 m2 g−1,which is 12 times lesser than SN-g-CN derived from silica nanospheres. Despiteits low surface area (10 m2 g−1), SBA–g-CN showed the H2 production activityof 11.9 μmol h−1,which is 30 times more active than bulk g-CN, and has similaractivity to that of larger surface area mesoporous g-CN(122 m2 g−1) (in both

Table 12.1 Surface area and photocatalytic performances of mesoporous g-CN using differenttemplates.

Precursor Template Pore type

Surfacearea(m2 g−1)

Photocatalyticperformance(than pristine) References

Cyanamide TEOS/Silica Mesoporous 273 20 times higherH2 production

[61, 99]

Dicyandiamide SBA-15 Mesoporous 122 34 times higherH2 production

[100]

Melamine P123 Mesoporous 90 2.45 times higherH2 evolution

[101]

Melamine Copolymer-F68 Mesoporous 185.4 6.66 times higherH2 evolution

[102]

Dicyandiamide Template free — 306 5.4 times higherH2 evolution

[103]

Dicyandiamide Solvothermaltreatment

Mesoporous 331 5.5 times higherH2 evolution

[104]

Melamine Bubbling fromammoniumpersulfate

Mesoporous 55 Six times higherH2 evolution

[105]

Melamine Template free — 35.6 4.8 times higherH2 evolution

[106]

Melamine Bubbling fromsulfur

Mesoporous 46 5.96 times higherH2 evolution

[107]


systems in situ photodeposited Pt was used as cocatalyst). A mesoporous carbonnitride has derived from advanced sol–gel-derived g-CN pyrolyzed at differenttemperatures and tested for photocatalytic H2 production [61]. The schematicdiagram of porous carbon nitride and silica using sol–gel route is given inFigure 12.4. Compared to g-CN, SG-CN-550 showed the higher surface areaof 273 m2 g−1 and showed the higher H2 production rate of 770 μmol h−1 undervisible irradiation with the presence of 5 wt% Pt as cocatalyst. SG-CN samplesshow about 20 times higher H2 production rates than bulk CN. This is due totheir porous structure, partial disorder, and high surface area, which favor shorttravel distances and fast trapping of separated electrons on the surface wherethey are available for reaction with protons.

The mesoporous carbon nitrides were synthesized by combining sol–geland thermal condensation approach using silica and evaluated for photocat-alytic hydrogen evolution [99]. By varying the composition of cyanamide andTetraethyl orthosilicate (TEOS), the maximum surface area of 270 m2 g−1 wasobtained. Photocatalytic activity of the resulting CN compounds for waterreduction has been investigated using platinum as the water reduction catalystand triethanolamine (TEOA) as the sacrificial reductant. The photocatalyticactivity of the mesoporous carbon nitrides prepared by this approach is muchhigher than observed for bulk and even mesoporous carbon nitrides preparedusing preformed silica templates.

A recent study showed that confined thermal condensation of cyanamideinside channels of porous anodic alumina oxide (AAO) membrane templatescan efficiently increase the crystallinity, extending the domain size and loweringthe HOMO position of g-CN-based materials (Figure 12.5). The photocat-alytic H2 evolution activity of CNRs is improved by three times than bulkg-CN. The enhanced H2evolution rate of CNR was attributed to the improve-ment of condensation and orientation within the high-aspect-ratio nanowirestructures [109].

Si(OEt)4

4M NH 4

HF 2

pH = 2

80 °C &

550 °C Ar650 °C Air

SiO2 CNx

NH2CN

+

Figure 12.4 The sol–gel route to obtain porous carbon nitride and silica by hard templateapproach. (Reproduced with permission from Ref. [99]. Copyright 2011, The Royal Society ofChemistry.)


(a)

(b)

½O2 + 2H+2H+

e–

h+

H2O

1 nm

H2

hv

1

2

3

300 nm 3 μm

20 μm

Figure 12.5 AAO templating approaches toward g-CN rods using cyanamide as precursor andSEM images of g-CN rods. (Reproduced with permission from Ref. [109]. Copyright 2011, TheAmerican Chemical Society.)

Compared with silica nanoparticles, the commercial calcium carbonatenanoparticles are of low cost and can be easily removed using diluted hydrochlo-ric acid, making the preparation processes much simpler. It is used as a hardtemplate to synthesize mesoporous g-CN by calcination with melamine and theresulting Brunauer–Emmett–Teller (BET) surface area of 31.8 m2 g−1 was threetimes higher than that of pristine g-CN [110]. Recently, mesoporous g-CN wassynthesized with commercial calcium carbonate particles as hard template withthe surface area of 59.7 m2 g−1 with higher photocatalytic performance of 12.3times than bulk g-CN [111].

12.3.2.2 Soft TemplatingCompared with the hard-template approach, the soft-template route not onlysimplifies the entire synthetic procedure but also allows for easy tuning ofthe morphology through the choice of different soft templates. Soft structure-directing agents, such as surfactants, amphiphilic block polymers or ionic liquidscan be utilized for the formation of nanostructured g-CN, thereby enablingthe rational synthesis of materials with desired porous structures and surfacemorphologies by using different soft templates for specific uses.

Pluronic P123 surfactant [101] was applied as a soft template to synthesizemesoporous g-CN with worm-like pores and a narrow pore size distribution.The BET surface area of porous g-CN (90 m2 g−1) is significantly 10 times higher


than that of g-CN (9 m2 g−1) without using the surfactant. The photocatalytic H2production was evaluated under visible light with 0.5 wt% Pt as cocatalyst usingtriethanolamine as scavenger. The porous g-CN showed the increased H2 evo-lution rate of 148.2 μmol h−1 than that of bare g-CN (60.5 μmol h−1) and alsoextended the visible-light absorption up to 800 nm. Wang et al. [102] demon-strated the comparative study of the photocatalytic performance of hierarchicallyporous g-CN. An amphiphilic block copolymer-F68 was used as the soft templatefor the synthesis of mpg-CN. The mesoporous CNT-2 (from melamine) showedthe higher H2 evolution rate of 1518 μmol h−1 g−1 than CNT-1 which is obviouslyrelated to the higher surface area (185 m2 g−1). Although CNT-1 and CNT-2 bothpossessed tri-s-triazine-based structure and worm-like porous morphology, thehierarchical pores and high surface area of CNT-2 result in a higher activity forphotocatalytic H2 evolution. It can be seen that all the soft or hard templating andtemplate-free approaches can be applied to synthesize porous g-CN with highsurface area and pore volume. The improved porous structure from the structuralmodification will enhance the light absorption and photocatalytic H2 productionof g-CN materials.

12.3.2.3 Template-FreeTemplate-free porous g-CN synthesis is another effective method, and fewnotable studies are detailed below. Han et al. [112] reported that the BET surfacearea of g-CN can be increased to about 210 m2 g−1 by a facile template-freemethod controlling the polymer reaction according to Le Chatelier’s principle. Inthe synthesis, a semiclosed system was applied to partially expose the polymer-ization to air. Partial oxidation can be also employed as a posttreatment of g-CN.Niu et al. [103] reported that the BET surface area of g-CN could be increasedto 306 m2 g−1 when it was exfoliated to nanosheets by a simple top-downmethod of thermal oxidation etching of bulk g-CN in air. Manipulation of thesynthesis without a template appears to be effective for improving the texturalproperties of g-CN. Yang et al. [113] demonstrated the free standing g-CNnanosheets by liquid-phase exfoliation in various organic solvents (isopropanol,N-methyl-pyrrolidone, water, ethanol, and acetone) as the dispersion medium.They found that g-CN exfoliated using isopropanol showed a higher surfacearea of 384 m2 g−1 and associated with abundant active sites with nitrogenatoms. The average hydrogen evolution rate of g-CN nanosheets was found to be93 μmol h−1 g−1 which was much higher than that of bulk g-CN (10 μmol h−1 g−1).Recently, Han et al. [104] studied the delamination of layer-type g-CN into atom-ically thin mesoporous nanomesh g-CN by a combination of freeze-drying andsolvothermal exfoliation process and its scheme is shown in Figure 12.6. Thesurface area of mesoporous g-CN nanomesh (331 m2 g−1) showed 33 timeshigher than that of traditional bulk g-CN (10 m2 g−1) and also exhibiting higherphotocatalytic activity of 8510 μmol h−1 g−1, which is far high compared to thatof bulk g-CN (350 μmol h−1 g−1).

Introducing salts in the synthesis has proven to be effective for the creationof a porous structure of g-CN. Ma et al. [114] developed an in situ ion-assistedsynthesis of porous g-CN nanosheets, in which lithium chloride was used to


(a) (b) (c) (d)

(e)(f)(g)

(j)(i)(h)

DCDA solution

Freeze-drying

Self-assembly

Solvothermal

exfoliation

Intercalation

Carbon Nitrogen Oxygen Hydrogen = =IPA H2O

Exfoliation

Calcination

Dispersion

IPA + H2O

Nanostrucutred

DCDA

g-C3N4 nanomesh

g-C3N4 nanomeshMesoporous g-C3N4 bulk

Mesoporous g-C3N4 bulk Suspension

Figure 12.6 (a–g) Preparation procedure of monolayer of mesoporous g-CN nanomesh and(h–j) solvothermal exfoliation from bulk mpg-CN to nanomesh. (Reproduced with permissionfrom Ref. [104]. Copyright 2016, The American Chemical Society.)

produce a porous sample with 2–3 nm pores and 2–3 nm thickness, and a highsurface area of 186 m2 g−1. Lin et al. [115] reported that crystalline g-CN withtri-s-triazine subunits derived from a tri-s-triazine-based precursor and KCl andLiCl salts possessed improved crystallinity and enhanced charge carrier mobilityand showed higher photocatalytic hydrogen production.

12.3.3 Heteroatom Doping

12.3.3.1 Metal DopingDoping heteroatoms into g-CN has been extensively proven to be an effectivestrategy to extend the light absorption and to enhance the photocatalytic per-formance. Of them, Fe-doped g-CN nanosheets were prepared by Tonda et al.[105] using ferric chloride as the Fe-precursor. The Fe dopant appeared to be inthe +3 oxidation state and could significantly influence the electronic and opticalproperties of g-CN. It was reported that 2 mol% Fe-doped g-CN showed almost7 and 4.5 times higher photocatalytic activity compared to unmodified g-CN andg-CN nanosheets, respectively. Besides, Zr-doped and W-doped g-CN were alsodeveloped for enhanced photocatalysis [116, 117].


12.3.3.2 Nonmetal DopingIn consideration of the metal free nature of the modified g-CN, nonmetal dop-ing has attracted more extensive attention. A variety of dopants of nonmetalelements have demonstrated great effectiveness for g-CN-based photocatalysis.Notably, potassium-doped g-CN [106] was prepared by the thermal polymeriza-tion of dicyandiamide and KI. The doped potassium was found to enhance thephotocatalytic activity by lowering the valence band and increasing the chargeseparation rate.

Nitrogen Doping Nitrogen-doped g-CN was prepared by the polycondensationof the precursor of melamine with a nitrogen-rich additive of hydrazine hydrate[118]. N-doping can lower the bandgap of 2.72 eV of pristine g-CN to 2.65 eVin N/g-CN and indicates that the valence band maximum of pristine g-CN isat 1.84 eV. Mott-Schottky plot gives the information about n-type characteris-tics and a flat band potential value of −0.98 and −1.13 eV for N/g-CN and g-CN,respectively. The hydrogen evolution activity of N-doped g-CN (44.28 μmol h−1)showed 4.6 times higher than that of pristine g-CN (7.86 μmol h−1) with 3 wt%Pt as cocatalyst under visible light. Zhou et al. [119] reported the synthesis ofN-doped g-CN using citric acid and urea as starting precursors for the ther-mal polymerization. It was suggested that the lone pair electron on the graphiticN-atom can result in the aromatic π-conjugated system being extended and delo-calized. As a result, the photocatalytic performance of N-doped g-CN showed 4.3times higher hydrogen evolution than g-CN.

Oxygen Doping Oxygen-doped g-CN was prepared by a hydrothermal routeusing H2O2 as the dopant precursor [120] and was observed from the XPS stud-ies (N—C—O).The photocatalytic H2 evolution was tested under visible lightwith 1.2 wt% Pt as a cocatalyst and the average H2 evolution rate (37.5 mmol h−1)on the O-doped g-CN is 2.5 times higher than that (15.2 mmol h−1) on theg-CN. Huang et al. [121] reported that porous O-doped g-CN was prepared by aprecursor pretreatment method, forming hydrogen-bond-induced supramolec-ular aggregates for the creation of the porous structure and tailored O-doping.The combination of porous structure and O-doped g-CN showed the higherhydrogen evolution of 6.1 and 3.1 times compared to bulk and porous g-CN(nondoping), respectively. Guo et al. [122] prepared holey-structured g-CN withdoped oxygen at the edges via photo-Fenton reactions. A lesser bandgap energyof 2.43 eV and an increased BET surface area of 348 m2 g−1 were achieved. TheH2 evolution rate of O/g-CN was 2.86 times higher than the bare one. Theeffect of oxygen doping on the electronic and geometric structure of g-CN wasinvestigated by first principles [123]. It was theoretically confirmed that oxygendoping can improve the visible-light absorption, increase the carrier mobility,produce more active sites and reduce the recombination of electron/hole pairs.

Sulfur Doping Zhang et al. [124] developed sulfur-mediated condensation withtrithiocyanuric acid as the precursor for S doped g-CN, in which the –SH groupswere supposed to play a key role in adjusting the physicochemical propertiesof the prepared g-CN. The photocatalytic activity of S doped g-CN shows 12.5


N

N

N

650 °C, N2, 2 hH2N

NH2

NH2

N

N

N 650 °C, N2, 2 h

H2NNH2

NH2

S

SS S

S

SSS

500 nm

500 nm

Figure 12.7 Photograph and TEM images of pure g-CN and sulfur-modified g-CN.(Reproduced with permission from Ref. [76]. Copyright 2012, Elsevier.)

times higher H2 evolution than pristine CN. The schematic representation andtransmission electron microscopy (TEM) image of sulfur loaded on g-CN andbare g-CN are given in Figure 12.7. Liu et al. [7] reported that sulfur-doping caninduce a unique electronic structure that shows an increased VB along with anelevated CB minimum and a minor declined absorbance. Significant changes inthe optical properties and electronic structures would lead to enhanced photo-catalysis in hydrogen evolution over the sulfur-doped g-CN, with rates 7.2 and 8.0times, respectively. He et al. [107] synthesized sulfur-bubble-template-mediatedS/g-CN using sublimed sulfur having the surface area of 46 m2 g−1 and enhancedthe photocatalytic performances up to 5.96 times than bare g-CN. Lin et al. [125]applied DFT calculations to determine the influence of sulfur-doping on the reac-tion mechanism of photocatalytic water oxidation. They found that sulfur-dopingof g-CN not only induces a different reaction mechanism but also decreases theoverpotential in the water-splitting process.

Phosphorus Doping Ran et al. [126] prepared porous P-doped g-CN nanosheetsby combining P doping and thermal exfoliation of a bulk material. The P-dopingand nanosheet morphology significantly increased the visible-light photocat-alytic H2 production compared to pristine g-CN. Mesoporous P-doped g-CNnanostructured flowers were also prepared by a cocondensation method in theabsence of any templates and attained 9.29 times higher H2 evolution than CN[127]. Guo et al. [128] applied a hexagonal rod-like supramolecular assembly ofmelamine-cyanuric acid adduct as a precursor of g-CN and phosphorous acid asa P source to prepare P-doped carbon nitride tubes. It was observed that, afterP-doping, the bandgap energy decreased to 2.55 from 2.67 eV and a seven timeshigher photocatalytic H2 evolution rate was achieved. A computational studyreveals that due to phosphorous doping, the conduction band and valence bandgets shifted downward, N2 and C1 sites in g-CN reduces to 2.03 and 2.22 eV,which is more suitable for visible-light photocatalysis [129].


Doping by Halogens Iodine-modified g-CN was synthesized by the cocondensa-tion of dicyandiamide and ammonium iodide. It showed the bandgap of 2.69 eVand two times higher H2 evolution than bare CN, which indicates the introduc-tion of iodine, would induce an effective extension of the aromatic carbon nitridesby I− ions [130]. In another study, iodinated g-CN nanosheets with higher surfacearea of 80 m2 g−1were prepared by a simple and scalable ball-milling techniqueand have the bandgap of 2.37 eV due to the effect of iodine doping. H2 evolu-tion studies show 9.1 times better activity when compared to the pristine CN[131]. Fluorinated g-CN was prepared by directly incorporating NH4F into thethermal condensation process in a g-CN synthesis. In photocatalytic hydrogenevolution, F-g-CN with 3 wt% Pt as the cocatalyst demonstrated about 2.7 timeshigher activity than the unmodified g-CN [132]. Bromine was also doped withurea-derived g-CN using NH4Br. The urea-derived g-CN-Br doped showed 2.4times higher H2 production and six times higher O2 evolution than pristine CNmaterial [133].

12.3.4 Metal Oxides/g-CN Nanocomposites

g-CN usually produce hydrogen under visible-light irradiation from an aqueoussolution of triethanolamine or methanol. Since the valence band potential isnot far positive when compared to the oxidation potential to produce oxygen,certain structural modifications are needed, for example, using compositestructure, of which, metal oxide composite with g-CN provides advantagesof each component by creating a heterojunction, mostly Z-scheme, which isgenerally adopted to produce hydrogen and oxygen simultaneously. Overallwater splitting using Z-scheme method with visible-light semiconductors arenot discussed in this chapter, and detailed discussions can be found in thereview [134]. It should be mentioned that the metal-oxide-g-CN compositewas also employed for the enhanced hydrogen generation itself, for example,notably like they carried out with g-CN-WO3 heterojunctions [135–138]. Thus,coupling one semiconductor to another is an effective technique to createheterojunctions for extended absorbance and improved charge separation,which could enhance the photocatalytic performance. Kailasam et al. [135]synthesized the mpg-CN/WO3 composites by a simple mixing method for theoverall water-splitting process. The surface area of WO3 loaded on mpg-CNderived using sol–gel (SG, 106 m2 g−1) and silica nanoparticles (SNP, 122 m2 g−1)was far high compared to that of bulk g-CN (10 m2 g−1). The H2 generation rateof W-SNP-CN was 2.4 times than SNP-CN, and the W-SG-CN showed theincreased H2 generation rate of 1.6 times than SG-CN.

However, W-TEOS-CN showed only a small amount of O2 evolution (about2.2 mmol h−1), whereas bare WO3 shows 17 mmol h−1 O2 evolution. This is due tothe location of the valence band (VB) potential of the porous CN being just belowthe oxidation potential of water to generate O2, which makes it relatively difficultfor the oxidation reaction to occur (Figure 12.8). Cheng et al. [136] approachedthe new type of CuFe2O4/g-CN, where little amount of CuFe2O4 species can dis-perse in the g-CN matrix, which leads to enhance the visible-light absorbance,surface area and also the charge carrier separation. They attained the H2 evolution


Figure 12.8 Schematicdiagram of the solid-stateZ-scheme photocatalyticmechanism in g-CN/WO3composites (SHE= standardhydrogen electrode).

–1

E (V versus SHE) g-CN

e–

e–

CB

H+/H2

H2O/O2

VB

VB

WO3

TEOSox

TEOS

h+

h+

CBhv

PtH2

H+

0

+1

+2

+3

rate of ∼76 μmol h−1 for heterostructure, which is about three times higher thanthe pure g-CN. Recently, a novel direct Z-scheme CoTiO3/g-CN was synthesizedusing a facile in-situ growth method and studied its photocatalytic activity [138].They showed the higher H2 evolution rate of∼858 μmol h−1 g−1 under the optimalweight percentage of CT/CN owing to the formation of close interface contact inthe heterojunction between CoTiO3 and g-CN.

12.3.5 Graphene and CNT-Based g-CN Nanocomposites

As g-CN has lower electrical conductivity, making composites with carbonnanomaterials is the obvious choice for the effective charge separation forenhanced photocatalytic activity. We tried to showcase some of the interestingstudies in this section. Xiang et al. [139] reported that the g-CN/graphenehybrid with an optimum graphene content (1.0 wt%) shows a H2 evolutionrate of 451 μmol h−1 g−1, which was 3.07 times more than that of pure g-CNand showed activity with recycling for at least four times. In another study,visible-light-induced photocatalytic activity of the multiwalled carbon nanotubes(MWCNT)/g-CN composite shows a hydrogen evolution rate of 42 mmol g−1

for the composite (0.5% MWCNT/g-CN). The nanocomposite of graphiticcarbon nitride with multiwalled carbon nanotubes shows 100% enhancement inits photocatalytic activity toward water splitting [140]. Chen et al. [141] studiedthe photocatalytic activity and the optimal CNT loading with g-CN showed theH2 production rate of 39.4 mmol h−1, which is about 2.4 times higher than thaton g-CN.

12.3.6 Structural Modification with Organic Groups

Irrespective of the modifications shown earlier, organic modification of theg-CN covalently and noncovalently proved to be an effective way mainly toharvest more visible light, thus by tuning the bandgap and positions withincreased charge separation for the better photocatalytic activity. Grafting offunctional groups onto g-CN can be achieved by copolymerization with organic


compounds having amino cyano moieties. Zhang et al. [43] employed variousmonomer building blocks (amino/cyano groups, 2-aminobenzonitrile) withdesired compositions and electronic structures for the modification of g-CN.Enhanced photocatalytic activity was observed for the modified photocatalysts.Chu et al. [142] reported a simple bottom-up approach to prepare g-CN witha desired band structure by incorporating an electron-deficient pyromelliticdianhydride (PMDA) monomer. The photocatalytic H2 evolution reaction wastested with methanol and Pt (1 wt%) as a cocatalyst under visible-light irradiationand the H2 evolution rate of Polyimide, PI (20.6 μmol h−1) showed about threetimes as high as that of g-CN (7.0 μmol h−1). The modified g-CN has a loweredVB (Figure 12.9), ensuring stronger photooxidation ability for the O2 evolutionalong with the H2 evolution.

Zhang et al. [143] demonstrated the enhanced optical absorption and photocat-alytic activity using 2D conjugated polymers. Here they used four typical organicagents, barbituric acid (BA), 2-aminobenzonitrile (ABN), 2-aminothiophene-3-carbonitrile (ATCN), and diaminomaleonitrile (DAMN), were selected basedon their unique chemical functions as the comonomers for condensation with acarbon nitride precursor as urea. Significantly, an overall enhanced H2 evolution

300

4

3

2

1C

–2

(1

09cm

4 F

–2)

1

0

(a) (b)

(c) (d)

400 500

Wavelength (nm) Potential (V vs Ag/AgCl)

Ab

so

rba

nce

(a

.u.)

600 700 800 –1.0

10

–1.0

–1.44 V

–0.70 V

1.26 V

g-C3N4

2.10 V

Pl

VB

CBH+/H2, –0.59 V

O2/H2O, 0.64 V

V versus Ag/AgCl (pH = 6.6)

0.0

1.0

2.0

Inte

nsity (

a.u

.)

8 6 4

Binding energy (eV)

2 0

g-C3N4, 1.44 eV

Pl, 2.05 eV

–2 –4

–0.5

–0.70 v

0.0 0.5 1.0

Figure 12.9 (a) UV–vis absorption spectrum and photograph (inset) of polyimide (PI),(b) Mott-Schottky plot of PI, (c) band structure and (d) VB XPS of g-CN and PI. (Reproducedwith permission from Ref. [142]. Copyright 2013, The American Chemical Society.)


rate was found for all the modified samples, especially for the CNU–ATCN,which enhances its H2 evolution rate by a factor of about 9 compared to thesheet-like sample and about 74 compared to dense carbon nitride dots (CND)(10 mmol h−1).

Xing et al. [144] developed a new type of g-CN-based composite photo-catalysts loaded with Pt and poly(3,4-ethylenedioxythiophene) (PEDOT). Theas-prepared CN–PEDOT–Pt composites showed drastically enhanced activityfor visible-light-driven photocatalytic H2 production compared to those ofCN–PEDOT (32 μmol h−1) and CN–Pt (6.4 μmol h−1), possibly due to thespatial separation of the reduction and oxidation reaction sites. These findingsdemonstrate the crucial role of the hole-conducting polymer for the genera-tion of specially separated reductive and oxidative sites toward the efficientphotocatalysis, and this idea could be applied to other semiconductor-basedphotocatalyst systems. Dye-sensitized mesoporous CN was developed withmagnesium phthalocyanine (MgPc) by Takanabe et al. [145], for photocatalyticH2 evolution. A monolayer of dye on the Pt/mpg-CN showed the highest rate ofH2 evolution, suggesting that charge transfer predominantly occurred throughthe conduction band of mpg-CN to Pt, and the further accumulation of dye ledto a decrease in the efficiency of charge transfer from MgPc to the conductionband of mpg-CN (Figure 12.10).

Furthermore, to improve the photocatalytic activity of g-CN, Zhang et al.[146] used zinc phthalocyanine derivative as sensitizer and extended the lightabsorption from 450 to >800 nm. They studied the effect of coadsorbent,chenodeoxycholic acid (CDCA) on the photocatalytic H2 production. Especially,Zn-tri-PcNc/g-CN with CDCA as coadsorbent exhibits a H2 production effi-ciency of 125.2 μmol h−1 under visible-light (𝜆≥ 500 nm) irradiation; moreover,it gives an extremely high apparent quantum yield (AQY) of 1.85% at 700 nmmonochromatic light irradiation. The above results show the promising appli-cation of phthalocyanines in photocatalytic H2 production system for moreefficiently utilizing the solar radiation with wavelength longer than 600 nm.

12.3.7 Crystalline Carbon Nitrides

Crystallinity of the g-CN photocatalysts is also one of the important factors toenhance the photocatalytic activity due to the defects in the structure beingminimized. Usually, the pristine g-CN synthesized by the condensation ofnitrogen containing precursors exhibits the lower crystallinity and moderatephotocatalytic activity. The existence of hydrogen bonds in the covalent carbonnitride framework may block electron conduction across the plane and leadto low conductivity. Therefore, the synthesis of a fully condensed, crystallineg-CN is desirable [134, 147]. As a notable example, recently, photocatalyticactivity of tri-s-triazine-based crystalline g-CN has synthesized using KCl/LiClsalts as subunits of tri-s-triazine [115]. The as-obtained sample exhibited highcrystallinity than bulk g-CN, and the crystal structure is quite different fromthat of the triazine-based poly(triazine imide), PTI. The H2 evolution wascarried out for different sacrificial agents like triethanolamine, ethanol andmethanol to examine the photocatalytic activity. The g-CN-1 sample showed


(a)

–2

–1

0

Po

ten

tia

l ve

rsu

s N

HE

(V

)

Pt

C3N4 MgPc

(b)

e–

LUMO

HOMO

H+/H2

H2

CB

VB

O2/H2O

h+

>600 nm

TEOA

1.8 eV

hν

e–

H+

+1

+2

N

N N

N

N Mg

N

N

N

N

N

N

N

N N

N

N N

N

N

N

N

N

NN

N

N

N

N

N

N N

N

N

N

N

N

N

N

NN

N

N

N

N

N

N

NN

N

N

N

N

N N

N

N

N

N

N

N

N

NN

Figure 12.10 (a) Chemical structure and (b) band position of g-CN and MgPc. (Reproducedwith permission from Ref. [145]. Copyright 2010, The Royal Society of Chemistry.)

the higher H2 evolution of 770 μmol h−1 under visible-light irradiation, usingTEOA as sacrificial agent with 3 wt% of Pt as cocatalyst. And the O2 evolutionrate (∼7.0 μmol h−1) of crystalline g-CN-1 without the use of any cocatalyst wasmuch higher than that of bulk g-CN (∼1.8 μmol h−1).

12.3.8 Overall Water Splitting and Large-Scale Hydrogen ProductionUsing Carbon Nitrides

The overall water splitting of g-CN is difficult to achieve without using any sacrifi-cial agents, because it depends on the preparation conditions to tune the texturalproperties of the polymer and the composite to control the reaction kinetics on

12.4 Novel Materials 349

the polymer surface. Notably, Zhang et al. [148] demonstrated the overall watersplitting of g-CN synthesized from different precursors like urea, dicyandiamideand ammonium thiocyanate without sacrificial agents. Liu et al. [149] fabricatedthe carbon dot-CN nanocomposites to split water into H2 and O2. They mea-sured the quantum efficiencies of 16% for wavelength 𝜆= 420± 20 nm, 6.29% for𝜆= 580± 15 nm, and 4.42% for 𝜆= 600± 10 nm, and determined an overall solarenergy conversion efficiency of 2.0%.

Schwarze and coworkers [150] successfully demonstrated the hydrogenevolution reaction in a large-scale reactor using sol–gel-derived mpg-CN as aphotocatalyst under natural sunlight irradiation. They fabricated the mpg-CNthin films from mpg-CN powder by drop-casting method; resultant stable thinfilm was obtained. The fabricated stable films produced approximately 18 lgaseous H2 in one month time on stream at an average H2 evolution reaction(HER) rate of 0.22 l kWh−1. The rate could be nicely predicted from prelimi-nary lab-scale experiments under well-defined conditions. Furthermore, thehydrogen production correlates with the sunlight intensity and a maximumsolar-to-hydrogen (STH) conversion of 0.12% was obtained.

12.4 Novel Materials

12.4.1 Triazine and Heptazine-Based Organic Polymers

g-CN is composed of amine-bridged heptazine (C6N7) units, one possiblepathway would be the use of functionalized heptazines that can be polymerizedto generate new polymeric CN structures. The amine-functionalized heptazine,also called “melem,” is an intermediate in the thermal condensation of N-richmolecules towards producing CN materials and is insoluble in nature [151].Cyameluric chloride (C6N7Cl3), which can be synthesized from melem in twosimple steps, seems to be a much more feasible monomer for the synthesis ofheptazine-based polymers as it is easily soluble in organic solvents, and thechloride groups are highly reactive for nucleophilic substitutions [59, 152].Recently, Kailasam and Thomas et al. introduced a heptazine-based micro-porous polymer networks (HMPs) synthesized at room temperature by thepolycondensation of aryl diamines with cyameluric chloride [153]. The result-ing HMPs (HMP-1 and HMP-2) showed better activity as photocatalysts forhydrogen evolution compared to g-CN under visible light. However, in HMP-3another strong electron acceptor, benzothiadiazole, used frequently in organicelectronics, was introduced into the structure to investigate the influence on thephotocatalytic activity for water splitting [154]. With the use of HMP-3_2 : 3 and4 : 3 as photocatalysts, a high and stable H2 evolution of 32 and 31 mmol H2 h−1,respectively, is observed which is significantly higher than that of HMP-1 andbulk g-CN. However, even though these are high values, especially comparedto other conjugated polymers prepared at low temperatures, we refrain from adirect comparison with other reported materials.

In the case of triazine-based CN polymers, Lotsch and coworkers [155]synthesized 2D triazine-based CN in a two-step ionothermal synthesis by using


4-amino-2,6-dihydroxypyrimidine as the dopant. A rather low level of structuraldefinition and the introduction of defects up to a certain doping level (16% for4-amino-2,6-dihydroxypyrimidine) leads to enhance the quantum efficiency of3.4%, which rivals the benchmark of heptazine-derived photocatalysts undervisible light. Bhunia et al. [156] obtained the triazine-based crystalline g-CNthrough the combination of supramolecular aggregation and polycondensationin an ionic melt by using melamine as a precursor and 2,4,6-triaminopyrimidineas a dopant. The improved condensation facilitates the high crystallinity and aremarkably increased H2 evolution, with a quantum yield of about 7% at 420 nm.

12.4.2 Covalent Organic Frameworks (COFs) and Beyond

COFs are highly crystalline porous polymers comprising lightweight elementsand exhibit high surface area [157, 158]. Accordingly, the ordered structure,well-accessible pore walls, tunable electrical and optical properties becomingCOFs as promising candidates for photocatalytic H2 evolution. Lotsch andcoworkers [159] developed a hydrazone-based COF (TFPT-COF; TFPT= 1,3,5-tris(4-formylphenyl)triazine) with a mesoporous 2D network and honeycomb-type in-plane structure with a larger surface area of 1603 m2 g−1. This COFsshowed the H2 evolution of 230 μmol h−1 g−1 after Pt deposited cocatalystsunder visible-light irradiation. When a sacrificial donor of TEOA was addedand achieved the higher H2 evolution of 1970 μmol h−1 g−1 for first 5 h. Lotschand coworkers [160] further synthesized a series of water- and photostable2D azine-linked COFs from hydrazine and triphenylarene aldehydes, having avarying numbers of nitrogen atoms, by using a solvothermal method at 120 ∘C.These synthesized 2D azine-linked COFs were crystalline with high surface areasof 1537 m2 g−1 and showed the H2 production of 1703 μmol h−1 g−1.

Here are some of the conjugated microporous polymers (CMPs) recentlyreported for photocatalytic water-splitting applications, which are mentionedhere and for the detailed studies refer to Chapter 9 contributed by Dr Ghosh[161–163]. Notably, Cooper and coworkers [161] synthesized a series of pyrene-based CMPs with tunable optical gap and the best one was found generatingvisible-light H2 evolution rate of 17.4 mmol h−1 in the presence of sacrificialagents. Again, Cooper and coworkers [162] developed spirobifluorene CMPswhich showed the larger surface area of ∼895 m2 g−1 and the hydrogen evo-lution rate of 8 μmol h−1. Li et al. [163] reported the importance of acceptorcomonomer using donor–acceptor porous conjugated polymers (PCP) forphotocatalytic hydrogen production with the best photocatalytic performanceof ∼106 μmol h−1.

Fan et al. [164] constructed a series of g-CN-based intramolecular donor-acceptor copolymers (2-4-dibromo quinoline), that is, aromatics incorporatedg-CN via nucleophilic substitution reactions. The copolymer showed theremarkably enhanced hydrogen evolution of 436 μmol h−1, which showed themost excellent g-CN-based visible-light photocatalysts. The intramolecularcharge transfer transition from the HOMO of donor (N) to the LOMO ofacceptor (aromatic ring) is believed to play a significant role in their remarkablyimproved hydrogen evolution activity.

12.5 Conclusions and Perspectives 351

The facile syntheses and latest developments of various photoactive organicpolymers, including poly-triazine/heptazine, covalent triazine frameworks(CTFs), COFs and CMPs and HMPs with a particular focus on the modificationof syntheses, structures, and properties will pave the way for developing betterphotocatalysts other than typical g-CN-based photocatalysts. The intimatestructure–performance relationships of these polymers are clearly mentioned inrecent papers and perspectives [165, 166].

12.5 Conclusions and Perspectives

Recent years have witnessed a fast growing interest in designing polymericg-CN-based photocatalysts. As a metal-free polymeric photocatalyst with abandgap of 2.7 eV, pristine g-CN suffered from some shortcomings comparedto inorganic photocatalysts, including high exciton binding energy, limitedlight-harvesting capability, and fast recombination of charge. In addition, thepoor crystallinity and many surface defects in polymeric g-CN restrict itsphotocatalytic applications. The development of diverse synthetic techniquesand physicochemical strategies to endow g-CN solids with an optimizedelectronic structure, nanostructure, crystal structure and heterostructurehas become an urgent necessity to increase the photocatalytic performanceof g-CN. Accordingly several improvements were made: First, doping is aneffective procedure to adjust the redox potentials of charge carriers and enhancethe optical absorption by introducing foreign impurities into g-CN, whereascopolymerization is desirable to extend the delocalization of the π-electronsand change the intrinsic semiconductor properties by grafting aromatic groupsonto the surface of the g-CN. These two methods enable the modulation ofthe molecular structure, electronic structure, and photocatalytic activity ofg-CN. Second, numerous nanostructured g-CN materials have been preparedby various synthetic pathways, including the exfoliation strategy, the hard/softtemplating strategy, solvothermal technology, the supramolecular preorganiza-tion approach, and other methods. In general, nanoarchitectured g-CN tendsto exhibit an outstanding photocatalytic performance compared to its bulkcounterpart, probably because of the favorable surface properties, optimizedelectronic structure, accelerated charge separation, as well as promoted massdiffusion during photoredox reactions.

Third, modulating the crystal structure of g-CN by polycondensation in anionic melt remarkably enhances the photocatalytic activity of g-CN. Finally, cre-ating heterostructure photocatalysts by combining g-CN and other semiconduc-tors with energetically matching band structures is an effective method to realizefast separation of photoinduced charge carriers which results in high photocat-alytic activities. Thus, with the reasonable design of the structure of g-CN atdifferent scales, the photocatalytic applications of g-CN would be significantlyenriched in a more rational manner. Although significant effort has already beendevoted to the modification of g-CN materials and optimizing their photocat-alytic activity, the potential of g-CN materials has yet to be exploited fully. In addi-tion, texture engineering brings the idea of creating membrane and core–shell


structures for the controlled deposition of oxidation and reduction cofactors forthe control of electron and hole transfer at the interface pave way for overall watersplitting. There will be extensive opportunities related to the utilization of g-CNin sustainable catalysis, solar energy conversion, and devices, which will requiremuch effort from researchers and scientists worldwide.

In the meantime, looking beyond g-CN polymers have attracted manyresearchers to work toward photocatalytic water splitting. Especially, the emer-gence of porous organic polymers in the form of metal organic frameworks(MOFs), COFs and CMPs and their application in generating solar fuels set thefield ignited with many reports coming recently and are expected to largelyboom to achieve high photocatalytic activity especially for H2 production.

With the significant development in organic photovoltaic applications bythe employment of various donor- and acceptor-based molecules with everenhancing performance will provide a variety of options to develop a series oforganic semiconductors for water-splitting applications. But the real challengelies in applying them for large-scale hydrogen production by utilizing naturalsunlight including the NIR part of the solar radiation. With initial efforts alreadyinto it, time is not too far to achieve this (im)probable task.

References

1 Hoffmann, M.R., Martin, S.T., Choi, W.Y., and Bahnemann, D.W. (1995)Environmental applications of semiconductor photocatalysis. Chem. Rev., 95,69–96.

2 Tong, H., Ouyang, S.X., Bi, Y.P., Umezawa, N., Oshikiri, M., and Ye, J.H.(2012) Nano-photocatalytic materials: possibilities and challenges. Adv.Mater., 24, 229–251.


4 Yu, J.G. and Ran, J.R. (2011) Facile preparation and enhanced photocatalyticH2-production activity of Cu(OH)2 cluster modified TiO2. Energy Environ.Sci., 4, 1364–1371.

5 Chen, D., Tang, L., and Li, J. (2010) Graphene-based materials in electro-chemistry. Chem. Soc. Rev., 39, 3157–3180.

6 Wu, N., Wang, J., Tafen, D.N., Wang, H., Zheng, J.G., Lewis, J.P., Liu, X.,Leonard, S.S., and Manivannan, A. (2010) Shape-enhanced photocatalyticactivity of single-crystalline anatase TiO2 (101) nanobelts. J. Am. Chem. Soc.,132, 6679–6685.

7 Liu, G., Niu, P., Sun, C., Smith, S.C., Chen, Z., Lu, G.Q., and Cheng,H.M. (2010) Unique electronic structure induced high photoreactivity ofsulfur-doped graphitic C3N4. J. Am. Chem. Soc., 132, 11642–11648.


9 Ahmad, H., Kamarudin, S.K., Minggu, L.J., and Kassim, M. (2015) Hydrogenfrom photo-catalytic water splitting process: a review. Renewable SustainableEnergy Rev., 43, 599–610.

References 353

10 Zhang, G., Liu, G., Wang, L., and Irvine, J.T.S. (2016) Inorganic perovskitephotocatalysts for solar energy utilization. Chem. Soc. Rev., 45, 5951–5984.


12 Xing, J., Fang, W.Q., Zhao, H.J., and Yang, H.G. (2012) Inorganic photocata-lysts for overall water splitting. Chem. Asian. J., 7, 642–657.

13 Jiang, X., Nisar, J., Pathak, B., Zhao, J., and Ahuja, R. (2013) Grapheneoxide as a chemically tunable 2-D material for visible-light photocatalystapplications. J. Catal., 299, 204–209.

14 Yeh, T.F., Siju, J.M., Cheng, C., Chang, T.H., and Heng, H. (2010) Graphiteoxide as a photocatalyst for hydrogen production from water. Adv. Funct.Mater., 20, 2255–2262.

15 Yeh, T.F., Feng, C.Y., Chen, S.J., and Teng, H. (2014) Nitrogen-dopedgraphene oxide quantum dots as photocatalysts for overall water-splittingunder visible light illumination. Adv. Mater., 26, 3297–3303.


17 Maeda, K. and Domen, K. (2010) Photocatalytic water splitting: recentprogress and future challenges. J. Phys. Chem. Lett., 1, 2655–2661.

18 Maeda, K. and Domen, K. (2007) New non-oxide photocatalysts designedfor overall water splitting under visible light. J. Phys. Chem. Lett., 111,7851–7861.

19 Liu, P., Nisar, J., Ahuja, R., and Pathak, B. (2013) Layered perovskiteSr2Ta2O7 for visible light photocatalysis: a first principles study. J. Phys.Chem. C, 117, 5043–5050.

20 Maeda, K., Teramura, K., Saito, N., Inoue, Y., and Domen, K. (2007) Photo-catalytic overall water splitting on gallium nitride powder. Bull. Chem. Soc.Jpn., 80, 1004–1010.

21 Pan, C., Takata, T., Nakabayashi, M., Matsumoto, T., Shibata, N., Ikuhara,Y., and Domen, K. (2015) A complex perovskite-type oxynitride: the firstphotocatalyst for water splitting operable at up to 600 nm. Angew. Chem. Int.Ed., 127, 2998–3002.

22 Zou, Z., Ye, J., Sayama, K., and Arakawa, H. (2001) Direct splitting of waterunder visible light irradiation with an oxide semiconductor photocatalyst.Nature, 414, 625–627.

23 Tian, J., Cheng, N., Liu, Q., Xing, W., and Su, X. (2015) Cobalt phosphidenanowires: efficient nanostructures for fluorescence sensing of biomoleculesand photocatalytic evolution of dihydrogen from water under visible light.Angew. Chem. Int. Ed., 54, 5493–5497.


25 Ni, M., Leung, M.K.H., Leung, D.Y.C., and Sumathy, K. (2007) A reviewand recent developments in photocatalytic water-splitting using TiO2 forhydrogen production. Renewable Sustainable Energy Rev., 11, 401–425.

26 Liang, S., Shen, L., Zhu, J., Zhang, Y., Wang, X., Li, Z., Wu, L., and Fu, X.(2011) Morphology-controlled synthesis and efficient photocatalytic perfor-mances of a new promising photocatalyst. Sr0.25H1.5Ta2O6⋅H2O. RSC Adv., 1,458–467.


27 Liang, S., Wang, X., Chen, Y., Zhu, J., Zhang, Y., Wang, X., Li, Z., and Wu,L. (2010) Sr0.4H1.2Nb2O6⋅H2O nanopolyhedra: an efficient photocatalyst.Nanoscale, 2, 2262–2268.

28 Maeda, K. (2011) Photocatalytic water splitting using semiconductor par-ticles: history and recent developments. J. Photochem. Photobiol. C, 12,237–268.

29 Moriya, Y., Takata, T., and Domen, K. (2013) Recent progress in the devel-opment of (oxy)nitride photocatalysts for water splitting under visible-lightirradiation. Coord. Chem. Rev., 257, 1957–1969.

30 Hisatomi, T., Kuboto, J., and Domen, K. (2014) Recent advances in semicon-ductors for photocatalytic and photoelectrochemical water splitting. Chem.Soc. Rev., 43, 7520–7535.

31 Yanagida, S., Kabumoto, A., Mizumoto, K., Pac, C., and Yoshino, K. (1985)Poly(p-phenylene)-catalysed photoreduction of water to hydrogen. J. Chem.Soc., Chem. Commun., 8, 474–475.

32 Takata, T., Pan, C., Nakabayashi, M., Shibata, N., and Domen, K. (2015)Fabrication of a core–shell-type photocatalyst via photodeposition of groupIV and V transition metal oxyhydroxides: an effective surface modificationmethod for overall water splitting. J. Am. Chem. Soc., 137, 9627–9634.

33 Ma, G., Chen, S., Kuang, Y., Akiyama, S., Hisatomi, T., Nakabayashi, M.,Shibata, N., Katayama, M., Minegishi, T., and Domen, K. (2016) Visiblelight-driven Z-scheme water splitting using oxysulfide H2 evolution photo-catalysts. J. Phys. Chem. Lett., 7, 3892–3896.

34 Takata, T., Pan, C., and Domen, K. (2015) Recent progress in oxynitridephotocatalysts for visible-light-driven water splitting. Sci. Technol. Adv.Mater., 16, 033506–033524.

35 Siritanaratkul, P., Maeda, K., Hisatomi, T., and Domen, K. (2011) Synthe-sis and photocatalytic activity of perovskite niobium oxynitrides with widevisible-light absorption bands. ChemSusChem, 4, 74–78.

36 Maeda, K., Higashi, M., Lu, D., Abe, R., and Domen, K. (2010) Efficient non-sacrificial water splitting through two-step photoexcitation by visible lightusing a modified oxynitride as a hydrogen evolution photocatalyst. J. Am.Chem. Soc., 132, 5858–5868.

37 Wang, X., Maeda, K., Thomas, A., Takanabe, K., Xin, G., Carlson, J.M.,Domen, K., and Antonietti, M. (2009) A metal-free polymeric photocata-lyst for hydrogen production from water under visible light. Nat. Mater., 8,76–80.

38 Shibata, T., Kabumoto, A., Shiragami, T., Ishitani, O., Pac, C.,and Yanagida, S. (1990) Novel visible-light-driven photocatalystpoly(p-phenylene)-catalyzed photoreductions of water, carbonyl compounds,and olefins. J. Phys. Chem., 94, 2068–2076.

39 Zhang, J., Chen, X., Takanabe, K., Maeda, K., Domen, K., Epping, J.D., Fu,X., Antonietti, M., and Wang, X. (2010) Synthesis of a carbon nitride struc-ture for visible-light catalysis by copolymerization. Angew. Chem. Int. Ed.,49, 441–444.

40 Jiang, D.L., Choi, C.K., Honda, K., Li, W.S., Yuzawa, T., and Aida, T. (2004)Photosensitized hydrogen evolution from water using conjugated polymerswrapped in dendrimeric electrolytes. J. Am. Chem. Soc., 126, 12084–12089.

References 355

41 Schwab, M.G., Hamburger, M., Feng, X., Shu, J., Spiess, H.W., Wang, X.,Antonietti, M., and Mullen, K. (2010) Photocatalytic hydrogen evolutionthrough fully conjugated poly(azomethine) networks. Chem. Commun., 46,8932–8934.

42 Bojdys, M.J., Müller, J.-O., Antonietti, M., and Tomas, A. (2008) Ionothermalsynthesis of crystalline, condensed, graphitic carbon nitride. Chem. Eur. J.,14, 8177–8182.

43 Zhang, J., Zhang, G., Chen, X., Lin, S., Möhlmann, L., Dołega, G., Lipner,G., Antonietti, M., Blechert, S., and Wang, X. (2012) Co-monomer control ofcarbon nitride semiconductors to optimize hydrogen evolution with visiblelight. Angew. Chem. Int. Ed., 51, 3183–3187.

44 Butchosa, C., Guiglion, P., and Zwijnenburg, M.A. (2014) Carbon nitridephotocatalysts for water splitting: a computational perspective. J. Phys.Chem. C, 118, 24833–24842.

45 Pan, Z., Zheng, Y., Guo, F., Niu, P., and Wang, X. (2017) Decorating CoP andPt nanoparticles on graphitic carbon nitride nanosheets to promote overallwater splitting by conjugated polymers. ChemSusChem, 10, 87–90.

46 Zheng, D., Cao, X.N., and Wang, X.C. (2016) Precise formation of a hollowcarbon nitride structure with a Janus surface to promote water splitting byphotoredox catalysis. Angew. Chem. Int. Ed., 55, 11512–11516.

47 Zhang, J., Zhang, M., Sun, R.Q., and Wang, X. (2012) A facile band align-ment of polymeric carbon nitride semiconductors to construct isotypeheterojunctions. Angew. Chem. Int. Ed., 51, 10145–10149.

48 Su, J., Geng, P., Li, X., and Chen, G. (2016) Graphene-linked graphitic car-bon nitride/TiO2 nanowire arrays heterojunction for efficient solar-drivenwater splitting. J. Appl. Electrochem., 46, 807–817.

49 Lewerenz, H.J., Peter, L., and Zhao, T.S. (2016) Photocatalysis Fundamentalsand Perspectives, RSC Energy & Environment Series.

50 Striegler, K. (2015) Modified Graphitic Carbon Nitrides for PhotocatalyticHydrogen Evolution from Water: Copolymers, Sensitizers and Nanoparticles,Springer.

51 Zhang, Y., Mon, T., and Ye, J. (2012) Polymeric carbon nitrides: semi-conducting properties and emerging applications in photocatalysis andphotoelectrochemical energy conversion. Sci. Adv. Mater., 4, 282–291.

52 Lou, S., Zhou, Z., Shen, Y., Zhan, Z., Wang, J., Li, S., and Zhang, Y. (2016)Comparison study of the photoelectrochemical activity of carbon nitridewith different photoelectrode configurations. ACS Appl. Mater. Interfaces, 8,22287–22294.

53 Xu, Y., Kraft, M., and Xu, R. (2016) Metal-free carbonaceous electrocatalystsand photocatalysts for water splitting. Chem. Soc. Rev., 45, 3039–3052.

54 Maeda, K., Wang, X., Nishihara, Y., Lu, D., Antonietti, M., and Domen,K. (2009) Photocatalytic activities of graphitic carbon nitride powder forwater reduction and oxidation under visible light. J. Phys. Chem. C, 113,4940–4947.

55 Ortega, J. and Sankey, O.F. (1995) Relative stability of hexagonal and planarstructures of hypothetical C&N4 solids. Phys. Rev. B, 51, 2624.

56 Lowther, J.E. (1999) Relative stability of some possible phases of graphiticcarbon nitride. Phys. Rev. B, 59, 11683–11686.


57 Teter, D.M., and Hemley, R.J. (1996) Low compressibility carbon nitrides.Science, 271, 53–55.

58 Liu, A.Y., and Wentzcovitch, R.M. (1994) Stability of carbon nitride solids.Phys. Rev. B, 50, 10362–10365.

59 Kroke, E., Schwarz, M., Horath-Bordon, E., Kroll, P., Noll, B., and Norman,A.D. (2002) Tri-s-tri-s-triazine to graphitic C3N4 structures. New J. Chem.,26, 508–512.

60 Kroke, E., and Schwarz, M. (2004) Novel group 14 nitrides. Coord. Chem.Rev., 248, 493–532.

61 Hollmann, D., Karnahl, M., Tschierlei, S., Kailasam, K., Schneider, M.,Radnik, J., Grabow, K., Bentrup, U., Junge, H., Beller, M., Lochbrunner,S., Thomas, A., and Brückner, A. (2014) Structure–activity relationships inbulk polymeric and sol–gel-derived carbon nitrides during photocatalytichydrogen production. Chem. Mater., 26, 1727–1733.

62 Jürgens, B., Irran, E., Schneider, J., and Schnick, W. (2000) Trimerizationof NaC2N3 to Na3C6N9 in the solid: ab initio crystal structure determina-tion of two polymorphs of NaC2N3 and of Na3C6N9 from X-ray powderdiffractometry. Inorg. Chem., 39, 665–670.

63 Groenewolt, M. and Antonietti, M. (2005) Synthesis of g-C3N4 nanoparticlesin mesoporous silica host matrices. Adv. Mater., 17, 1789–1792.

64 Cui, Y., Ding, Z., Liu, P., Antonietti, M., Fu, X., and Wang, X. (2012)Metal-free activation of H2O2 by g-C3N4 under visible light irradiationfor the degradation of organic pollutants. Phys. Chem. Chem. Phys., 14,1455–1462.

65 Wang, X., Blechert, S., and Antonietti, M. (2012) Polymeric graphitic carbonnitride for heterogeneous photocatalysis. ACS Catal., 2, 1596–1606.

66 Ji, H., Chang, F., Hu, X., Qin, W., and Shen (2013) Photocatalytic degra-dation of 2,4,6-trichlorophenol over g-C3N4 under visible light irradiation.Chem. Eng. J., 218, 183–190.

67 Yan, S.C., Li, Z.S., and Zou, Z.G. (2009) Photodegradation perfor-mance of g-C3N4 fabricated by directly heating melamine. Langmuir, 25,10397–10401.

68 Zhang, G., Zhang, J., Zhang, M., and Wang, X. (2012) Polycondensation ofthiourea into carbon nitride semiconductors as visible light photocatalysts. J.Mater. Chem., 22, 8083–8091.

69 Dong, F., Wang, Z., Sun, Y., Ho, W.K., and Zhang, H. (2013) Engineering thenanoarchitecture and texture of polymeric carbon nitride semiconductor forenhanced visible light photocatalytic activity. J. Colloid Interface Sci., 401,70–79.

70 Jorge, A.B., Martin, D.J., Dhanoa, M.T.S., Rahman, A.S., Makwana, N., Tang,J., Sella, A., Corà, F., Firth, S., Darr, J.A., and McMillan, P.F. (2013) H2 andO2 evolution from water half-splitting reactions by graphitic carbon nitridematerials. J. Phys. Chem. C, 117, 7178–7185.

71 Liu, J., Zhang, T., Wang, Z., Dawson, G., and Chen, W. (2011) Simple pyroly-sis of urea into graphitic carbon nitride with recyclable adsorption and pho-tocatalytic activity. J. Mater. Chem., 21, 14398–14401.

References 357

72 Dong, F., Wu, L., Sun, Y., Fu, M., Wu, Z., and Lee, S.C. (2011) Efficient syn-thesis of polymeric g-C3N4 layered materials as novel efficient visible lightdriven photocatalysts. J. Mater. Chem., 21, 15171–15174.

73 Sakata, Y., Yoshimoto, K., Kawaguchi, K., Imamura, H., and Higashimoto, S.(2011) Preparation of a semiconductive compound obtained by the pyrol-ysis of urea under N2 and the photocatalytic property under visible lightirradiation. Catal. Today, 161, 41–45.

74 Zhang, Y., Liu, J., Wu, G., and Chen, W. (2012) Porous graphitic car-bon nitride synthesized via direct polymerization of urea for efficientsunlight-driven photocatalytic hydrogen production. Nanoscale, 4,5300–5303.

75 Yuan, Y.P., Xu, W.T., Yin, L.S., Cao, S.W., Liao, Y.S., Tng, Y.Q., and Xue,C. (2013) Large impact of heating time on physical properties and photo-catalytic H2 production of g-C3N4 nanosheets synthesized through ureapolymerization in Ar atmosphere. Int. J. Hydrogen Energy, 38, 13159–13163.

76 Yan, H., Chen, Y., and Xu, S. (2012) Synthesis of graphitic carbon nitride bydirectly heating sulfuric acid treated melamine for enhanced photocatalyticH2 production from water under visible light. Int. J. Hydrogen Energy, 37,125–133.

77 Zhang, J., Zhang, M., Zhang, G., and Wang, X. (2012) Synthesis of carbonnitride semiconductors in sulfur flux for water photoredox catalysis. ACSCatal., 2, 940–948.

78 Long, B., Lin, J., and Wang, X. (2014) Thermally-induced desulfurization andconversion of guanidine thiocyanate into graphitic carbon nitride catalystsfor hydrogen photosynthesis. J. Mater. Chem. A, 2, 2942–2951.

79 Liu, J., Li, W., Duan, L., Li, X., Ji, L., Geng, Z., Huang, K., Lu, L., Zhou, L.,Liu, Z., Chen, W., Liu, L., Feng, S., and Zhang, Y. (2015) A graphene-likeoxygenated carbon nitride material for improved cycle-life lithium/sulfurbatteries. Nano Lett., 15, 5137–5142.

80 Shalom, M., Inal, S., Fettkenhauer, C., Neher, D., and Antonietti, M. (2013)Improving carbon nitride photocatalysis by supramolecular preorganizationof monomers. J. Am. Chem. Soc., 135, 7118–7121.

81 Jun, Y.S., Lee, E.Z., Wang, X.C., Hong, W.H., Stucky, G.D., and Thomas, A.(2013) From melamine-cyanuric acid supramolecular aggregates to carbonnitride hollow spheres. Adv. Funct. Mater., 23, 3661–3667.

82 Jun, Y.S., Park, J., Lee, S.U., Thomas, A., Hong, W.H., and Stucky, G.D.(2013) Three-dimensional macroscopic assemblies of low-dimensional car-bon nitrides for enhanced hydrogen evolution. Angew. Chem. Int. Ed., 52,11083–11087.

83 Jun, Y.S., Park, J., Lee, S.U., Thomas, A., Hong, W.H., and Stucky, G.D.(2013) Three-dimensional macroscopic assemblies of low-dimensional car-bon nitrides for enhanced hydrogen evolution. Angew. Chem. Int. Ed., 125,11289–11293.

84 Martin, D.J., Qiu, K., Shevlin, S.A., Handoko, A.D., Chen, X., Guo, Z., andTang, J. (2014) Highly efficient photocatalytic H2 evolution from water usingvisible light and structure-controlled graphitic carbon nitride. Angew. Chem.Int. Ed., 53, 9240–9245.


85 Xu, J., Zhang, L., Shi, R., and Zhu, Y. (2013) Chemical exfoliation ofgraphitic carbon nitride for efficient heterogeneous photocatalysis. J. Mater.Chem. A, 1, 14766–14772.

86 Pawar, R.C., Kang, S., Park, J.H., Kim, J., Ahn, S., and Lee, C.S. (2016)Room-temperature synthesis of nanoporous 1D microrods of graphiticcarbon nitride (g-C3N4) with highly enhanced photocatalytic activity andstability. Sci. Rep., 6. doi: 10.1038/srep31147

87 Zhao, F., Cheng, H., Hu, Y., Song, L., Zhang, Z., Jiang, L., and Qu, L.(2014) Functionalized graphitic carbon nitride for metal-free, flexible andrewritable nonvolatile memory device via direct laser-writing. Sci. Rep., 4.doi: 10.1038/srep05882

88 Xu, H., Yan, J., She, X., Xu, L., Xia, J., Xu, Y., Song, Y., Huang, L., and Li, H.(2014) Graphene-analogue carbon nitride: novel exfoliation synthesis and itsapplication in photocatalysis and photoelectrochemical selective detection oftrace amount of Cu2+. Nanoscale, 6, 1406–1415.

89 Lu, Y., Chu, D., Zhu, M., Du, Y., and Yang, P. (2015) Exfoliated carbonnitride nanosheets decorated with NiS as an efficient noble-metal-freevisible-light-driven photocatalyst for hydrogen evolution. Phys. Chem. Chem.Phys., 17, 17355–17361.

90 Du, X., Zou, G., Wang, Z., and Wang, X. (2015) A scalable chemical routeto soluble acidified graphitic carbon nitride: an ideal precursor for isolatedultrathin g-C3N4 nanosheets. Nanoscale, 7, 8701–8706.

91 Lee, J.H., Park, M.J., Yoo, S.J., Jang, J.H., Kim, H.J., Nam, S.W., Yoon, C.W.,and Kim, J.Y. (2015) A highly active and durable Co–N–C electrocatalystsynthesized using exfoliated graphitic carbon nitride nanosheets. Nanoscale,7, 10334–10339.

92 Wang, Z., Chen, L., Du, X., Zou, G., and Wang, X. (2016) A “pillared” pro-cess to construct graphitic carbon nitride based functionalized mesoporousmaterials. RSC Adv., 6, 15605–15609.

93 Svec, F., Germain, J., and Frechet, J.M.J. (2009) Nanoporous polymers forhydrogen storage. Small, 5, 1098–1111.

94 Thomas, A., Kuhn, P., Weber, J., Titirici, M.M., and Antonietti, M. (2009)Porous polymers: enabling solutions for energy applications. Macromol.Rapid Commun., 30, 221–236.

95 Thomas, A., Goettmann, F., and Antonietti, M. (2008) Hard templates forsoft materials: creating nanostructured organic materials. Chem. Mater., 20,738–755.

96 Wan, Y. and Yan, D. (2007) On the controllable soft-templating approach tomesoporous silicates. Chem. Rev., 107, 2821–2860.

97 Hamley, I.W. (2003) Nanostructure fabrication using block copolymers.Nanotechnology, 14, R39–R54.

98 Ryoo, R., Joo, S.H., Kruk, M., and Jaroniec, M. (2001) Ordered mesoporouscarbons. Adv. Mater., 13, 677–681.

99 Kailasam, K., Epping, J.D., Thomas, A., Losseb, S., and Junge, H. (2011)Mesoporous carbon nitride–silica composites by a combined sol–gel/thermalcondensation approach and their application as photocatalysts. EnergyEnviron. Sci., 4, 4668–4674.

References 359

100 Li, X., Masters, A.F., and Maschmeyer, T. (2015) Photocatalytic hydrogenevolution from silica-templated polymeric graphitic carbon nitride–is thesurface area important. ChemCatChem, 7, 121–126.

101 Yan, H.J. (2012) Soft-templating synthesis of mesoporous graphitic carbonnitride with enhanced photocatalytic H2 evolution under visible light. Chem.Commun., 48, 3430–3432.

102 Wang, Y., Ibad, M.F., Kosslick, H., Harloff, J., Beweries, T., Radnik, J., Schulz,A., Tschierlei, S., Lochbrunner, S., and Guo, X. (2015) Synthesis and com-parative study of the photocatalytic performance of hierarchically porouspolymeric carbon nitrides. Microporous Mesoporous Mater., 211, 182–191.

103 Niu, P., Zhang, L.L., Liu, G., and Cheng, H.M. (2012) Graphene-like carbonnitride nanosheets for improved photocatalytic activities. Adv. Funct. Mater.,22, 4763–4770.

104 Han, Q., Wang, B., Gao, J., Cheng, Z., Zhao, Y., Zhang, Z., and Qu, L. (2016)Atomically thin mesoporous nanomesh of graphitic C3N4 for high-efficiencyphotocatalytic hydrogen evolution. ACS Nano, 10, 2745–2751.

105 Tonda, S., Kumar, S., Kandula, S., and Shankar, V. (2014) Fe-doped and-mediated graphitic carbon nitride nanosheets for enhanced photocatalyticperformance under natural sunlight. J. Mater. Chem. A, 2, 6772–6780.

106 Zhang, M., Bai, X.J., Liu, D., Wang, J., and Zhu, Y.F. (2015) Enhanced cat-alytic activity of potassium-doped graphitic carbon nitride induced by lowervalence position. Appl. Catal., B, 164, 77–81.

107 He, F., Chen, G., Yu, Y.G., Zhou, Y.S., Zheng, Y., and Hao, S. (2015) Thesulfur-bubble template-mediated synthesis of uniform porous g-C3N4 withsuperior photocatalytic performance. Chem. Commun., 51, 425–427.

108 Wang, X.C., Maeda, K., Chen, X., Takanabe, K., Domen, K., Hou, Y., Fu, X.,and Antonietti, M. (2009) Polymer semiconductors for artificial photosynthe-sis: hydrogen evolution by mesoporous graphitic carbon nitride with visiblelight. J. Am. Chem. Soc., 131, 1680–1681.

109 Li, X.H., Zhang, J., Chen, X., Fischer, A., Thomas, A., Antonietti, M.,and Wang, X.C. (2011) Condensed graphitic carbon nitride nanorods bynanoconfinement: promotion of crystallinity on photocatalytic conversion.Chem. Mater., 23, 4344–4348.

110 Wang, J.H., Zhang, C., Shen, Y.F., Zhou, Z.X., Yu, J.C., Li, Y., Wei, W., Liu,S.Q., and Zhang, Y.J. (2015) Environment-friendly preparation of porousgraphite-phase polymeric carbon nitride using calcium carbonate as tem-plates, and enhanced photoelectrochemical activity. J. Mater. Chem. A, 3,5126–5131.

111 Chen, D., Yang, J., and Ding, H. (2017) Synthesis of nanoporous carbonnitride using calcium carbonate as templates with enhanced visible-lightphotocatalytic activity. Appl. Surf. Sci., 391, 384–391.

112 Han, K.K., Wang, C.C., Li, Y.Y., Wan, M.M., Wang, Y., and Zhu, J.H. (2013)Facile template-free synthesis of porous g-C3N4with high photocatalyticperformance under visible light. RSC Adv., 3, 9465–9469.

113 Yang, S., Gong, Y., Zhang, J., Zhan, L., Ma, L., Fang, Z., Vajtai, R., Wang, X.,and Ajayan, P.M. (2013) Exfoliated graphitic carbon nitride nanosheets as


efficient catalysts for hydrogen evolution under visible light. Adv. Mater., 25,2452–2456.

114 Ma, L.T., Fan, H.Q., Wang, J., Zhao, Y.W., Tian, H.L., and Dong, G.Z. (2016)Water-assisted ions in situ intercalation for porous polymeric graphitic car-bon nitride nanosheets with superior photocatalytic hydrogen evolutionperformance. Appl. Catal., B, 190, 93–102.

115 Lin, L.H., Ou, H.H., Zhang, Y.F., and Wang, X.C. (2016) Tri-s-triazine-basedcrystalline graphitic carbon nitrides for highly efficient hydrogen evolutionphotocatalysis. ACS Catal., 6, 3921–3931.

116 Rong, X.S., Qiu, F.X., Rong, J., Zhu, X.L., Yan, J., and Yang, D.Y. (2016)Enhanced visible light photocatalytic activity of W-doped porous g-C3N4and effect of H2O2. Mater. Lett., 164, 127–131.

117 Wang, Y.G., Wang, Y.Z., Li, Y.G., Shi, H.C., Xu, Y.L., Qin, H.F., Li, X., Zuo,Y.H., Kang, S.F., and Cui, L.F. (2015) Simple synthesis of Zr-doped graphiticcarbon nitride towards enhanced photocatalytic performance under simu-lated solar light irradiation. Catal. Commun., 72, 24–28.

118 Fang, J.W., Fan, H.Q., Li, M.M., and Long, C.B. (2015) Nitrogen self-dopedgraphitic carbon nitride as efficient visible light photocatalyst for hydrogenevolution. J. Mater. Chem. A, 3, 13819–13826.

119 Zhou, Y.J., Zhang, L.X., Huang, W.M., Kong, Q.L., Fan, X.Q., Wang, M., andShi, J.L. (2016) N-doped graphitic carbon-incorporated g-C3N4 for remark-ably enhanced photocatalytic H2 evolution under visible light. Carbon, 99,111–117.

120 Li, J.H., Shen, B.A., Hong, Z.H., Lin, B.Z., Gao, B.F., and Chen, Y.L. (2012)A facile approach to synthesize novel oxygen-doped g-C3N4with superiorvisible-light photoreactivity. Chem. Commun., 48, 12017–12019.

121 Huang, Z.F., Song, J.J., Pan, L., Wang, Z.M., Zhang, X.Q., Zou, J.J., Mi, W.B.,Zhang, X.W., and Wang, L. (2015) Carbon nitride with simultaneous porousnetwork and O-doping for efficient solar-energy-driven hydrogen evolution.Nano Energy, 12, 646–656.

122 Guo, S.N., Zhu, Y., Yan, Y.Y., Min, Y.L., Fan, J.C., and Xu, Q.J.H. (2016)Structured graphitic carbon nitride thin sheets with edge oxygen doping viaphoto-Fenton reaction with enhanced photocatalytic activity. Appl. Catal., B,185, 315–321.

123 Cui, J., Liang, S.H., Wang, X.H., and Zhang, J.M. (2016) First principle mod-eling of oxygen-doped monolayer graphitic carbon nitride. Mater. Chem.Phys., 161, 194–200.

124 Zhang, J.S., Sun, J.H., Maeda, K., Domen, K., Liu, P., Antonietti, M., Fu,X.Z., and Wang, X.C. (2011) Sulfur-mediated synthesis of carbon nitride:band-gap engineering and improved functions for photocatalysis. EnergyEnviron. Sci., 4, 675–678.

125 Lin, S., Ye, X.X., Gao, X.M., and Huang, J. (2015) Mechanistic insight intothe water photooxidation on pure and sulfur-doped g-C3N4 photocatalystsfrom DFT calculations with dispersion corrections. J. Mol. Catal. A Chem.,406, 137–144.

126 Ran, J.R., Ma, T.Y., Gao, G.P., Du, X.W., and Qiao, S.Z. (2015) PorousP-doped graphitic carbon nitride nanosheets for synergistically enhanced

References 361

visible-light photocatalytic H2 production. Energy Environ. Sci., 8,3708–3717.

127 Zhu, Y.P., Ren, T.Z., and Yuana, Z.Y. (2015) Mesoporous phosphorus-dopedg-C3N4 nanostructured flowers with superior photocatalytic hydrogen evolu-tion performance. ACS Appl. Mater. Interfaces, 7, 16850–16856.

128 Guo, S.E., Deng, Z.P., Li, M.X., Jiang, B.J., Tian, C.G., Pan, Q.J., andFu, H.G. (2016) Phosphorus-doped carbon nitride tubes with a layeredmicro-nanostructure for enhanced visible-light photocatalytic hydrogenevolution. Angew. Chem. Int. Ed., 55, 1830–1834.

129 Liu, J.J. (2016) Effect of phosphorus doping on electronic structure and pho-tocatalytic performance of g-C3N4: insights from hybrid density functionalcalculation. J. Alloys Compd., 672, 271–276.

130 Zhang, G.G., Zhang, M.W., Ye, X.X., Qiu, X.Q., Lin, S., and Wang, X.C.(2014) Iodine modified carbon nitride semiconductors as visible light photo-catalysts for hydrogen evolution. Adv. Mater., 26, 805–809.

131 Han, Q., Hu, C.G., Zhao, F., Zhang, Z.P., Chen, N., and Qu, L.T. (2015)One-step preparation of iodine-doped graphitic carbon nitride nanosheets asefficient photocatalysts for visible light water splitting. J. Mater. Chem. A, 3,4612–4619.

132 Wang, Y., Di, Y., Antonietti, M., Li, H.R., Chen, X.F., and Wang, X.C. (2010)Excellent visible-light photocatalysis of fluorinated polymeric carbon nitridesolids. Chem. Mater., 22, 5119–5121.

133 Lan, Z.A., Zhang, G.G., and Wang, X.C. (2016) A facile synthesis ofBr-modified g-C3N4 semiconductors for photoredox water splitting. Appl.Catal., B, 192, 116–125.

134 Maeda, K. (2013) Z-scheme water splitting using two different semiconduc-tor photocatalysts. ACS Catal., 3, 1486–1503.

135 Kailasam, K., Fischer, A., Zhang, G., Zhang, J., Schwarze, M., Schrçder,M., Wang, X., Schomacker, R., and Thomas, A. (2015) Mesoporous carbonnitride-tungsten oxide composites for enhanced photocatalytic hydrogenevolution. ChemSusChem, 8, 1404–1410.

136 Cheng, R., Fan, X., Wang, M., Li, M., Tian, J., and Zhang, L. (2016) Facileconstruction of CuFe2O4/g-C3N4 photocatalyst for enhanced visible-lighthydrogen evolution. RSC Adv., 6, 18990–18995.

137 Akple, M.S., Low, J., Wageh, S., Al-Ghamdi, A.A., Yu, J., and Zhang, J.(2015) Enhanced visible light photocatalytic H2-production of g-C3N4/WS2composite heterostructures. Appl. Surf. Sci., 358, 196–203.

138 Ye, R.Q., Fang, H.B., Zheng, Y.Z., Li, N., Wang, Y., and Tao, X. (2016) Fabri-cation of CoTiO3/g-C3N4 hybrid photocatalysts with enhanced H2 evolution:Z-scheme photocatalytic mechanism insight. ACS Appl. Mater. Interfaces, 8,13879–13889.

139 Xiang, Q., Yu, J., and Jaroniec, M. (2011) Preparation and enhancedvisible-light photocatalytic H2-production activity of graphene/C3N4composites. J. Phys. Chem. C, 115, 7355–7363.

140 Xu, Y., Xu, H., Wang, L., Yan, J., Li, H., Song, Y., Huang, L., and Cai, G.(2013) The CNT modified white C3N4 composite photocatalyst withenhanced visible-light response photoactivity. Dalton Trans., 42, 7604–7613.


141 Chen, Y., Li, J., Hong, Z., Shen, B., Lin, B., and Gao, B. (2014) Origin of theenhanced visible-light photocatalytic activity of CNT modified g-C3N4 forH2 production. Phys. Chem. Chem. Phys., 16, 8106–8113.

142 Chu, S., Wang, Y., Guo, Y., Feng, J.Y., Wang, C.C., Luo, W.J., Fan, X.X., andZou, Z.G. (2013) Band structure engineering of carbon nitride: in search ofa polymer photocatalyst with high photooxidation property. ACS Catal., 3,912–919.

143 Zhang, M. and Wang, X. (2014) Two dimensional conjugated polymerswith enhanced optical absorption and charge separation for photocatalytichydrogen evolution. Energy Environ. Sci., 7, 1902–1906.

144 Xing, Z., Chen, Z., Zong, X., and Wang, L. (2014) A new type of carbonnitride-based polymer composite for enhanced photocatalytic hydrogenproduction. Chem. Commun., 50, 6762–6764.

145 Takanabe, K., Kamata, K., Wang, X., Antonietti, M., Kubota, J., and Domen,K. (2010) Photocatalytic hydrogen evolution on dye-sensitized mesoporouscarbon nitride photocatalyst with magnesium Phthalocyanine. Phys. Chem.Chem. Phys., 12, 13020–13025.

146 Zhang, X., Yu, L., Zhuang, C., Peng, T., Li, R., and Li, X. (2014) Highlyasymmetric phthalocyanine as a sensitizer of graphitic carbon nitride forextremely efficient photocatalytic H2 production under near-infrared light.ACS Catal., 4, 162–170.

147 Diebold, U. (2011) Photocatalysts: closing the gap. Nat. Chem., 3, 271.148 Zhang, G., An Lan, Z., Lin, L., Lin, S., and Wang, X. (2016) Overall water

splitting by Pt/g-C3N4 photocatalysts without using sacrificial agents. Chem.Sci., 7, 3062–3066.

149 Liu, J., Liu, Y., Liu, N., Han, Y., Zhang, X., Huang, H., Lifshitz, Y., Lee, S.T.,Zhong, J., and Kang, Z. (2015) Water splitting. Metal-free efficient photo-catalyst for stable visible water splitting via a two-electron pathway. Science,347, 970–974.

150 Schröder, M., Kailasam, K., Borgmeyer, J., Neumann, M., Thomas, A.,Schomäcker, R., and Schwarze, M. (2015) Hydrogen evolution reaction ina large-scale reactor using a carbon nitride photocatalyst under naturalsunlight irradiation. Energy Technol., 3, 1014–1017.

151 Jürgens, B., Irran, E., Senker, J., Kroll, P., Mîller, H., and Schnick, W. (2003)Melem (2,5,8-Triamino-tri-s-triazine), an important intermediate duringcondensation of melamine rings to graphitic carbon nitride: synthesis, struc-ture determination by X-ray powder diffractometry, solid-state NMR, andtheoretical studies. J. Am. Chem. Soc., 125, 10288–10300.

152 Schwarzer, A. and Kroke, E. (2011) A series of aryl-/alkyl-substitutedmelems. New J. Chem., 35, 953–958.

153 Kailasam, K., Schmidt, J., Bildirir, H., Zhang, G., Blechert, S., Wang, X.C.,and Thomas, A. (2013) Room temperature synthesis of heptazine-basedmicroporous polymer networks as photocatalysts for hydrogen evolution.Macromol. Rapid Commun., 34, 1008–1013.

154 Kailasam, K., Mesch, M.B., Mçhlmann, L., Baar, M., Blechert, S., Schwarze,M., Schrçder, M., Schomacker, R., Senker, J., and Thomas, A. (2016)

References 363

Donor–acceptor-type heptazine-based polymer networks for photocatalytichydrogen evolution. Energy Technol., 4, 744–750.

155 Schwinghammer, K., Tuffy, B., Mesch, M.B., Wirnhier, E., Martineau, C.,Taulelle, F., Schnick, W., Senker, J., and Lotsch, B.V. (2013) Triazine-basedcarbon nitrides for visible-light-driven hydrogen evolution. Angew. Chem.Int. Ed., 52, 2435–2439.

156 Bhunia, M.K., Yamauchi, K., and Takanabe, K. (2014) Harvesting solarlight with crystalline carbon nitrides for efficient photocatalytic hydrogenevolution. Angew. Chem. Int. Ed., 53, 11001–11005.

157 Gu, C., Huang, N., Chen, Y., Zhang, H., Zhang, S., Li, F., Ma, Y., and Jiang,D. (2016) Porous organic polymer films with tunable work functions andselective hole and electron flows for energy conversions. Angew. Chem. Int.Ed., 128, 3101–3105.

158 Wang, Z.J., Ghasimi, S., Landfester, K., and Zhang, K.A.I. (2015) Molecularstructural design of conjugated microporous poly(benzooxadiazole) net-works for enhanced photocatalytic activity with visible light. Adv. Mater., 27,6265–6270.

159 Stegbauer, L., Schwinghammer, K., and Lotsch, B.V. (2014) A hydrazone-based covalent organic framework for photocatalytic hydrogen production.Chem. Sci., 5, 2789–2793.

160 Vyas, V.S., Haase, F., Stegbauer, L., Savasci, G., Podjaski, F., Ochsenfeld, C.,and Lotsch, B.V. (2015) A tunable azine covalent organic framework plat-form for visible light-induced hydrogen generation. Nat. Commun., 6, 8508.

161 Sprick, R.S., Jiang, J.X., Bonillo, B., Ren, S., Ratvijitvech, T., Guiglion, P.,Zwijnenburg, M.A., Adams, D.J., and Cooper, A.I. (2015) Tunable organicphotocatalysts for visible-light-driven hydrogen evolution. J. Am. Chem. Soc.,137, 3265–3270.

162 Sprick, R.S., Bonillo, B., Sachs, M., Clowes, R., Durrant, J.R., Adams, D.J.,and Cooper, A.I. (2016) Extended conjugated microporous polymersfor photocatalytic hydrogen evolution from water. Chem. Commun., 52,10008–10011.

163 Li, L., Lo, W., Cai, Z., Zhang, N., and Yu, L. (2016) Donor–acceptor porousconjugated polymers for photocatalytic hydrogen production: the importanceof acceptor comonomer. Macromolecules, 49, 6903–6909.

164 Fan, X., Zhang, L., Cheng, R., Wang, M., Li, M., Zhou, Y., and Shi, J. (2015)Construction of graphitic C3N4-based intramolecular donor–acceptorconjugated copolymers for photocatalytic hydrogen evolution. ACS Catal., 5,5008–5015.

165 Zhang, G., AnLan, Z., and Wang, X.C. (2016) Conjugated polymers:catalysts for photocatalytic hydrogen evolution. Angew. Chem. Int. Ed.,55, 15712–15727.

166 Vyas, V.S., Lau, V.W., and Lotsch, B.V. (2016) Soft photocatalysis: organicpolymers for solar fuel production. Chem. Mater., 28, 5191–5204.

365

13

Solar Photochemical Splitting of WaterSrinivasa Rao Lingampalli and C. N. R. Rao

New Chemistry Unit, CSIR Centre of Excellence in Chemistry and International Centre for Materials Science,Sheikh Saqr Laboratory, Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur P.O.,Bangalore 560064, India

13.1 Introduction

Solar-to-chemical energy conversion provides a means to overcome the energycrisis and environmental issues presently faced by humankind. Increase in energyconsumption and rise in the atmospheric levels of CO2 due to the burning ofcarbon-based fuels are the problems faced today [1]. There have been effortsto develop clean, reliable, environmentally friendly, and sustainable alternativeenergy sources. Nature provides us a strategy to overcome the energy crisisemploying the photosynthesis process which uses earth-abundant water andCO2 in the conversion of solar energy to chemical fuels. Production of a cleanenergy source such as hydrogen from water, using solar energy by mimickingphotosynthesis would be a worthwhile task. It should be noted that hydrogen isa green-energy source with a chemical energy density of 142 MJ kg−1.

Numerous efforts have been made to achieve direct conversion of solarenergy to chemical fuels. In solar-driven or photocatalytic water splitting, waterdecomposes to yield hydrogen and oxygen [2–4]. Several materials that can drivewater splitting under solar irradiation have been identified, and mechanism,role of different components, as well as physical and optical properties of thephotocatalysts for water splitting have been elucidated. However, even thepresent state-of-the-art materials are not entirely suitable for practical usage andthe goal of developing efficient, low-cost, scalable, and stable materials for thisprocess is being pursued actively.

In recent years, half reactions in water splitting have received attention. Halfreactions of water splitting are generally carried out in the presence of sacrificialelectron donors or acceptors. The sacrificial agents allow one to study one halfreaction independent of the counter half reaction. The sacrificial agents, whichare consumed in the process, help to optimize water reduction or water oxida-tion. Photocatalysts can be used for photochemical water splitting via two typesof configurations: photoelectrochemical cells and particulate photocatalytic sys-tems. In this chapter, we briefly discuss the fundamentals of photocatalytic water


366 13 Solar Photochemical Splitting of Water

splitting and use of the visible-light-responsive photocatalysts for overall watersplitting, water oxidation, and water reduction. We also describe the possibilityof combining water reduction along with the oxidation of other molecules.

13.2 Photocatalytic Water Splitting

13.2.1 Fundamentals of Water Splitting

Photocatalysis is based on principles of photochemistry and redox catalysis.Splitting of water to hydrogen and oxygen is an endothermic reaction with apositive change in Gibbs free energy of +237 kJ mol−1 (Eq. (13.1)). It is an uphillreaction and external driving energy is, therefore, necessary to drive the reaction.For example, external electrical energy is employed in the case of electrolysisof water. Solar energy can similarly be used in the photolysis of water (solar orphotocatalytic water splitting). Splitting of water is a combination of the twohalf reactions: reduction of water and oxidation of water (Eqs. (13.2) and (13.3))[5]. Reduction of water to hydrogen is a two electron-transfer process, whereasoxidation of water to oxygen is a four electron-transfer process involving sluggishkinetics. Oxidation of water is a bottleneck reaction in determining the overallrate of the reaction.

H2O → H2 + 1∕2O2(ΔG = +237 kJ mol−1, ΔE∘ = 1.23 V) (13.1)

Reduction half-reaction ∶ 4H+ + 4e− → 2H2(0 V vs NHE) (13.2)Oxidation half-reaction ∶ 2H2O → 4H+ + O2 + 4e−(1.23 V vs NHE)

(13.3)Light-induced splitting of water is possible with a photon of wavelength less

than 1000 nm (equivalent to 1.23 eV) [6], and four such photons are involved inthe formation of one molecular oxygen (O2). Photocatalytic water splitting pro-cess involves three steps (Figure 13.1a) [5]:1) Absorption of light by a light-harvesting unit2) Generation and separation of electron–hole pairs3) Redox reactions at the surface of the photocatalysts.

The overall efficiency (𝜂c) of solar energy conversion is, therefore, the cumula-tive result of the efficiencies of all these processes such as light absorption (𝜂abs),charge generation (𝜂cg), separation and migration (𝜂csm), and charge utilization(𝜂cu) as shown in Eq. (13.4) [4].

Overall efficiency, 𝜂c = 𝜂abs × 𝜂cg × 𝜂csm × 𝜂cu (13.4)In semiconductor-based photocatalysis, upon light irradiation on the semi-

conductor, electrons get excited to the conduction band (CB) while the holesremain in the valence band (VB). The excited electrons are utilized in thereduction of water, whereas the holes are utilized in the oxidation of water(Figure 13.1b). In order to reduce the proton, the energy of the excited electronshould be sufficiently large. The CB minimum (CBM) should be more negativethan the water reduction potential. Similarly, VB maximum (VBM) should bemore positive than the water oxidation potential. Thus, semiconductors that

13.2 Photocatalytic Water Splitting 367

(a) (b)

(c)

3.0

2.0

1.0

0.0

–1.0

E vs NHE

3.0

3.2

3.2

3.2

5.0 3

.4

3.4

2.2

2.8

3.2

3.8

1.1

1.7

2.4 1

.3

H+/H2

TiO

2 (A

)

SrT

iO3

WO

3

Sn

O2

Mo

S2

TiO

2(R

)

Zn

O

ZrO

2

Nb

2O

5

KTa

O3

Fe

2O

3

Ba

TiO

3

Si

Cd

Se

Cd

S

Zn

2N

F2

.8

O2/H2O

–8.0

–7.0

–6.0

–5.0

–4.0

–3.0

0.0

Vacuum level (eV)

I II III

e– h+ h+e–

(i)(ii)(ii)

hv > Eg

H+

H2

O2

OH–(iii)

(iv)

(iii)

Po

ten

tia

l e

ne

rgy

CB

VB

e–

e–

e–

h+

H+

H2

O2

OH–

H+/H2

O2/H2O

Figure 13.1 Schematic illustration of (a) the processes involved and (b) relative energy levelsand mechanism of photocatalytic water splitting. (c) Representation of band positions ofsemiconductors relative to the redox potentials of water. Dashed lines indicate the waterreduction and oxidation potentials. (Reproduced with permission from Ref. [5]. Copyright2015, John Wiley & Sons.)

straddle the water reduction and oxidation potentials between CB and VB arethermodynamically suitable for the splitting of water (Figure 13.1c).

We can classify semiconductors into three groups based on their band edgepositions relative to the water reduction and oxidation potentials. In group I,TiO2, SrTiO3, KTaO3, ZnO, and so on straddle the water reduction and oxidationpotentials and are thermodynamically suitable for the direct splitting of water.Materials that belong to group II (such as Si) and III (such as BiVO4, WO3,and Fe2O3) are only capable of reduction and oxidation of water, respectively.However, construction of Z-scheme photocatalysts with a combination of groupII and III materials or group III and group I materials can result in a suitableconfiguration.

13.2.2 Light-Harvesting Units

Light-harvesting units or sensitizers play the primary role of absorption oflight, which triggers the entire process. There are three main ways of using thelight-harvesting units in photocatalysis [6]:1) Combination of sensitizer–electron relay2) Combination of sensitizer–semiconductor3) Semiconductors.


S*+ R

S + R

S+ + R–

H2O

O2

H2O

H2

(a)

S*

S

hvhv

S+

H2O

O2

H2O

H2 CBCAT

1

CAT

2

CAT2CAT

1

Back

reac

tion

Forwardtransfer

EF

Semiconductor

(b)

O2

H2O

H2

e

CB

EF

VB

e–

h+H2O

Semiconductor

(c)

hvCAT

2

CAT

1

Figure 13.2 Schematic representation of light-harvesting units in (a) sensitizer-relay(b) sensitizer-semiconductor, and (c) semiconductor photocatalytic systems. (Reproduced withpermission from Ref. [6]. Copyright 1981, American Chemical Society.)

Figure 13.2 shows the mechanism of water splitting by using these systems [6].Upon light irradiation, sensitizer (S) gets excited to S*, and subsequently, transferselectrons to the electron relay (R) producing the charged species, S+ and R−. R−

donates electron to water and produces hydrogen, and S+ accepts the electronsfrom water producing oxygen (Eqs. (13.5)–(13.8)).

Sh𝜈−−→ S∗ (13.5)

S∗ + R → S+ + R−(electron-transfer) (13.6)R− + H2O → 0.5H2 + OH− + R (13.7)2S+ + H2O → 0.5O2 + 2H+ + 2S (13.8)

In order to drive reduction and oxidation of water (Eqs. (13.7) and (13.8)), theenergy levels of S+ and R− should satisfy the conditions given in Eq. (13.9). In thesemiconductor-sensitizer system, a semiconductor with suitable band positionsacts as an electron relay and accepts the electron, reducing water while S+ oxi-dizes water to oxygen. Where a colloidal semiconductor particle acts as the pho-tosensitizer, the electron in the CB reduces water and the hole in the VB oxidizeswater. The CBM and VBM should satisfy the Eq. (13.10). Thus, the thermody-namic requirements of water splitting under standard conditions are as follows:

E∘(S+∕S) > 1.23 V (vs NHE) and E∘(R∕R−) < 0 V(vs NHE) (13.9)EVBM > 1.23 V (vs NHE) and ECBM < 0 V (vs NHE) (13.10)

Ruthenium-based dyes (such as Ru(bpy)32+) are the most popular and efficient

photocatalysts for dye-assisted water splitting [7]. They also possess good

13.2 Photocatalytic Water Splitting 369

visible-light absorption; however, the excited state of the dye is too short. Thesedyes have been modified with various functional groups such as carboxylic andphosphate groups to assist the adsorption of dye to semiconductors, to facilitatequick transfer of charge carriers. Typically, Ru(bpy)3

2+ and its derivativesporphyrin derivatives, acridine dyes, and Eosin Y are employed as sensitizers.Viologens, Eu3+, V3+, Ru(bpy)3

3+, and cobalt complexes are employed as electronrelay materials. Semiconductors such as CdS and C3N4 as well as plasmonicmetal nanoparticles (Au and Ag) are also employed for visible-light harvesting.

13.2.3 Photocatalytic Activity

In order to compare the activities of photocatalysts, it is necessary to employappropriate units for the activity. In most of the studies, the activities are reportedin terms of μmol h−1 g−1, μmol h−1 and apparent quantum yield (AQY), turnovernumber (TON) or turnover frequency (TOF). The activity of photocatalyst is nota linear function of the weight of the photocatalyst. Representing activity in theunits of μmol h−1 g−1 or TON, therefore, is misleading. Representing activities inother units such as μmol-h−1 (weight independent) is, therefore, necessary. AQYor QY (Eq. (13.11)) represents the efficiency of conversion of the photoexcitedcharges to the products and it is necessary to mention. AQY should be mea-sured using the same irradiation source that has been employed for measuringthe activity.

AQY (%) = (No. of reacted e−∕No. of incident photons) × 100 (13.11)TON is often used to represent the activity and is useful in distinguishing catalyticreaction from sacrificial reactions (Eq. (13.12)). TOF is the TON achieved per unittime (Eq. (13.13)), given in units of h−1 or s−1 [3].

TON = (No. of reacted electrons in moles)∕(No. of moles of photocatalyst)(13.12)

TOF = TON∕Duration of the reaction (13.13)At present scenario, it is difficult to compare the activities of any two pho-

tocatalysts especially in the case of sacrificial-assisted reactions, carried outunder different conditions (varying amount of catalyst, nature of the electrondonors or sacrificial agents, wavelength range and power of the illuminationsource, etc.). In some of the studies, the position of the reduction potential ofthe “sacrificialox/sacrificial” is ignored. It should not be more negative than thewater reduction potential. Otherwise, ΔG of the reaction would be negative andthe reaction becomes exergonic indicating the production of hydrogen fromthese sources to be thermodynamically feasible in the absence of light. However,photocatalytic water splitting is an endergonic reaction (ΔG is positive). Thesepoints should be considered before choosing Na2SO3, triethanolamine (TEOA),hydrazine, and so on as sacrificial agents.

13.2.4 Effect of Size of Nanostructures

Photocatalysts with nanodimensions possess large surface-to-volume ratios,resulting in enhanced photocatalytic activities. Low dimensions also favor quick


charge (electron and hole) transfer to the surface. Therefore, nanomaterials havea greater tendency to promote surface reactions compared to corresponding bulkmaterials [8]. On the other side, bending of bands at the solid–liquid interfaceis significant, while the semiconductor particle is present in the electrolyte.Charge equilibrium across the semiconductor and electrolyte interface causes abuilt-in potential and separates the charges, therefore, reducing recombination(Figure 13.3). This built-in potential favors the separation of charge carriersand allows only one type of charge carriers to reach the surface. In most ofthe particulate photocatalysts, both the charge carriers need to reach surface.The space charge region exists when the size of the particle is above a criticalvalue (d> dsc), here, d is the diameter and dsc is width of the space charge region(Figure 13.3). If the size of the particles is below the width of the space chargeregion (d< dsc), there would be no bending of bands to separate the charges [9,10]. Typical widths of the space charge region are in micrometers. It is essentialto create an alternative way to separate the charges in the case of semiconductornanomaterials. Heterostructures of semiconductors and p–n junctions are wellexplored for their use in separation of charge carriers. Relative band offsetsof the semiconductors determine the properties of heterostructures. Type IIheterostructures (such as ZnO/CdS) drives the migration of electrons to onecomponent and holes to other component. Therefore, charge carriers life timesare prolonged and are used in the photocatalysis. It also spatially separates thereactive intermediates.

A

A–

D+

D

d

d

hv

dsc

A

E

D

d > dsc

hv

A

E

D

d << dsc

hv

A

A–

D+

D

hv

(a) (b)

Figure 13.3 Transfer of charge carriers on (a) large and (b) small semiconductor particles inthe presence of an electron acceptor (A) and a donor (D). (Reproduced with permission fromRef. [10]. Copyright 2009, Springer.)

13.3 Overall Water Splitting 371

13.3 Overall Water Splitting

Splitting of water to hydrogen and oxygen in the stoichiometric ratio of 2 : 1 isreferred to as overall water splitting. Water splitting photocatalysts are classifiedas (i) one-step and (ii) two-step photoexcitation photocatalysts (Figure 13.4). Inthe one-step process, one light-harvesting unit is employed to generate one activee− and h+, whereas in the two-step or the Z-scheme process, two light-harvestingunits simultaneously generate two e− and two h+, wherein only one e− and oneh+ are active and rest of them recombine.

13.3.1 One-Step Photocatalytic Process

In the one-step photocatalytic process, the CB should be more negative thanwater reduction potential and VB should be more positive than the water

Electron donor

(a)

(b)

Redoxrelay

A

A∗

B∗

B

2H2O

Pote

ntial energ

yP

ote

ntial energ

y

Electron acceptor

hv

hv

hv

4H+ + O2

Electron donor

Electron acceptor

A

A∗

2H2O 4H+ + O2

H2 2H+/H2O

H2 2H+/H2O

Figure 13.4 Schematic illustration of mechanism of water splitting in (a) two-step and(b) one-step excitation processes. (Reproduced with permission from Ref. [5]. Copyright 2016,John Wiley & Sons.)


oxidation potential. Often, it becomes necessary to use more than onecomponent in the one-step photoexcitation process. Therefore, CBs of all thesemiconductors should be more negative than the water reduction potential andVBs should be more positive than the water oxidation potential. It should benoted that the mechanism of water splitting by the Z-scheme photocatalysts isdifferent from that of the two-component-based heterostructures.

Wide bandgap semiconductors, such as NaTaO3 (Eg = 4 eV), SrTiO3(Eg = 3.75 eV), TiO2 (Eg = 3.2 eV), 𝛽-Ge3N4 (Eg = 3.8 eV), GaN (Eg = 3.4 eV),and ZnO (Eg = 3.2 eV), possess suitable band positions for water splitting. Thesematerials exhibit evolution of hydrogen and oxygen from water in the stoi-chiometric ratio of 2 : 1, but only under ultraviolet (UV, 4% of total sun energy)irradiation due to the large bandgaps. To use the larger fraction of solar irradia-tion via utilizing visible-light (∼43%), narrowing the bandgap of these materialsis important. Visible-light absorption in such materials has been achieved bythe modification of the electronic structures by doping with foreign elements orby the formation of solid solutions. Moreover, dye sensitization, small bandgapsemiconductor quantum dot sensitization, and so on have also been employed.

Solid solutions are effective in altering the electronic structures for photo-catalytic water splitting. Here, we restrict our discussion to the most efficientphotocatalysts obtained by the formation of solid solutions of ZnO and GaN(Ga1−xZnxN1−xOx) [11, 12]. Solid solutions of ZnO and GaN have reducedbandgaps compared to individual ZnO and GaN. Insight into the origin of thereduction in bandgap is elucidated by comparing the electronic structure of solidsolutions with those of the parent ZnO and GaN. In the electronic structure ofZnO, CB consists of Zn 4s, VB higher energy bands consist of O 2p states, andlower energy bands consist of Zn 3d states. On the other hand, in the electronicstructure of GaN, CB consists of 4s and 4p states of Ga and VB consists of 2pstates of N. Density functional theory (DFT) calculations on Ga1−xZnxN1−xOxshown that CB consists of 4s and 4p states of Ga and VB consists of 2p states ofN and 3d states of Zn. Repulsion between N 2p states and Zn 3d states causes thedecrease in the bandgap. Figure 13.5 shows the UV-visible absorption spectrum

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0300 350 400


Quantu

m e

ffic

iency (

%)

Absorb

ance (

a.u

.)

450 500 0

2.0

1.5

1.0

0.5

05 10 15 20

Reaction time (h)

Am

ount of evo

lved g

ases (

mm

ol)

25 30 35

Figure 13.5 (a) Comparison of quantum efficiencies of water splitting with the absorptionspectrum of ZnO:GaN. (b) Rate of evolution of hydrogen and oxygen with ZnO:GaN.(Reproduced with permission from Ref. [12]. Copyright 2006, The American Chemical Society.)


of Ga1−xZnxN1−xOx. The absorption edge extends up to 500 nm (Eg = 2.5 eV)covering a significant fraction of the visible region. It should also be noted thatthe bandgap is a function of the composition. RuO2 deposited Zn1−xGaxO1−xNxproduces H2 and O2 of 320 and 160 μmol h−1, respectively, from water (pH 3,H2SO4) under visible-light irradiation with an AQY of 0.14% in the range of300–480 nm (note that the parent compounds, ZnO and GaN are not suitablefor water splitting under visible-light irradiation). The highest AQY of 5.1% at410 nm on Ga1−xZnxN1−xOx has been achieved upon loading of Rh2−yCryO3 onGa1−xZnxN1−xOx. Most of the water reduction and oxidation cocatalysts alsocatalyze the recombination of H2 and O2 or reactive intermediates to yield water.It is necessary to suppress the recombination of these reaction intermediates.Hence, cocatalysts should drive the forward reaction and suppress the backwardreactions. Maeda et al. [13] have introduced core–shell structures of Rh-Cr2O3on these solid solutions. The Rh core catalyzes water reduction and the Cr2O3shell (∼2 nm) prevents oxygen molecules from reaching Rh, suppressing therecombination of reactive intermediates and yielding H2 and O2 of 180 and90 μmol h−1, respectively, from pure water.

Codeposition of H2- and O2 evolution catalysts together on single particleeffectively drives both water reduction and oxidation reactions. The codepositionof Pt and RuO2 on TiO2 exhibits overall water splitting under UV irradiation[6]. Here, Pt and RuO2 act as H2- and O2 evolution cocatalysts, respectively.Similarly, when both these cocatalysts are deposited on CdS, there is resistanceto photocorrosion resulting in the splitting of water [6]. Deposition of RuO2effectively removes holes from CdS causing reduction in photocorrosion ofCdS. Coloading of Pt and CoOx on g-C3N4 significantly affects the photocat-alytic activity of g-C3N4, with hydrogen and oxygen evolution rates of 1.2 and0.6 μmol h−1, respectively, under visible-light irradiation (λ > 420 nm) with anAQY of 0.3% at 405 nm [14]. Here, Pt acts as the H2 evolution catalyst and CoOxacts as the O2 evolution cocatalyst.

Many photocatalysts have been investigated to generate hydrogen and oxygenfrom pure water without any additives. Reducing the size of the semiconductorparticles (near to the exciton Bohr radius) leads to the quantum-size effects,which affect the positions of the VB and the CB. Considering the special casewherein the bulk semiconductor does not possess suitable band positions, butthe smaller size possesses suitable band positions. This is due to the elevationof the CB when the size is smaller. It has been experimentally shown thatCo3O4 quantum dots yield 79 and 40 μmol h−1 g−1 hydrogen and oxygen uponirradiation, but the bulk Co3O4 is not suitable [15]. Similarly, CoO nanoparticlesobtained by laser ablation and ball milling of microparticles of CoO exhibitsplitting of water with a solar-to-hydrogen conversion efficiency of ∼5% underthe irradiation of AM1.5G simulated solar light (100 mW cm−2) [16]. The bandpositions of bulk and nanoparticles relative to the water reduction and oxida-tion potentials are shown in Figure 13.6. The elevation in the band positions(obtained from the electrochemical measurements) in the case of nanoparticlesof CoO causes these materials to be thermodynamically suitable for watersplitting.


E vs NHE

Nanoparticles

(a) (b)

pH = 7

0

CB

CBH+/H2

12

10

8

6

4

2

0

0 200 400

H2

O2

Incident power (mW)

Gas g

enera

tion (

ml)

600 800 1000

O2/H2O

Micropowder

VB

VB

EF

EF2

3

1

–

+

Figure 13.6 (a) Band positions of CoO nanocrystals (right) and micropowders (left) relative tothe water reduction and oxidation potentials. (b) Production of hydrogen and oxygen fromCoO nanoparticles as a function of incident laser power (𝜆= 532 nm). (Reproduced withpermission from Ref. [16]. Copyright 2014, Macmillan Publishers.)

13.3.2 Two-Step (Z-Scheme) Photocatalytic Process

Z-scheme photocatalysis or two-step photoexcitation process is a replica ofnatural photosynthesis wherein two photoexcitation units are coupled withan electron-transfer channel. As in natural photosynthesis, one unit of thephotocatalyst (PSI) involves in the reduction of water and another one (PSII)involves in the oxidation of water (Figure 13.4). Upon light irradiation, bothPSI and PSII absorb light, get excited, and generate charges simultaneously.Typical photocatalysts employed as PSI are SrTiO3:Rh, SrTiO3 (Cr, Ta) and soon and PSII are WO3, BiVO4, Fe3O4, and so on. These are also known as H2- andO2-evolution photocatalysts, respectively. There are three possible ways of thedriving charge transfer from one component to the other:

1) Use of redox shuttles (e.g., IO3−/I−, Fe3+/Fe2+)

2) Metals and other conducting channels (e.g., Ag and graphene)3) Direct transfer.

In the redox-shuttle-mediated Z-scheme photocatalysis, both photocatalystcomponents (PSI and PSII) are dispersed in water in an appropriate ratio alongwith a redox mediator. The choice of the redox mediator depends on the relativeenergy levels. The reduction potential of IO3

−/I− couple or Fe3+/Fe2+shouldstraddle between VB of PSI and CB of PSII. The redox potentials of IO3

−/I− andFe3+/Fe2+ are 0.67 and 0.77 V vs NHE, respectively. The excited electrons in PSIreduce H+ to H2 and the holes in PSII oxidize water to oxygen. Holes in PSIoxidize I− to IO3

− and I− ions are regenerated from IO3− by receiving the excited

electrons from PSII. Similarly in the case where Fe3+/Fe2+ is used as the redoxshuttle, Fe2+ gets oxidized by the holes in PSI and Fe3+ gets reduced back to Fe2+

by the electrons in PSII. Upon irradiation, two photons generate two electronsand two holes, among which only one electron and one hole participate in watersplitting.


–0.41

PS2[O2] PS1[H2]

Pt-SrTiO3(Cr-Ta doped)

0.09

0.67

0.82

O2

H2O h+

h+

e–

e–

Pt

Pt

IO3–

IO3–

H2O

H2H+/H2

IO3–/I–

O2/H2O

I–

I–

Pt-WO3

visible light

visible light

hν

hν

Po

ten

tia

l (V

) vs N

HE

(p

H =

7)

Figure 13.7 Schematic illustration of redox-shuttle meditated two-step (Z-scheme)photoexcitation process. (Reproduced with permissions from Ref. [17]. Copyright 2001, TheRoyal Society of Chemistry.)

We discuss here an example in which IO3−/I− has been utilized as a redox

shuttle. Relative positions of the energy levels are shown in Figure 13.7. Pt-WO3and Pt-SrTiO3 (Cr-Ta-doped) are dispersed in an aqueous solution of NaI(2 mM). Here, Pt-WO3 is the oxygen evolution catalyst and Pt-SrTiO3 (Cr, Ta) isthe hydrogen evolution catalysts. Pt-WO3:Pt-SrTiO3(Cr, Ta) yields hydrogen andoxygen evolution rates of 80 and 40 μmol h−1, respectively, under visible-lightirradiation (Figure 13.8). The presence of only H2 evolution catalyst yields asmaller quantity of hydrogen, without any O2 evolution. The H2 evolution reac-tion is terminated within few hours due to the lack of regeneration of I− ions [17].

100

Am

ou

nt

of

ga

s e

volu

tio

n (μm

ol)

Evacuated

Light Dark

H2

H2

O2

(b)(b)

(b)

(a)O2

Light LightDark

80

60

40

20

0

0 50 100

Time (h)

150 200 250

Figure 13.8 Generation of hydrogen and oxygen as a function of time in (a) the presence ofonly the H2-evolution catalyst (Pt-SrTiO3: Cr, Ta) and (b) the presence of both H2- andO2-evolution ((Pt-SrTiO3: Cr, Ta) and Pt-WO3) catalysts. (Reproduced with permission fromRef. [17]. Copyright 2001, The Royal Society of Chemistry.)


Effective charge separation is achieved by the formation of heterostructures ofMgTa2O6−xNy/TaON. Pt loaded MgTa2O6−xNy/TaON along with O2 evolutioncatalysts (PtOx-WO3) in aqueous solution of 1 mM NaI, resulting in the hydrogenand oxygen evolution of 108.3 and 55.8 μmol h−1, respectively, with an estimatedAQY of 6.8% at 420 nm. MgTa2O6−xNy and TaON alone or physical mixture ofMgTa2O6−xNy yields very low activities [18].

Metals are employed as charge transfer channels between the two light-harvesting units. Typically, Au, Ag, Pt, Cu, and so on are used for this purpose.It is assumed that metal deposition at the interface enhances the recombinationof electron in CB of PSII with holes in VB of PSI. Metals with Fermi level thatexists between the Fermi levels of both water oxidation and water reductioncocatalysts are suitable for this purpose. Conducting graphene oxide also actsas an efficient solid state electron-transfer mediator. In (Ru/SrTiO3:Rh)-PRGO(BiVO4), RGO acts as the electron-transfer mediator, yielding H2- and O2evolution rates of 11 and 5.5 μmol h−1, respectively [19]. The yields correspondto TON of 3.2 in 24 h indicating the reaction to be catalytic. Superior activitiesare obtained with the graphene oxide photoreduced on BiVO4, probably due tothe good contact between BiVO4 and graphene oxide. In the absence of RGO,Ru/SrTiO3:Rh-BiVO4 exhibits H2 and O2 evolution rates of 3.7 and 1.9 μmol h−1,respectively.

13.4 Oxidation of Water

Oxidation of water to oxygen is a four-electron-transfer reaction, associated withsluggish kinetics (Eq. (13.3)). In this section, we discuss the notable catalysts thathave been developed for the oxidation of water. Studies on water oxidation havebeen generally carried out in the presence of AgNO3 or Na2S2O8 solutions. Thesesacrificial electron acceptors consume the excited electrons. The concentrationof the sacrificial agent significantly affects the activity and the activities reportedso far are measured at different sacrificial conditions. For the convenience of thereaders, in addition to activities, we also mention the effect of concentration ofthe sacrificial agent.

In natural photosynthesis, water oxidation in photosystem II (PSII) is catalyzedby an oxygen-bridged Mn4O5Ca cluster, called the water oxidizing complex(WOC) [1]. Upon illumination of light on P680 (Pigment which absorbs light of680 nm), the electrons excited are transferred to pheophytin and successively toPSI through an electron-relay chain, resulting in the formation of P680∙+. P680∙+

extracts electrons from WOC through a single-electron gate tyrosine. Afterfour such steps, WOC releases water and four protons restoring to its restingstate [1]. Mn4O5Ca cluster is in a chair form in the resting state and has thecubane structure in the intermediate state. The TOF and life time of PSII-WOCare 5× 102 s−1and 30 min, respectively [6]. It is to be noted that nature uses theearth-abundant Mn in variable oxidation states.

In recent years, considerable research has been carried out to oxidize waterto oxygen in the laboratory. Unlike water reduction, role of cocatalysts is moreimportant in the oxidation of water. In the water oxidation process, it was

13.4 Oxidation of Water 377

assumed that the metal–oxocubane structure core was critical for oxygenevolution reaction (OER). Researchers, therefore, systematically investigatedthe effect of the cubane structure on OER activity. Brimblecombe et al. [20]and McCool et al. [21] have employed compounds containing cubane Mn4O4and Co4O4, respectively, for OER. It is found that 𝜆-MnO2 obtained by thedelithiation of Li2MnO4 also exhibits superior activity compared to Li2MnO4[22]. However, it has been shown recently that the e1

g configuration of thetrivalent metal ions (Mn3+ and Co3+) is crucial for water oxidation (Figure 13.9)[23]. Mn2O3 or LaMnO3 with Mn3+ in t3

2ge1g configuration and LaCoO3 with

Co(III) the intermediate spin in t52g e1

g configuration show the superior oxygen

500

400

300

200

100

(a)

(b)

0

800(i)

(ii)

(iv)

(iii)

0 2 4 6 8 10 12 14 16

(i)

(ii)

(iii)

(iv)

45

30

15

(mm

ol m

ol

–1 m

in–

1)

O2 e

volv

ed

(mm

ol/

mo

le o

f tr

an

sitio

n m

eta

l)

600

400

200

0

0 5 10

Time (min)

15

0 5

50 10

Li2Co2O4

Li2Co2O4

Li1.1Co2O4

15

0

50

100(m

mo

l/m

ole

of

tra

nsitio

n m

eta

l. m

2)

Time (min)

O2 e

vo

lve

d (

mm

ol/m

ole

of

Co

)

Incre

asin

g L

i

10 15

Figure 13.9 (a) Amount of oxygen evolved per mole of Co by LixCo2O4 with different amountof Li. (○) Li2Co2O4 and (◽) Li1.1Co2O4. (b) Amount of oxygen evolved per mole of transitionmetal in (i) LaCoO3, (ii) Li2Co2O4, (iii) Mn2O3, and (iv) LaMnO3. (Reproduced with permissionfrom Ref. [23]. Copyright 2001, PNAS.)


evolution activity. The eg orbitals can form sigma bonds with the reactive anionicadsorbates; in addition, an electron in the eg orbital gives the necessary strengthof interaction between the catalyst and oxygen. LaCoO3 exhibits the highestTOF of 1.4× 10−3 s−1. TOFs of Mn3+ containing compounds are of the order of10−4 s−1, whereas TOF of solid solutions of Co2O3 in the rare earth sesquioxidesexhibits 8× 10−4 − 1.3× 10−3 s−1.

Semiconductors with suitable VB positions have been extensively studied aslight-harvesting units. From Figure 13.1, it is clear that the VB of the BiVO4,WO3, and Fe3O4 are suitable for the oxidation of water. Introducing cocatalystssuch as IrO2, RuO2, and CoOx remarkably improves the activity. For example,studies on BiVO4 demonstrate that the use of cocatalysts such as CoOx, Co-Piand IrOx results in oxygen evolution activity of 63, 78, and 52 μmol h−1 g−1,respectively, superior to the activity BiVO4 alone (33 μmol h−1 g−1). In situdeposition of CoOx on Sm2Ti2S2O5 yields 1630 μmol h−1 g−1 (163 μmol h−1) withan AQY 5% at 420 nm, which is nearly 6 and 16 times superior compared toCoOx-Sm2Ti2S2O5 (impregnation method) and bare Sm2Ti2S2O5, respectively[24]. Ultrathin nanosheets of 𝛼-Fe3O4 show an oxygen evolution activity of70 μmol h−1 g−1, whereas nanoparticles of 𝛼-Fe3O4 yield 767 μmol h−1 g−1 undervisible-light irradiation. The method of deposition of the cocatalysts and themorphology of the photocatalysts play an important role.

More interesting is the multicomponent structures, which facilitate spatialseparation of photogenerated charges as in nature photosynthesis. Li et al. [25]have selectively photodeposited water reduction and water oxidation catalystson (010) and (011) facets, respectively (Figure 13.10). Such site-selectivity isnot observed while deposited by the impregnation method which results in therandom deposition causing poor activity. BiVO4-Pt-MnOx with two cocatalysts,Pt (H2 evolution) and MnOx (O2 evolution), selectively deposited on (010) and(011) facets of BiVO4, yield an oxygen evolution activity of 650 μmol h−1 g−1. Theactivity of these structures is superior to those obtained with either Pt or MnOxbeing photodeposited on BiVO4. The activity is also superior to the activitiesof structures obtained by the either one of them or both cocatalysts beingdeposited by impregnation method. Similarly, BiVO4-graphene with compoundscontaining Co4O4 units gives rise to oxygen evolution activity of 12 100 μmol g−1

under visible-light irradiation. Here, graphene acts as a conducting channel andcompounds containing Co4O4 units catalyze the oxidation of water [26].

Heterostructures or composites of the two or more materials have creatednotable impact on the water oxidation. For example, combination of 𝛼-Fe3O4 andrGO as in 𝛼-Fe3O4-rGO enhances the activity two times compared to 𝛼-Fe3O4[27]. 𝛼-Fe3O4/Mn3O4/graphene yields 81.7 μmol h−1 g−1 under visible-lightirradiation (λ > 400 nm) [28]. It has been shown that Ag3PO4 oxidizes of water,with an activity of 1272 μmol h−1 g−1 with an AQY of 80% at 480 nm [29]. Depo-sition of semiconductors with suitable band positions (including wide bad gapmaterials) results in the effective charge separation. For example, deposition ofSrTiO3 on Ag3PO4 enhances the charge separation and increases the activity to1316 μmol h−1 g−1 [30]. SrTiO3 facilitates the hole transfer causing the reductionin recombination.

13.4 Oxidation of Water 379

MnOx

MnOx

Pt

M.m+

M.Ox

Pt

{010}

(a) (b)

(c)

{110}

Photo-oxidation

Pt(imp)/MnOx(imp)/BiVO4

Pt(imp)/MnOx(P.D.)/BiVO4

Pt(P.D.)/MnOx(imp)/BiVO4

MnOx(P.D.)/BiVO4

Pt(P.D.)/BiVO4

BiVO4

0 100 200 300

O2 evolution amount (μmol h–1 g–1)

400 500 600 700

Pt(P.D.)/MnOx(P.D.)/BiVO4

e– e–

h+ h+deposition

M

Mn+

Photo-reductiondeposition

Figure 13.10 (a) SEM images of Pt and MnOx deposited BiVO4. (b) Schematic illustration ofmechanism of water splitting on Pt and MnOx deposited BiVO4. (c) Comparison ofphotocatalytic activity of Pt and MnOx photodeposited BiVO4, with the activities of otherphotocatalysts. (Reproduced with permission from Ref. [25]. Copyright 2013, MacmillanPublishers.)

Where metal oxides have large bandgaps, the corresponding oxynitrides havesignificantly lower bandgaps [31]. Here, we examine the case of TaON. TaON is apartial nitridation product of Ta2O5 (bandgap ∼3.8 eV) with a bandgap of 2.5 eV.DFT calculations on TaON show that the bottom of the CB consists of Ta 5d statesas in Ta2O5, and top of the VB consists of O 2p and N 2p (predominately) states.The higher potential energy of the hybridization orbital results in a decrease inthe bandgap to 2.5 eV. TaON exhibits an oxygen evolution rate of 660 μmol h−1

with an AQY of 34% (420–500 nm) [32]. Ta3N5 with 2 wt% of CoOx exhibits anoxygen evolution rate of 450 μmol h−1 with an AQY of 5.2% (500–600 nm) [33].Heterostructures of Ta3N5-TaON yield oxygen evolution of 208 μmol h−1 with anAQY of 67% 420 nm [34]. Codoping of both N and F in TiO2 drastically reducesthe bandgap of the oxide. N- and F-doped TiO2 with exposed (001) facets exhibitsan oxygen evolution rate of 500 μmol h−1 g−1 under visible-light irradiation [35].Similarly, bandgaps of oxide materials can also be controlled by the formation


of oxy-sulfide solid solutions [3]. For example, partial substitution of oxygen inSm2Ti2O7 with S (as in Sm2Ti2O5S2) decreases the bandgap from 3.5 to 2.0 eV.Similarly, other oxy-sulfides (Ln2Ti2O5S2; Ln=Gd, Tb, Dy, Ho, and Er) also pos-sess lower bandgaps compared to corresponding oxides. The decrease in bandgapis due to the existence of S 3p level above the O 2p level. These materials havealso shown oxygen evolution activity in the visible-light irradiation due to narrowbandgap compared to corresponding oxides.

13.5 Reduction of Water

In photocatalytic hydrogen generation, electrons are utilized for the reduction ofprotons, while holes are utilized by the sacrificial electron donors for oxidation.Semiconductors which possess a more negative CB edge relative to the waterreduction potential are thermodynamically suitable. Since water oxidation hasbeen replaced by a sacrificial agent oxidation, the position of the VB relative tooxidation potential of the sacrificial agent determines the activity. We shall brieflypresent the performance of known photocatalysts such as graphene, C3N4, andMoS2 as well as oxides and sulfides.

13.5.1 C3N4 and Related Materials

Graphene has recently gained great attention due to its unique properties, includ-ing high specific surface area, high electrical conductivity, as well as high mobilityof charge carriers. It can act as a good channel for charge carriers. Graphene hasbeen employed as a photocatalyst in a few reports. However, intimate contactof graphene with the other materials can suppress the recombination of chargecarriers [19]. Li et al. [36] have employed graphene nanosheets decorated withCdS. The presence of a small amount of graphene (graphene-CdS) increases theactivity. Graphene-CdS-Pt (0.5 wt% Pt and 1.0 wt% graphene) yields a hydro-gen evolution rate of 1.12 mmol h−1 in the presence of 10 vol% lactic acid withan AQY of 22.5% at 420 nm. The high yield of hydrogen is attributed to tworeasons: (i) increase in the specific surface area of the composite; (ii) grapheneaccepts the electrons and decreases the recombination of electron and hole pairs(Figure 13.11).

In an effort to eliminate the use of noble metal catalysts, C3N4 and othercarbon-based materials have been employed as photocatalysts. Wang et al.[37] reported the photocatalytic activity of g-C3N4. g-C3N4 is chemically andthermally stable in air up to 600 ∘C and possesses a bandgap of 2.7 eV (yellowin color). DFT calculations on ideal infinite sheet of g-C3N4 show a bandgapvalue (HOMO–LUMO) of 2.1 eV, though it is slightly underestimated. VB ofthe g-C3N4 is derived from the N pz levels, whereas CB is derived from the Cpz levels. Therefore, N acts as the site of oxidation, whereas C acts as the site ofreduction. Absolute positions of VB and CB straddle the water reduction andwater oxidation potentials (Figure 13.12). Water reduction is more favorablecompared to water oxidation due to the larger gap between CBM and the waterreduction potential compared to the smaller gap between VBM and the wateroxidation potential. g-C3N4 modified with Pt (3 wt%) yields an hydrogenevolution rate of 10.7 μmol h−1 under visible-light irradiation with an AQY of0.1% in the range of 420–460 nm in the presence of TEOA.

13.5 Reduction of Water 381

1.40

1.20

1.00

0.80

0.60

0.40

0.20

(a) (b)

H2-p

roduction r

ate

(m

mol h

–1)

Samples

0.23

0.38

1.12

0.55

0.23

0.020

0GC0 GC0.5 GC1.0 GC2.5GC5.0 GC40 G

H2O

H2O

H2

H+

H2

H2

Pt

Pt

Visible light

C

CB

VB

CdS

Figure 13.11 (a) Schematic illustration of mechanism of hydrogen generation onCdS-graphene-Pt and (b) comparison of activities of photocatalysts with different amountgraphene loading. (Reproduced with permission from Ref. [36]. Copyright 2011, The AmericanChemical Society.)

Ab

so

rba

nce

(a

.u.)

300 400

(a) (b)

(c)

500

Wavelength (nm)

600 700 800

4

3

2

1

0

V–

VN

HE

(V)

–0.1ΓY X0

k (Å–1)

0.1

H+/H2

O2/H2O

0.2

2

1

0

–1

–2

Figure 13.12 (a) UV–visible absorption spectrum, (b) electronic structure, and (c) molecularstructure of g-C3N4. (Reproduced with permission from Ref. [37]. Copyright 2009, SpringerNature.)

Several researchers have modified the g-C3N4, by developing various methodsof combining g-C3N4 with other compounds. g-C3N4 in combination with NiS(g-C3N4/NiS) yields an hydrogen evolution of 48.2 μmol h−1 in the presence of15 vol% TEOA in water. The estimated quantum efficiency is 1.9% at 440 nm[38]. TEOA is used as an electron donor, but one should note that it undergoesphotodecomposition even in the absence of any catalyst yielding hydrogen.Photochemical activities of g-C3N4-based catalysts are lower than those of metaloxide and sulfide-based semiconductors.


Edge sites of the inorganic graphene analogues MoS2 and WS2 are catalyticallyactive for the reduction of water. Zong et al. have employed the colloidal MoS2for dye-assisted hydrogen generation and showed that they exhibit an hydrogenevolution activity of 77.7 μmol h−1 [39]. Here, Ru(bpy)3

2+ (bpy= 2,2′-bipyridine)is used as photosensitizer and ascorbic acid is used as electron donor. Compositeof MoS2 with RGO as in MoS2-RGO shows superior activity compared to eitherMoS2 or physically mixed MoS2 and RGO (Eosin Y is used as dye) yielding4.4 mmol h−1 g−1 hydrogen [40]. 2D MoS2 and MoSe2 have been employed forphotocatalytic hydrogen generation [41, 42]. The composite of graphene (exfo-liated; EG)-2H MoS2 shows a superior activity (0.54 μmol h−1 g−1) compared to2H MoS2 (0.05 μmol h−1 g−1). Composite of N doped EG-MoS2 (2H) exhibitedhydrogen evolution activity of 0.83 mmol h−1 g−1 (TOF 0.45 h−1). Since grapheneis conducting, it enhances the life time of the photoexcited dye (EY1*) and theprobability of forming stable EY3*. EY receives electrons from the electron donorand forms EY−. Graphene collects the electrons from EY− and transfers to MoS2,and the edge sites of MoS2 catalyze the reduction of water. Nitrogen doping ingraphene (NG) enhances its electron donation ability. It causes enhancement inH2 evolution activity when NG formed composites with MoS2. The observedtrend in activity is as follows: NG-MoS2 >G-MoS2 >MoS2. Composite ofNRGO-MoS2 (2H) shows even more activity yielding 10.8 mmol h−1 g−1 (TOF2.9 h−1). Unlike 2H-MoS2, 1T MoS2 is metallic and the metallic nature of1T-MoS2 originated from the incompletely filled 4dxy,xz,yz orbitals under theoctahedral Oh field (Figure 13.13) and can be used as potential candidate forreplacing both EG-MoS2 (2H). 1T-MoS2 exhibits a hydrogen evolution activity of26 mmol h−1 g−1(TOF 6.2 h−1). 2H-MoSe2 shows superior activity compared to2H-MoS2. More interestingly, metallic 1T-MoSe2 shows the excellent hydrogenevolution activity yielding 75 mmol h−1 g−1 under visible-light irradiation [42].

13.5.2 Semiconductors

Since many of the semiconductors possess large bandgaps and can only beactive under UV light irradiation. It is, therefore, essential to alter the bandgap

2502H-MoS21T-MoS2

EY3*

EY1*

EY

EY–.

TEOA

TEOA+

1T-MoS2

(a) (b)

H2O/H+

H2

dxz, yz

dx2−y2,xy

dx2−y2,z2

dx2−y2,z2

dxy, xz, yz

dxy, xz, yzdz2

ISC

Light

200

150

100

50

0

0 6 8 102 4

Time (h)

30 mm

ol g–1 h

–1

H2 e

volv

ed

(m

mo

l g

−1)

Figure 13.13 (a) The crystal-field-splitting induced electronic configuration of 2H-MoS2 and1T-MoS2 and proposed mechanism for photocatalytic activity of 1T-MoS2. (b) Time course ofphotocatalytic H2 evolved by freshly prepared 1T-MoS2. (Reproduced with permission fromRef. [41]. Copyright 2013, John Wiley & Sons.)


to achieve visible absorption. Generally changes in the electronic structure areobtained by the doping of foreign elements. For example, doping of Cr, Sb, Ni,Ta, and Rh in the wide bandgap SrTiO3 results in visible absorption. Superioractivity of 117 μmol h−1 has been achieved by doping rhodium [3]. Codopingof metal ions is effective in altering the optical properties. For example, Sb andCr codoped TiO2 and SrTiO3 have an intense visible absorption and exhibithydrogen evolution activity under visible-light irradiation. Here, Cr3+ and Sr5+

ions are cosubstituted in place of Ti4+, which results in reduced recombinationcenters [3]. Rh-doped SrTiO3 is extensively used as hydrogen evolution catalystsin Z-scheme photocatalytic water splitting [3].

Anion substitution predominately alters the VB and, therefore, it is preferredover cation substitution. Doping of N in ZnO and TiO2 also results in visible-lightabsorption. However, N doping alone in these compounds increases the oxy-gen vacancies. Cosubstitution of N and F in place of O in ZnO and TiO2 showsremarkable decrease in bandgap and renders them colored. N and F codopedTiO2 exhibited an H2 evolution of 60 μmol h−1 g−1 under visible-light irradiation(λ > 400 nm) [43].

The structure and properties of ZnO analogue such as Zn2NF have beenreported recently [44]. Zn2NF has a bandgap of 2.7 eV and exhibits a hydrogenevolution of 220 μmol h−1 g−1 in the presence of Na2S-Na2SO3 as sacrificialagents under visible light. N and F codoped ZnO (ZnO0.2N0.5F0.3/Pt) exhibitsa hydrogen evolution rate of 114 μmol h−1 g−1 in the presence of Na2S-Na2SO3as sacrificial agents. P and Cl doped CdS compositions also have reducedbandgaps and exhibit hydrogen evolution under visible light in the presence ofNa2S-Na2SO3 as sacrificial agents [45]. Complete substitution of S in CdS withP and Cl should result in Cd2PCl, but this composition has not been prepared.However, properties of its close composition, Cd4P2Cl3, have been recentlyexplored [46]. In contrast to CdS, Cd4P2Cl3/Pt exhibits a hydrogen evolution inthe absence of any sacrificial agent in basic medium.

Photocatalytic activities obtained with single-particulate photocatalysts aregenerally lower due to the absence of a charge-separating mechanism and closeproximity of the reaction sites.

13.5.3 Multicomponent Heterostructures

The best results have been obtained with the multicomponent photocatalysts,where two or more than two components are involved in obtaining the efficientactivity. Nearly 86% of the studies in the literature have used multicomponentphotocatalysts. The best photocatalytic activities obtained so far are with het-erostructures of the type CdSe/CdS/Pt [47, 48], CdS-Ni [49], ZnO/Pt/CdS [50],and so on. Bao et al. [51] have employed porous CdS obtained by ion-exchangeof Cd(OH)2 as a photocatalyst. Upon 10 wt% of Pt loading, it yielded hydrogenevolution activity of 4.1 mmol h−1 (27.3 mmol h−1 g−1) in the presence of 0.35 MNa2S and 0.25 M Na2SO3. Porous structures of CdS are responsible for the highactivity. The estimated AQY is 60.3% at 420 nm.

Alivisatos and coworkers have shown the role of spatial separation of electronand hole in a multicomponent heterostructure [47]. Here, the CdSe seed embed-ded CdS nanorods with Pt tips (CdSe/CdS/Pt) are employed as photocatalysts for


Samples of Pt-tipped seeded

and unseeded rods(a) (b)

60 nm 20 nm 27 nm 40 nm 70 nm 60 nm

QE = 20%D = 2.3 nm

D = 3.1 nm

CdSe CdS

Pt

2 H+ H2

CdSe

CdS

Pt

e–

h+

Re

lative

QE

fo

r H

2 P

rod

uctio

n

No seed

Figure 13.14 (a) Schematic illustration of structure and relative energy levels and(b) comparison of photocatalytic activity of different configuration of CdSe/CdS/Pt.(Reproduced with permission from Ref. [47]. Copyright 2010, The American Chemical Society.)

hydrogen evolution. Figure 13.14 shows the scheme, representing the mechanismof hydrogen evolution. Upon irradiation, photogenerated electrons migrate tothe Pt, whereas holes are confined to the CdSe quantum dots. Increasing thelength of the nanorod results in an increase in the distance between the holesand electrons and increases the photocatalytic activity. These heterostructuresexhibit hydrogen evolution of 40 mmol h−1 g−1 in the presence of methanolunder visible-light irradiation (AQY of 20% at 450 nm).

Simon et al. [49] have discovered two-step redox shuttle mechanism forhydrogen production on Ni decorated cysteine stabilized CdS nanorods. Pho-tocatalytic hydrogen evolution activity increases drastically at higher pH (14.7)(Figure 13.15). At pH 14.7, these structures exhibited an H2 evolution rate of withan AQY 53% at 447 nm under the illumination intensity 7 mW cm−2. However,the VBM is more negative than E∘(O∙−/−OH) at pH 14. At pH 14.7, the photogen-erated holes oxidize OH− to ∙OH (which is impossible at pH 7). The concentrationof OH− is abruptly increased at high pH. Therefore, at high pH, photogeneratedholes oxidize OH− to ∙OH. Then ∙OH leaves the surface of the photocatalystsand oxidizes the ethanol to acetaldehyde or further to acetic acid. Kalisman

–1

00 7 14

20

15

10

5

0

Efficie

ncy

3.0 6.0

(a) (b)

11.0 12.0

pH

14.0 14.7

pH

E°

vers

us N

HE

(V

)

CdS

H2 fo

rmation r

ate

(m

mol g

–1 h

–1)

1

2

•OH/–OH

EVB(CdS)

ECB(CdS)

O2/H2

O

H+/H2

EtOH/CH3CHO

Figure 13.15 (a) Function of water redox potentials, hydroxyl anion and ethanol oxidationpotentials with pH. (b) Variation efficiency with the pH. (Reproduced with permission from Ref.[49]. Copyright 2014, Springer Nature.)


et al. [48] have recently employed the hydroxyl anion/radical redox couple forCdSe/CdS/Pt heterostructures. Activity of these heterostructures increases withincrease in pH and reached AQY of 100%. However, placing Pt particles at boththe tips of CdS nanorods results in a drastic drop of AQY to 58.5%.

Solution-processed ZnO/Pt/CdS heterostructures for hydrogen evolutionhave been investigated in detail [50]. Upon light illumination (visible), CdSabsorbs the light and generates electron–holes pairs. Excited electrons areinjected to CB of ZnO and eventually reach Pt on ZnO and utilized in thereduction of water. Photogenerated holes in the VB of CdS are consumed bysacrificial (Figure 13.16). The presence of Pt on ZnO is more efficient than thepresence on CdS due to favorable band positions for vectorial electron transfer.Partial substitution of Zn in CdS as in ZnO/Pt/Cd1−xZnxS further increasesthe hydrogen evolution. Despite the increase in bandgap, hydrogen evolutionincreased, probably due to the efficient transfer of electrons arising from theincrease in the energy gap between CBs of CdS and ZnO. These heterostruc-tures are stable and exhibit consistent photocatalytic activity (Figure 13.17).The AQY values follow the absorption edge of the heterostructures, further

3.0

2.0

1.0

0.0

–1.0

E vs NHE

–8.0

CH2OH CHO–7.0

–6.0

–5.0

–4.0

–3.0

0.0Vacuum level (eV)

ZnOCdS

Pt

H2

H+

H+/H2

O2/H2O

e– e–

h+ h+

e–

e– e– e–

Figure 13.16 Schematic representation of process of hydrogen generation on ZnO/Pt/CdSheterostructures. (Reproduced with permission from Ref. [50]. Copyright 2013, The RoyalSociety of Chemistry.)

40

30

20

10

0

0 2 4

Time (h)(a) (b) Wavelength (nm)

30

25

20

15

10

5

AQ

Y (

%)

λ > 395 nm

H2 e

volv

ed

(m

mo

l g

–1)

Ab

so

rba

nce

(a

.u.)

6 8 400 500 600 700

Figure 13.17 (a) Visible light (𝜆> 395 nm) induced H2 evolution with ZnO/Pt/Cd0.8Zn0.2S as afunction of time and (b) comparison of the AQYs obtained upon irradiation of selectedwavelengths of light, with the absorption spectrum of ZnO/Pt/Cd0.8Zn0.2S. (Reproduced withpermission from Ref. [50]. Copyright 2013, The Royal Society of Chemistry.)


confirming photo-induced activity (Figure 13.17). N and F cosubstitutedZnO have been examined in these heterostructures in place of ZnO [52].In these heterostructures, both N and F cosubstituted ZnO and CdS arecapable of visible-light harvesting. The increased amount of hydrogen evo-lution is probably due to the synergistic effect of both these components.Pt in these heterostructures is replaced by NiO, to obtain comparable inactivity [53].

13.6 Coupled Reactions

Most of the photocatalysts are active only in the presence of electron donors(sacrificial agents). Coupling of other oxidation half-reactions with the reduc-tion of water appears to be useful strategy. Thus, amination and hydroxylation ofbenzene have been coupled with the simultaneous production of hydrogen fromNH3 and H2O, respectively. Ni modified CdS is also employed for the hydrogenproduction and simultaneous oxidation of alcohols to carbonyl compounds. Oxi-dation of alcohols to carboxylic acids or to aldehydes is an important industrialreaction. A bipyridene-based ruthenium complex has been employed for the pho-tocatalytic production of hydrogen along with oxidation of aliphatic as well asaromatic alcohols to the corresponding acids [54].

Oxidation of benzyl alcohol is coupled with hydrogen generation undervisible-light irradiation (Figure 13.18) [50]. The amount of hydrogen producedmarkedly increases in these coupled reactions. Photogenerated electronsreduce water to hydrogen, whereas the holes oxidize benzyl alcohol to ben-zaldehyde. The photocatalytic activities are superior to those obtained withconventional sacrificial agents such as Na2S-Na2SO3. TiO2/Pt/Cd0.8Zn0.2Sheterostructures also show simultaneous production of hydrogen with theoxidation of benzyl alcohol [55]. It should be noted that in addition tohydrogen, these oxidation products such as benzaldehyde are of commercialimportance.

40 ZnO/Pt/Cd0.8Zn0.2S

Na2 S,Na2 SO3 PhCH2 OH

λ > 395 nm

ZnO/Pt/CdS

Activity (

mm

ol h

–1 g

–1)

35

30

25

20

15

10

5

0

Figure 13.18 Comparison ofphotocatalytic activities of ZnO/Pt/CdSand ZnO/Pt/Cd0.8Zn0.2S obtained witha coupled organic reaction (PhCH2OHoxidation) and inorganic sacrificialreactions. (Reproduced withpermission from Ref. [50]. Copyright2013, The Royal Society of Chemistry.)

References 387

13.7 Summary and Outlook

The previous sections should clearly indicate that photochemical reduction andoxidation of water are readily accomplished. The problem that remains is findingthe best catalyst that can generate hydrogen on a commercial scale. Consider-ing the excellent results obtained with semiconductor heterostructures and 2Dsheets of MoS2, there is every reason to believe that the goal will be reached.

There are many catalyst materials that can be explored for their photocatalyticactivity. N and F codoped oxides and P and Cl codoped sulfides, nitride-fluorides,phosphide–chlorides are potential materials for water splitting, that are to beexamined. Cd4P2Cl3 is worth mentioning here. Cd4P2Cl3 is close to the hypo-thetical compound that would be obtained upon complete replacement of S by Pand Cl. It is a direct bandgap semiconductor with a bandgap value of 2.3 eV. Moreimportantly, Cd4P2Cl3 is resistant to photocorrosion unlike CdS. Cd4P2Cl3 givesevolution of hydrogen even in the absence of any sacrificial agent. Compositesof C3N4 with MoS2 and other materials are also promising due to their uniqueelectronic and optical properties, capability of charge separation along with thecatalytic edge sites of MoS2.

Acknowledgments

SRL is thankful to the Council for Scientific and Industrial Research (CSIR),India, for the senior research fellowship (SRF) and Ras Al Khaimah Centreof Advanced Materials (RAK-CAM) and Sheikh Saqr Laboratory for the SSLstudent fellowship.

References

1 Cox, N., Pantazis, D.A., Neese, F., and Lubitz, W. (2015) Artificial photosyn-thesis: understanding water splitting in nature. Interface Focus, 5, 20150009.

2 Rao, C.N.R., Lingampalli, S.R., Dey, S., and Roy, A. (2016) Solar photochem-ical and thermochemical splitting of water. Philos. Trans. R. Soc. London, Ser.A, 374, 20150088.


4 Li, X., Yu, J., Low, J., Fang, Y., Xiao, J., and Chen, X. (2015) Engineering het-erogeneous semiconductors for solar water splitting. J. Mater. Chem. A, 3,2485–2534.

5 Rao, C.N.R. and Lingampalli, S.R. (2015) Generation of hydrogen byvisible-light induced water-splitting with the use of semiconductors anddyes. Small, 12, 16–23.

6 Grätzel, M. (1981) Artificial photosynthesis: water cleavage into hydrogen andoxygen by visible light. Acc. Chem. Res., 14, 376–384.


7 Zhang, X., Peng, T., and Song, S. (2016) Recent advances in dye-sensitizedsemiconductor systems for photocatalytic hydrogen production. J. Mater.Chem. A, 4, 2365–2402.

8 Andrew Frame, F., Carroll, E.C., Larsen, D.S., Sarahan, M., Browning, N.D.,and Osterloh, F.E. (2008) First demonstration of CdSe as a photocatalyst forhydrogen evolution from water under UV and visible light. Chem. Commun.,2206–2208.

9 Memming, R.d. (1994) Electron Transfer I, Springer Berlin Heidelberg, Berlin,Heidelberg, pp. 105–181.

10 Zhang, H., Chen, G., and Bahnemann, D.W. (2009) Photoelectrocatalyticmaterials for environmental applications. J. Mater. Chem., 19, 5089–5121.

11 Maeda, K., Takata, T., Hara, M., Saito, N., Inoue, Y., Kobayashi, H.,and Domen, K. (2005) GaN:ZnO solid solution as a photocatalyst forvisible-light-driven overall water splitting. J. Am. Chem. Soc., 127, 8286–8287.

12 Maeda, K., Teramura, K., Lu, D., Takata, T., Saito, N., Inoue, Y., and Domen,K. (2006) Photocatalyst releasing hydrogen from water. Nature, 440,295–295.

13 Maeda, K., Teramura, K., Lu, D., Saito, N., Inoue, Y., and Domen, K. (2006)Noble-metal/Cr2O3 core/shell nanoparticles as a cocatalyst for photocatalyticoverall water splitting. Angew. Chem. Int. Ed., 45, 7806–7809.

14 Zhang, G., Lan, Z.-A., Lin, L., Lin, S., and Wang, X. (2016) Overall watersplitting by Pt/g-C3N4 photocatalysts without using sacrificial agents. Chem.Sci., 7, 3062–3066.

15 Zhang, N., Shi, J., Mao, S.S., and Guo, L. (2014) Co3O4 quantum dots: reversemicelle synthesis and visible-light-driven photocatalytic overall water splitting.Chem. Commun., 50, 2002–2004.

16 Liao, L., Zhang, Q., Su, Z., Zhao, Z., Wang, Y., Li, Y., Lu, X., Wei, D., Feng,G., Yu, Q., Cai, X., Zhao, J., Ren, Z., Fang, H., Robles-Hernandez, F., Baldelli,S., and Bao, J. (2014) Efficient solar water-splitting using a nanocrystallineCoO photocatalyst. Nat. Nanotechnol., 9, 69–73.

17 Sayama, K., Mukasa, K., Abe, R., Abe, Y., and Arakawa, H. (2001) Stoi-chiometric water splitting into H2 and O2 using a mixture of two differentphotocatalysts and an IO3

-/I- shuttle redox mediator under visible lightirradiation. Chem. Commun., 2416–2417.

18 Chen, S., Qi, Y., Hisatomi, T., Ding, Q., Asai, T., Li, Z., Ma, S.S.K., Zhang, F.,Domen, K., and Li, C. (2015) Efficient visible-light-driven Z-scheme overallwater splitting using a MgTa2O6−xNy/TaON heterostructure photocatalyst forH2 evolution. Angew. Chem. Int. Ed., 54, 8498–8501.


20 Brimblecombe, R., Koo, A., Dismukes, G.C., Swiegers, G.F., and Spiccia, L.(2010) Solar driven water oxidation by a bioinspired manganese molecularcatalyst. J. Am. Chem. Soc., 132, 2892–2894.

21 McCool, N.S., Robinson, D.M., Sheats, J.E., and Dismukes, G.C. (2011) ACo4O4 “Cubane” water oxidation catalyst inspired by photosynthesis. J. Am.Chem. Soc., 133, 11446–11449.

References 389

22 Robinson, D.M., Go, Y.B., Greenblatt, M., and Dismukes, G.C. (2010) Wateroxidation by 𝜆-MnO2: catalysis by the cubical Mn4O4 subcluster obtained bydelithiation of spinel LiMn2O4. J. Am. Chem. Soc., 132, 11467–11469.

23 Maitra, U., Naidu, B.S., Govindaraj, A., and Rao, C.N.R. (2013) Impor-tance of trivalency and the eg

1 configuration in the photocatalytic oxida-tion of water by Mn and Co oxides. Proc. Natl. Acad. Sci. U. S. A., 110,11704–11707.

24 Li, R., Chen, Z., Zhao, W., Zhang, F., Maeda, K., Huang, B., Shen, S., Domen,K., and Li, C. (2013) Sulfurization-assisted cobalt deposition on Sm2Ti2S2O5photocatalyst for water oxidation under visible light irradiation. J. Phys. Chem.C, 117, 376–382.

25 Li, R., Zhang, F., Wang, D., Yang, J., Li, M., Zhu, J., Zhou, X., Han, H., andLi, C. (2013) Spatial separation of photogenerated electrons and holes among{010} and {110} crystal facets of BiVO4. Nat. Commun., 4, 1432.

26 Wang, Y., Li, F., Li, H., Bai, L., and Sun, L. (2016) Photocatalytic water oxida-tion via combination of BiVO4-RGO and molecular cobalt catalysts. Chem.Commun., 52, 3050–3053.

27 Meng, F., Li, J., Cushing, S.K., Bright, J., Zhi, M., Rowley, J.D., Hong, Z.,Manivannan, A., Bristow, A.D., and Wu, N. (2013) Photocatalytic wateroxidation by Hematite/reduced graphene oxide composites. ACS Catal., 3,746–751.

28 Yin, S., Wang, X., Mou, Z., Wu, Y., Huang, H., Zhu, M., Du, Y., and Yang, P.(2014) Synergistic contributions by decreasing overpotential and enhancingcharge-transfer in 𝛼-Fe2O3/Mn3O4/graphene catalysts with heterostruc-tures for photocatalytic water oxidation. Phys. Chem. Chem. Phys., 16,11289–11296.

29 Yi, Z., Ye, J., Kikugawa, N., Kako, T., Ouyang, S., Stuart-Williams, H., Yang,H., Cao, J., Luo, W., Li, Z., Liu, Y., and Withers, R.L. (2010) An orthophos-phate semiconductor with photooxidation properties under visible-lightirradiation. Nat. Mater., 9, 559–564.

30 Guan, X. and Guo, L. (2014) Cocatalytic effect of SrTiO3 on Ag3PO4 towardenhanced photocatalytic water oxidation. ACS Catal., 4, 3020–3026.

31 Ahmed, M. and Xinxin, G. (2016) A review of metal oxynitrides for photo-catalysis. Inorg. Chem. Front., 3, 578–590.

32 Hitoki, G., Takata, T., Kondo, J.N., Hara, M., Kobayashi, H., and Domen, K.(2002) An oxynitride, TaON, as an efficient water oxidation photocatalystunder visible light irradiation (λ ≤ 500 nm). Chem. Commun., 1698–1699.

33 Ma, S.S.K., Hisatomi, T., Maeda, K., Moriya, Y., and Domen, K. (2012)Enhanced water oxidation on Ta3N5 photocatalysts by modification withalkaline metal salts. J. Am. Chem. Soc., 134, 19993–19996.

34 Wang, Z., Hou, J., Jiao, S., Huang, K., and Zhu, H. (2012) In situ chemicalreduction of the Ta3N5 quantum dots coupled TaON hollow spheres hetero-junction photocatalyst for water oxidation. J. Mater. Chem., 22, 21972–21978.

35 Zong, X., Xing, Z., Yu, H., Chen, Z., Tang, F., Zou, J., Lu, G.Q., and Wang, L.(2011) Photocatalytic water oxidation on F, N co-doped TiO2 with dominantexposed {001} facets under visible light. Chem. Commun., 47, 11742–11744.


36 Li, Q., Guo, B., Yu, J., Ran, J., Zhang, B., Yan, H., and Gong, J.R. (2011)Highly efficient visible-light-driven photocatalytic hydrogen productionof CdS-cluster-decorated graphene nanosheets. J. Am. Chem. Soc., 133,10878–10884.

37 Wang, X., Maeda, K., Thomas, A., Takanabe, K., Xin, G., Carlsson, J.M.,Domen, K., and Antonietti, M. (2009) A metal-free polymeric photocatalystfor hydrogen production from water under visible light. Nat. Mater., 8, 76–80.

38 Hong, J., Wang, Y., Wang, Y., Zhang, W., and Xu, R. (2013) Noble-metal-freeNiS/C3N4 for efficient photocatalytic hydrogen evolution from water. Chem-SusChem, 6, 2263–2268.

39 Zong, X., Na, Y., Wen, F., Ma, G., Yang, J., Wang, D., Ma, Y., Wang, M., Sun,L., and Li, C. (2009) Visible light driven H2 production in molecular sys-tems employing colloidal MoS2nanoparticles as catalyst. Chem. Commun.,4536–4538.

40 Min, S. and Lu, G. (2012) Sites for high efficient photocatalytic hydrogen evo-lution on a limited-layered MoS2 cocatalyst confined on graphene sheets therole of graphene. J. Phys. Chem. C, 116, 25415–25424.

41 Maitra, U., Gupta, U., De, M., Datta, R., Govindaraj, A., and Rao, C.N.R.(2013) Highly effective visible-light-induced H2 generation by single-layer1T-MoS2 and a nanocomposite of few-layer 2H-MoS2 with heavily nitro-genated graphene. Angew. Chem. Int. Ed., 52, 13057–13061.

42 Gupta, U., Naidu, B.S., Maitra, U., Singh, A., Shirodkar, S.N., Waghmare,U.V., and Rao, C.N.R. (2014) Characterization of few-layer 1T-MoSe2 and itssuperior performance in the visible-light induced hydrogen evolution reaction.APL Mat., 2, 092802.

43 Kumar, N., Maitra, U., Hegde, V.I., Waghmare, U.V., Sundaresan, A., and Rao,C.N.R. (2013) Synthesis, characterization, photocatalysis, and varied prop-erties of TiO2 cosubstituted with nitrogen and fluorine. Inorg. Chem., 52,10512–10519.

44 Lingampalli, S.R., Manjunath, K., Shenoy, S., Waghmare, U.V., and Rao,C.N.R. (2016) Zn2NF and related analogues of ZnO. J. Am. Chem. Soc., 138,8228–8234.

45 Kouser, S., Lingampalli, S.R., Chithaiah, P., Roy, A., Saha, S., Waghmare, U.V.,and Rao, C.N.R. (2015) Extraordinary changes in the electronic structure andproperties of CdS and ZnS by anionic substitution: co-substitution of P andCl in place of S. Angew. Chem. Int. Ed., 54, 8149–8153.

46 Roy, A., Shenoy, U.S., Manjunath, K., Vishnoi, P., Waghmare, U.V., and Rao,C.N.R. (2016) Structure and properties of Cd4P2Cl3, an analogue of CdS.J. Phys. Chem. C, 120, 15063–15069.

47 Amirav, L. and Alivisatos, A.P. (2010) Photocatalytic hydrogen productionwith tunable nanorod heterostructures. J. Phys. Chem. Lett., 1, 1051–1054.

48 Kalisman, P., Nakibli, Y., and Amirav, L. (2016) Perfect photon-to-hydrogenconversion efficiency. Nano Lett., 16, 1776–1781.

49 Simon, T., Bouchonville, N., Berr, M.J., Vaneski, A., Adrovic, A., Volbers, D.,Wyrwich, R., Döblinger, M., Susha, A.S., Rogach, A.L., Jäckel, F., Stolarczyk,J.K., and Feldmann, J. (2014) Redox shuttle mechanism enhances photo-catalytic H2 generation on Ni-decorated CdS nanorods. Nat. Mater., 13,1013–1018.

References 391

50 Lingampalli, S.R., Gautam, U.K., and Rao, C.N.R. (2013) Highly efficientphotocatalytic hydrogen generation by solution-processed ZnO/Pt/CdS,ZnO/Pt/Cd1-xZnxS and ZnO/Pt/CdS1-xSex hybrid nanostructures. EnergyEnviron. Sci., 6, 3589–3594.

51 Bao, N., Shen, L., Takata, T., and Domen, K. (2008) Self-templated synthe-sis of nanoporous CdS nanostructures for highly efficient photocatalytichydrogen production under visible light. Chem. Mater., 20, 110–117.

52 Lingampalli, S.R. and Rao, C.N.R. (2014) Remarkable improvement invisible-light induced hydrogen generation by ZnO/Pt/Cd1-yZnyS het-erostructures through substitution of N and F in ZnO. J. Mater. Chem. A,2, 7702–7705.

53 Lingampalli, S.R., Roy, A., Ikram, M., and Rao, C.N.R. (2014) Visible-lightinduced hydrogen generation with ZnO/NiO/Cd1-xZnxS (x= 0.0, 0.2) het-erostructures. Chem. Phys. Lett., 610-611, 316–320.

54 Balaraman, E., Khaskin, E., Leitus, G., and Milstein, D. (2013) Catalytic trans-formation of alcohols to carboxylic acid salts and H2 using water as theoxygen atom source. Nat. Chem., 5, 122–125.

55 Roy, A., Lingampalli, S.R., Saha, S., and Rao, C.N.R. (2015) Effects of mor-phology and surface area of the oxide nanostructures on the visible-lightinduced generation of hydrogen in ZnO (TiO2)/Cd1-x Znx S and ZnO(TiO2)/Pt/Cd1-x Znx S heterostructures (x= 0.0, 0.2). Chem. Phys. Lett., 637,137–142.

393

14

Recent Developments on Visible-Light PhotoredoxCatalysis by Organic Dyes for Organic SynthesisShounak Ray, Partha Kumar Samanta, and Papu Biswas

Indian Institute of Engineering Science and Technology, Department of Chemistry, College Road,Botanic Garden, Shibpur, Howrah 711 103, India

14.1 Introduction

Visible-light photoredox catalysis utilizes visible light as a renewable and greenenergy source to develop sustainable synthetic routes involving electron trans-fers (ETs) [1–20]. Visible-light photoredox processes occur under mild conditionsand do not involve radical initiators or stoichiometric reagents as compared tothermal reactions. Unlike specialized UV reactors employed in classical photo-chemistry, typical irradiation sources such as LEDs or household compact flu-orescent lamps (CFLs) which are much cheaper and easier to handle are used.The recent works on photoredox catalysis by MacMillan and Nicewicz, Yoon,Stephenson, König, and Rueping groups have received significant attention fromthe organic synthesis community [1–20]. Commonly, polypyridine complexes ofruthenium and iridium have been utilized as visible-light photocatalysts in mostof these reactions. Although these ruthenium and iridium polypyridyl complexesexhibit excellent photophysical properties in visible-light photocatalysis, theseare expensive and potentially toxic. Recently, organic dyes (Scheme 14.1) havebeen used as an attractive alternative to transition metal polypyridine complexesin visible-light-driven photoredox catalysis as these are inexpensive, nontoxic,easy to modify and handle, and even outperform metal-based catalysts in somecases [21–24]. In this chapter, we discuss general mechanism, key photophysicalproperties of dyes, and recent applications of some common organic dyes as avisible-light photocatalyst in organic synthesis.

14.2 General Mechanism

The photoinduced electron transfer (PET) as a key step in photochemical reac-tions has long been recognized. The rich photophysical properties of organic dyesand their ability to participate in PET processes are well known. Simplified stateenergy diagram given in Scheme 14.2 demonstrates the range of excited state


394 14 Recent Developments on Visible-Light Photoredox Catalysis

O

Br

Br

O

Br

Br

HO

COOH

Eosin Y (EY)

S

N

NNMe

Me

Me

Me

Methylene blue (MB)

S

NCNCN

X X

X = Br, H, OMe

Phenothiazine dye (PHENZ)

N N

NN

3,6-Di(pyridin-2-yl)

-1,2,4,5-tetrazine (PYTZ)

O

I

O

I

I

HO

I

HOOC Cl

Cl

Cl

Rose bengal (RB)

Scheme 14.1 Common organic dyes for visible-light photoredox catalysis.

S0

S1

+hν –hν

–hν E0,0 E0,0 Eox

Eox

Eox

S1 S1 S1T1

T1 T1

E*ox (cat / cat*)

E*ox (cat / cat*) < 0 E*ox (cat* / cat) > 0

E*ox (cat / *cat)

or

E1/2 (cat / cat)

cat EredEred

Eredt

or

E1/2 (cat / cat)

ca

ISC

Cat

Cat is an excited

state reductant if

Cat is an excited

state oxidant ifNon radiative

Radiative

IC

IC

Scheme 14.2 Photophysical and electrochemical processes in organic dyes.

energies and feasibility of different PET processes, which helps us to understandthe reactivity of an organic photoredox catalyst. Upon irradiation with visiblelight, promotion of an electron to a higher energy level occurs from a groundstate singlet (S0) to a singlet excited state (S1). Different photophysical pathwaysare now possible in the electronically excited molecule: S1 can come back toS0 either by a radiative transition (fluorescence) or by a nonradiative transition(internal conversion, IC), or it can proceed to T1 by a spin-forbidden nonra-diative process (intersystem crossing, ISC). Both S1 and T1 excited states canundergo energy transfer (EnT) and ET to participate in bimolecular reactions(i.e., reactions with a substrate). Thus, the term PET is used here to refer to the

14.2 General Mechanism 395

S or O

S or O

Cat Cat

Cat *

Cat

S or R

S or R

S or R S or O

S or R

S or O

Oxid

ative

quenchin

g

cycle

Reductive

quenchin

g

cycle

S = Substrate

O = Oxidant

R = Reductant

Scheme 14.3 Oxidative and reductive quenching cycles of a photoredox catalyst.

overall process of excitation and ET between the excited-state molecule and aground-state molecule.

In general, photoredox catalytic reactions follow one of the two mecha-nistic pathways illustrated in Scheme 14.3. First, organic photocatalyst, cat,is excited under visible-light irradiation to produce its excited-state speciescat*. This species can act as either a single-electron oxidant or reductant via asingle-electron transfer (SET) mechanism. In an oxidative quenching cycle, the

R1

R2

H

R1R2

A

hν

Ox. quench. Red. quench.

ox

ox

ox

ox

hν

–e

–

R1

R2

H

Cat

Cat∗

Cat∗

Cat Cat∗

Cat

CatCat

Cat∗

Cat∗

Cat∗

Cat

Cat

Cat

Cat

Red. quench. Ox. quench. Cat

Cat

Red

Red

hν

+e

–

CatRed

Redhν

R1

R2

R1

R2

AB+ A-B ±e–

Visible light

Ox. quench

Red. quench

Visible light

Net oxidative

Net reductive

Net redox-neutral

Scheme 14.4 Net redox outcomes in photoredox transformations.


excited state of catalyst, cat*, is quenched through donation of an electron eitherto substrate, S, or to an oxidant, O, present in the reaction mixture. In a reductivequenching cycle, cat* is quenched by accepting an electron from substrate, S,or from a reductant, R. Finally, the catalyst turnover step involves reductionof the oxidized cat∙+ in the oxidative cycle and oxidation of the reduced cat∙−in the reductive cycle. Overall, three general redox outcomes are possible forthe substrate in either quenching cycles: net oxidative, net reductive and netredox-neutral (Scheme 14.4). In net oxidative reactions, external oxidant acceptselectrons in either the PET step or the turnover step. Similarly, in net reductivereactions, external reductant donates electrons during the PET or turnover step.Net redox-neutral processes involve return electron transfer with the oxidizedor reduced catalyst, sometimes mediated by a redox active cocatalyst.

14.3 Recent Application of Organic Dyesas Visible-Light Photoredox Catalysts

14.3.1 Photocatalysis by Eosin Y

Eosin Y (EY) exhibits a characteristic peak at 539 nm with a molar extinctioncoefficient 𝜀= 60 803 M−1 cm−1 and absorbs green light. Upon exposure to visiblelight, Eosin Y is excited to S1 state and then undergoes rapid ISC to the lowestenergy triplet state having a life time of 24 μs [22, 25–27]. At excited state, EosinY becomes more reducing and more oxidizing compared to in its ground state[28, 29]. Scheme 14.5 shows the redox potentials of Eosin Y in CH3CN—H2O(1:1) in ground and corresponding excited states.

14.3.1.1 Perfluoroarylation of ArenesThe König group recently reported a direct arylation of simple arenes with fluo-rinated aryl bromides [30]. This work builds on that of Chen and coworkers, whopreviously disclosed the same reaction employing UV irradiation [31, 32]. How-ever, in the König work, 5 mol% of the common organic dye, Eosin Y, is employedas the photocatalyst under inert environment with triethylamine (TEA) as sacri-ficial electron donor to effect the arylation of simple arenes with fluorinated arylbromides (Scheme 14.6). The reaction was tolerant to a number of functionalgroup substitutions, including halides, methoxy, and nitro group as well as vari-ous heteroaromatic partners. The substitution of a methoxy group in the complexstructure of the alkaloid brucine was also reported.

The mechanism for the perfluoroarylation of arenes is shown in Scheme 14.7.Authors proposed both reductive and oxidative quenching cycles for thereaction. In the reductive quenching cycle, initially PET from TEA to the EY*occurs. The re-oxidation of the generated radical anion EY∙− involves SET tobromopentafluorobenzene. Subsequently, the reduced fluorinated arene cleavesthe CAr—Br bond generating the pentafluorophenyl radical, which reacts witharene to yield the product. The oxidative quenching cycle involves a PET fromEY* to bromopentafluorobenzene. The regeneration of the Eosin Y from theradical cation EY∙+ involves SET to TEA. In both pathways, hydrobromic acid

14.3 Recent Application of Organic Dyes as Visible-Light Photoredox Catalysts 397

EY

3EY∗

1EY∗

1.89 ev

+0.83 V–1.11V

hν

Eosin Y–1.06 V+0.78 V

EY EY

Scheme 14.5 The redox potentials of Eosin Y in CH3CN–H2O (1:1) in ground andcorresponding excited states.

Br

Fn

+

Reosin Y (5 mol%)

Et3N (2 equiv.)

535 nm, 72 h

MeCN (dry), N2, 40 °CFn

R

Scheme 14.6 Perfluoroarylation of arenes.

EY

EY EY

EY* EY*

C6F5Br

C6F5Br

TEA+

+

+

–

–

–

TEA

–Br–C6F5

Ar-H

C6F5-Ar

+HBr TEA

TEA

Reductive

quenching

cyclehν

Oxidative

quenching

cycle

C6F5Br–Br–

C6F5Ar-H C6F5-Ar + HBr

C6F5Br

EY

hν

Scheme 14.7 Perfluoroarylation of arenes mechanistic proposal.

is produced, which is neutralized by equivalent amount of TEA present in thereaction mixture.

14.3.1.2 Synthesis of Benzo[b]phosphole OxidesSynthesis of highly functionalized benzo[b]phosphole oxides using organic pho-toredox catalysis was recently described by Lakhdar and coworkers [33]. Reac-tions of arylphosphine oxides with alkynes in the presence of 4 mol% Eosin Y asthe catalyst and N-ethoxy-2-methylpyridinium tetrafluoroborate as the oxidantunder irradiation of 5 W LED (green light) afforded benzo[b]phosphole oxides(Scheme 14.8). The reaction proceeds through an oxidative C—H/P—H func-tionalization reaction of secondary phosphine oxides with alkynes and has broadfunctional group compatibility.


PR3

R4

XH +

R1

R2

eosin Y (4mol%)

Green LED (525 nm)

N-Ethoxy-2-methylpyridinium

tetrafluoroborate (2 equiv.)

NaHCO3 (1.2 equiv.)

DMF, 35 °C, 48 h

POR4

R2

R1

R3

4 5 6

Scheme 14.8 Synthesis of benzo[b]phosphole oxides.

POPh

Ph

PhHP

OPh

PhPh

POPh

Ph

PhH

N

OEt

N

+ OEt

HCO3P

OPh

Ph

Ph

SETEYEY

[EY-EMP]

[EY-EMP]*

N

OEt

+ OEt

P

Ph

O

Ph

Ph

Ph

hν

Chain

propagation

Photoredox

cycleP

Ph

H

Ph

O

EtOH

8 9 10 6a

5a7

4a

SE

T

N

Scheme 14.9 Mechanism for synthesis of benzo[b]phosphole oxides.

Control experiments lend support for the mechanism depicted in Scheme 14.9.The reaction initiates with the formation of the electron donor−acceptor (EDA)complex between the EY and the N-ethoxy-2-methylpyridinium (EMP). In thepresence of green light, the EY-EMP complex generates an unstable ethoxy radi-cal through single ET. The ethoxy radical abstracts a hydrogen from the secondaryphosphine oxide 4a to give rise to the corresponding phosphinoyl radical 7. Theradical produced then reacts with the alkyne 5a to generate the alkenyl radical8. The radical generated consequently attacks the phenyl ring of the phosphineoxide to give the cyclohexadienyl radical 9. The cyclohexadienyl radical 9 is read-ily oxidized by EY∙+ to generate the Wheland intermediate 10 and release thephotocatalyst. This species is instantaneously rearomatized through deprotona-tion by HCO3

− to yield the desired benzophosphole oxide 6a.

14.3.1.3 Direct C—H Arylation of HeteroarenesKönig and coworker utilize Eosin Y as a photoredox catalyst to initiate photocat-alyzed SET-mediated direct C—H bond arylation of heteroarenes (Scheme 14.10)


N2BF4

R +

O O

R

eosin Y (1 mol%)

DMSO, 20 °C

530 nm LED, 2 h11 12 13

Scheme 14.10 Direct C—H arylation of heteroarenes.

EY* EY

EY

X

R

X

R+

H

BF4– –HBF4

X

R

N2 + BF4– +

N2BF4

R R

14

N2BF4

R

R

SET

15

16

13

X

12

Radical propagation11

14

Scheme 14.11 Proposed mechanism for direct C—H arylation of heteroarenes.

with aryl diazonium salts [34]. The substrate scope of the reaction was extensivelyexplored and was found to exhibit a broad scope toward diazonium salts andheterocycles with a wide range of functional group tolerance including halides,alcohols, esters, nitro, and cyano groups. This methodology offers a mild and effi-cient alternative to transition-metal-catalyzed and tBuOK-promoted methods aswell evades the use of copper salts necessary in the classical Meerwein arylationprotocol [22, 23].

The proposed mechanism proceeds (Scheme 14.11) through reduction of thediazonium salt by the excited catalyst EY*. Aryl diazonium salts are known fortheir high reduction potential and easily accept electron from excited catalyst.Addition of aryl radical 14 to heteroarene 12 yields radical intermediate 15. Theallyl radical (15) can then undergo oxidation either by radical cation EY∙+ orthrough radical propagation by another species of aryl diazonium salt.

14.3.1.4 Synthesis of 1,2-Diketones from AlkynesSun and coworkers utilized the photoredox chemistry of Eosin Y under visiblelight for the synthesis of 1,2-diketones by the oxidation of corresponding


RAr

eosin Y

4-chlorobenzenethiol

MeCN, Air, rt, 8 h

blue LED

R

O

O

17 18

Scheme 14.12 Synthesis of 1,2-diketones from alkynes.

alkynes using air as the oxidant and 4-chlorobenzenethiol as sacrificial reagent(Scheme 14.12) [35]. Different photocatalysts, such as Ru(bpy)3Cl2⋅6H2O,Ir(ppy)3, rose bengal (RB), fluorescein, and rhodamine B, were examinedbut exhibited inferior catalytic efficiency than that of Eosin Y. The reactiondemonstrates good selectivity and tolerates to a wide range of functional groupsubstituents on the aryl ring, including some oxidation-sensitive groups such asformyl and a carbon−carbon double bond.

The authors proposed a SET mechanism for this synthesis based on controlexperiments done and literature reports (Scheme 14.13). At first ET to EY* fromthiophenol occurs to afford radical cation 19 and form EY∙−. The radical anionEY∙− is oxidatively quenched to the ground state by aerobic oxygen to completethe photoredox cycle through generation of superoxide radical anion (O2

∙−). Theresulting radical cation 19 is subsequently deprotonated by O2

∙− to produce thethiophenyl radical 20. The thiylperoxyl radical 21 was then formed through thereversible trapping of 20 with molecular oxygen. The addition of radical 21 to1,2-diphenylethyne (17) generated vinyl radical 22. The radical intermediate 23subsequently formed by rearrangement of 22 via the homolytic O—O bond cleav-age and the radical transfer. Desired product 1,2-diphenylethane-1,2-dione 18was finally produced by the elimination of thiophenyl radical 20 from 23.

EY

EY

EY∗

O2

O2

Ar-SH

Ar-SH–

–

–

+

Ar-S

Ar

S S

Ar O2 Ar-SOOPh Ph

Ph

OOSAr

Ph

O

PhPh

O

SAr

−ArS

PhPh

O

O

O2

HO2 H2O2 + O2

19

20

24

2122 23

18

Blue light

Scheme 14.13 Plausible mechanism for synthesis of 1,2-diketones from alkynes.


14.3.1.5 Thiocyanation of ImidazoheterocyclesHajra and coworkers developed a visible-light-mediated, metal-free process forthe thiocyanation of imidazoheterocycles using Eosin Y as a photoredox catalystunder ambient air at room temperature (Scheme 14.14) [36]. The protocol showsbroad substrates scope applicability and a wide range of functional groups toler-ance. A plausible mechanism was suggested on the basis of the controlled exper-iments and the literature reports (Scheme 14.15). Initially, thiocyanate anion isbeing oxidized to thiocyanate radical by SET from thiocyanate anion to EosinY* via a reductive quenching cycle. The resulting thiocyanate radical reacts with24a to generate the radical intermediate 26. Consequently, the intermediate 27is formed by oxidation of 26 and the product 25a obtained via deprotonation.Authors proposed involvement of molecular oxygen to complete the photoredoxcycle by oxidation of the radical anion EY∙− to the ground state.

N

N

R + NH4SCN

eosin Y

MeCN

Blue LED (525 nm)

3 h, ambient air

N

N

R

SCN24 25

Scheme 14.14 Thiocyanation of imidazoheterocycles.

EY

EY∗

EYO2

O2

Blue LED

EEY∗/EY = 0.83 V vs. SCE

SCN

SCN

ESCN/SCN = 0.62 V vs. SCE

NN

NN

NCS

NN

NCS

NN

NCS

24a

262725a

[O]–H+

Scheme 14.15 Suggested mechanism for thiocyanation of imidazoheterocycles.


14.3.2 Photocatalysis by Rose Bengal

RB displays unusual spectroscopic and photochemical properties including alarge absorption coefficient (90 400 M−1 cm−1 at 550 nm in water) in the visibleregion and a high tendency for ISC to produce a photochemically active tripletexcited state. It is also well documented that RB could react with molecularoxygen under visible-light irradiation to generate two reactive species – singletoxygen (1O2) through EnT or superoxide radical anion via single ET [37]. Dyessuch as RB and methylene blue (MB) are very efficient photosensitizers, as theypossess triplet states of suitable energies for sensitization of oxygen (Scheme14.16 and Table 14.1).

14.3.2.1 Aerobic Indole C-3 Formylation ReactionThe Li group recently developed an aerobic visible-light-promoted indole C-3formylation reaction catalyzed by RB [38]. This protocol utilizes molecularoxygen as the terminal oxidant and tetramethylethylenediamine (TMEDA) as

Fluorescence T1

∼150 KJ

Chemical reations

Free radical, redox

τ ∼10–8 s

τ ∼10–3 s

95 kJ

Chemical reactions

1O2

S1

O2

Superoxide

Dioxygen

Organic dye

O2

T0

∼200 kJ

S0

S1

+hν IC

ISC

Scheme 14.16 Generation of excited photosensitizer states and reactive dioxygen species.

Table 14.1 Photophysical properties of rose bengal and methylene blue.

DyeTriplet state energy,ET (kJ mol−1) 𝜱𝚫

a)𝜱𝚫

a)𝜱𝚫

a)

Water Ethanol Methanol

Rose bengal 176 0.75 0.68 0.76Methylene blue 134 0.52 0.50

a) 𝛷Δ is the quantum yield of singlet oxygen generation.


N

R2

R1 + NN

Rose bengal (5 mol%)

KI (4 equiv.), O2, hνMeCN–H2O, 60 °C

N

R2

R1

CHO

28 29 30

Scheme 14.17 Aerobic indole C-3 formylation reaction.

the one-carbon source through C—N bond cleavage (Scheme 14.17). Variety ofphotosensitizers were screened including Ru(bpy)3(PF6)2, Ru(bpy)3Cl2, Ir(ppy)3,rhodamine B, Alizarin Red S and Eosin Y but RB turned out to be the mostefficient catalyst. The reaction is compatible with a variety of substitution groupson the indole nitrogen as well as functional groups on the carbon skeleton ofN-methylindole ring with very-good-to-moderate yield.

The reaction proceeds through oxidative quenching of the visible-light-excitedrose bengal (RB*) by TMEDA (29) resulted in the generation of radical anionRB∙− and radical cation 31 (Scheme 14.18). The radical cation 31 either gives up ahydrogen atom to the superoxide radical anion to yield hydrogen peroxide anionand iminium ion 32 or gives up a proton and form an α-amino carbon radical,

NN

NN

NN

NN

–e

N

N

NN

N

NN

N

N

CHOHN

N

H2O

RB*

RB

RB Photoredox

catalysis

Visible

light

–H

–H

29

31

28a

32

33

34

35

30a

36

Photoredox

catalysis

O2

O2

Scheme 14.18 A plausible mechanism for aerobic indole C-3 formylation reaction.


which possibly goes through a second oxidation to afford iminium ion 32. Then,electrophilic addition of iminium ion 32 to N-methylindole (28a) produce indoleiminium ion 33. A C-3-substituted N-methylindole intermediate 34 is subse-quently formed through rearomatization by loss of proton. Presumably, a secondvisible-light photoredox cycle occurred to generate iminium ion 35 as intermedi-ate 34 is more facile to form iminium ion than TMEDA. Hydrolysis of 35 cleavedthe C—N bond and afforded 3-formyl-N-methylindole (30a). Transfer of an elec-tron to O2 to form the superoxide radical anion from RB∙−, regenerate RB.

14.3.2.2 Decarboxylative/Decarbonylative C3-Acylation of IndolesThe Wang group and the Li group collaboratively developed a simple andefficient approach for the synthesis of 3-acylindoles via visible-light-promotedC3-acylation (Scheme 14.19) of free (NH)- and N-substituted indoles withα-oxocarboxylic acids [39]. The reactions were carried out in ethanol at ambienttemperature in air using a green LED as visible-light source. The reactionexhibits high regioselectivity and good functional group compatibility in bothindole and α-oxocarboxylic acids.

Proposed mechanism for this C3-acylation of indoles is shown in Scheme 14.20.RB* produced in the presence of green LED interacts with molecular oxygento generate singlet oxygen 1O2 via the EnT [37], along with the generationof RB to its ground state. Consequently, the produced 1O2 reacts with 38awhich undergoes decarboxylation to afford a hydroperoxyl radical (40) anda benzoyl radical (41) with the release of CO2. Authors proposed differenttransformation pathways simultaneously for reaction of the benzoyl radical (41).In path I, the benzoyl radical (41) is added to the carbon–carbon double bondof indole (37a) at the C3-position to yield the intermediate radical (42), whichundergoes one-electron oxidation in the presence of the formed hydroperoxylradical (40) to produce intermediate 43. Intermediate 43 is rearomatized bydeprotonation to give the final product 3-benzoylindole (39a). Whereas, in pathII, in the presence of the 40, the free indole (37a) is oxidized to an indoliniumradical cation (44) via a SET process. Subsequently, benzoyl radical (41) isadded to the carbon–carbon double bond of the indolinium radical cation(44) at the C3-position to give an intermediate 45. Then, intermediate 45undergoes an aromatization process to generate 3-benzoylindole (39a) as thefinal product.

N

H

R1

R2

R3 +R4

O

OH

O


4 A molecular sieve (80 mg)

EtOH, air, rt, 10 h NR1

R2

R3

OR4

37 38 39

Scheme 14.19 Decarboxylative/decarbonylative C3-acylation of indoles.


R4

O

NH

OR4

NH

OR4

NH

H

OR4

NH

OR4

HOOH2O2

OOH

HOO–

HOO

H2O2

NH

N

HOOSET

Path IPath IINH

R4 TEMPO

O

RB*

RB

hν

O2

1O2

R4

O

O

O

R4

O

OH

O

OOH

38a

MinorMajor

Ph

CO2CO2

TEMPO

39a

37a

40

41

37a

43

44

4245

46

Scheme 14.20 Proposed mechanism for decarboxylative/decarbonylative C3-acylation ofindoles.

14.3.2.3 Oxidative Annulation of ArylamidinesFast catalytic synthesis of multisubstituted quinazolines from commonlyavailable amidines via visible-light-mediated oxidative C(sp3)—C(sp2) bondformation has been developed by the Tang group [40]. Reactions were per-formed with RB (1 mol%) as photocatalyst, CBr4 (1.2 equiv.) as an oxidant, andCs2CO3 as the base under irradiation from a 18 W fluorescent bulb in DMSO at100 ∘C (Scheme 14.21). Several commonly used photocatalysts, such as Eosin Y,rhodamine B, Ir(ppy)3, Ru(bpy)3Cl2, and CdS were also investigated but found toshow inferior activity than RB. The reaction was found to tolerate a wide rangeof functional groups.

On the basis of the control experiments and literature reports, a plausiblemechanism is proposed as shown in Scheme 14.22. Authors have suggested morethan one reaction pathway. In path I, without RB, CBr3, and a bromide radicalcould be generated from CBr4 in the presence of visible light. The generated CBr3radical then abstracts a hydrogen atom of another molecule of benzimidamideto produce the α-amino radical intermediate 49. The intermediate 49, onceagain enters the radical chain process with CBr4 to generate iminium ion 51.Iminium ion 51 undergoes an intramolecular Friedel−Craft reaction to pro-duce intermediate 52 which after dehydrogenation afforded hydroquinone 53.


NR1

HN

R2Rose bengal (1 mol%)

CBr4 (1.2 equiv.)

K2CO3 (1.5 equiv.)

DMSO (0.05 M)

Visible light

NR1

N

R2

47 48

Scheme 14.21 Oxidative annulation of arylamidines.

NR1

HN

R2

NR1

HN

R2

NR1

HN

R2

NR1

HN

R2

R3 R3

R3 R3

NR1

HN

R2 H

R3

NR1

HN

R2

R3

NR1

N

R2

R3

CBr3 HCBr3

CBr4

CBr3 + Br–

CBr3 HCBr3

–H+

[O]

RB*

RB

RB

CBr4

CBr3 + Br–

Visible

light

–H+

47 49

51

525348

50

Path I

Path II

Scheme 14.22 Mechanistic proposal for oxidative annulation of arylamidines.

Aromatization of 53 gives the stable product quinazoline. In path II, authors haveproposed the oxidative quenching of RB* by benzimidamide to form RB∙− andradical cation 50. Transfer of an electron to CBr4 to form the CBr3 radical fromRB∙− regenerated the catalyst RB. Radical cation 50 gives up a hydrogen atom,apparently to the radical anion to afford the CBr3 radical, bromide anion, andradical intermediate 49. There may be an additional possibility in which CBr3radical abstracts a hydrogen atom from radical cation 50 to produce iminium ion51 directly following the same procedure to give the stable product quinazoline.

14.3.2.4 Cross-Dehydrogenative Coupling of Tertiary Amines with DiazoCompoundsVisible-light-induced cross-dehydrogenative coupling between tertiary aminesand diazo compounds has recently been explored by Zhou and coworkers[41]. The reaction proceeds smoothly under mild conditions by utilizingair or O2 as the oxidant and RB as photocatalyst under 5 W green LEDirradiation to afford various β-amino-α-diazo adducts (Scheme 14.23). Theresulting adducts were successfully employed for the synthesis of 4- or5-ester N-aryl-2,3-dihydrobenzo[d]azepines with high regioselectivity. The


ArN H

R1

R2

+ R3

O

N2

H

Rose bengal (0.5 mol%)

DCM (1 ml)

5 W green LED

O2, rt, 12 h

ArN

R1

R2

N2

R3

O

54 55 56

Scheme 14.23 Cross-dehydrogenative coupling of tertiary amines with diazo compounds.

ArN H

R1

R2

ArN H

R1

R2

ArN

R1

R2

[O2]

HO2

R3

O

N2

H

ArN

R1

R2

N2+

O

R3H

H2O2

ArN

R1

R2

N2

R3

O

RB* RB

RB

SET

Visible light

O2

O2

54

55

56 59

5857

HO2

Scheme 14.24 Suggested cross-dehydrogenative coupling of tertiary amines with diazocompounds.

reaction showed a wide range of functional group tolerance. The substratesbearing methyl, tertiary butyl, methoxyl, chloro, fluoro, trifluoromethyl atthe para-position of the benzene ring connected to the nitrogen atom in theN-aryltetrahydroisoquinolines (THIQ) underwent smooth reactions. ThoughN-aryl THIQ bearing methoxyl, methyl at the o-, m-position produced thecorresponding diazo esters in slightly low yields.

The proposed reaction mechanism is shown in Scheme 14.24. A reductivequenching cycle for RB* by tertiary amine via a SET process to form intermediate57 was proposed. The regeneration of photocatalyst occurs through oxidation ofRB∙− by molecular oxygen to generate superoxide radical anion. Abstraction ofhydrogen atom from 57 by O2∙− provides the iminium ion 58, which undergoesnucleophilic attack by a diazo compound to yield intermediate 59. Finally,deprotonation of intermediate 59 by hydroperoxide anion leads to generation ofthe desired α-amino diazo compound 56.

14.3.2.5 C—H Functionalization and Cross-Dehydrogenative CouplingReactionsRueping and coworkers recently reported the application of RB in variouscontinuous flow C—C and C—P bond-forming reactions, including unprece-dented visible-light organophotoredox catalyzed multicomponent reactions[16]. Few organic dyes, such as Eosin Y and rhodamine B, were tested alongwith RB in the multicomponent reaction in flow, but lower conversions wereobserved. This environmentally friendly, metal-free, photoorganocatalytic, con-tinuous flow methodology was successfully applied in the α-functionalizationof tertiary amines (Scheme 14.25). Different substrates such as nitroalkanes,


NAr

R1

R1

+ NuRose bengal (5 mol%)

Green LED

solvent 0.03 ml min–1

NAr

R1

R1

Nu

60 61

Scheme 14.25 α-Functionalization of tertiary amines.

N

R1 + CN-R2+ H2O


Green LED

N

R1

NH

O

R2

62 63 64

Scheme 14.26 Ugi-multicomponent reaction in flow.

TMSCN, dialkyl malonates, and dialkyl phosphites were reacted with variousN-aryl tetrahydroisoquinolines to get the corresponding products in mod-erate to excellent yields. Moreover, N ,N-dimethylanilines were successfullyreacted with different isocyanides in the Ugi-multicomponent reaction in flow(Scheme 14.26), resulting in highly valuable α-amino amides in good yields.

14.3.2.6 Oxidative Cross-Coupling of Thiols with P(O)H CompoundsThe direct S—P(O) coupling between thiols and P(O)H compounds in the pres-ence of air as oxidant and RB as photocatalyst (Scheme 14.27) has been reportedby the Li and the Zhang groups [42]. Preliminary, several photocatalysts such asRu(bpy)3Cl2, fac-Ir(ppy)3, Mes-Acr+, Eosin Y, Eosin B, rhodamine B, and RB wereexamined as photocatalysts but RB was found to be the most effective. The reac-tion exhibits excellent chemoselectivity and good functional-group tolerance.

The proposed mechanism shows that photogenerated RB* interacts withO2 to generate 1O2 via the EnT (Scheme 14.28). Concomitantly, excited-stateRB* returns to its ground state. Consequently, the generated 1O2 abstractshydrogen atom from thiol 65 to produce the thiyl radical 68. The thiyl radical68 then undergoes homocoupling to form the disulfide 69. Simultaneously,the P-centered radical 70 is also generated from the P(O)H compound 66 by a

R1-SH + P R2

R3

O

HRose bengal (5 mol%)

DMF, rt, air

10 W blue LED

P R2

R3

O

SR1

65 66 67

Scheme 14.27 Oxidative cross-coupling of thiols with P(O)H compounds.


R1-SH R1-S R1

SS

R1

P

OH

R3

R2

P

O

R3

S R2R1

P

O

R3

R2

Path I

Path III

Path II

P

O

R3

R2 H

RB

RB*

O2

1O2

Visible light

HO2

R1

SS

R1

P

O

R3

R2 H

1O2

HO2

65

66 66

67

6869

66a

70

69

Scheme 14.28 Suggested mechanism for oxidative cross-coupling of thiols with P(O)Hcompounds.

similar oxidation route. Finally, the formed P-centered radical 70 couples withthe thiyl radical 68 (path I) or reacts with the disulfide 69 (path II) to yieldthe desired product 67. Alternatively, the product 67 could also be formed bynucleophilic attack of the P(O)H compound 66 with the disulfide 69 (path III).

14.3.3 Photocatalysis by Methylene Blue

MB is another organic photosensitizer, which also known for its ability to gener-ate singlet oxygen [43]. MB exhibits two peaks in the visible region at about 610and 670 nm. Upon irradiation by visible light, it also has high tendency to gen-erate a photochemically active triplet excited state. Substrates generally interactwith active triplet excited state to form radicals or, to generate singlet oxygen inthe case of molecular oxygen (Scheme 14.16 and Table 14.1) [37].

14.3.3.1 Oxidative Hydroxylation of Arylboronic AcidsThe photocatalytic hydroxylation of boronic acids with MB as photosensi-tizer has been recently reported by the Scaiano group [44]. The reaction waspreviously explored by Xiao and coworkers, employing [Ru(bpy)3]2+ as thephotoredox catalyst [45]. The Scaiano group demonstrated efficiency of theenvironmentally benign and low-cost organic dye as sensitizers for the samereaction. The reaction was performed in presence of 1% MB and air using


BOH

OH

OH1 mole % methylene blue

5 eq iPr2NEt

MeCN:H2O (4:1)

O2, hν71 72

R R

Scheme 14.29 Oxidative hydroxylation of arylboronic acids.

MB

MB

MB*

iPr2NEt

iPr2NEt

O2

ArB(OH)2

hydrolysisAr

N

O2

Slow

Fast

BH(OH)2

O O Disproportionation

NHR2

H

–OH–

O B

Ar BH(OH)2

O OH

OH

OHArHydrolysis

ArOH

73

Scheme 14.30 A plausible mechanism for oxidative hydroxylation of arylboronic acids.

diisopropylethylamine (iPr2NEt), as a sacrificial electron donor (Scheme 14.29).They observed that MB was more efficient than [Ru(bpy)3]2+ at catalyzing theconversion through a series of excited state quenching studies. The yields of thephenol adducts were all quite high; however, the scope of the protocol was notnearly as extensive as Xiao’s work.

Scaiano and coworkers concluded that the catalytic cycle likely involves reduc-tive quenching of the triplet state of the methylene blue (MB*) via single ET fromiPr2NEt (Scheme 14.30) to produce the semireduced form of methylene blue(MB∙). The generated MB∙ then acts as an electron donor to the molecular oxygento form O2

∙−. This O2∙− reacts with the boron center to produce 73. The interme-

diate 73 abstracts hydrogen atom from the amine cation radical iPr2NEt⋅+ and arearrangement followed by hydrolysis yielded the desired phenol.

14.3.3.2 Radical TrifluoromethylationThe Scaiano group later on reported the catalytic radical trifluoro- and hydrotri-fluoromethylation of electron-rich heterocycles as well as terminal alkenes andalkynes under visible-light irradiation using MB [46]. The reaction proceedswith moderate to good yields at low catalyst concentration (1 mol%) and shortirradiation times using TMEDA as the electron donor and Togni’s reagent asthe electrophilic CF3 source (Scheme 14.31). A plausible mechanistic pathwayas proposed by authors for the catalytic formation of CF3 radicals is shown inScheme 14.32. The triplet excited state MB* reductively quenched by TMEDA toform the semireduced MB∙ radical and an α-amino radical. Both of these specieshave potential to reduce Togni’s reagent, resulting in the release of a CF3 radicaland the formation of 2-iodobenzoate.


X

R2

R1

+

X = N, S

IO

O

CF3

MB (2 mol%)

TMEDA (2 equiv)

DMF, visible light X

R2

R1

CF3

74 75 76

Scheme 14.31 Radical trifluoromethylation.

IO

O

CF3

IO

O

CF3

SET

IO

O

CF3

MB MBhν

R1

N R3

R2

R1

N R3

R2

–H+

R1

N R3

R2

Scheme 14.32 Proposed mechanism for radical trifluoromethylation.

14.3.4 Photocatalysis by 3,6-Di(pyridin-2-yl)-1,2,4,5-tetrazine

s-Tetrazine molecules are highly colored (because of a low-lying π* orbitalleading to a n–π* transition in the visible region) and electroactive heterocyclesthat display a very high electron affinity [47], which makes them easily reducible(actually they are the electron poorest C—N heterocycles). This reducingnature of the s-tetrazines is even more prominent in its first excited state,which therefore has a relatively strong oxidizing power [47]. Consequently,s-tetrazines interact with various electron donor substrates at an excited state.3,6-Di(pyridin-2-yl)-1,2,4,5-tetrazine (Pytz) exhibits maximum absorptionpeaks at 535 nm (𝜀= 235 M−1 cm−1) and a quasi-reversible reduction peak(ΔE 1/2 = 90 mV) at 160 mV in ethanol [48].

14.3.4.1 Synthesis of 2-Substituted Benzimidazole and BenzothiazoleRecently, we reported synthesis of 2-substituted benzimidazoles and ben-zothiazoles from the reaction of aldehyde with o-phenylenediamine or


RCHO +

H2N

XH Visible light, O2, rt, ethanol

N

XR

X = NH or S

77 78 79

Pytz

Scheme 14.33 Synthesis of 2-substituted benzimidazole and benzothiazole.

pytz

pytz*

pytz

XH

NH2

XH

N

–H+

R H

O

N

X

R

H

N

X

R

H

H+

+

R

H

N

X

R

H

H

O2

X

N

R

X = S or NH

81

82

83

84

XH

N

R

H

80

–H2O

–H2

79

Scheme 14.34 Suggested mechanism for synthesis of 2-substituted benzimidazole andbenzothiazole.

o-aminothiophenol under visible-light irradiation at ambient temperature(Scheme 14.33) in the presence of air using 3,6-di(pyridin-2-yl)-1,2,4,5-tetrazine[48]. The reaction demonstrates excellent catalytic activity for alkyl, aryl, andorganometallic substituted aldehydes as well as reducing sugar. This reactionfulfilled many of the requirements of green chemistry: (a) a very low E-factor wasachieved as minimal waste is generated; (b) high atom-economy was achieved.

We proposed that the reaction proceeds through the imine intermediate,80 (Scheme 14.34). Upon irradiation, pytz gets excited to pytz*, which isreductively quenched by imine intermediate 1 to produce pytz radical anionand monoaldimine radical cation 81 via SET oxidation. The radical cation 81then yields another radical species 82 after deprotonation. Now, intramolecularnucleophilic attack on C=N carbon atom takes place, followed by regeneration


of the catalyst pytz through oxidation. Subsequent proton uptake produceshydrogenated cyclized intermediate 84. Then, oxidative dehydrogenation by airleads to desired benzimidazoles or benzothiazoles.

14.3.4.2 Oxidation of Alcohols to Carbonyl DerivativesWe further explored the photoredox activity of pytz and reported efficientoxidation of alcohols to the corresponding carbonyl compounds undervisible-light irradiation using pytz as catalyst (Scheme 14.35) using O2 ortert-butylnitrite/acetic acid mixture as sacrificial co-oxidant [49]. Conversionof various primary and secondary alcohols to the corresponding aldehydes andketones were carried out efficiently under mild conditions with high yields. Onthe basis of experimental results and the previous studies, we proposed that thereaction may proceed through the formation of radical cation, ROH∙+ (87). Inthe presence of visible light, pytz is converted to pytz*, which is transformed in topytz radical anion through SET from alcohol to form radical cation, ROH∙+, 87(Scheme 14.36). Subsequently, hydrogen atom abstraction by stable pytz radical

R1 R2

OHMethod A: pytz (5 mol%), CH3CN, rt, O2

visible light irradiation, 10–12 h

Method B: pytz (5 mol%), CH3CN, rt

t-BuONO (1.1 equiv), AcOH (1.1 equiv)

visible light irradiation, 4–5 h

R1 R2

O

85 86

Scheme 14.35 Oxidation of alcohols to carbonyl derivatives.

R1 R2

OH

pytz*

R1 R2

OH

pytz

pytz H2pytzR1 R2

OOxidant

Vis

ible

lig

ht

+ e–

+ e–

+ 2 H+

87

Scheme 14.36 A plausible mechanism oxidation of alcohols to carbonyl derivatives.


anion from the radical cation 87 followed by deprotonation gives the carbonylcompound. Oxidation of the generated H2pytz to reproduce catalyst, pytz, wasdone in presence of molecular oxygen or tert-butylnitrite/acetic acid mixture.

14.3.5 Photocatalysis by Phenothiazine Dyes: Oxidative Couplingof Primary Amines

Highly efficient photocatalytic oxidative coupling of primary amines at ambienttemperature using phenothiazine dyes (Scheme 14.1) as photoredox catalyst hasbeen reported recently (Scheme 14.37) [50]. The synthesized dyes (Scheme 14.1)showed maximum absorption peaks in the 402–447 nm range and reversibleone-electron oxidation potentials in the 0.45–0.53 V range. The phenothiazinedye with electron-donating —OMe group attached at 4-position of the sidephenyl groups exhibited highest efficiency for oxidative coupling of primaryamines. The electronic effect of substituents on arenes of benzyl amines wasinsignificant in conversion. The photocatalytic system also showed good activityin the simple oxidation of the secondary amine but alkylamines were not con-verted to the imines. The mechanism of oxidative coupling of primary amineswas proposed based on the literature reports (Scheme 14.38). The formationof benzenemethanimine (90) and successive addition of amine produced thecoupled imine and H2O2. A quantitative amount of H2O2 was detected at10.2 ppm in 1H NMR spectroscopy.

R1 NH2

Phenothiazine dyes

O2, Blue LED

MeCN, rt, 20 h

R1 N R1

88 89

Scheme 14.37 Oxidative coupling of primary amines.

R1NH2

+

O2

R1 NH2+

HO2 R1 NH2

OOH

or

R1NH + HO2

R1 NH + H2O2

R1 NH2

HN R1R1 N R1

+ NH3

R1 NH2

90

Scheme 14.38 Oxidative coupling of primary amines mechanistic proposal.

References 415

14.4 Conclusion

Visible-light-photoredox catalysis with ruthenium and iridium polypyridylcomplexes has already received a lot of attention as a tool for organic synthetictransformations. In the recent past, organic dyes emerge as a suitable alternativeto metal-polypyridine-based catalysts, and the synthetic organic chemistscommunity has begun to realize the immense potential of these dyes in catalysis.These catalysts are able to deliver unique chemical reactivity under mild con-ditions and are also tolerant to wide range of complex functionality. A varietyof astonishing organic synthetic transformations have been made possible byorganic photoredox catalysis via PET reactions including perfluoroarylationof arenes, direct C—H arylation of heteroarenes, 1,2-diketones from alkynes,aerobic indole C-3 formylation reaction, and so on. For numerous organic trans-formations organic dyes exhibited enhanced efficiency and serve as an attractivealternative to redox active metal based catalysts. Moreover, the use of organicphotocatalysts has been extended in continuous flow technology. Therefore,metal-free photocatalytic reactions via the PET of organic photocatalysts offernovel ways to accomplish environmentally benign organic transformations.Thus, the scope and the strategy of using organic photoredox catalysts couldbe expanded to the utilization of solar energy in various metal-free catalyticmethodologies in organic synthesis, chemical industry, and pharmaceuticalindustry and are expected to expand much further in the future.

Abbreviations

CFL compact fluorescent lampEY eosin YIC internal conversionISC intersystem crossingLED light-emitting diodeMB methylene bluePET photoinduced electron transferPYTZ 3,6-di(pyridin-2-yl)-1,2,4,5-tetrazineRB rose bengalSET single electron transferTEA triethyl amineTHIQ N-aryltetrahydroisoquinolinesTMEDA tetramethylethylenediamineTMSCN trimethylsilyl cyanide

References

1 Niceicz, D.A. and MacMillan, D.W.C. (2008) Merging photoredox catalysiswith organocatalysis: the direct asymmetric alkylation of aldehydes. Science,322, 77–80.


2 Ischay, M.A., Anzovino, M.E., Du, J., and Yoon, T.P. (2008) Efficient visiblelight photocatalysis of [2+2] enone cycloadditions. J. Am. Chem. Soc., 130,12886–12887.

3 Narayanam, J.M.R., Tucker, J.W., and Stephenson, C.R.J. (2009)Electron-transfer photoredox catalysis: development of a tin-free reductivedehalogenation reaction. J. Am. Chem. Soc., 131, 8756–8757.

4 Reckenthäler, M. and Griesbeck, A.G. (2013) Photoredox catalysis for organicsyntheses. Adv. Synth. Catal., 355, 2727–2744.

5 Hari, D.P. and Konig, B. (2013) The photocatalyzed Meerwein arylation: clas-sic reaction of aryl diazonium salts in a new light. Angew. Chem. Int. Ed., 52,4734–4743.



8 Hari, D.P., Hering, T., and König, B. (2013) Synthetic applications of aryl dia-zonium salts enabled by visible light photoredox catalysis. Chem. Today, 31,59–63.

9 Hari, D.P., Hering, T., and König, B. (2014) The photoredox-catalyzed Meer-wein addition reaction: intermolecular amino-arylation of alkenes. Angew.Chem. Int. Ed., 53, 725–728.

10 Xuan, J. and Xiao, W.-J. (2012) Visible-light photoredox catalysis. Angew.Chem. Int. Ed., 51, 6828–6838.

11 Xi, Y., Yi, H., and Lei, A. (2013) Synthetic applications of photoredox catalysiswith visible light. Org. Biomol. Chem., 11, 2387–2403.

12 Shi, L. and Xia, W. (2012) Photoredox functionalization of C—H bondsadjacent to a nitrogen atom. Chem. Soc. Rev., 41, 7687–7697.

13 Mo, F., Dong, G., Zhang, Y., and Wang, J. (2013) Recent applications of arenediazonium salts in organic synthesis. Org. Biomol. Chem., 11, 1582–1593.

14 Schroll, P., Hari, D.P., and König, B. (2012) Photocatalytic arylation of alkenes,alkynes and enones with diazonium salts. Chem. Open, 1, 130–133.

15 Hering, T., Hari, D.P., and König, B. (2012) Visible-light-mediated α-arylationof enol acetates using aryl diazonium salts. J. Org. Chem., 77, 10347–10352.

16 Rueping, M., Vila, C., and Bootwicha, T. (2013) Continuous flow organocat-alytic C—H functionalization and cross-dehydrogenative coupling reactions:visible light organophotocatalysis for multicomponent reactions and C—C,C—P bond formations. ACS Catal., 3, 1676–1680.

17 Prier, C.K., Rankic, D.A., and MacMillan, D.W.C. (2013) Visible light pho-toredox catalysis with transition metal complexes: applications in organicsynthesis. Chem. Rev., 113, 5322–5363.

18 Tucker, J.W. and Stephenson, C.R.J. (2012) Shining light on photoredox catal-ysis: theory and synthetic applications. J. Org. Chem., 77, 1617–1622.

19 Yoon, T.P., Ischay, M.A., and Du, J. (2010) Visible light photocatalysis as agreener approach to photochemical synthesis. Nat. Chem., 2, 527–532.

20 Angnes, R.A., Li, Z., Correia, C.R.D., and Hammond, G.B. (2015) Recent syn-thetic additions to the visible light photoredox catalysis toolbox. Org. Biomol.Chem., 13, 9152–9167.

References 417

21 Romero, N.A. and Nicewicz, D.A. (2016) Organic photoredox catalysis. Chem.Rev., 116, 10075–10166.

22 Hari, D.P. and König, B. (2014) Synthetic applications of eosin Y in photore-dox catalysis. Chem. Commun., 50, 6688–6699.

23 Nicewicz, D.A. and Nguyen, T.M. (2014) Recent applications of organic dyesas photoredox catalysts in organic synthesis. ACS Catal., 4, 355–360.

24 Fukuzumi, S. and Ohkubo, K. (2014) Organic synthetic transformations usingorganic dyes as photoredox catalysts. Org. Biomol. Chem., 12, 6059–6071.

25 Penzkofer, A., Beidoun, A., and Daiber, M. (1992) Intersystem-crossing andexcited-state absorption in eosin Y solutions determined by picosecondsdouble pulse transient absorption measurements. J. Lumin., 51, 297–314.

26 Penzkofer, A., Beidoun, A., and Speiser, S. (1993) Singlet excited-state absorp-tion of eosin Y. Chem. Phys., 170, 139–148.

27 Penzkofer, A. and Beidoun, A. (1993) Triplet–triplet absorption of eosin Y inmethanol determined by nanosecond excimer laser excitation and picosecondlight continuum probing. Chem. Phys., 177, 203–216.

28 Lazarides, T., McCormick, T., Du, P., Luo, G., Lindley, B., and Eisenberg, R.(2009) Making hydrogen from water using a homogeneous system withoutnoble metals. J. Am. Chem. Soc., 131, 9192–9194.

29 Neumann, M., Füldner, S., König, B., and Zeitler, K. (2011) Metal-free, coop-erative asymmetric organophotoredox catalysis with visible light. Angew.Chem. Int. Ed., 50, 951–954.

30 Meyer, A.U., Slanina, T., Yao, C.-J., and König, B. (2016) Metal-free perfluo-roarylation by visible light photoredox catalysis. ACS Catal., 6, 369–375.

31 Chen, Q.Y. and Li, Z.T. (1993) Pentafluorophenylation of aromatics withpentafluorophenyl perfluoro- and polyfluoroalkanesulfonates: a photoin-duced electron-transfer cation diradical coupling process. J. Org. Chem., 58,2599–2604.

32 Chen, Q.-Y. and Li, Z.-T. (1993) Photoinduced electron transfer reactionsof pentafluoroiodobenzene with aromatic compounds. J. Chem. Soc., PerkinTrans. 1, 14, 1705–1710.

33 Quint, V., Morlet-Savary, F., Lohier, J., Lalevée, J., Gaumont, A., andLakhdar, S. (2016) Metal-free, visible light-photocatalyzed synthesis ofbenzo[b]phosphole oxides: synthetic and mechanistic investigations. J. Am.Chem. Soc., 138, 7436–7441.

34 Hari, D.P., Schroll, P., and König, B. (2012) Metal-free, visible-light-mediateddirect C—H arylation of heteroarenes with aryl diazonium salts. J. Am. Chem.Soc., 134, 2958–2961.

35 Liu, X., Cong, T., Liu, P., and Sun, P. (2016) Synthesis of 1,2-diketones via ametal-free, visible-light-induced aerobic photooxidation of alkynes. J. Org.Chem., 81, 7256–7261.

36 Mitra, S., Ghosh, M., Mishra, S., and Hajra, A. (2015) Metal-free thiocyana-tion of imidazoheterocycles through visible light photoredox catalysis. J. Org.Chem., 80, 8275–8281.

37 (a) Tung, C.-H., Wu, L.-Z., Zhang, L.-P., and Chen, B. (2003) Supramolecularsystems as microreactors: control of product selectivity in organic pho-totransformation. Acc. Chem. Res., 36, 39–47; (b) Ravelli, D., Fagnoni, M.,


and Albini, A. (2013) Photoorganocatalysis. What for? Chem. Soc. Rev., 42,97–113; (c) Ghogare, A.A. and Greer, A. (2016) Using singlet oxygen tosynthesize natural products and drugs. Chem. Rev., 116, 9994–10034; (d)DeRosa, M.C. and Crutchley, R.J. (2002) Photosensitized singlet oxygen andits applications. Coord. Chem. Rev., 233–234, 351–371.

38 Xiang, L., Xiangyong, G., Yongjuan, L., and Pixu, L. (2014) Aerobictransition-metal-free visible-light photoredox indole C-3 formylation reaction.ACS Catal., 4, 1897–1900.

39 Shi, Q., Li, P., Zhua, X., and Wang, L. (2016) Decarboxylative/decarbonylativeC3-acylation of indoles via photocatalysis: a simple and efficient route to3-acylindoles. Green Chem., 18, 4916–4923.

40 Shen, Z.-C., Yang, P., and Tang, Y. (2016) Transition metal-free visiblelight-driven photoredox oxidative annulation of arylamidines. J. Org. Chem.,81, 309–317.

41 Xiao, T., Li, L., Lin, G., Mao, Z.-W., and Zhou, L. (2014) Metal-freevisible-light induced cross-dehydrogenative coupling of tertiary amines withdiazo compounds. Org. Lett., 16, 4232–4235.

42 Sun, J.-G., Yang, H., Li, P., and Zhang, B. (2016) Metal-freevisible-light-mediated oxidative cross-coupling of thiols with P(O)H com-pounds using air as the oxidant. Org. Lett., 18, 5114–5117.

43 (a) Wilkinson, F., Helman, W.P., and Ross, A.B. (1995) Rate constants for thedecay and reactions of the lowest electronically excited singlet state of molec-ular oxygen in solution: an expanded and revised compilation. J. Phys. Chem.,24, 663–677; (b) Tanielian, C. and Wolff, C. (1995) Determination of theparameters controlling singlet oxygen production via oxygen and heavy-atomenhancement of triplet yields. J. Phys. Chem., 99, 9831–9837; (c) Turro, N.J.,Ramamurthy, V., and Scaiano, J.C. (2010) Modern Molecular Photochemistryof Organic Molecules, University Science Publishers, New York.

44 Pitre, S.P., McTiernan, C.D., Ismaili, H., and Scaiano, J.C. (2013) Mechanisticinsights and kinetic analysis for the oxidative hydroxylation of arylboronicacids by visible light photoredox catalysis: a metal-free alternative. J. Am.Chem. Soc., 135, 13286–13289.

45 Zou, Y.Q., Chem, J.R., Liu, X.P., Lu, L.Q., Davis, R.L., Jorgensen, K.A., andXiao, W.J. (2012) Highly efficient aerobic oxidative hydroxylation of aryl-boronic acids: photoredox catalysis using visible light. Angew. Chem. Int. Ed.,51, 784–788.

46 Pitre, S.P., McTiernan, C.D., Ismaili, H., and Scaiano, J.C. (2014) Metal-freephotocatalytic radical trifluoromethylation utilizing methylene blue and visiblelight irradiation. ACS Catal., 4, 2530–2535.

47 (a) Clavier, G. and Audebert, P. (2010) s-Tetrazines as building blocks for newfunctional molecules and molecular materials. Chem. Rev., 110, 3299–3314;(b) Saracoglu, N. (2007) Recent advances and applications in 1,2,4,5-tetrazinechemistry. Tetrahedron, 63, 4199–4236.

References 419

48 Samanta, S., Das, S., and Biswas, P. (2013) Photocatalysis by 3,6-disubstituted-s-tetrazine: visible-light driven metal-free green synthesis of 2-substitutedbenzimidazole and benzothiazole. J. Org. Chem., 78, 11184–11193.

49 Samanta, S. and Biswas, P. (2015) Metal free visible light driven oxidation ofalcohols to carbonyl derivatives using 3,6-di(pyridin-2-yl)-1,2,4,5-tetrazine(pytz) as catalyst. RSC Adv., 5, 84328–84333.

50 Park, J.H., Ko, K.C., Kim, E., Park, N., Ko, J.H., Ryu, D.H., Ahn, T.K., Lee, J.Y.,and Son, S.U. (2012) Photocatalysis by phenothiazine dyes: visible-light-drivenoxidative coupling of primary amines at ambient temperature. Org. Lett., 14,5502–5505.

421

15

Visible-Light Heterogeneous Catalysts for PhotocatalyticCO2 ReductionSanyasinaidu Boddu, S.T. Nishanthi, and Kamalakannan Kailasam

Institute of Nano Science & Technology, Habitat Centre, Sector 64, Phase X, 160062 Mohali, Punjab, India

15.1 Introduction

Carbon dioxide is a major part of the world’s greenhouse gas emissions. Accord-ing to the International Panel on Climate Change (IPCC) prediction, CO2 levelsin the atmosphere could reach up to 590 ppm by 2100, and the global meantemperature would rise by 1.9 ∘C, which may cause disastrous consequencessuch as ice melting at the Earth’s pole, fast rising sea level, and increasingprecipitation across the globe [1, 2]. Hence, there is a growing need to mitigateCO2 emissions for the sustainable development of human beings. Importantstrategies to mitigate CO2 emissions are capturing, storage, and converting CO2into simple C1/C2 fuels such as CO, CH4, HCOOH, HCHO, CH3OH, C2H5OH,and other hydrocarbon compounds. Several methods have been developed toreduce CO2 into useful chemicals, namely, thermochemical reduction [3–5],photocatalytic reduction [6–8], photoelectrochemical reduction [9, 10], andelectrocatalytic reduction [11, 12].

Reduction of CO2 is thermodynamically uphill as illustrated by its standardfree energy of formation (ΔG∘ =−394.359 kJ mol−1). Thus, it requires veryhigh energy to reduce CO2 into useful chemicals. Solar energy is an abundant,renewable form of energy. One of the best ways to convert CO2 into fuels isthrough photocatalysis in the presence of sunlight. Currently, the search forsustainable and stable photocatalytic systems for CO2 reduction by visiblelight is being actively pursued. It is also a rapidly developing research area asthis technology provides possible solutions to the environmental and energyproblems that we are facing today. Despite other methods, photocatalytic CO2reduction has many advantages as follows:

1) Reaction will be driven by inexhaustible solar energy.2) The reactants will be untreated water and CO2.3) It is carried out at room temperature and normal pressure.4) The hydrocarbon fuels converted from CO2 can decrease the current energy

demand.5) There will not be any generation of secondary pollutants.


422 15 Visible-Light Heterogeneous Catalysts for Photocatalytic CO2 Reduction

Honda and coworkers first reported the photocatalytic reduction of CO2on various semiconductors in 1979 [13]. Since then numerous efforts havebeen directed toward photocatalytic conversion of CO2 into useful chemicalsusing semiconductors as photocatalyst. There are several reviews, perspectiveson photocatalytic reduction of CO2, which discusses on possibilities, basicrequirements, design of photocatalyst, design strategies for reactors, and thechallenges ahead to achieve this task [6–10, 14–19].

15.2 Basic Principles of Photocatalytic CO2 Reduction

In semiconductor-based photocatalysis, three main steps are involved as shownin Figure 15.1. In the first step, electron–hole pairs are generated when a semicon-ductor photocatalyst is irradiated with a suitable light having energy equal to orgreater than the bandgap energy (Eg) of the semiconductor. In the second step,a fraction of generated electrons and holes migrate to the surface of the semi-conductor or a cocatalyst in contact with the semiconductor. Other fractions ofelectron–hole pairs recombine together and release the energy in the form ofheat or photons. In the third step, the generated electrons and holes involve inthe reduction and oxidation of the species adsorbed on catalyst surfaces, respec-tively. In case of photocatalytic reduction of CO2, electrons reduce CO2 into CO,HCOOH, CH3OH, or CH4, whereas holes oxidize H2O to O2. The reduction ofH2O is a competitive process to the reduction of CO2. The efficiency of photo-catalytic CO2 reduction depends on the efficiency of the light harvesting, chargeseparation, and the surface reaction.

Reduction

Oxidation

D

D+

CB

VB

Volumerecombination

3

hν

hν

1 4 3

+

+

++ –

+

–

–

–

–

+–

2

4

Sur

face

reco

mbi

natio

n

Surface

reco

mbinatio

n

A

A–

Figure 15.1 Photoinduced formation of an electron–hole pair in a semiconductorphotocatalyst with possible decay paths. A= electron acceptor, D= electron donor.(Reproduced with permission from Ref. [20]. Copyright 1995, The American Chemical Society.)

15.2 Basic Principles of Photocatalytic CO2 Reduction 423

15.2.1 Thermodynamic Favorability of the Reactions

The redox potentials (E0) for the reduction of CO2 into various products can beobtained from thermodynamic data and these values are given below [21, 22].

CO2 + e− → CO.−2 E0

redox = −1.90 VCO2 + 2H+ + 2e− → HCOOH E0

redox = −0.61 VCO2 + 2H+ + 2e− → CO + H2O E0

redox = −0.53 VCO2 + 4H+ + 4e− → HCHO + H2O E0

redox = −0.48 VCO2 + 6H+ + 6e− → CH3OH + H2O E0

redox = −0.38 VCO2 + 8H+ + 8e− → CH4 + 2H2O E0

redox = −0.24 V2H+ + 2e− → H2 E0

redox = −0.41 V

All the redox potential values given here are with reference to normal hydrogenelectrode (NHE) and at pH 7. From these values, one electron reduction of CO2 isthermodynamically highly unfavorable due to the high negative redox potentialof CO2/CO2 (−1.90 V vs NHE, at pH 7) but the proton-assisted multielectronreduction of CO2 reactions are much more favorable by considering the relativelylower redox potential (vs NHE, at pH 7).

Thermodynamically, the reduction of CO2 is possible only if the flat bandpositions of the semiconductor are suitable, that is, semiconductor should haveconduction band (CB) located at higher or more negative potential than thereduction potential of CO2 and valence bands (VB) located at lower or morepositive potential than the oxidation potential of H2O. VB and CB potential andbandgap energies of various semiconductor photocatalysts with respect to CO2reduction potentials have been given in Figure 15.2.

Extensive studies are present on UV-light-induced photocatalytic CO2reduction using large bandgap (>3 eV) semiconductors in order to achieve bothoxidizing and reducing power. But, sunlight (AM1.5G) contains 4% of ultraviolet

–2.0Fe2O3 WO3 TiO2 Si Cu2O TaON CdSe ZnO SnO2 CdS

CO2–/CO2 (–1.90 V)

2.1

eV2.6

eV

3.2

eV

1.1

eV2.4

eV 1.7

eV 3.3

eV3.6

eV

2.4

eV

2.0

–2.2

eV

HCOOH/CO2 (–0.61 V)HCHO/ CO2 (–0.48 V)CH3OH/ CO2 (–0.38 V)CH4/ CO2 (–0.24)

O2/ H2O (0.82 V)

–1.0

0.0

E0

red

ox /

V v

s. N

HE

1.0

2.0

3.0

Figure 15.2 Conduction and valence band potentials and bandgap energies of varioussemiconductors relative to the redox potentials of compounds involved in CO2 reduction atpH 7. (Reproduced with permission from Ref. [23]. Copyright 2012, The Royal Society ofChemistry.)


rays (𝜆< 400 nm), 53% of visible light (𝜆= 400–800 nm), and 43% of infrared rays(𝜆> 800 nm) [24]. For utilization of sunlight efficiently, one needs to developvisible-light-active photocatalysts. Several studies were going on for effectivesunlight utilization such as size distribution, dye sensitization, anion doping,and heterostructure formation. It should be noted that there are limited reportson the single semiconductor-based photocatalyst for CO2 reduction. In thischapter, we discuss the visible-light-induced photocatalytic CO2 reduction.

15.3 Inorganic Semiconductors

15.3.1 Metal Oxides

WO3 is known to be an excellent visible-light-active catalyst for water oxida-tion. But bulk form of WO3 is not active catalyst for CO2 reduction. However,nanosheets of WO3 become active catalyst for CO2 reduction [25]. The photo-generated electrons in the CB of bulk WO3 cannot be used for the reduction ofCO2 or H2O due to its lower band-edge position compared to reduction poten-tials of CO2 or H2O as shown in Figure 15.3. Commercially available microcrys-tals of WO3 do not show any CH4 evolution in the photocatalytic reduction ofCO2 under visible-light irradiation, whereas 2D WO3 nanosheets with thicknessof 4–5 nm show 16 μmol g−1 of CH4 evolution for 14 h with an evolution rate of1.14 μmol g−1 h−1. From commercial sample to nanosheets, UV–vis absorptionspectra show a blue shift and bandgap has been increased from 2.63 to 2.79 eV dueto quantum size effect in nanosheets. The CB edge position of commercial WO3and nanosheets of WO3 has been estimated to be 0.05, −0.42 V (vs NHE), respec-tively. Thus, the CB potential of commercial WO3 is more positive than CO2/CH4reduction potential (−0.24 V), whereas CB potential of WO3 nanosheets is morenegative than CO2/CH4 reduction potential. This makes WO3 nanosheets activephotocatalyst for CO2 reduction. Similar kind of results also has been observedby Xie et al. with WO3 having faceted cubes and rectangular sheet [26]. Theyfound that faceted cubes are not active but rectangular sheet like WO3 is activefor photocatalytic CO2 reduction with CH4 evolution rate of 0.34 μmol g−1 h−1.

–1.0

CommercialWO3

2.63 eV 2.71 eV 2.79 eV 2.79 eV

CO2/CH4 (–0.24 eV)

O2/H2O (0.82 eV)

CB 0.05 eV

VB 2.68 eV 2.61 eV2.37 eV 2.39 eV

–0.1 eV–0.42 eV –0.4 eV

Cubic-likeWO3

WO3nanosheet

Sheet-likeWO3

0

+1.0

+2.0

+3.0

E, V

vs.

NH

E (

pH

= 7

)

Figure 15.3 Bandgap and band-edge potentials of WO3 with different morphologies.(Reproduced with permission from Ref. [25]. Copyright 2012, The Royal Society of Chemistry.)

15.3 Inorganic Semiconductors 425

Composite of graphene–WO3 shows much better photocatalytic activity for CO2reduction to methane [27].

TiO2 is an active photocatalyst for CO2 reduction under UV-light irradiation.But it becomes active under visible light upon suitable modifications such as dop-ing with anions, cations, sensitization with dye, or other semiconductors [28–31].Ong et al. reported that CNT@Ni/TiO2 nanocomposites were active for the pho-toreduction of CO2 into CH4 under visible-light irradiation and the bandgap ofthe nanocomposites is 2.22 eV [32]. The photocatalytic studies show the high-est CH4 yield of 0.145 μmol g−1 h−1 using CNT@Ni/TiO2 nanocomposites com-pared to Ni/TiO2 and pure anatase TiO2 due to the synergistic combination ofthe CNTs and TiO2. TiO2 becomes visible-light active when doped with Ti3+ andthe bandgap narrow down to 2.9 eV [33]. It shows activity for photoreduction ofCO2 to CH4 under visible-light illumination. The activity increases by employing1 wt% Cu, Pd as cocatalyst, and it further enhances by using both these metalstogether as cocatalyst. Wang et al. synthesized ordered mesoporous Co-dopedTiO2 by multicomponent self-assembly process [34]. These materials are activefor photocatalytic reduction of CO2 to CO and CH4 under visible-light illumina-tion. The relative ratio of CO and CH4 highly depends on the amount of cobaltdoped in TiO2. Highest methane evolution rate was observed when Co:Ti ratiois 0.2 and it gives highest CO evolution when the ratio is 0.025. The photographsand catalytic activity of Co doped TiO2 are shown in Figure 15.4.

Ultrathin Bi2WO6 nanoplates with 9.5 nm thickness have been prepared in thepresence of oleylamine using a hydrothermal route by Zhou et al. [35] Nanoplatescan be seen in Figure 15.5a. These nanoplates have a bandgap of 2.75 eV withCB edge at −0.31 V (vs NHE), which is more negative than the CO2/CH4 redoxpotential (−0.24 V). These nanoplates could be able to produce 6 μmol of CH4 in5.5 h under visible-light irradiation with an evolution rate 1.1 μmol g−1 h−1. ButBi2WO6 prepared by solid-state synthesis method shows very poor photocat-alytic activity with CH4 evolution rate of 0.045 μmol g−1 h−1. The photocatalyticactivity of these materials has been given in Figure 15.5b.

Wang and coworkers reported that cobalt-based spinel oxides show good pho-tocatalytic activity for reduction of CO2 to CO in the presence of [Ru(bpy)3]Cl2 ⋅6H2O dye under visible-light irradiation [36]. Triethanolamine (TEOA) was usedas hole scavenger. They observed the generation of CO and H2 with generationrates of 0.45 and 0.13 μmol g−1 h−1. But after that, the generation rate decreasesand reaches a saturation point due to photo bleaching of Ru-dye after limitedcatalytic operations (Figure 15.6a). The authors also studied the effect of differ-ent wavelengths of light irradiation and found that the catalytic activity decreasesas wavelength increases (Figure 15.6b). Similar results also have been observedwith ZnCo2O4 and NiCo2O4 spinel oxides [37, 38].

Zhou and coworkers synthesized Na2V6O16⋅xH2O nanoribbons of 5 nm thick-ness and 500 μm length by hydrothermal method. The bandgap of these nanorib-bons has been estimated to be 1.93 eV with CB position being more negativethan reduction potential of CO2 to CH4 [39]. These nanoribbons show photo-catalytic CO2 reduction activity with methane evolution rate 0.008 μmol g−1 h−1

under visible-light irradiation and they exhibit 0.2 μmol g−1 h−1 methane evolu-tion rate upon coloading with 1 wt% Ru and 1 wt% Pt. This group also reported


OMT Co-OMT-1 Co-OMT-2 Co-OMT-3 Co-OMT-4

Co-OMT-5

(a)

(b)

0.30

CH4

CO

3.0

CO

evo

lution (μm

ol g

–1 h

–1)

2.5

2.0

1.5

1.0

0.5

0.0

0.25

0.20

0.15

0.10

CH

4 e

volu

tion (μm

ol g

–1 h

–1)

0.05

0.00

OM

TCo-

OM

T-2

Co-

OM

T-4

Co-

OM

T-7

Co-

OM

T-8

Au-O

MT

N-d

oped

TiO

2

NaO

H-W

O 3

C 3N 4

Co 3

O 4Co-OMT-6 Co-OMT-7 Co-OMT-8 Co3O4

Figure 15.4 (a) Photographs of TiO2 with different amount of Co-doping (b) methane andcarbon monoxide evolution rates by various catalysts under visible-light irradiation. Co:Ti ratiois 0, 0.002, 0.005, 0.01, 0.025, 0.05, 0.1, 0.15, 0.2 for OMT, Co-OMT-1, Co-OMT-2, Co-OMT-3,Co-OMT-4, Co-OMT-5, Co-OMT-6, Co-OMT-7, Co-OMT-8, respectively. (Reproduced withpermission from Ref. [34]. Copyright 2015, The Royal Society of Chemistry.)

Fe2V4O13 nanoribbons with a bandgap 1.78 eV [40]. These nanoribbons with Ptloading show photocatalytic CO2 reduction activity with methane formation rateof 0.5 μmol g−1 h−1 under visible-light irradiation.

BiVO4 is one of the well-known visible-light-active water oxidation catalysts.Liu et al. synthesized BiVO4 by a microwave-assisted hydrothermal methodusing cetyltrimethyl ammonium bromide (CTAB) or polyethylene glycol (PEG)as surfactant [41] They obtained two forms of BiVO4, namely, monoclinic and


100 nm

7

6

5

4

3

2

1

0

0

(a) (b)

1 2 3 4

Time (h)

Nanoplates

SSR

CH

4 (μm

ol g

–1)

5 6 7 8 9 10

Figure 15.5 (a) FESEM image of Bi2WO6 nanoplates and (b) CH4 generation over nanoplatesand the SSR sample as a function of visible-light irradiation times (𝜆> 420 nm). (Reproducedwith permission from Ref. [35]. Copyright 2011, The American Chemistry Society.)

40COH2

COH2

Pro

du

ce

d g

as (μm

ol)

Pro

du

ce

d g

as (μm

ol)

30

20

10

0

0

(a) (b)1 2

40

30

20

10

0

Time (h)

Ab

s. (a

.u.)

Ab

s. (a

.u.)

Wavelength (nm)

3 4 400 450 500 550 600

200 300 400 500

Wavelength (nm)

600 700 800

Figure 15.6 Photocatalytic CO2 reduction activity by MnCo2O4 (a) under visible-lightirradiation, (b) under different wavelengths of light irradiation (line spectrum indicateabsorption of Ru-dye and inset shows the DR UV–vis spectrum of MnCo2O4) (Reproduced withpermission from Ref. [36]. Copyright 2015, The American Chemical Society.)

tetragonal zircon type by choosing CTAB or PEG. The photocatalytic CO2 reduc-tion studies reveal that monoclinic form shows superior catalytic activity forthe production of ethanol than the tetragonal form. The ethanol formation ratesare 21.5 and 1.1 μmol h−1 for monoclinic and zircon type BiVO4, respectively,whereas these values are 406.6 and 4.9 μmol h−1 under full-arc irradiation.

Indium-based oxides are other notable ones, active for CO2 reduction. InTaO4and InNbO4 have optical bandgap of 2.6, 2.5 eV, respectively [42]. Both theseoxides have a wolframite structure, which contains corner-shared InO6 andTaO6/NbO6 octahedra arranged in a zigzag manner. Pan and Chen studied pho-tocatalytic CO2 reduction on InTaO4 under the visible-light illumination using500 W halogen lamp and KHCO3 was employed as an absorbent of CO2. Theobserved methanol formation rate was ∼1.0 μmol h−1 g−1 and the methanol for-mation rate was further increased to 1.39 μmol h−1 g−1 upon loading with 1 wt%NiO cocatalyst on the surface of NaTaO4 [43]. Tsai et al. reduced the bandgap ofthe InTaO4 to 2.28 eV by doping with nitrogen in order to get more absorption in


visible light [44]. Ni@NiO cocatalyst has been used to trap electrons in the nickelto provide catalytic centers on the surface of NiO. Photocatalytic CO2 reductionstudies performed in water medium in visible-light irradiation (390–770 nm)with 100 mW intensity. Methanol formation rates are in the following order:as-prepared InTaO4 < InTaO4—N<Ni@NiO/InTaO4—N. InNbO4 is also usedfor photocatalytic CO2 reduction to methanol using light in the wavelengthrange from 500 to 900 nm [45]. Methanol formation rate 1.5 μmol g−1 h−1 wasachieved, which shows that the addition of 1 wt% of NiO or Co3O4 as a cocatalystmarginally increased the yields.

15.3.2 Sulfides

Sulfides are known to have narrow bandgap compared to their oxide counterpartsdue to the contribution of sulfur 3p orbitals to the VB. There are many sulfidessuch as CdS, Bi2S3, Cu2S, CdIn2S4, and ZnIn2S4, which are reported for the pho-tocatalytic reduction of CO2. The major advantage of these sulfides is that theyhave narrow bandgap with very strong visible-light absorption. The major disad-vantage is that they are generally not stable under light illumination because ofthe oxidation of lattice sulfide (S2−) ions to elemental sulfur in the absence of oxy-gen and SO4

2− in the presence of oxygen by photogenerated holes [46]. In orderto prevent this photocorrosion, hole scavengers such as sulfite (SO3

2−), thiosul-fate (S2O3

2−), hypophosphite (H2PO2−) anions, tertiary amines (e.g., trimethy-

lamine), and alcohols (e.g., isopropanol) are added to the reaction mixture.CdS is one of the widely studied sulfide semiconductors for photocatalytic CO2

reduction. The bandgap of CdS is 2.4 eV. Eggins et al. reported the photocatalyticreduction of CO2 on the CdS suspension in water in the presence of tetramethy-lammonium chloride (TMACl) as an electron donor [47]. The authors observedthe formation of glyoxlate, acetate, formate, formaldehyde, and methanol. Thequantum efficiency and the relative ratio of these products are highly dependenton pH of the solution. Yanagida et al. prepared CdS nanoparticles with an averageparticle size of 4 nm in N ,N-dimethylformamide (DMF) [48]. These nanoparti-cles show a promising photocatalytic activity for the reduction of CO2 to CO inthe presence of an electron donor, triethylamine (TEA) under visible-light irra-diation. It shows a quantum yield of 9.8% at 405 nm. Kisch and Lutz studied thephotocatalytic bicarbonate reduction on silica-supported cadmium sulfide [49].In the presence of sodium sulfite scavenger, the silica-supported cadmium sul-fide catalyses the reduction of bicarbonate to formate, formaldehyde, and oxalate.Composite of montmorillonite and CdS nanoparticles were synthesized to pro-tect the material against photooxidation in an aqueous dispersion [50]. Hydrogen,methane, and CO were reported with an overall efficiency 4–8 times larger thanthe analogous system containing P25 particles.

Chen et al. synthesized Bi2S3 nanomaterials with different shapes such asnanoparticles, urchin-like spheres, and microspheres by template-free solvother-mal method [51]. Among these, microspheres show the highest activity followedby urchin-like sphere and nanoparticles for the photocatalytic reduction ofCO2 to methyl formate in methanol. This highest activity has been attributed to


their special hierarchical structure, good permeability, and high light-harvestingcapacity. ZnIn2S4 nanosheets with hexagonal and cubic structures have beensynthesized through liquid ultrasonic exfoliation method by Chen et al. [52]These nanosheets photoreduced CO2 to methyl formate in methanol. The activityof hexagonal ZnIn2S4 shows better activity than cubic ZnIn2S4. Both hexagonaland cubic ZnIn2S4 nanosheet show much higher activity than ZnIn2S4 micro-spheres prepared by the hydrothermal method. The methyl formate formationactivity after 4 h light illumination is as follows: Hexagonal ZnIn2S4 nanosheets(762 μmol g−1)> cubic ZnIn2S4 nanosheets (629 μmol g−1)>ZnIn2S4 micro-spheres (200 μmol g−1). In another study, Jiang et al. have synthesized CdIn2S4microspheres with different sulfur sources such as l-cysteine, thioacetamide, andthiourea [53]. Photocatalytic CO2 reduction to methyl formate in methanol hasbeen studied over these microspheres under visible light, and the methyl formateformation activities are as follows: l-cysteine (2857 μmol h−1 g−1), thioacetamide(3604 μmol h−1 g−1), thiourea (5258 μmol h−1 g−1). CuxAgyInzZnkSm solid solu-tions (solid solutions between ZnS, Cu2S, and AgInS2) were synthesized withRuO2 or Rh1.32Cr0.66O3 cocatalysts [54]. The bandgap of the solid solutiondepends on the composition. Photocatalytic CO2 reduction studies were con-ducted in the presence of Na2S sacrificial reagent under visible-light irradiation,and methanol was found as a final product with the highest formation rate of34.3 μmol g−1 h−1 for Cu0.12Ag0.30In0.38Zn1.22S2/RuO2⋅Cu2S/Pt nanorods withan average 30 nm in length and 5 nm in width has been synthesized throughcation exchange from CdS by Manzi et al. [55]. This nanorod acts as efficientcatalyst for photocatalytic reduction of CO2 to carbon monoxide and methaneand the formation rates are 3.02 and 0.13 μmol h−1 g−1, respectively. In this study,Na2CO3 was used as carbon source and Na2SO3 as a hole scavenger. A Cu2Snanorod with Pt nanoparticles at tip shows much better catalytic activity thanCu2S nanorods with Pt nanoparticle randomly distributed on their surface.

15.3.3 Oxynitrides

Oxynitrides are one of the important classes of semiconducting materials withnarrow bandgap. Several reports are available on oxynitrides for photocatalyticCO2 reduction [56–59]. Zhang et al. synthesized mesoporous zinc germaniumoxynitride by treating microporous Zn2GeO4 with ammonia at 800 ∘C [56].The bandgap of the resulting oxynitride increases from 2.59 to 2.79 eV withincreasing reaction time from 1 to 15 h. Samples treated for 10 h shows thehighest activity (2.7 ppm g−1 h−1) for photocatalytic CO2 conversion to CH4. Liuet al. synthesized zinc germanium oxynitride hyperbranched nanostructuresby nitriding Zn2GeO4 in ammonia atmosphere at 700 ∘C [57]. The CO2 to CH4conversion rate is dependent on the heating time under ammonia atmosphereand samples heated for 6 h show optimum conversion rate of 1.35 μmol g−1 h−1.Single-crystalline zinc gallium oxynitride (ZnGaNO) nanorods with opticalbandgap of 2.54 eV were synthesized by a molten salt ion-exchange route[58]. Carbon dioxide photoreduction into CH4 over ZnGaNO nanorods with0.5 wt% Pt as the cocatalyst in the presence of water vapor under visible-light


–2

–1

0

1

2

0.6 5

4

3

2

1

0

0.5

ZGNO-SSRZGNO-tubeZGNO-nanotubeZGNO-nanorod

UV–vis

Vis

0.4

0.3

Pro

ducts

yie

ld (μm

ol g

–1)

Pro

ducts

yie

ld (μm

ol g

–1)

0.2

0.1

0.00

(a)

(b) (c)2 4 6

Time (h) Time (h)8 10 0 2 4 6 8 10

Po

ten

tia

l /V

(vs. S

HE

at

PH

= 7

)

CB

–1.24V –1.26V

–0.24V

+0.82VO2/H2O

CH4

–1.28V –1.27V

+1.37V

2.6

1e

V

2.5

0e

V

2.5

8e

V

2.5

4e

V

+1.24V +1.23V +1.27V

ZGNO-SSR ZGNO-tube ZGNO-nanotube ZGNO-nanorod

VB

Figure 15.7 (a) The band structures of ZGNO-SSR, ZGNO-tube, ZGNO-nanotube andZGNO-nanorod at pH 7. The time course evolution of methane yields by CO2 photoreductionunder visible-light irradiation over ZGNO-nanotube, ZGNO-SSR, ZGNO-tube, and ZGNO-nanorod (b) a comparison of full-arc and visible-light irradiation over ZGNO-nanotube (c).(Reproduced with permission from Ref. [59]. Copyright 2016, The Royal Society of Chemistry.)

irradiation was studied and the CH4 evolution rate is 0.019 μmol h−1 g−1. Zhouet al. synthesized zinc gallium oxynitride (ZGNO) nanotubes via the Kirkendalleffect with ZnO nanorods and Ga2O3 nanosheets as precursors [59]. The bandstructure and methane evolution rates of different ZGNO materials are given inFigure 15.7.

15.4 Organic Semiconductors

15.4.1 Carbon Nitride and their Composites

During the last few years, CO2 photoreduction using g-CN has received increas-ing attention and an emerging research topic due to their high surface area, poresize distribution, and surface state that may offer some opportunities in this area.In 2012, Dong and Zhang [60] synthesized porous g-CN* by heating melamine ormelamine hydrochloride and evaluated the photocatalytic CO2 reduction activity

15.4 Organic Semiconductors 431

in the presence of water vapor under visible light. Under these reaction condi-tions, CO was obtained as the reduction product. In another work, mesoporousCN has been used as a catalyst for CO2 activation to CO after coupling with ben-zene oxidation [61].

*g-C3N4 is most commonly used to denote polymeric graphitic carbonnitrides, which is incorrect as always around 1–3% of hydrogen is left in thecarbon nitride structure depending on the temperature used during the thermalcondensation process, for example, at 550 ∘C, it is just polymeric melon [62].But for the reader’s understanding polymeric graphitic carbon nitrides will bedenoted as g-CN throughout chapter.

The use of g-CN for CO2 reduction is that couples organic basic functionality tophotocatalytic functionality and allows for activation/adsorption and reductionof CO2 [63–65]. In the previous system for CO2 to CO photocatalytic conversion,the quantum yield is less than 1 [66, 67]. In the continued detailed study by Maedaet al. [68], g-CN and a ruthenium complex as light-harvesting units and cat-alytic active sites, respectively, developed a promising heterogeneous system forthe reduction of CO2 into formic acid under visible-light irradiation by mergingorganometallic chemistry with polymer photocatalysis. By carefully optimizingthe heterogeneous catalyst and the reaction conditions, the apparent quantumyield (AQY) was remarkably enhanced to 5.7 with a high turnover number (TON)of >1000, which are the highest values for g-CN and are better than that of otherheterogeneous photocatalysts working with visible light.

Wang et al. [69] tuned the electronic structure of g-CN by doping S into thepristine g-CN and can enhance its optical adsorption as well as CO2 reductionactivity. S-doped g-CN was prepared by the condensation of thiourea and theCH3OH yield is 1.4 times more over the pristine g-CN. Gao et al. [70] investigatedthe single atom supported on g-CN, including Pd/g-CN and Pt/g-CN, for thephotocatalytic CO2 reduction into hydrocarbon fuels based on density functionaltheory (DFT) calculations. As evaluated by the reaction barriers, the preferredproduction of CO2 reduction on the Pd/g-CN catalyst was HCOOH with abarrier of 0.66 eV, while the Pt/g-CN catalyst was able to reduce CO2 to CH4 effi-ciently with a barrier of 1.16 eV. To construct more efficient systems, one strategyis to couple the highly active homogeneous catalyst with g-CN. For instance, aRu-complex, cis- and trans-[Ru{4,4′-(CH2PO3H2)2-2,2′-bipyridine}-(CO)2Cl2])(Ru) was adsorbed on the surface of mpg-CN with a high surface area of180 m2 g−1. The Ru/mpg-CN was able to reduce CO2 into formic acid under vis-ible light while a small amount of H2 and CO was also detected in acetonitrile inthe presence of TEOA as the sacrificial reagent [71, 72]. A schematic illustrationof photocatalytic CO2 reduction on the Ru/CN composite under visible-lightillumination is given in Figure 15.8. Isotopic measurement results indicatedthat formic acid entirely came from CO2 reduction while 77% of the evolvedCO originated from the carbonyl ligand unit of the Ru catalyst. Subsequently,the same group studied the effect of the pore-wall structure of mpg-CN andthe effect of the Ru-complex structure on CO2 photoreduction [73, 74]. With—PO3H2 used as the linker group, RuP/mpg-CN efficiently reduced CO2 toHCOOH under visible light in N ,N-dimethylacetamide with TEOA as the


Visible lightCB

VB

H2O3P

Tri-s-triazine (melem) unit

H2O3PCl

N

N

Cl

Ru

CO

CO

CO2

HCOOH

C3N4

D

D+

e–

h+

N

N

N

N

N

N N

N

N N

Figure 15.8 Schematic diagram of photocatalytic CO2 reduction of Ru/CN composite undervisible-light illumination. (Reproduced with permission from Ref. [71]. Copyright 2013, TheRoyal Society of Chemistry.)

sacrificial reagent and showed a high TON of greater than 1000 and a quantumyield (QE) of 5.7% at 400 nm.

In another study, Lin et al. [75] prepared a Co(bpy)3Cl2/g-CN hybrid mate-rial by self-assembly as the photocatalyst for the reduction of CO2 in acetonitrileunder visible light in the presence of TEOA. CO and H2 were the main products.A TON of 4.3 with a relatively high selectivity of 88.4% for CO production wasobtained by the optimized hybrid system. The surface of g-CN or mpg-CN wasalso modified with cobalt species as oxidative promoters to enhance CO2 pho-toreduction. Co(bpy)3Cl2/CoOx/mpg-CN gave the highest TON of 13 and theselectivity of CO to H2 was 78.5%. Under UV–vis light, CO2 can be reduced tohydrocarbons (mainly CH4, CH3OH, and HCHO) using a Pt/g-CN photocata-lyst [76]. Pt acts as an electron sink to enrich the surface of g-CN with electronsfor efficient CO2 reduction. The maximum yield can be obtained when the load-ing amount of Pt was 0.75%. Pt/g-CN was also prepared via a polyol process andused for photoreduction of CO2 in the presence of water vapor under day lightlamp irradiation [77]. CH4 was the main product for CO2 reduction and a 5.1-foldenhancement of CH4 production was obtained after 2% Pt was loaded on g-CN.

The composites of g-CN and metal oxides have been investigated by variousresearch groups for CO2 photoreduction. As in the case of photocatalytic watersplitting, coupling g-CN with a suitable semiconductor enhances the chargeseparation via band alignment, which leads to increased activity. Shi et al. [78]synthesized the composite of NaNbO3 nanowires and g-CN by annealing themixture of NaNbO3 nanowires and melamine at 520 ∘C in air. The suitableband alignment between NaNbO3 and g-CN facilitates the charge separationin the composite. After photodeposition of 0.5% Pt, the composite was capable


to reduce CO2 to CH4 and the activity was much higher than those of theindividual components loaded with Pt. Cao et al. [79] prepared an In2O3/g-CNphotocatalyst by a solvothermal method in dimethyl sulfoxide. In2O3/g-CNexhibited similar optical adsorption properties to the pristine g-CN but thetransient photoresponse showed an increased photocurrent for In2O3/g-CN.After loading Pt as an electron sink over 10% In2O3/g-CN, 159.2 ppm CH4 canbe evolved for 4 h. In another study, a g-CN/TiO2 heterojunction was preparedby an in situ growth method [80]. The surface area of the composite increasedwith the percentage of TiO2 in the composite. When the photoreduction of CO2was carried out with water vapor without a cocatalyst under UV–vis irradiation,CO was found to be the main product although a small amount of CH4 was alsoproduced.

A solvothermal process was used to grow Bi2WO6 in situ to form ag-CN/Bi2WO6 composite [81, 82]. The measured CB and VB positions ofg-CN and Bi2WO6 were used to explain the possible mechanism for the pho-toreduction of CO2 to CO. Compared to the pure Bi2WO6, the activity of thecomposite was significantly enhanced. ZnO, a wide bandgap semiconductor,was also used to make a composite with g-CN by an impregnation method forCO2 reduction. The charge separation and transportation were promoted by thesuitable band alignment between g-CN and ZnO, which leads to an enhancedactivity [83]. Besides oxides, carbon materials have been coupled with g-CNfor CO2 photoreduction. For example, a sandwich-like graphene/g-CN hybridnanostructure was fabricated using graphene oxide as a structure-directingagent [84]. The hybrid material shows enhanced activity for the conversion ofCO2 to CH4 in the presence of water vapor under a daylight lamp. The enhancedactivity was attributed to the improved electron transfer induced by graphene.

Wang et al. [65] coupled g-CN with a Co-containing zeolitic imidazole frame-work (Co-ZIF). Co-ZIF-9 has a high CO2 adsorption capacity of 2.7 mmol g−1 andaffords a high microporous surface area of 1607 m2 g−1. As a result, Co-ZIF-9 cancapture and concentrate CO2 in its pores. After the addition of electron mediator,bipyridine, the photoexcited electrons can be transferred from g-CN to Co-ZIF-9for CO2 reduction as revealed by photoluminescence (PL) quenching study. COwas the main product in this system and a QE of 0.9% can be obtained, evenwithout the loading of a cocatalyst. The native g-CN has a sheet-like structure,and the surface carries negative charges, which make this material suitable to beassembled with the positive layered double hydroxide (LDH) nanosheets [85, 86].Hong et al. [67] constructed the self-assembly of CN and LDH by electrostaticinteraction and reported the photocatalytic reduction of CO2 in the presence ofPd as a cocatalyst. The enriched CO2 in the form of interlayer CO3

2− in LDHcan be reduced more efficiently by the photogenerated electrons from g-CN atthe active site of the Pd cocatalyst. The highest QE of 0.093% was obtained at440 nm over the optimized Pd/LDH/g-CN assembly. The QE is still low but thisexample shows that the concept of coupling g-CN with a CO2 capturing materialis promising in photocatalytic CO2 reduction.

Nanostructure and surface engineering are also used to improve the pho-tocatalytic performance of g-CN [87–89]. Bulk MCN was exfoliated in waterresulting in monolayer surface hydroxyl group on g-CN [87]. Due to the


monolayer structure and surface hydroxyl, the surface area enhanced ∼3 timesand the CB/VB lifted ∼0.80 eV, resulting in increased activity of CO2 reduc-tion to CH4. Moreover, mixing with other organic blocks as the comonomercan also enhance the activity of g-CN. For instance, barbituric acid (BA),2-aminothiophene-3-carbonitrile (ATCN), 2-aminobenzonitrile (ABN), anddiaminomaleonitrile (DAMN) are mixed to fabricate modified carbon nitridederived from urea (CNU) [88]. The more porous and framework structure ofCNU–BA, resulting in improved optical absorption, reduced charge recombi-nation, and enhanced charge transfers, which shows ∼15 times better activityof CO2 reduction under visible light. Zheng et al. [89] synthesized helicalnanorod-like g-CN (HR-CN) based on chiral mesoporous silica (CMS) as thetemplate and cyanamide (CY) as the precursor. HR-CN with surface area of∼56 m2 g−1, which is ∼14 times than that of mesoporous carbon nitride (MCN),and it shows ∼22-fold enhancement compared to MCN in CO2 reductionto CO under visible-light irradiation. Mao et al. [90] synthesized g-CN fromurea (UCN) with thinner mesoporous flake-like structures and larger surfacearea than that synthesized from melamine (MCN), which results in ∼2.5 timesenhancement of AQY (from 0.08% up to 0.18%) in CO2-saturated NaOH solutionunder visible light.

15.4.2 Metal Organic Frameworks (MOFs)

MOFs, as a class of newly inorganic–organic hybrid porous materials wereapplied in CO2 reduction systems. They are not water tolerant due to theweak connection between metals and organic linkers, which must react inCO2-soluted organic solutions [91–93]. Moreover, poor stability has long beena major obstacle to the practical applications of MOFs even though it can bepartially overcome by using sacrificing reagents such as TEA and TEOA toreverse the electron deficiency for better stability. Some of the notable exampleswere discussed in detail. Lin et al. [94, 95] first demonstrated a MOF-basedheterogeneous catalytic system by doping different molecular complexes intoUiO-67(Zr) for water oxidation and photocatalytic CO2 reduction undervisible-light irradiation. Lin and coworkers reported a MOF photocatalystdoped with Re(bpy)(CO)3Cl complexes that reduced CO2 to CO under UV lightirradiation. Fu et al. [96] synthesized visible-light-sensitive NH2-MIL-125(Ti)with an amine-functionalized organic linker. This material reduced CO2 toHCOOH in the presence of TEOA under visible-light irradiation. In addi-tion, Li et al. [97] developed a nonporous coordination polymer consisting ofY metal ions and Ir(ppy)2(dcbpy) metalloligands and this material reduced CO2to HCOOH under visible-light irradiation.

In MOFs, the metal clusters are always regarded as reactive centers of CO2reduction, while the organic linkers are excited by irradiation and then providechannels for excited electrons migrating to the metal canters [98–104]. Theschematic representation of photocatalytic CO2 reduction over NH2-MIL-101(Fe) under visible-light irradiation is shown in Figure 15.9 [99]. Li et al.’s[100] reported that NH2-UIO-66(Zr) exhibits CO2 reduction activity in thepresence of TEOA as sacrificial agent under visible-light irradiation. This report


O

O O

O

TEOAHCOO–

Fe–Oclusters

CO2

H

H2N

CO2 HCOO–

H, e–e–

e –

e–

e–

hυ

hυ

Figure 15.9 Diagrammatic representation of photocatalytic CO2 reduction overNH2-MIL-101(Fe) under visible-light irradiation. (Reproduced with permission from Ref. [99].Copyright 2014, The American Chemical Society.)

explains that both metal clusters and organic linkers could be modified toimprove the reduction activity. Kang and Cohen group [101] reported MOFscontaining mixed metals (Zr/Ti) and mixed ligands show highly efficient androbust CO2 reduction to HCOOH under visible-light irradiation. In addition, itwas reported that a porphyrin-based MOF (PCN-222) can selectively captureand further photoreduce CO2 with high efficiency under visible-light irradiation[102]. To further overcome the obstacle of stability in the practical applicationsof MOFs, Luo and coworkers [103, 104] synthesized Ru–polypyridine-basedMOFs with noninterpenetrated and interpenetrated structures and they foundthat both showed CO2 reduction activity in acetonitrile solution. Moreover, theunique flower-like 3D hierarchical nanostructure not only highly improves thephotostability of Ru-MOF but also enhances the activity of this MOF materialfor visible-light-driven CO2 reduction.

15.4.3 Covalent Organic Frameworks

Covalent organic frameworks (COFs) are a class of polymer networks thatpossess long-range order due to structural regularity. In spite of their largecontribution in many applications including H2 fuel production by watersplitting, it is highly surprising that despite their enormous potential, COFshave hardly been explored for visible-light-active photocatalytic CO2 reduction.Very recently, Yadav et al. [105] developed a triazine-based covalent organicframework (2D-CTF) as an inexpensive and highly efficient visible-light-activeflexible film photocatalyst for solar fuel production from CO2. In this chapter,the condensation polymerization between cyanuric chloride and perylenedi-imide has been exploited for the first time as a new synthetic approach tothe construction of 2D-CTFs. The current study is a benchmark example ofCOF-based photocatalysts for solar fuel production from CO2 and is expectedto trigger further interest in potential solar energy conversion applications suchas wearable devices. The schematic diagram is shown in Figure 15.10.


Figure 15.10 Schematic illustration of CTF film photocatalyst-enzyme coupled systeminvolved in exclusive production of formic acid from CO2. Rhox= [Cp*Rh(bipy)H2O]2+,Rhred1 =Cp*Rh(bipy), Rhred2 = [Cp*Rh(bipy)H]+ ; Cp*=pentamethylcyclopentadienyl,bpy= 2,2′-bipyridine. (Reproduced with permission from Ref. [105]. Copyright 2016, The RoyalSociety of Chemistry.)

15.5 Semiconductor Heterojunctions

Semiconductor heterojunctions with appropriate band-edge positions arebelieved to be capable by enhancing the visible-light absorption and efficientseparation of electron–hole pairs thus by enhancing the photocatalytic per-formance. There are different types of semiconductor heterojunction based ontheir relative band-edge positions and bandgaps where photocatalytic activitycan be enhanced (Figure 15.11). By choosing suitable semiconductors, stabilityof narrow bandgap semiconductors can be enhanced against photocorrosion.

Often, wide bandgap semiconductors are combined with narrow bandgapsemiconductors in order to get sensitized with visible light as well as effi-cient e−–h+ separation. For example, TiO2—C3N4 [80], Cu2O—TiO2 [106],CdS—TiO2 [107], ZnO—C3N4 [83], ZnO—ZnTe [108], and SrTiO3—ZnTe[109] are reported in the recent literature. In all these heterojunctions, e−–h+

pairs are created in narrow bandgap semiconductors upon visible-light absorp-tion. The excited electrons will be transferred to the CB of the wide bandgapsemiconductor where they reduce the CO2. The holes will be remaining inthe VB of the narrow bandgap semiconductors and they oxidize water or holescavengers.

15.6 Conclusion and Perspectives 437

H2O

CO2 + H2O

(a) (b) (c)

CH4,CO, H2

e–e–

O2

h+

H2O

CO2 + H2O

CH4,CO, H2

e–e–

O2

h+h+

H2O

CO2 + H2O

CH4,CO, H2

e–

e–

O2h+ h+

Figure 15.11 Schematic illustrations of three kinds of charge-transfer mechanisms incomposite semiconductors (a) sensitization mechanism, (b) p–n junction mechanism,(c) Z-scheme mechanism. (Reproduced with permission from Ref. [10]. Copyright 2016,The Royal Society of Chemistry.)

In the case of p-n junction mechanism and Z-scheme mechanism, bothsemiconductors absorb the light and the CO2 reduces to solar fuels on one semi-conductor and oxidation of water or electron donor occurs on the othersemiconductor. Based on the relative band-edge positions, the p–n junctionmechanism or Z-scheme mechanism occurs and these two mechanisms com-plement each other. Generally, when the CB edge of one semiconductor andthe VB edge of the other semiconductor are close enough, Z-scheme typemechanistic pathway occurs. Whereas when energy gap is increasing, p-njunction mechanism becomes more prominent. There are several reports onthese both mechanisms, for example, SnO2−x/g-C3N4 [110], 𝛼-Fe2O3/Cu2O[111], CdS—WO3 [112], Ag3PO4/g-C3N4 [71], g-C3N4/Bi2WO6 [81, 82], andg-C3N4/NaNbO3 [78].

15.6 Conclusion and Perspectives

With increase in pollution affecting the humankind with severe climate changes,tapping the greenhouse gases such as CO2 became a huge challenge for theresearchers. As described in this chapter, photocatalytic reduction of CO2 is oneof the ways forward to synthesize solar fuels. Many classes of inorganic materialssuch as metal oxides, sulfides, oxynitrides have been studied for photocatalyticCO2 reduction. Activities of these materials have been enhanced by utilizingcocatalysts such as Ag, Au, Pt, Pd, Ru, NiO, Co3O4, and Ru. Composites of theseaforementioned materials with electron transfer agents (graphene, GO, andrGO) further improve the catalytic activity. Wide bandgap semiconductors aresensitized by dye molecules or narrow bandgap semiconductors to catalyze invisible light. Narrow bandgap semiconductors are very good absorbers of visiblelight but they suffer with fast e−–h + recombination. The recombination can betackled up to a certain level by designing semiconductor heterojunction wherethey can efficiently separate the e−–h+ pairs.

With the recent surge in graphitic carbon nitrides, different g-CN-based photo-catalysts were synthesized including organic–inorganic hybrids, metal-depositedg-CN, nanocomposites of g-CN with oxides or carbon materials, and compos-ites of g-CN with CO2-adsorbing materials. The photocatalytic activity of


g-CN-based photocatalysts depends on many factors and continuous efforts areneeded for the development of more efficient and stable photocatalysts for CO2reduction. With the emergence of recent reports on MOFs and COFs , the vastscope of tuning the structure and porosity through various functional groupscould also be appended to bind and reduce CO2 more effectively. Thus, there aremany opportunities in this field and advances are likely.

Especially, most of the semiconductors reported absorb only certain portion ofthe visible light, a variety of composite systems, sensitizers, and Z-scheme meth-ods are employed for entire visible-light absorption. Stability and efficiency arethe major issues for the large-scale utilization of the above-mentioned systems.Another way to enhance the performance is to increase the surface area of thesemiconductors with high CO2 absorption and subsequent photocatalytic reduc-tion of CO2 to fuels. Synthesis of higher order light hydrocarbons, though noteasy to achieve, target toward it provide a great potential for this photoreduc-tion process. Thus, the challenge lies in developing semiconductors with tunablebandgap and band positions for the photoreduction of CO2. With the emergingscenario, various oxynitrides, g-CN-based composites, MOFs and COFs couldbe the promising candidates for the sustainable utilization of CO2 directly fromCO2, water, and light.

References

1 Li, K.F., An, X.Q., Park, K.H., Khraisheh, M., and Tang, J.W. (2014) A criticalreview of CO2 photoconversion: catalysts and reactors. Catal. Today, 224,3–12.

2 Marszewski, M., Cao, S.W., Yu, J.G., and Jaroniec, M. (2015)Semiconductor-based photocatalytic CO2 conversion. Mater. Horiz., 2,261–278.

3 Dey, S., Naidu, B.S., and Rao, C.N.R. (2016) Beneficial effects of substitut-ing trivalent ions in the B-site of La0.5Sr0.5Mn1−xAxO3 (A=Al, Ga, Sc) onthe thermochemical generation of CO and H2from CO2 and H2O. DaltonTrans., 45, 2430–2435.

4 Dey, S., Naidu, B.S., and Rao, C.N.R. (2015) Ln0.5A0.5MnO3 (Ln= lanthanide,A=Ca, Sr) perovskites exhibiting remarkable performance in the thermo-chemical generation of CO and H2 from CO2 and H2O. Chem. Eur. J., 21,7077–7081.

5 Dey, S., Naidu, B.S., Govindaraj, A., and Rao, C.N.R. (2015) Noteworthyperformance of La1−xCaxMnO3 perovskites in generating H2 and CO by thethermochemical splitting of H2O and CO2. Phys. Chem. Chem. Phys., 17,122–125.

6 Li, K., Peng, B., and Peng, T. (2016) Recent advances in heterogeneousphotocatalytic CO2 conversion to solar fuels. ACS Catal., 6, 7485–7527.

7 Habisreutinger, S.N., Schmidt-Mende, L., and Stolarczyk, J.K. (2013) Pho-tocatalytic reduction of CO2 on TiO2 and other semiconductors. Angew.Chem. Int. Ed., 2013 (52), 7372–7408.

References 439

8 Roy, S.C., Varghese, O.K., Paulose, M., and Grimes, C.A. (2010) Towardsolar fuels: photocatalytic conversion of carbon dioxide to hydrocarbons.ACS Nano, 4, 1259–1278.

9 Kumar, B., Llorente, M., Froehlich, J., Dang, T., Sathrum, A., andKubiak, C.P. (2012) Photochemical and photoelectrochemical reductionof CO2. Annu. Rev. Phys. Chem., 63, 541–569.

10 Xie, S., Zhang, Q., Liu, G., and Wang, Y. (2016) Photocatalytic and pho-toelectrocatalytic reduction of CO2 using heterogeneous catalysts withcontrolled nanostructures. Chem. Commun., 52, 35–59.

11 Kondratenko, E.V., Mul, G., Baltrusaitis, J., Larrazabalc, G.O., andPerez-Ram𝚤rez, J. (2013) Status and perspectives of CO2 conversion intofuels and chemicals by catalytic, photocatalytic and electrocatalytic pro-cesses. Energy Environ. Sci., 6, 3112–3135.

12 Martín, A.J., Larrazábal, G.O., and Pérez-Ramírez, J. (2015) Towards sus-tainable fuels and chemicals through the electrochemical reduction of CO2:lessons from water electrolysis. Green Chem., 17, 5114–5130.


14 Hong, J., Zhang, W., Ren, J., and Xu, R. (2013) Photocatalytic reductionof CO2: a brief review on product analysis and systematic methods. Anal.Methods, 5, 1086–1097.


16 Wang, W.-H., Himeda, Y., Muckerman, J.T., Manbeck, G.F., and Fujita, E.(2015) CO2 hydrogenation to formate and methanol as an alternative tophoto- and electrochemical CO2 reduction. Chem. Rev., 115, 12936–12973.

17 Mao, J., Li, K., and Peng, T. (2013) Recent advances in the photocatalyticCO2 reduction over semiconductors. Catal. Sci. Technol., 3, 2481–2498.

18 Naval, S., Dhakshinamoorthy, A., Alvaro, M., and Garcia, H. (2013) Pho-tocatalytic CO2 reduction using non-titanium metal oxides and sulphides.ChemSusChem, 6, 562–577.

19 White, J.L., Baruch, M.F., Pander, J.E. III, Hu, Y., Fortmeyer, I.C., Park, J.E.,Zhang, T., Liao, K., Gu, J., Yan, Y., Shaw, T.W., Abelev, E., and Bocarsly, A.B.(2015) Light-driven heterogeneous reduction of carbon dioxide: photocata-lysts and photoelectrodes. Chem. Rev., 115, 12888–12935.

20 Linsebigler, A.L., Lu, G., and Yates, J.T. (1995) Photocatalysis on TiO2 sur-faces: principles, mechanisms, and selected results. Chem. Rev., 95, 735–758.

21 Surdhar, P.S., Mezyk, S.P., and Armstrong, D.A. (1989) Reduction poten-tial of the carboxyl radical anion in aqueous solutions. J. Phys. Chem., 93,3360–3363.

22 Qiao, J., Liu, Y., Hong, F., and Zhang, J. (2014) A review of catalysts for theelectroreduction of carbon dioxide to produce low-carbon fuels. Chem. Soc.Rev., 43, 631–675.

23 Tran, P.D., Wong, L.H., Barber, J., and Loo, J.S.C. (2012) Recent advancesin hybrid photocatalysts for solar fuel production. Energy Environ. Sci., 5,5902–5918.


24 Zhang, X.H., Peng, T.Y., and Song, S.S. (2016) Recent advances indye-sensitized semiconductor systems for photocatalytic hydrogen pro-duction. J. Mater. Chem. A, 4, 2365–2402.

25 Chen, X., Zhou, Y., Liu, Q., Li, Z., Liu, J., and Zou, Z. (2012) Ultrathin,single-crystal WO3 nanosheets by two-dimensional oriented attachmenttoward enhanced photocatalystic reduction of CO2 into hydrocarbon fuelsunder visible light. ACS Appl. Mater. Interfaces, 4, 3372–3377.

26 Xie, Y., Liu, G., Yin, L., and Cheng, H. (2012) Crystal facet-dependent pho-tocatalytic oxidation and reduction reactivity of monoclinic WO3 for solarenergy conversion. J. Mater. Chem., 22, 6746–6751.

27 Wang, P.-Q., Bai, Y., Luo, P.-Y., and Liu, J.-Y. (2013) Graphene–WO3nanobelt composite: elevated conduction band toward photocatalytic reduc-tion of CO2 into hydrocarbon fuels. Catal. Commun., 38, 82–85.

28 Phongamwong, T., Chareonpanicha, M., and Limtrakul, J. (2015) Role ofchlorophyll in Spirulina on photocatalytic activity of CO2 reduction undervisible light over modified N-doped TiO2 photocatalysts. Appl. Catal., B,168, 114–124.

29 Songa, G., Xina, F., Chena, J., and Yin, X. (2014) Photocatalytic reduction ofCO2 in cyclohexanol on CdS—TiO2 heterostructured photocatalyst. Appl.Catal., A, 473, 90–95.

30 Kumar, P., Joshi, C., Labhsetwar, N., Boukherroub, R., and Jain, S.L. (2015)A novel Ru/TiO2 hybrid nanocomposite catalysed photoreduction of CO2 tomethanol under visible light. Nanoscale, 7, 15258–15267.

31 Xu, H., Ouyang, S., Liu, L., Wang, D., Kako, T., and Ye, J. (2014)Porous-structured Cu2O/TiO2 nanojunction material toward efficient CO2photoreduction. Nanotechnology, 25, 165402.

32 Ong, W.-J., Gui, M.M., Chai, S.-P., and Mohamed, A.R. (2013) Direct growthof carbon nanotubes on Ni/TiO2 as next generation catalysts for photore-duction of CO2 to methane by water under visible light irradiation. RSCAdv., 3, 4505–4509.

33 Sasan, K., Zuo, F., Wang, Y., and Feng, P. (2015) Self-doped Ti3+—TiO2 as aphotocatalyst for the reduction of CO2 into a hydrocarbon fuel under visiblelight irradiation. Nanoscale, 7, 13369–13372.

34 Wang, T., Meng, X., Liu, G., Chang, K., Li, P., Kang, Q., Liu, L., Li, M.,Ouyang, S., and Ye, J. (2015) In situ synthesis of ordered mesoporousCo-doped TiO2 and its enhanced photocatalytic activity and selectivityfor the reduction of CO2. J. Mater. Chem. A, 3, 9491–9495.

35 Zhou, Y., Tian, Z., Zhao, Z., Liu, Q., Kou, J., Chen, X., Gao, J., Yan, S., andZou, Z. (2011) High-yield synthesis of ultrathin and uniform Bi2WO6 squarenanoplates benefitting from photocatalytic reduction of CO2 into renew-able hydrocarbon fuel under visible light. ACS Appl. Mater. Interfaces, 3,3594–3601.

36 Wang, S., Hou, Y., and Wang, X. (2015) Development of a stable MnCo2O4cocatalyst for photocatalytic CO2 reduction with visible light. ACS Appl.Mater. Interfaces, 7, 4327–4335.

37 Wang, S., Ding, Z., and Wang, X. (2015) A stable ZnCo2O4 cocatalyst forphotocatalytic CO2 reduction. Chem. Commun., 51, 1517–1519.

References 441

38 Wang, Z., Jiang, M., Qin, J., Zhou, H., and Ding, Z. (2015) Reinforced pho-tocatalytic reduction of CO2 to CO by a ternary metal oxide NiCo2O4. Phys.Chem. Chem. Phys., 17, 16040–16046.

39 Feng, S., Chen, X., Zhou, Y., Tu, W., Li, P., Li, H., and Zou, Z. (2014)Na2V6O16⋅xH2O nanoribbons: large-scale synthesis and visible-light pho-tocatalytic activity of CO2 into solar fuels. Nanoscale, 6, 1896–1900.

40 Li, P., Zhou, Y., Tu, W., Liu, Q., Yan, S., and Zou, Z. (2013) Direct growthof Fe2V4O13 nanoribbons on a stainless-steel mesh for visible-light photore-duction of CO2 into renewable hydrocarbon fuel and degradation of gaseousisopropyl alcohol. ChemPlusChem, 78, 274–278.

41 Liu, Y., Huang, B., Dai, Y., Zhang, X., Qin, X., Jiang, M., and Whangbo, M.H.(2009) Selective ethanol formation from photocatalytic reduction of carbondioxide in water with BiVO4 photocatalyst. Catal. Commun., 11, 210–213.

42 Ye, J., Zou, Z., Arakawa, H., Oshikiri, M., Shimoda, M., Matsushita, A., andShishido, T. (2002) Correlation of crystal and electronic structures withphotophysical properties of water splitting photocatalysts InMO4 (M=V5+,Nb5+, Ta5+). J. Photochem. Photobiol., A, 148, 79–83.

43 Pan, P.W. and Chen, Y.W. (2007) Photocatalytic reduction of carbon diox-ide on NiO/InTaO4 under visible light irradiation. Catal. Commun., 8,1546–1549.

44 Tsai, C.-W., Chen, H.M., Liu, R.-S., Asakura, K., and Chan, T.-S. (2011)Ni@NiO core–shell structure-modified nitrogen-doped InTaO4 forsolar-driven highly efficient CO2 reduction to methanol. J. Phys. Chem. C,115, 10180–10186.

45 Lee, D.-S., Chen, H.-J., and Chen, Y.-W. (2012) Photocatalytic reductionof carbon dioxide with water using InNbO4 catalyst with NiO and Co3O4cocatalysts. J. Phys. Chem. Solids, 73, 661–669.

46 Meissner, D., Memming, R., and Kastening, B. (1988) Photoelectrochemistryof cadmium sulfide. 1. Reanalysis of photocorrosion and flat-band potential.J. Phys. Chem., 92, 3476–3483.

47 Eggins, B.R., Irvine, J.T.S., Murphy, E.P., and Grimshaw, J. (1988) Formationof two-carbon acids from carbon dioxide by photoreduction on cadmiumsulphide. J. Chem. Soc., Chem. Commun., 1123–1124.

48 Yanagida, S., Kanemoto, M., Ishihara, K.-I., Wada, Y., Sakata, T., andMori, H. (1997) Visible-light induced photoreduction of CO2 with CdSnanocrystallites – importance of the morphology and surface structurescontrolled through solvation by N ,N-dimethyl formamide. Bull. Chem.Soc. Jpn., 70, 2063–2070.

49 Kisch, H. and Lutz, P. (2002) Photoreduction of bicarbonate catalyzed bysupported cadmium sulphide. Photochem. Photobiol. Sci., 1, 240–245.

50 Praus, P., Kozk, O., Koc, K., Pancek, A., and Dvorsky, R. (2011) CdSnanoparticles deposited on montmorillonite: preparation, characterizationand application for photoreduction of carbon dioxide. J. Colloid InterfaceSci., 360, 574–579.

51 Chen, J., Qin, S., Song, G., Xiang, T., Xin, F., and Yin, X. (2013)Shape-controlled solvothermal synthesis of Bi2S3 for photocatalytic


reduction of CO2 to methyl formate in methanol. Dalton Trans., 42,15133–15138.

52 Chen, J., Xin, F., Yin, X., Xiang, T., and Wang, Y. (2015) Synthesis of hexag-onal and cubic ZnIn2S4 nanosheets for the photocatalytic reduction of CO2with methanol. RSC Adv., 5, 3833–3839.

53 Jiang, W., Yin, X., Xin, F., Bi, Y., Liu, Y., and Li, X. (2014) Preparation ofCdIn2S4 microspheres and application for photocatalytic reduction of carbondioxide. Appl. Surf. Sci., 14 (288), 138–142.

54 Liu, J.-Y., Garg, B., and Ling, Y.-C. (2011) CuxAgyInzZnkSm solid solu-tions customized with RuO2 or Rh1.32Cr0.66O3 co-catalyst display visiblelight-driven catalytic activity for CO2 reduction to CH3OH. Green Chem.,13, 2029–2031.

55 Manzi, A., Simon, T., Sonnleitner, C., Döblinger, M., Wyrwich, R., Stern, O.,Stolarczyk, J.K., and Feldmann, J. (2015) Light-induced cation exchange forcopper sulfide based CO2 reduction. J. Am. Chem. Soc., 137, 14007–14010.

56 Zhang, N., Ouyang, S., Kako, T., and Ye, J. (2012) Mesoporous zinc ger-manium oxynitride for CO2 photoreduction under visible light. Chem.Commun., 48, 1269–1271.

57 Liu, Q., Xu, M., Zhou, B., Liu, R., Tao, F., and Mao, M. (2017) Unique zincgermanium oxynitride hyperbranched nanostructures with enhanced vis-ible light photocatalytic activity for CO2 reduction. Eur. J. Inorg. Chem.doi: 10.1002/ejic.201700044

58 Zhou, P., Yan, S.C., and Zou, Z.G. (2015) Molten salt ion exchange route toZnGaNO single crystal nanorods for improved CO2 photoreduction to CH4.CrystEngComm, 17, 992–995.

59 Zhou, P., Gao, H.L., Yan, S.C., and Zoua, Z.G. (2016) The Kirkendall effecttowards oxynitride nanotubes with improved visible light driven conversionof CO2 into CH4. Dalton Trans., 45, 3480–3485.

60 Dong, G. and Zhang, L. (2012) Porous structure dependent photoreac-tivity of graphitic carbon nitride under visible light. J. Mater. Chem., 22,1160–1166.

61 Goettmann, F., Thomas, A., and Antonietti, M. (2007) Metal-free activationof CO2 by mesoporous graphitic carbon nitride. Angew. Chem. Int. Ed., 46,2717–2720.

62 Bojdys, M.J., Müller, J.-O., Antonietti, M., and Tomas, A. (2008) Ionothermalsynthesis of crystalline, condensed, graphitic carbon nitride. Chem. Eur. J.,14, 8177–8182.

63 Zhang, G., Zhang, M., Ye, X., Qiu, X., Lin, S., and Wang, X. (2014) Iodinemodified carbon nitride semiconductors as visible light photocatalysts forhydrogen evolution. Adv. Mater., 26, 805–809.

64 Wang, X., Maeda, K., Thomas, A., Takanabe, K., Xin, G., Carlsson, J.M.,Domen, K., and Antonietti, M. (2009) A metal-free polymeric photocata-lyst for hydrogen production from water under visible light. Nat. Mater., 8,76–80.

65 Wang, S., Lin, J., and Wang, X. (2014) Semiconductor–redox catalysis pro-moted by metal–organic frameworks for CO2 reduction. Phys. Chem. Chem.Phys., 16, 14656–14660.

References 443

66 Niu, P., Yang, Y., Yu, J.C., Liu, G., and Cheng, H.M. (2014) Switching theselectivity of the photoreduction reaction of carbon dioxide by control-ling the band structure of a g-C3N4photocatalyst. Chem. Commun., 50,10837–10840.

67 Hong, J., Zhang, W., Wang, Y., Zhou, T., and Xu, R. (2014) Photocatalyticreduction of carbon dioxide over self-assembled carbon nitride and layereddouble hydroxide: the role of carbon dioxide enrichment. ChemCatChem, 6,2315–2321.

68 Kuriki, R., Sekizawa, K., Ishitani, O., and Meada, K. (2015)Visible-light-driven CO2 reduction with carbon nitride: enhancing theactivity of ruthenium catalysts. Angew. Chem. Int. Ed., 54, 2406–2409.

69 Wang, K., Li, Q., Liu, B., Cheng, B., Ho, W., and Yu, Y. (2015) Sulfur-dopedg-C3N4 with enhanced photocatalytic CO2-reduction performance. Appl.Catal., B, 176–177, 44–52.

70 Gao, G., Jiao, Y., Waclawik, E.R., and Du, A. (2016) Single atom (Pd/Pt)supported on graphitic carbon nitride as an efficient photocatalyst forvisible-light reduction of carbon dioxide. J. Am. Chem. Soc., 139, 6292–6297.

71 Maeda, K., Sekizawa, K., and Ishitani, O. (2013) A polymeric-semiconductor–metal-complex hybrid photocatalyst for visible-light CO2reduction. Chem. Commun., 49, 10127–10129.

72 Onga, W.J., Tana, L.L., Chaia, S.P., Yonga, S.T., and Mohamed, A.R. (2015)Surface charge modification via protonation of graphitic carbon nitride(g-C3N4)for electrostatic self-assembly construction of 2D/2D reducedgraphene oxide (rGO)/g-C3N4 nanostructures toward enhanced photocat-alytic reduction of carbondioxide to methane. Nano Energy, 13, 757.

73 Kuriki, R., Sekizawa, K., Ishitani, O., and Maeda, K. (2015)Visible-light-driven CO2 reduction with carbon nitride: enhancing theactivity of ruthenium catalysts. Angew. Chem., 127, 2436–2439.

74 Maeda, K., Kuriki, R., Zhang, M., Wang, X., and Ishitani, O. (2014) Theeffect of the pore-wall structure of carbonnitride on photocatalytic CO2reduction under visible light. J. Mater. Chem. A, 2, 15146–15151.

75 Lin, J., Pan, Z., and Wang, X. (2014) Photochemical reduction of CO2 bygraphitic carbon nitride polymers. ACS Sustainable Chem. Eng., 2, 353–358.

76 Yu, J., Wang, K., Xiao, W., and Cheng, B. (2014) Photocatalytic reduction ofCO2 into hydrocarbon solar fuels over g-C3N4—Pt nanocomposite photocat-alysts. Phys. Chem. Chem. Phys., 16, 11492–11501.

77 Ong, W., Tan, L., Chai, S., and Yong, S. (2015) Heterojunction engineer-ing of graphitic carbon nitride (g-C3N4) via Pt loading with improveddaylight-induced photocatalytic reduction of carbon dioxide to methane.Dalton Trans., 44, 1249–1257.

78 Shi, H., Chen, G., Zhang, C., and Zou, Z. (2014) Polymeric g-C3N4 coupledwith NaNbO3 nanowires toward enhanced photocatalytic reduction of CO2into renewable fuel. ACS Catal., 4, 3637–3643.

79 Cao, S., Liu, X., Yuan, Y., Zhang, Z., Liao, Y., Fang, J., Loo, S.C.J., Sum, T.C.,and Xue, C. (2014) Solar-to-fuels conversion over In2O3/g-C3N4 hybridphotocatalysts. Appl. Catal., B, 147, 940–946.


80 Zhou, S., Liu, Y., Li, J., Wang, Y., Jiang, G., Zhao, Z., Wang, D., Duan, A.,Liu, J., and Wei, Y. (2014) Facile in situ synthesis of graphitic carbon nitride(g-C3N4)-N-TiO2 heterojunction as an efficient photocatalyst for the selec-tive photoreduction of CO2 to CO. Appl. Catal., B, 158–159, 20–29.

81 Sun, Z., Yang, Z., Liu, H., Wang, H., and Wu, Z. (2014) Visible-light CO2photocatalytic reduction performance of ball-flower-like Bi2WO6 synthesizedwithout organic precursor: effect of post-calcination and water vapour. Appl.Surf. Sci., 315, 360–367.

82 Li, M., Zhang, L., Fan, X., Zhou, Y., Wu, M., and Shi, J. (2015) Highly selec-tive CO2 photoreduction to CO over g-C3N4/Bi2WO6 composites undervisible light. J. Mater. Chem. A, 3, 5189–5196.

83 He, Y., Wang, Y., Zhang, L., Teng, B., and Fan, M. (2015) High-efficiencyconversion of CO2 to fuel over ZnO/g-C3N4 photocatalyst. Appl. Catal., B,168–169, 1–8.

84 Ong, W., Tan, L., Chai, S., and Yong, S. (2015) Graphene oxide as astructure-directing agent for the two-dimensional interface engineeringof sandwich-like graphene–g-C3N4 hybrid nanostructures with enhancedvisible-light photoreduction of CO2 to methane. Chem. Commun., 51,858–861.

85 Zhang, Y.J., Thomas, A., Antonietti, M., and Wang, X.C. (2009) Activation ofcarbon nitride solids by protonation: morphology changes, enhanced ionicconductivity, and photoconduction experiments. J. Am. Chem. Soc., 131,50–51.

86 Zhang, X., Xie, X., Wang, H., Zhang, J., Pan, B., and Xie, Y. (2013) Enhancedphotoresponsive ultrathin graphitic-phase C3N4 nanosheets for bioimaging.J. Am. Chem. Soc., 135, 18–21.

87 Huang, Y., Wang, Y.J., Bi, Y.Q., Jin, J.R., Ehsan, M.F., Fu, M., and He, T.(2015) Preparation of 2D hydroxyl-rich carbon nitride nanosheets for photo-catalytic reduction of CO2. RSC Adv., 5, 33254–33261.

88 Qin, J.N., Wang, S.B., Ren, H., Hou, Y.D., and Wang, X.C. (2015) Photocat-alytic reduction of CO2 by graphitic carbon nitride polymers derived fromurea and barbituric acid. Appl. Catal., B, 179, 1–8.

89 Zheng, Y., Lin, L.H., Ye, X.J., Guo, F.S., and Wang, X.C. (2014) Helicalgraphitic carbon nitrides with photocatalytic and optical activities. Angew.Chem. Int. Ed., 53, 11926–11930.

90 Mao, J., Peng, T.Y., Zhang, X.H., Li, K., Ye, L.Q., and Zan, L. (2013) Effectof graphitic carbon nitride microstructures on the activity and selectivityof photocatalytic CO2 reduction under visible light. Catal. Sci. Technol., 3,1253–1260.

91 Wang, C.C., Zhang, Y.Q., Li, J., and Wang, P. (2015) Photocatalytic CO2reduction in metal–organic frameworks: a mini review. J. Mol. Struct., 1083,127–136.

92 Zhang, T. and Lin, W.B. (2014) Metal–organic frameworks for artificial pho-tosynthesis and photocatalysis. Chem. Soc. Rev., 43, 5982–5993.

93 Kumar, S., Wani, M.Y., Arranja, C.T., Silva, J.D.A.E., Avula, B., and Sobral, A.(2015) Porphyrins as nanoreactors in the carbon dioxide capture and conver-sion: a review. J. Mater. Chem. A, 3, 19615–19637.

References 445

94 Wang, C., Xie, Z., deKrafft, K.E., and Lin, W. (2011) Doping metal organicframeworks for water oxidation, carbon dioxide reduction, and organicphotocatalysis. J. Am. Chem. Soc., 133, 13445–13454.

95 Wang, J.L., Wang, C., and Lin, W. (2012) Metal–organic frameworks for lightharvesting and photocatalysis. ACS Catal., 2, 2630–2640.

96 Fu, Y., Sun, D., Chen, Y., Huang, R., Ding, Z., Fu, X., and Li, Z. (2012) Anamine-functionalized titanium metal–organic framework photocatalyst withvisible-light-induced activity for CO2 reduction. Angew. Chem. Int. Ed., 51,3364–3367.

97 Li, L., Zhang, S., Xu, L., Wang, J., Shi, L.X., Chen, Z.N., Hong, M., andLuo, J. (2014) Effective visible-light driven CO2 photoreduction via a promis-ing bifunctional iridium coordination polymer. Chem. Sci., 5, 3808–3813.

98 Sun, D.R., Fu, Y.H., Liu, W.J., Ye, L., Wang, D.K., Yang, L., Fu, X.Z., andLi, Z.H. (2013) Studies on photocatalytic CO2 reduction overNH2–Uio–66(Zr) and its derivatives: towards a better understanding ofphotocatalysis on metal-organic frameworks. Chem. Eur. J., 19, 14279–14285.

99 Wang, D.K., Huang, R.K., Liu, W.J., Sun, D.R., and Li, Z.H. (2014) Fe-basedMOFs for photocatalytic CO2 reduction. ACS Catal., 4, 4254–4260.

100 Sun, D.R., Liu, W.J., Fu, Y.H., Fang, Z.X., Sun, F.X., Fu, X.Z., Zhang, Y.F., andLi, Z. (2014) Noble metals can have different effects on photocatalysis overmetal–organic frameworks (MOFs): a case study on M/NH2-MIL-125(Ti)(M=Pt and Au). Chem. Eur. J., 20, 4780–4788.

101 Lee, Y., Kim, S., Kang, J.K., and Cohen, S.M. (2015) Photocatalytic CO2reduction by a mixed metal(Zr/Ti), mixed ligand metal–organic frameworkunder visible light irradiation. Chem. Commun., 51, 5735–5738.

102 Xu, H.Q., Hu, J.H., Wang, D.K., Li, Z.H., Zhang, Q., Luo, Y., Yu, S.H., andJiang, H.L. (2015) Visible–light photoreduction of CO2 in a metal–organicframework: boosting electron–hole separation via electron trap states. J. Am.Chem. Soc., 137, 13440–13443.

103 Zhang, S.Q., Li, L.N., Zhao, S.G., Sun, Z.H., and Luo, J.H. (2015) Construc-tion of interpenetrated ruthenium metal–organic frameworks as stablephotocatalysts for CO2 reduction. Inorg. Chem., 54, 8375–8379.

104 Zhang, S.Q., Li, L.N., Zhao, S.E., Sun, Z.H., Hong, M.C., and Luo, J.H.(2015) Hierarchical metal–organic framework nanoflowers for effective CO2transformation driven by visible light. J. Mater. Chem. A, 3, 15764–15768.

105 Yadav, R.K., Kumar, A., Park, N.J., Kong, K.J., and Baeg, J.O. (2016) Highlyefficient covalent organic framework film photocatalyst for selective solarfuel production from CO2. J. Mater. Chem. A, 4, 9413–9418.

106 Wang, J.Y., Ji, G.B., Liu, Y.S., Gondal, M.A., and Chang, X.F. (2014)Cu2O/TiO2 heterostructure nanotube arrays prepared by an electrode posi-tion method exhibiting enhanced photocatalytic activity for CO2 reductionto methanol. Catal. Commun., 46, 17–21.

107 Pan, X.Y. and Xu, Y.J. (2015) Graphene-templated bottom-up fabricationof ultra large binary CdS—TiO2 nanosheets for photocatalytic selectivereduction. J. Phys. Chem. C, 119, 7184–7194.


108 Ehsan, M.F. and He, T. (2015) In situ synthesis of ZnO/ZnTe common cationheterostructure and its visible-light photocatalytic reduction of CO2 intoCH4. Appl. Catal., B, 166–167, 345–352.

109 Ehsan, M.F., Ashiq, M.N., Bi, F., Bi, Y.Q., Palanisamy, S., and He, T. (2014)Preparation and characterization of SrTiO3—ZnTe nanocomposites for thevisible-light photoconversion of carbon dioxide to methane. RSC Adv., 4,48411–48418.

110 He, Y., Zhang, L., Fan, M., Wang, X., Walbridge, M., Nong, Q., Wu, Y., andZhao, L. (2015) Z-scheme SnO2−x/g-C3N4 composite as an efficient photo-catalyst for dye degradation and photocatalytic CO2 reduction. Sol. EnergyMater. Sol. Cells, 137, 175–184.

111 Wang, J.-C., Zhang, L., Fang, W.-X., Ren, J., Li, Y.-Y., Yao, H.-C., Wang, J.-S.,and Li, Z.-J. (2015) Enhanced photoreduction CO2 activity over directZ-scheme 𝛼-Fe2O3/Cu2O heterostructures under visible light irradiation.ACS Appl. Mater. Interfaces, 7, 8631–8639.

112 Jin, J., Yu, J., Guo, D., Cui, C., and Ho, W. (2015) A hierarchical Z-schemeCdS—WO3 photocatalyst with enhanced CO2 reduction activity. Small, 11,5262–5271.

447

Part IV

Mechanistic Studies of Visible Light Active Photocatalysis

449

16

Band-gap Engineering of Photocatalysts: SurfaceModification versus DopingEwa Kowalska1, Zhishun Wei1, and Marcin Janczarek1, 2

1Hokkaido University, Institute for Catalysis, N21 W10, 001-0021 Sapporo, Japan2Gdansk University of Technology, Department of Chemical Technology, Narutowicza 11/12, 80-233 Gdansk,Poland

16.1 Introduction

An application of metal oxide semiconductors as photocatalysts is dependent onthe character of their intrinsic properties such as a range of light absorption,electronic band structure, specific surface area, particle size, and morphology.A vision of photocatalyst with an efficient visible-light response mostly deter-mines the research in the material engineering field to find the most proper semi-conductor to realize this aim. Generally, it is difficult to find a metal oxide inits unchanged form to fulfill all prerequisites (price, photocatalytic efficiency,and stability) to obtain photocatalytic material with high applicability. There-fore, there is a necessity to modify metal oxides to adjust their intrinsic proper-ties toward defined requirements. The introduction of visible-light activity (whilemaintaining other valuable properties) to metal oxides has become a main aim ofsuch modification during the last decade.

Semiconductors, such as titanium(IV) oxide (TiO2), exhibit low photochemicalquantum yield due to their relatively high recombination rate of photogener-ated electron–hole pairs. In addition, the large band-gap of these semiconduc-tors results in limitation of light absorption to UV light, which accounts for avery short range of solar spectrum, resulting in very low solar energy conversion.These challenges serve as the motivation to engineer an environmentally benignand efficient solar photocatalytic material by modifying the surface-electronicstructure of metal oxides while retaining their advantageous catalytic proper-ties. Modification of the band-gap structure of metal oxide semiconductors forenhancement of visible-light response is also often connected with the improve-ment of photocatalytic activity under UV irradiation. In some cases, the mech-anism of both actions (under vis and UV) is shown by one complex drawing.However, for simplicity, only the mechanism under visible light (vis) will be dis-cussed in detail in this chapter. (In brief, under UV irradiation, charge carri-ers are formed, which either recombined [no reaction] or are transferred to the


450 16 Band-gap Engineering of Photocatalysts: Surface Modification versus Doping

surface of semiconductor where they react with adsorbed species [often withoxygen and water].) Usually, the modification aims to inhibit the charge carrierrecombination, but also to increase specific surface area, separation abilities, andstability.

Two types of modification could be considered, that is, surface modificationand bulk modification. Surface modification means modification of the surfaceof semiconductor, which was prepared earlier, whereas bulk modificationmeans modification during semiconductor synthesis. Usually, the latter is alsoconsidered as titania doping since anions or cations could easily replace atomsin the crystalline structure during semiconductor synthesis. However, it must bepointed out that bulk modification is not equivalent to doping since bulk modifi-cation could also result in a preferential synthesis of pure semiconductor crystalswith modifiers deposited on its surface. Interestingly, surface modificationcould also result in either surface-modified semiconductor (most often), dopedsemiconductor (e.g., formation of oxygen vacancies known as “self-doping”), oreven formation of new semiconductor due to crystal phase transformation, forexample, titania treated with lithium hydroxide solution and calcined at 700 ∘Ctransformed to Li4Ti5O12 (such examples, although very interesting, will not bediscussed in this chapter) [1].

The distinction between the terms “surface modification” and “doping” isnecessary. In many research publications, the word “doping” is used too fre-quently describing a majority of semiconductor modifications. In this chapter,for simplicity, “surface modification” will be used as description for the resul-tant structure, in which the surface of the semiconductor was modified (byeither bulk modification, surface modification, or physical operations such asgrinding and calcination), and thus the resultant material can be named as“surface-modified semiconductor.” Whereas “doping” will be used for nanos-tructures, in which semiconductor was either doped with other cations/anionsor “self-doped” (during synthesis or by other treatment operations). Moreover,some mixed nanostructures have been also reported, in which dual functionof one modifier (as dopant and surface modifiers) is proposed, for example,carbon-modified titania resulting in band-gap narrowing (doping) and enhancedelectron scavenging and transfer (carbon deposits on the surface).

In this chapter, the photocatalysis mechanisms of visible-light-active pho-tocatalysts will be discussed, in which activities arise from the changes in theband-gap structure of the semiconductor resulted from adsorbed modifiers(surface modification) or band-gap narrowing (doping). Heterojunction betweentwo semiconductors resulting in visible-light response will be also shortly pre-sented. Finally, hybrid nanostructures comprising different types of modificationwill be also discussed. Semiconductor doping can be realized by metal and non-metal ions, as a codoping of metal and/or nonmetal elements and by so-calledself-doping described as introduction of oxygen vacancies and reduced metalcations to metal oxide structure. Surface modification can mean modificationwith metals, nonmetals, organic compounds, inorganic compounds, as wellas other semiconductors (heterojunctions). Since the mechanism of action fortwo combined semiconductors is different, heterojunctions are presented in aseparate section.

16.2 Doping 451

It should be pointed out that selection of chemical compounds for testing ofphotocatalytic activity is very important. For example, dyes must not be useddue to sensitization of titania by them, which phenomenon is well known andeven commercially used for dye-sensitized solar cells (as shortly described inSection 3.3). Although there are many reports discussing this issue [2, 3], thereare still many studies that have used dyes for activity study due to simplicity andavailability of analysis (UV/vis spectrophotometry). Discussion on mechanismin such a case is more complex and often it is impossible to decide by whichpathway dyes are decomposed, that is, by (i) self-degradation during sensitizationor by (ii) photogenerated charge carriers (e−/h+) and/or formed reactive oxygenspecies (ROS). Therefore, use of dyes should be avoided for mechanism studiesand for proving the visible response of novel photocatalysts. The use of dyes isonly acceptable in two cases: (i) under UV irradiation for highly active photocat-alysts such as anatase-rich titania (e.g., P25), where content of sensitization in theoverall activity is negligible, and (ii) for purification of dyes containing wastew-ater (but in this case, drawing the conclusion on the mechanism of action andproving vis activity are impossible).

16.2 Doping

The doping of metal oxide materials (substitutional or interstitial) with metal ornonmetal elements is responsible for a change of the chemical nature in the solid,which is intrinsically connected with attractive and repulsive forces combinedinto a chemical bond. For example, a substitutional doping of anatase titania bycation or anion species modifies the chemical bonds of TiO6 octahedral, resultingin a local electronic density rearrangement. This fact influences the improvementin electronic properties described as band-gap engineering [4]. Recently, manytheoretical and experimental studies have been developed to investigate the metal(Fe, Cr, V, Mo, Re, Ru, Mn, Co, Rh, Bi, etc.) and nonmetal (N, S, B, C, F) dopingeffects on the electronic structure of TiO2 photocatalysts.

Based on an example of TiO2, its vis activity incorporated by doping can resultfrom (a) a lower shift of conduction band minimum (CBM), (b) a higher shiftof valence band maximum (VBM), and (c) impurity states in the band-gap, asshown in Figure 16.1. The following requirements were defined for the dopedphotocatalysts:i) The CBM should be higher than the H2/H2O level, while the VBM lower than

the O2/H2O level to ensure photoreduction and photo-oxidation activities forwater splitting, respectively.

ii) The states in the gap should be shallow or mixed, with the band states of TiO2enough to transfer photo- or thermally excited carriers to reactive sites at thephotocatalyst surface within their lifetime [5].

16.2.1 Metal Ion Doping

Regarding the metal doping process, the theoretical and experimental reportssuggest that a redshift of the band-gap occurs due to the insertion of a new


TiO2

CBM

VBM

Eg

(a) (b) (c)

Figure 16.1 Three schemes of the band-gap modifications for visible-light sensitization withlower shift of CBM (a), a higher shift of VBM (b), and impurity states (c). (Reprinted withpermission from Ref. [5]. Copyright 2014, The American Chemical Society.)

band closer to the CBM, improving the photocatalytic properties. First, themetal ion implantation method was successfully used for modification ofphysical–chemical properties of TiO2 with metal doping [6]. Ion implantationinto TiO2 revealed that various ions occupy substitutional sites by replacing Tiatoms. The replacement of Ti atoms depends on the size mismatch energy ofthe implanted ions [6, 7]. Anpo et al. prepared metal ion-doped titania photo-catalysts with visible-light photocatalytic activity [8–10]. They compared titaniadoped chemically with Cr by impregnation and physically by ion-implantation.The study of absorption bands of such prepared photocatalytic materials showedthat in case of chemically doped titania powders, no shift in the absorptionband was observed; however, a new absorption shoulder appeared at around420 nm, due to the formation of the impurity energy level within the band-gap,and its intensity increased with the amount of Cr ions chemically doped [8].For Cr-implanted titania, shifts in the absorption bands toward the visible-lightregion were observed. The method of doping influences the electronic propertiesof titania in completely different ways. The order of the effectiveness in theredshift was found to be V>Cr>Mn> Fe>Ni implanted ions [10]. Visible-lightphotocatalytic activity of Cr-implanted TiO2 has been reported, for example, forthe NO decomposition in the gas phase.

It was proposed that for the metal ion-implanted TiO2, the overlap of the con-duction band (CB) due to Ti(d) orbital of TiO2 and the metal(d) orbital of theimplanted metal ions can decrease the band-gap of TiO2 to enable vis absorption.Substitution of Ti ions with the isolated metal ions implanted into the lattice posi-tion of the bulk of TiO2 is the determining factor for the utilization of visible light[9]. Yen et al. prepared titania thin films modified by Fe plasma ion implantation[11]. Fe3+ ions were substituted into the original Ti4+ in the TiO2 lattice to formFe—O in the implanted layer, decreasing the band-gap energy and thus increas-ing the optical absorption range to vis (450–800 nm). The results indicated thatshallow impurity levels were created near the CB during the implantation pro-cess and that impurity states (Fe 3d) were formed below the Ti 3d states. Anothervis active iron-doped titania photocatalysts were prepared by the combinationof sol–gel process with hydrothermal treatment [12, 13]. Iron ions were mostlyincorporated into the anatase crystal lattice. It was found that Fe3+ doping content

16.2 Doping 453

(CB)

Ti 3d–2

hν

hν

Rh 4d

O2

CB

VB

AcH, H2O, D

AcH•+, OH•, D•+

TiO2

Rh3+ / Rh4+

–

+ +

–

Ti 3d

O 2p

–1

0

Pote

ntial/

V v

s. A

g/A

gN

O3

1

2

3

2,4-DCP

(a) (b)

OH•

OH–

O2

O–2

O 2p

(VB)

V t2g levele–

++

e–

Visible light

Visible light

X

X

O–2

Figure 16.2 Schematic diagram to illustrate the mechanism of (a) photocatalytic degradationof 2,4-dichlorophenolunder visible-light irradiation on V-doped titania (Reprinted withpermission from Ref. [14]. Copyright 2009, Elsevier.), and (b) light-induced electron transferprocesses presented for Rh—TiO2. (Kuncewicz, http://pubs.rsc.org/en/Content/ArticleHtml/2016/RA/c6ra09364g. Licensed under CC BY 3.0.)

decreased from the surface to the core. This distribution of dopants may be infavor of the interfacial charge transfer reactions. Fe3+ ions help the separation ofphotogenerated electrons and holes by trapping them temporarily and shallowly.Fe-doped titanium dioxide photocatalysts absorb and utilize visible light due tothe excitation of 3d electron of Fe3+ to the CB of TiO2 and the electron transfertransition between Fe3+ themselves. Similarly, the excitation behavior of V-dopedTiO2 under visible irradiation was related to V 3d orbital (Figure 16.2a) [14]. Dueto the fact that the t2g level of V 3d orbital is located a little below the CB edge ofTiO2, electrons can be excited from the valence band (VB) of TiO2 to the t2g levelof V 3d orbital under visible-light irradiation, and further migrate to adsorbedO2 to form O2

−. Simultaneously, holes migrate to the surface hydroxyl group toproduce hydroxyl radicals (∙OH). Owing to this mechanism, organic compoundscan be degraded under visible-light irradiation [14].

Kuncewicz and Ohtani prepared visible-light-active rutile titania dopedwith rhodium ions [15]. Low concentration of rhodium resulted in dopedTiO2 particles. The photosensitization based on two-step band-gap excitationwas proposed, in which excitation of the electrons from Rh3+ to the CB wasfollowed by the light-induced electron transfer from the VB to photogeneratedRh4+(Figure 16.2b). Moreover, the mechanism assured not only generation ofboth electrons and holes within titania bands but also recovery of photoactiveRh3+ ions during the photocatalytic process [15].

16.2.2 Nonmetal Ion Doping

Taking into consideration the problems with thermal stability of metal-dopedsemiconductors and introduction of additional recombination centers caused bymetal ion doping, and the associated rather high modification costs, in recentyears, a large number of nonmetal elements have been used as dopants to mod-ify the band-gap of TiO2. A redshift of the band-gap by modifying the VBM canbe realized by substitution of anion species for the doping rather than cationic


–2

B

C

S

hνVisible light

O2

O2(surf)

Adsorption

Activation

Radicals

formation

CB

VB

h+ h+

h+ h+

h+

h+h+

A B C E

D

F

Ov

Ni

Ns

e–

e– e– e–

e– e–

e–

e–

Mineralization

Toluene

H2O

H2O

H2O

CO2

+

OH·(surf)

OH–(surf)

–

Radicals

formation

Adsorption

N

O

F

P

pπ

TiO2Ti–4

–6

–8

Ato

mic

level (e

V)

–10

–12

(a) (b)Atomic states

p d

–

O2(surf)–

Figure 16.3 (a) Comparison of atomic p levels among anions. The band-gap of TiO2 is formedbetween O 2p𝜋 and Ti 3d states (Reprinted with permission from Ref. [5]. Copyright 2014, TheAmerican Chemical Society.) (b) Proposed band structure of N-doped doped TiO2 undervisible-light irradiation. (Reprinted with permission from Ref. [16]. Copyright 2009, Elsevier.)

metals. On the basis of the band structure of TiO2 where the VBM reflects thenonbonding p

𝜋state of O, it is possible to compare atomic p levels among anions,

as shown in Figure 16.3a [5]. Nonmetal doping mainly consists of N, C, F, B,and some other elements having an atomic radius similar to that of the O atom.Among them, nitrogen has attracted much attention and has been widely studied.For example, Dong et al. prepared N-doped titania by thermal decomposition ofthe mixture of titanium hydroxide and urea [16]. The as-prepared samples exhib-ited strong vis absorption due to nitrogen doping in the form of substitutional(N—Ti—O and Ti—O—N) and interstitial (𝜋* character NO) states, which were0.14 and 0.73 eV above the top of VB, respectively. Then, the CB electrons couldreduce O2 molecules to superoxide anions (O2−), which facilitated the forma-tion of oxidant species such as H2O2 and hydroxyl radicals (∙OH), as shown inFigure 16.3b.

The activities for the photocatalytic oxidation of organic compounds in airunder vis irradiation are strongly dependent on the N content in TiO2, whichis determined by the balance of the amount of doped N ions and O vacanciesinduced in N-doped TiO2. Controlling this balance is essential to achievinghigher photocatalytic activity under vis irradiation. Yamanaka and Morikawaanalyzed the charge separation and trapping dynamics for visible-light-activeN—TiO2 by femtosecond time-resolved diffuse reflectance spectroscopy (TDR)[17]. As shown in Figure 16.4a, the TDR spectrum for N-doped TiO2 afterUV (360 nm) light excitation revealed that the surface-trapped electrons andholes were generated immediately after excitation, similar to that for TiO2.The number of surface-trapped electrons for N-doped TiO2 decreased morerapidly than that for TiO2 due to deep trapping by additionally induced oxygenvacancies. The TDR spectrum for N-doped TiO2 after 450 nm light excitationclearly indicated the generation of charge carriers (Figure 16.4b). As comparedto 360 nm excitation, time evolution for 450 nm excitation showed a significant

16.2 Doping 455

Equilibrium

(a) (b)

Deep trapping ∼300 ps

N2p 2.48 eV N2p 2.48 eV

Vo 0.75 eV

Vo 1.18 eV

Vo 0.75 eV

Vo 1.18 eV

Surface Surface

hυ360 nm

hυ 450 nm Trap directly <1 ps

Ti3d

0 eV

Ti3d

0 eV

O2p

3.2 eV

O2p

3.2 eV

e–

h+

h+

h+

e– e–

e–

e–

e–

e–

e–

e–

CB

VB VB

CB

Figure 16.4 Spatial and energetic distribution of electrons and holes in N-doped TiO2 powderafter weak femtosecond laser excitation at (a) 360 nm, and (b) 450 nm. VO indicates an oxygenvacancy. (Reprinted with permission from Ref. [17]. Copyright 2012, The American ChemicalSociety.)

decrease in charge carriers just after excitation because of the deep trapping ofelectrons within 1 ps (nearly equivalent to time resolution of the detector).

16.2.3 Codoping

A large number of studies reported that metal oxide semiconductors codopedwith different elements could exhibit much higher photocatalytic activity thansingly doped photocatalysts because of the synergistic effect between the dopingelements promoting the vis absorption and facilitating the separation efficiencyof photogenerated electron–hole pairs.

For example, Samsudin and Abd Hamid prepared mono- and codoped titaniawith nitrogen and fluorine dopants [18]. The kind of anion (N and F) governsdifferent electronic structures of TiO2 and is determined by its chemical statebinding energy, impurities, defects, and VB shifting. Nevertheless, the change inthe electronic structure is dominantly due to the formation of midband states,which results in the narrowing of band-gap and thus improved visible-lightabsorption.

The N—B—TiO2 (red anatase TiO2), prepared by heating anatase TiO2 micro-spheres (with a predoped interstitial boron shell) in a gaseous NH3 atmosphere,exhibited high photoelectrochemical water splitting activity under visible light[19]. The UV/vis absorption spectra of red anatase TiO2 show an extendedabsorption edge up to 700 nm, due to a band-gap gradient, which varies from1.94 eV on the surface to 3.22 eV in the core by gradually elevating the VBmaximum (Figure 16.5a). This strategy of codoping approach by a predopedinterstitial boron gradient improved the solubility of substitutional nitrogen inTiO2 bulk without introducing the Ti3+ impurity level. The interstitial borondopant effectively weakened the surrounding Ti—O bonds to facilitate easiernitrogen substitution and increased the chemical stability of codoped TiO2.

The tridoped titania with N, F, and Ta, prepared using NH3 and TaF5 as pre-cursors, was efficient for phenol degradation under vis due to synergistic effects


B/N rich

Conduction band

Valence band (core)

3.22 eV1.94 eV

Oxygen

vacancy

Ti3+ 3d

1.94 eV

Valence band

(shell)

Valence band

(shell)

(a) (b)

(c)

B/N free

CB

VB

N doped TiO2

N 2pπ∗ N–O

Vis Vis

Ti3+ 3dOxygen

vacancy

VB

CBAdsorbed O2

Surface

O2–

H2O, –OH

·OH

Nf

e–

Nn

Ta 5d

N–F–Ta tridoped TiO2

N-T

a in

tera

ctio

n

N 2pπ∗ N–O

Vis VisVis

Vis

Figure 16.5 Schematic drawings of the band structure of (a) boron and nitrogen-doped redTiO2 depicting band-gap gradient, and (central and right) N doped (Reprinted with permissionfrom Ref. [19]. Copyright 2012, The Royal Society of Chemistry.) (b) and N—F—Ta tridoped(c) TiO2. (Reprinted with permission from Ref. [20]. Copyright 2008, Elsevier.)

such as increased surface area, increased hydroxyl radical generation, and highvis absorption [20]. EPR and XPS studies revealed that N—Ta interaction inducedcharge compensation to form fully occupied continuum like the N 2p—Ta 5dhybridized states in VB edge, which promotes vis absorption and improves chargecarrier separation (Figure 16.5b). The fluorine dopant facilitated nitrogen incor-poration, which promoted the formation of N 2p—Ta 5d hybridized states. Thecoexistence of tantalum and nitrogen narrowed the band-gap by a charge transferfrom the positively charged tantalum to the negatively charged nitrogen, leadingto the formation of a stable N—Ta chemical bond.

Similarly, the photocatalytic activity of TiO2 codoped with Fe3+ and Ho3+ ionswas significantly improved due to the cooperative actions of the two dopants.Fe-doping broadened the absorption profile, improving the visible light utiliza-tion, and thus generating more electron–hole pairs. Ho-doping restrained theincrease in grain size (due to crystal expansion and matrix distortion), createdshallow energy states in the bottom of the CB, and retarded the recombinationof the photoexcited charge carriers [21].

16.2 Doping 457

16.2.4 Self-Doping

Different from impurity doping, self-doping that produces Ti3+ species in TiO2has also been proven to be an effective way for extending the optical absorptionof TiO2 toward the vis region. If a Ti4+ ion is changed to Ti3+ ion, then the localelectrostatic balance will be broken, and an oxygen vacancy will be introducedbecause of charge compensation [22]. It was believed that the introduced local-ized oxygen vacancy states with energies 0.75–1.18 eV below the CBM of TiO2are lower than the redox potential for hydrogen evolution, which, in combinationwith the low electron mobility in the bulk region due to this localization, makesthe photocatalytic activity of the reduced TiO2 negligible [23, 24]. However, the-oretical calculations showed that a high vacancy concentration could induce avacancy band of electronic states just below the CB and this statement was provedexperimentally by the preparation of reduced titania with improved vis activity[25, 26]. Chen et al. found that the hydrogenated black TiO2 nanoparticles (NPs)possessed high photocatalytic activities in oxidation of organic pollutants (e.g.,phenol) and hydrogen generation from a water–methanol solution [27]. Li et al.prepared the electrochemically reduced black anatase nanotubes with improvedphotocatalytic activity for decomposition of rhodamine [28].

Ti3+ ions and oxygen vacancies can cause visible-light absorption, due to theinduced mid-gap states within the band-gap of TiO2, resulting in the change ofpowder color from white to gray, yellow, blue, or black [22, 29]. These Ti3+ defects(abbreviated here as Vo—Ti3+) are occupied states and usually act as electrondonors, and the relative position of energy band of Ti3+ self-doped TiO2-x ispresented in Figure 16.6. Meanwhile, the electrons in the VB can be excited tothese unoccupied sites. Therefore, electronic transitions from Vo—Ti3+ localized

O2–

Ti4+

O 2

O 2

O 2

Ti3+

e–

e–

e–e–

e–

e–H+

CB

VB

h+

h+

H+

MB

CO2

H2O

H2O

·OH

h+

Vo–Ti3+

MB

H2OCO2·O2

–

H2

·OH

O2

Vo

Figure 16.6 Illustration of the energy bands for Ti3+ self-doped TiO2−x with high oxygenvacancy concentration and the photoinduced charge transfer processes. (Reprinted withpermission from Ref. [30]. Copyright 2014, The Royal Society of Chemistry.)


states to the CB and from the VB to Vo—Ti3+ are responsible for visible-lightabsorption [30].

16.3 Surface Modification

Surface modification is the process of modifying the surface of the material bybringing physical, chemical, or biological characteristics different from the onesoriginally found on the surface of a material. Various compounds could be used assurface modifiers, for example, metals (in the form of metallic NPs [zero-valentmetal] or chemical compounds), nonmetals (adsorbed anions or chemical com-pounds), and organic compounds (colorless and color such as dyes). Finally, het-erojunctions between semiconductors should be considered as another type ofsurface modification. All of these examples will be shortly presented in the fol-lowing sections.

16.3.1 Metals

Surface modification with metals includes modification with metal ions, com-pounds, and metallic NPs (zero-valent). One of the first studies showing visactivity of titania modified with metal compounds was presented by Macyk andKisch for platinum(IV) chloride as surface modifier [31, 32]. The proposed mech-anism (similar to sensitization of titania by dyes, presented in Section 16.3.3) isshown in Figure 16.7a. The excited platinum complex by visible light undergoeshemolytic cleavage of platinum–chloride bond affording the Pt(III) intermediateand a surface-bound chlorine atom. Then, electron transfer from the former tothe CB of titania and from the electron donor (here 4-chlorophenol, ArOH)to the chlorine atom regenerates the sensitizer. Reduction of adsorbed oxygenthrough several steps leads to hydroxyl radical.

A different mechanism was proposed for titania modified with [Fe(acac)3] by apartial ligand exchanged between acetylacetonate and surface hydroxyl groups,where the organic part was oxidized by postheating to prepare iron oxide clusters

CBLnPtIII …Cl0 ArOH

–0.48 –3.96

–6.84

–7.16

2.40

Vacant

surface

d sub-band

Surface

d sub-band

h+

Oxidation

Organic

pollutants

TiO2

p3

p4

p5 p2

p1

Vis

UV

e–

h+

e–

cb

vb

O2–

O2

2.72E/V

vs. S

HE

at pH

5.9

E/e

V v

s. vacuum

ArO•

–H+

LnPtIV – Cl

hν

VB

–

O2/O2–

•OH/OH–

(a) (b)TiO2

–

Figure 16.7 (a) Titania sensitization by platinum(IV) chloride complexes (Reprinted withpermission from Ref. [32]. Copyright 2003, The Royal Society of Chemistry.), (b) energy banddiagram for (FeOx)m/TiO2, where the position of the vacant d levels is assumed to be close tothat of iron doped into rutile TiO2. (Reprinted with permission from Ref. [33]. Copyright 2011,John Wiley & Sons.)

16.3 Surface Modification 459

on the titania surface (FeOx/TiO2), for which an energy-band diagram is shownin Figure 16.7b [33]. The electronic excitation from the surface d subband to theCB of titania (p1) results in strong absorption of visible light. The holes generatedin the surface d subband take part in the oxidation part (p2), while the excitedelectrons efficiently reduce oxygen (p3) (p4–p5 illustrate UV excitation).

Noble metals (Pt, Au, Ag, Ir, Pd) in the form of either adsorbed complexes[34–36] or metallic deposits [36–39] have been used for improvement of pho-tocatalytic activity of semiconductors for more than 40 years. Enhancement ofphotoactivity (under UV irradiation) originates from prolongation of lifetime ofcharge carriers (photogenerated electrons and holes) [36], since the noble metal(NM) serves as an electron sink, thus accelerating the transfer of electrons fromtitania to substrates, for example, protons to evolve hydrogen [40, 41].

Very recently another property of NM, that is, visible-light absorption due toplasmon resonance, has been used for activation of wide band-gap semiconduc-tors toward visible light, that is, mainly TiO2 [42, 43], but also other materialssuch as CeO2 [44, 45], Fe2O3 [46], ZnO [47], and KNbO3 [48]. Although plas-monic properties of noble metals were observed 100 years ago, explained morethan 30 years ago, and commercially used in many fields (e.g., surface enhancedRaman spectroscopy [SERS] [49, 50], medicine [50], optical data storage [51]), theexamination of their use for photocatalysis started about 10 years ago. Despite thenovelty of plasmonic photocatalysis, a large number of studies have already beenperformed to improve photoactivity and stability as well as to clarify the mech-anism under visible-light irradiation; a few reviews on plasmonic photocatalysishave also been published [52–56].

Applications of localized surface plasmon resonance (LSPR) to photocatalysisstarted at the beginning of this century, but plasmonic features were only usedfor characterization of gold NPs deposited on titania, that is, their formation,properties (size and shape), and stability (under UV irradiation) [57]. Finally,gold-modified titania was used as visible-light-responsive photocatalysis fordecomposition of MTBE (methyl tert-butyl ether) [58]. The first reports directlyshowing that plasmon resonance was responsible for the visible-light activitywere presented by action spectrum (AS) analyses for photocurrent generation[42] and for photo-oxidation of 2-propanol [43]. In these studies, AS resembledrespective absorption spectra, thus proving that plasmonic resonance of gold isresponsible for activity of Au/TiO2 under visible-light irradiation.

Three main mechanisms are proposed for plasmonic photocatalysis under vis-ible light:

i) Charge transfer (mainly electron transfer)ii) Energy transfer

iii) Plasmonic heating (thermal activation).

The two first mechanisms can be classified as “plasmon-assisted photocatal-ysis” since energy/electron transfer happens directly under visible irradiation,while the last one (iii) can be classified as “plasmon-assisted catalysis” since LSPRactivates catalytic (thermal) reaction, as shortly discussed further.

Electron transfer from Au NPs to the CB of titania (mechanism i) was pro-posed by Tian and Tatsuma in 2005 for photocurrent generation on Au/TiO2 [42].


Electron transfer

Energy transfer

(a) (b)

(c) (d)

CB

Oxidation products

520–

570

nm

450

nm

650

nm50

0 nm

Vis

ible

ligh

t

irrad

iatio

n

Transverse

longitudinal

LSPR

Adsorbed O2

CB

OCse–

e–

–

–– –

–

––––

+++ +

–

+ + +

– –

+++ +

EF

VBhν

hνhνVB

CB CB

VB

EF

LE

MF

RE

TEF

e–

e–e

h+ h+

e

h+

e

O2–

Figure 16.8 Electron transfer (a, b) and energy transfer (c, d) mechanisms for plasmonicphotocatalysts; (b) oxidative decomposition of organic compounds (OCs) by electron transfermechanism for titania modified with NPs/NRs of gold with different sizes. (Reprinted withpermission from Ref. [60]. Copyright 2010, The Royal Society of Chemistry.)

In general, the mechanism of decomposition of organic compounds (OCs) onAu/TiO2 under visible-light irradiation is similar to that of activation of sensi-tizers, such as metal complexes or dyes (discussed in Section 16.3.3) [32, 43, 59,60], and thus NM is also called “plasmonic photosensitizer.” [61] At first, inci-dent photons are absorbed by NM NPs through their LSPR excitation, and thenan electron (“hot electron”) is transferred from NM NPs into the CB of titania(Figure 16.8a,b). Then, the electron reduces molecular oxygen adsorbed on thetitania surface and the resultant electron-deficient NM NPs can oxidize OCs torecover its original metallic state (Figure 16.8b) [43, 60].

There are plenty of indirect and direct proofs for electron transfer mechanismin the literature, which involve different types of experiments, for example,

i) femtosecond transient absorption spectroscopy with an IR probe of inter-band absorption of electrons injected from gold nanodots into the CB oftitania [62, 63];

ii) comparison of the photocatalytic activity of gold-modified semiconductor(Au/TiO2) to gold-modified insulator (Au/ZrO2 and Au/SiO2) for the oxi-dation of OCs under visible-light irradiation [60, 64, 65];

iii) shift of electrode potential (negative or positive photopotential) and gener-ation of anodic or cathodic photocurrent depending on electrode configu-ration: ITO/TiO2/Au or ITO/Au/TiO2, respectively [66, 67];


iv) different mechanisms of OCs oxidation under UV and visible-light irradi-ation, that is, generation of electron–hole pairs and only electrons, respec-tively [68];

v) inhibition of vis activity in the presence of other molecules on thesurface of gold or between gold and titania (dodecanethiol [69],3-mercaptopropyltrimethoxysilane [60]);

vi) dissolution of metal due to its partial oxidation [70];vii) EPR experiments on Au/TiO2 photocatalyst resulting in detection of differ-

ent species under irradiation with UV and visible light [71–73];viii) time-domain density functional theory proving the electron injection into

TiO2 by the nonadiabatic mechanism due to high density of acceptorstates [74].

Besides “hot” electrons, participation of “hot” holes is also proposed in themechanism of photocurrent generation for gold NPs deposited into porous tita-nia film [69]. However, the origin of “hot” holes is not due to plasmonic excita-tion of gold NPs, but interband d–sp transition within the gold NPs. Therefore,“hot” electrons and “hot” holes are responsible for photocurrent generation atdifferent excitation wavelengths, that is, at plasmonic resonance range (about550–650 nm) and at shorter wavelengths of about 410–440 nm, respectively.

Modified mechanism of electron transfer without participation of semiconduc-tor (or with its negligible contribution) has been also reported. For example, “hot”electrons are responsible for dissociation of H2 under visible-light irradiation ofgold NPs deposited on titania [75]. However, titania practically does not partic-ipate in the mechanism of electron transfer (only “a small, additional negativecharge” transfer from titania to gold, due to oxygen vacancies). It is proposed that“hot” electrons with energy between the vacuum level and work function of metalcould be transferred into the resonance of the hydrogen molecule adsorbed onthe surface of gold NPs, and causes its dissociation. Similarly, the electron trans-fer is proposed for unsupported gold NPs in the presence of hydrogen peroxideas electron acceptor for selective oxidation of organic alcohols [76].

Energy transfer (mechanism ii) between two different compounds can takeplace when they have closely matched energy levels. This is not expected forAu/TiO2 since the plasmon energy of gold NPs (LSPR of about 2.2 eV) is lowerthan the band-gap of titania (about 3 eV). Therefore, the first reports suggestingenergy transfer from gold NPs to titania have been mentioned for premodifiedtitania in photocatalysts order to make it able to absorb visible light, thusbeing active under vis irradiation. For example, energy transfer (also calledplasmon energy transfer [PRET]) [77] was shown for photocurrent generationon 5-nm gold NPs on a titania film premodified with nitrogen and fluorine[78]. The photocurrent enhancement was attributed to the energy transferbetween two matching energy levels of F-/N-titania and Au. Other premodifiedtitania, for example, with nitrogen [79], and narrow band-gap semiconductorspossessing activity under visible-light irradiation like CuWO4 (2.0–2.5 eV) [77],show enhanced photocatalytic activity after addition of gold NPs, which isalso attributed to PRET mechanism. Interestingly, titania possessing crystallinedefects [80] and amorphous titania with disorders resulting in localized


electronic states inside the band-gap [81] (i.e., electron traps, ETs) are also goodmaterials for energy transfer due to energy matching of these states and LSPRof gold.

Cushing et al. deepened the concept of energy transfer and proposed two typesof mechanisms for energy transfer, that is, local electromagnetic field (LEMF) andresonant energy transfer (RET), as shown in Figure 16.8c,d [82]. LEMF appliesonly to the case of energy matching between gold NPs and semiconductor (asdiscussed above), whereas RET can also create electron–hole pairs in the semi-conductors with an energy lower than the band-gap energy (novel concept), dueto nonradiative dipole–dipole energy transfer.

The first report showing plasmon-assisted catalysis (mechanism iii) waspublished by Chen et al. in 2008, where it was presented that heated gold NPscould activate organic molecules to induce their oxidation by visible light atambient temperature [83]. Gold was deposited on different supports, that is,mainly on insulating oxides, ZrO2 and SiO2, with band-gaps of about 5 and 9 eV,respectively, for which neither visible-light excitation of electrons from the VBnor energy/electron transfer could be expected. It was proposed that gold NPscould be heated quickly (2–3 min) to 100 ∘C by light absorption, which led toformaldehyde oxidation. The reaction was attributed to the activation of polarmolecules (inactivity for cyclohexane) by the electromagnetic field, resulting inreaction with oxygen. However, deposition of gold on semiconducting supports(Fe2O3 and CeO2) resulted in significant enhancement of photocatalytic activity,suggesting that participation of other mechanisms (energy/electron transfer)could not be excluded for Au/Fe2O3 and Au/CeO2 nanomaterials. It was alsosuggested that the mechanism of plasmonic photocatalysis depended on semi-conductor support. For example, both Au/TiO2 and Au/ZnO were active undervisible-light irradiation, but Au/TiO2 heated in the dark was inactive [84], whileAu/ZnO was active [85]. It was deduced that the mechanism of photocatalysisis a plasmon-assisted photocatalysis one for Au/TiO2 and a plasmon-assistedcatalysis one (plasmonic heating) for Au/ZnO.

Plasmonic heating was proposed as the main mechanism for (i) catalyticreduction of CO2 on Au/ZnO photocatalyst [85], (ii) CO oxidation for gold NPsdeposited on titania incorporated in mesoporous silica [86], and (iii) hydrolysisof methyl parathion for gold capped with homogeneous catalysts (CuII(bpy))[87]. However, in the last case, two other mechanisms of plasmon-assistedphotocatalysis were also possible, that is, (i) energy transfer sensitizing the Cu(II)metal center, through the creation of d–d excited state, and (ii) electron transferresulting in the formation of a more active Cu(I) catalyst (reduction of Cu(II) by“hot electrons”).

On the contrary, there are plenty of reports rejecting the mechanism of plas-monic heating, due to inactivity of gold-modified insulators or unsupported goldNPs in comparison to highly active gold-modified semiconductors [60, 80, 82,88]. Plasmonic heating is not considered the main mechanism in the case of

i) water splitting, which requires much higher energy (1.32 eV) than the thermalenergy generated by plasmonic heating [78];

ii) photocurrent generation (“hot” electrons transfer) [89];


iii) hydrogen dissociation, where heating of gold NPs in the dark resulted ina negligible reaction, and direct photodissociation of hydrogen required10 times higher laser intensities than used for plasmon-assisted photocatal-ysis [75].

Studies on activation energy also excluded plasmonic heating in the case ofdecomposition of organic compounds [45] and photocurrent generation [90].For example, the apparent activation energy (2.4 kJ mol−1) for mineralization offormic acid on Au/CeO2 is much lower than that for the thermal activation pro-cess (24–36 kJ mol−1 on Pt/Al2O3, Pt/C, Pt/CexZr1−xO2). This indicates that therate-determining step of mineralization is different from that of oxygen activa-tion over supported metals, suggesting that a photoinduction step (by plasmonicactivation of Au) is involved in the mechanism [45].

Various structures of plasmonic photocatalysts have been proposed, whichdiffer significantly in physicochemical properties (adsorption properties of thereagents, light absorption properties, and resultant activities). It is thereforenot surprising that different photocatalytic mechanisms are proposed. More-over, several mechanisms may be involved in the (photo)catalytic reaction,for example, plasmon-assisted photocatalysis and catalysis (in the dark) orplasmon-assisted photocatalysis and plasmon-assisted catalysis. For example,Tsukamoto et al. reported aerobic oxidation of alcohols by two simultaneousmechanisms, that is, catalysis and plasmon-assisted photocatalysis, dependingon gold properties [73]. An interesting report has recently been published onthe influence of thickness of the titania layer on the reaction mechanism in thecase of trilayered gold(core)/silver/titania nanorods (as shown in Figure 16.9)[70]. Electron transfer was the predominant mechanism for thin titania layers(<10 nm) since silver was released in solution as Ag+ because of its oxidationafter electron transfer to the CB of titania. In contrast, energy transfer was thepredominant mechanism in the case of thick titania layers (≥10 nm). The optimalthickness of the titania shell (10 nm) was not consistent with the electron transfermechanism, where an increase in the thickness of the titania shell should resultin a decrease in photocatalytic activity as a result of an increase in the probabilityof trapping electrons before their arrival at the shell surface. Therefore, theoptimal thickness of the titania shell should be explained by the energy transfermechanism, that is, with the increase in the titania shell thickness, a larger

Electron

transfer

TiO2

Energy

transfer

Ag Ag

(a) (b)

TiO2

Au Au

Figure 16.9 Schemes showing core(gold)–shell(Ag) structures covered with titania layer ofdifferent thicknesses for which mechanism of electron (a) and energy (b) transfers areproposed under visible-light irradiation.


volume of the titania shell can receive the plasmonic energy up to optimal shellthickness, over which energy transfer should decrease (since the electric fielddecreases exponentially with the distance from the metal surface).

Au and Ag are mainly used for plasmonic photocatalysis, but also other NMshave been applied such as Pd [91], Pt [92], Cu [93]. It is also possible that themechanism of action of titania modified with other metals (nonnoble) could besimilar to plasmonic photocatalysis. For example, electron transfer, confirmed bythe time-resolved microwave conductivity (TRMC) method, has been proposedfor fine bismuth nanoclusters deposited on titania [94].

16.3.2 Nonmetals

Similar to modification with metals, various forms of nonmetallic modifiershave been applied for surface modification, such as adsorbed anions, inorganiccompounds, and organic compounds. Some modifications with inorganiccompounds are discussed in Section 16.4 on heterojunctions, while modificationwith organic compounds is mainly presented in the next section (Section 16.3.3).

For example, surface modification of NaTaO3 with glucose followed by ther-mal treatment resulted in enhanced visible-light absorption due to formationof carbonaceous species on the surface of the semiconductor. It was proposedthat under visible-light irradiation, carbonaceous species worked as sensitizersto absorb visible light. Accordingly, photogenerated electrons could be trans-ferred from carbonaceous species to the CB of NaTaO3, where they can reactwith adsorbed oxygen, forming superoxide ions, as shown in Figure 16.10 [95].

16.3.3 Organic Compounds (Colorless and Color)

Similar to modification with metals and nonmetals, various types of organic com-pounds have been used as surface modifiers. There are two main mechanisms

Visible light

e

e– e–

h+

h+

Carbonaceous

speciesNaTaO3

UV light

NO

OH–/H2O

N2 + O2

NO2 → NO3–

VB

CB

O2

•O2–

•OH

Figure 16.10 Mechanism of photocatalysis on C-modified NaTaO3 used for NO oxidation.(Reprinted with permission from Ref. [95]. Copyright 2014, The Royal Society of Chemistry.)


FTO glass

Dye sensitized

TiO2 film Sensitizer dyeRedox couple

in electrolyte

Pt counter

electrode

Energy vs.

vacuum (eV)

(2)

CB

(1)

(4)(5)

(3)

(D+/D∗)

(D+/D)

TiO2 Dye

hv

Electrolyte

–3.5

–4.0

–4.5

–5.0

–5.5

l+/l3–

(I–/I3–)

2

1

4

3

Load

Electron flow

S+/S•

S/S+

(a) (b)

Figure 16.11 Schematic diagram of electron transfer processes in DSSCs; forward processes:[P1] photoexcitation of dye (1), [P2] electron injection from dye to titania (2), [P3] redox coupleregeneration (3 [left]), [P4] dye regeneration (4 [left]) and 3 [right]); and backward processes:[P5] recombination of CB electrons with the oxidized dye (4 [right]) and [P6] recombination ofCB electrons with the oxidized electrolyte (5 [right]). (Reprinted with permission fromRefs [97, 98]. Copyright 2012, The Royal Society of Chemistry and Copyright 2011, The RoyalSociety of Chemistry.)

of their action, that is, (i) sensitization mechanism (usually for dyes, where thedye is excited with visible light) or (ii) formation of complex structures (organiccompound-semiconductor) with ability of vis absorption.

Modification of semiconductors with dyes has been extensively studied afterthe invention of dye-sensitized solar cells (DSSCs) by O’Regan and Graetzel. FirstDSSCs with high efficiency were reported by the aforementioned authors in 1991for 10-μm-thick and transparent film of titania NPs of a few nanometers in size,coated with a monolayer of a charge-transfer dye (trimeric ruthenium complex:RuL2(μ-(CN)Ru(CN)L′

2)2; where L – 2,2′-bipyridine-4,4′-dicarboxylic acid andL′ – 2,2′-bipyridine) [96].

In general, DSSCs (examples shown in Figure 16.11) are composed of (i) anode(wide band-gap semiconductor [TiO2, SnO2, ZnO] with adsorbed dye molecule[mainly polypyridyl complexes of ruthenium and osmium] on conducting glass[FTO:SnO2-coated glass]), (ii) electrolyte with redox mediator (usually iodineand tri-iodine [I−/I3

−] in acetonitrile), and (iii) cathode (conducting glass withcatalyst [deposited Pt]). The efficiency of a DSSC depends on four energy lev-els of the components: the excited state (approximately LUMO) and the groundstate (HOMO) of the photosensitizer, the Fermi level of the TiO2 electrode, andthe redox potential of the mediator (I−/I3

−) in the electrolyte.The main processes occurring in a DSSC are shown in Figure 16.11 and

described below [97]:(P1) The incident photon is absorbed by the dye (photosensitizer) adsorbed on

the TiO2 surface. The photosensitizers are excited from the ground state (S) tothe excited state (S*):

S + h𝜈 → S∗[P1]


(P2) The excited electrons are injected into the CB of the TiO2 electrode, result-ing in the oxidation of the photosensitizer (S+), as shown below:

S∗ → S+ + e−(TiO2) [P2]

(P3) The injected electrons in the CB of TiO2 are transported between TiO2NPs, with diffusion toward the back contact (TCO). And the electrons finallyreach the counter electrode through the circuit. The redox mediator, I3

−, diffusestoward the counter electrode and then is reduced to I− ions:

I3− + 2 e− → 3 I−[P3]

(P4) The oxidized photosensitizer (S+) accepts electrons from the I− ion redoxmediator, leading to regeneration of the ground state (S), and the I− is oxidizedto the oxidized state, I3

−:

S+ + e− → S [P4]

Unfortunately (P5 and P6), electrons in the nanocrystalline mesoporous TiO2can also recombine with either the oxidized dye (P5) or oxidized electrolyte (P6),as shown in Figure 16.11b (processes 4 and 5) [98], which limits the efficiency ofDSSCs.

S+ + e−(CB) → S [P5]I3

− + 2 e−(CB) → 3 I−[P6]

To improve the photocatalytic performance of DSSCs, various modificationsof their structure have been performed, for example, by synthesis of novel dyes(with ability of wide-range of visible-light photoabsorption [99]), modification ofthe titania nanostructure (mixture of small and large titania NPs [100] to achieveefficient adsorption of dye [101] and enhance light scattering, respectively), andselection of different redox couples. Since the sensitization process is ultrafast(ps), it is believed that the efficiency of DSSCs is controlled by the diffusion ofelectrons through the titania film. Therefore, various modifications of anodematerials have been proposed, for example, by an increase in crystallinity, betterconnection between particles, optimal width and composition (low recombi-nation of charge carriers and enhanced light harvesting), and nanoarchitecturedesign (e.g., titania nanotubes, TNT). A very interesting example of hinderingthe recombination of CB electrons with an oxidized electrolyte was proposedfor layered semiconductors (for another application, that is, enhanced hydrogenevolution reaction) [102]. In brief, the dye was adsorbed on the external surfaceof the H4Nb6O17-layered semiconductor, while fine Pt NPs between layers.Inhibited access of I3

− to the reduction site (Pt) by the electrostatic repulsionbetween I3

− anions and the negatively charged (NbO17)4− resulted in a preferablereduction of protons (but not I3

−) with a steady-state rate.Usually, the semiconductor is modified with molecules absorbing visible light

(e.g., dyes), but modifications with colorless compounds have also been reported.For example, modification of titania with toluene diisocynate (TDI) resulted inthe formation of a surface complex structure – NHCOOTi – between titaniaand TDI, which was excited under vis irradiation, causing electron injection into


the CB of titania [103]. The electron was scavenged by the adsorbed oxygen,forming superoxide anion radical and, via a series of redox steps, hydroxylradicals. Similarly, modification of titania with ascorbic acid resulted in visresponse due to specific bindings between both compounds, resulting in theformation of Ti—O—C bonds [104]. Another example has been shown for DDAT(S-1-dodecyl-S′-(𝛼,𝛼′-dimethyl-𝛼′′-acetic acid) trithiocarbonate), a colorlesscompound, and thus without photoabsorption of visible light [105]. However,surface complexation of titania particles by DDAT results in vis absorption dueto intramolecular ligand-to-metal charge transfer (LMCT) transition within theinner-sphere titanium(IV) surface complex (light promotes electron transferfrom the surface complexant to the CB of titania). DDAT molecules formingsurface complexes with titanol groups undergo a deep modification of theirelectronic transitions (HOMO to LUMO) that occur at the energies lower thanEg of titania. Accordingly, DDAT molecule can be excited with energy lowerthan Eg (visible light), which promotes an electron into the LUMO level and itssuccessive transfer into the CB of titania.

A very interesting example was shown for titania modification with urea (usu-ally such a modification was reported as resulting in titania N-doping) by Mitorajand Kisch [106]. It was proved that during thermal decomposition of urea, titaniaacted as a thermal catalyst for the conversion of intermediate isocyanic acid tocyanamide, and then trimerization of cyanamide to melamine followed by poly-condensation to melem- and melon-based poly(aminotri-s-triazine) derivatives,as shown in Figure 16.12. It was proposed that absorption at the vis range is acharge-transfer band, enabling an optical electron transfer from the polytriazinecomponent to titania. This material showed a unique example of a covalentlycoupled inorganic–organic (polytriazine was present as a crystalline layer) semi-conductor photocatalyst connected through Ti—N—C bond.

O

NH3

NH2

NH

OH

OH

“Dark” catalysis

[TiO2]

[TiO2]–OH

[TiO2]H2O

[TiO2]

[C(NH)(NH2)2]+

CO2

OC(NH2)2

C

C NH2N

C OHN

N

N

N

N

N N

N

N

H H

N

NH2

N

N

N

N

N N

N

NH

H

N

NH2

N N

N

NH2

NH2H2N

N N

N

OH

OHHO

Figure 16.12 Mechanism of urea-induced titania modification. (Reprinted with permissionfrom Ref. [106]. Copyright 2010, John Wiley & Sons.)


16.4 Heterojunctions

The development of coupled semiconductor systems (heterojunctions) hasbeen an active research field in recent years. In consequence of semiconductorcoupling, the inhibition of charge carrier recombination can be achieved. Thecharge transfer process in mixed semiconductor colloids, for example, CdS—TiO2, CdS—ZnO, Cd3P2—TiO2 and Cd3P2—ZnO, AgI—Ag2S, have alreadybeen reported [107, 108].

There are a lot of semiconductors with sufficient narrow band-gap for solarapplication, for example, WO3, Cu2O, CdTe, CuO, CdS, ZnSe, CdSe, and CdS.However, these semiconductors often do not have sufficient large band-gap foroverall water splitting (to oxidize and reduce water simultaneously). In addition,the CB level of semiconductors should be more negative than the potential forthe single-electron reduction of oxygen in order to allow efficient consumption ofphotoexcited electrons and subsequent oxidative decomposition of OC by holesto proceed in air [109].

O2 + e− = O2−(aq), −0.284 V vs NHE;

O2 + H+ + e− = HO2(aq), −0.046 V vs NHEThe CB levels of simple oxide semiconductors with visible-light absorption,

such as tungsten(VI) oxide, are generally more positive (+0.5 V vs NHE for WO3)than the reduction potentials of O2 due to the deeply positive level of VB, whichmainly consists of O 2p orbitals. This fact causes inactivity of bare WO3 for effi-cient oxidative decomposition of organic compounds in air. Thus, only modifiedWO3 photocatalysts (e.g., with Pt NPs as cocatalyst for multielectron O2 reduc-tion [109]) are active in OCs decomposition under visible-light irradiation.

The coupling of two/three semiconductors has been widely studied, due tothe possibility of inhibition of charge carrier recombination. The charge transferprocess in mixed semiconductor colloids has already been reported, for example,CdS—TiO2, CdS—ZnO, Cd3P2—TiO2, Cd3P2—ZnO, and AgI—Ag2S [107, 108].The enhanced photoactivity under vis has been studied also for particles ofmixed oxide nanocomposites, for example, Cu2O/TiO2 heterojunction [110] andBi2WO6 (composed of accumulated layers of corner-sharing WO6 octahedral[WO4]2− sheets and bismuth oxide (Bi2O2)2+ sheets) [111]. Selection of aproper semiconductor is very important and results in different mechanismsof interaction between semiconductors. Two mechanisms applied for wideband-gap semiconductors (or coupled wide- and narrow-semiconductors) havebeen reported, that is, (Section 16.4.1) transfer of only one charge carrier (e−or h+), and (Section 16.4.2) simultaneous transfer of both charge carriers. Thesimultaneous transfer of both charge carriers is more desirable, since the holesmove in the opposite direction to the electrons, making charge separation moreefficient. Examples of both systems are shown below.

16.4.1 Excitation of One Component

Heterojunctions containing titania belong to the first group in which titania isonly an acceptor of generated electrons (due to its wide band-gap), for example,

16.4 Heterojunctions 469

1D titania nanobelts (TNBs) decorated with 3D CdS nanospheres (NSPs). TheTNBs/CdS NSPs nanocomposite exhibited improved visible-light-driven photo-catalytic activities in comparison with the counterparts (TNBs and CdS NSPs)owing to the formation of well-defined heterojunctions between TNBs and CdSNSPs, thus allowing enhanced light absorption and prolonged lifetime of photo-generated charge carriers [112].

Heterojunctions can be also formed for organic–inorganic composites, forexample, photocurrent was generated under visible-light irradiation of titaniamodified with graphite-like C3N4 [113]. It was proposed that under visible-lightirradiation, g-C3N4 absorbed photons to induce 𝜋→𝜋* transition (exciting theelectrons from HOMO to the LUMO). Because the LUMO position of g-C3N4was more negative than the CB of titania, the photogenerated electrons fromg-C3N4 could be injected into the CB of titania.

16.4.2 Excitation of Both Components

Excitation of two components is only possible when both components can absorbvisible light. Therefore, bare titania cannot be used in such heterojunctions (wideband-gap). Since the position of the VB of titania is determined by O 2p orbitals,other compounds containing nitrogen have been used. Recently, large interestis focused on titanium oxynitride (TiON) [114–116] and nitrides, since N 2porbitals have more negative energy levels than O 2p orbitals, thus the tops ofVB are more negative than that of the corresponding oxide. For example, Ta3N5[116, 117], LaTiO2N [118], BaTaO2N [119], Ge3N4 [120], GaN:ZnO [121], andTiNxOyFz [122] have been reported as vis active photocatalysts.

Two semiconductors of different type (n- and p-types) with sufficient narrowband-gap (visible-light absorption) and with much different positions of CB andVB have been also reported. This p–n junction between n- and p-type semi-conductors of narrow band-gap was proposed for CaFe2O4(CFO)/WO3 [123].It was shown that the interfaces among the particles of the CFO/WO3 compos-ite played an important role in the photocatalytic reaction. The heat treatmentreduced the defect states such as dangling bonds and surface hydroxyl groupsat the surface of the particles. Figure 16.13 shows the schematic diagrams of theenergy levels of the composites under different conditions. The perfect p-n junc-tion structure between two semiconductors shown in Figure 16.13a is impossible

e e e e e ee e e e e e e e e

h

CFO CFO CFO

WO3

(a) (b) (c)

h h h h h

h h h

WO3 WO3

h h h h h h

Figure 16.13 Schematic diagram of the energy levels of the CFO/WO3 composites: (a) perfectjunction, (b) with defects at interface, (c) ohmic contact junction. (Reprinted with permissionfrom Ref. [123]. Copyright 2009, The American Chemical Society.)


to be achieved by simply mixing two components. In addition, holes or electronsin the perfect p–n junction structure do not have sufficient oxidative or reductivepotential for OCs decomposition. For the just mixed CFO/WO3 composite, theremay be intermediate levels at the interface formed by defect states, which couldact as traps of photogenerated holes or electrons and enhance the charge separa-tion in both n-type WO3 and p-type CFO (Figure 16.13b). Modifying the particlesurface using metallic Ag and ITO makes the CFO–WO3 connection more likeohmic contacting p-n junction. As shown in Figure 16.13c, photogenerated holesin p-type CFO and electrons in n-type WO3 recombine at the metallic contact-ing position and promote the charge separation, and as a result, increasing thephotocatalytic activity.

Quantum dots (QDs), for which optical and electronic properties differfrom those of larger particles, are another group of semiconductors with highinterest. Many types of QDs have already been reported and used for variousapplications since the properties (e.g., photoabsorption) depend on the dots’size, shape, and material. One of the examples of heterojunctions between visactive semiconductors and QDs has been proposed for TiON hollow spheresand Ta3N5 QDs [124]. The TiON—Ta3N5 heterojunctions showed high rate ofoxygen evolution (Φapp = 67% at 420 nm), which was 3.3 times higher than thatof pristine TaON hollow spheres. The enhancement of photocatalytic activity isattributed to electric field-assisted charge transfer at the heterojunction inter-face. The driving force of charge transfer came from matching band potentialsbetween two semiconductors resulting in efficient separation of charge carriersin these two materials. The flat-band potentials (Efb) of Ta3N5 QDs and TaONwere estimated to be −0.68 and −0.6 V versus saturated calomel electrode (SCE),respectively. Since the CB edge potential of Ta3N5 QDs is more negative thanthat of TaON, the photoinduced electrons can migrate from Ta3N5 to TaON viaa well-developed interface (heterojunction). Similarly, the photoinduced holestransfer in the opposite direction, that is, from TaON to Ta3N5. Moreover, largedifferences in VB edge potentials and CB potentials retard the recombination ofcharge carriers. Nowadays, various heterojunctions have been extensively stud-ied, since they are believed to allow efficient solar energy conversion, resultingin commercial application of those materials, for example, MgTa2O6-xNy/TaON[125] and CaFe2O4/TaON [126].

16.5 Z-Scheme

Photocatalytic reactions under vis through two-step photoexcitation betweentwo different photocatalysts known as the Z-scheme system have been pro-posed for overall water splitting, as shown in Figure 16.14a [127] (A review onZ-scheme systems for water splitting has already been written [128].). The firstreport showing overall water splitting by a Z-scheme photocatalytic systemwas presented for SrTiO3 doped with Cr and Ta for H2 evolution, WO3 for O2evolution, and iodate/iodine redox couple as an electron mediator (similar asin DSSCs) [129]. Different kinds of semiconductors and redox mediators havebeen widely investigated to obtain high efficiency and stability of water splitting,

16.6 Hybrid Nanostructures 471

Visible

light

Photocatalyst B

Photocatalyst A

Visible

light

e–

e–

D

Redox mediator

(a) (b)

Visible light Visible light

Electron donor

level formed

by Rh3+

Rh species existing at

the surface of SrTiO3:Rh

CB

VB VB

Rh4+/3+

BiVO4SrTiO3:Rh

h+

h+

e–CB

Ru H2

H2O

H2O

O2

e–

D: donorA: acceptor

A

h+

h+

H2O

O2

H2

H+

Figure 16.14 Z-scheme photocatalytic systems with (a) and without (b) redox mediator.(Panel (a): Reprinted with permission from Ref. [97]. Copyright 2013, The Royal Society ofChemistry. Panel (b): Reprinted with permission from Ref. [123]. Copyright 2009, The AmericanChemical Society.)

for example, (i) for H2 evolution: TaON, BaTaO2N, LaTiO2N, T3N5, Pt/TaON,mixed oxynitrides ATaO2N (A: Ca, Sr, Ba), TaON/ZrO2, Pt/ZrO2/TaON,SrTiO3:Rh, SrTiO3:Rh loaded with Au, Ni, Ru, and Pt, and (ii) for O2 evolution:Pt/WO3, Pt/BiVO4, TaON/RuO2, Ta3N5/TiO2/Ir, and redox mediator: NaIO3,NaIO3/NaI, Fe3+/Fe2+ [130–136].

At first, dissolved redox mediators in the suspension of two semiconductorswere used for Z-scheme water splitting, which unfortunately was connected withunwanted backward reactions (as has been discussed for DSSCs). Therefore,novel approaches were applied for water splitting by Z-scheme interparticleelectron transfer without electron mediator (similar as heterojunctions), forexample, Ru/SrTiO3:Rh (H2 evolution) with BiVO4 (O2 evolution), as shown inFigure 16.14b [137].

16.6 Hybrid Nanostructures

Various hybrid nanostructures have also been reported, in which (i) inorganicand organic surface modifiers (e.g., Ru(II) complexes and plasmonic NPs [101,138, 139]), (ii) surface modifiers and dopants, (iii) dopants and heterojunctions,and (iv) surface modifiers and heterojunctions have been applied. Some examplesare presented below.

One example of hybrid nanostructure, that is, doped–surface modified photo-catalyst, was reported for titania doped with nitrogen and surface modified withnickel chloride [140]. Authors claimed that enhanced activity under visible-lightirradiation was caused by co-modification. The mechanism of action underUV (A) and visible-light (B and C) irradiation is shown in Figure 16.15a. Itwas proposed that doped nitrogen existed as N—Ox surface species, for whichthe energy level was located 0.25 eV above the VB of titania. Therefore, thevis irradiation resulted in excitation of electrons from energy level of N—Oxin TiO2—N to CB of titania (process B). Whereas surface modification withnickel chloride resulted in electronic transition from VB of titania to the energylevel of O—Ni—O species, located 0.15 eV below the CB of titania (processC). Therefore, photogenerated electrons from the CB of titania (process B) and


CB CB

CarbonC-TiO2

C-doping level

VBVB

(A)(B)

O2

O2–

O–Ni–Cl

N-Ox

(a) (b)

OH•

OH–OH–

e–

h+ h+ h+ h+ h+ h+

e– e– e– e– e–

·OH

(C)

+ +

– –

– ––

+

+ +

hυ

Figure 16.15 (a) Schematic diagram for the band structure of titania doped with nitrogen andsurface modified with nickel chloride and the photocatalytic mechanism (Reprinted withpermission from Ref. [140], Copyright 2013, Elsevier.); (b) photocatalytic mechanism onC(bimodal)—TiO2. (Reprinted with permission from Ref. [141], Copyright 2012, Elsevier.)

from O—Ni—Cl surface species (process C) could be captured by the adsorbedoxygen to form O2

− species. Meanwhile, the photogenerated holes in VB oftitania and at the energy level of N—Ox species could directly oxidize organicmolecule adsorbed on the surface of the photocatalyst.

A very interesting example was shown for titania doped and surface modifiedwith carbon, that is, bimodal carbon modification effect, including carbondoping in the lattice of titania and carbon adsorption on titania surface [141].C—TiO2 photocatalysts were prepared by ethanol supercritical solvothermalmethod involving tetrabutyl titanate and raw rice. The simplified mechanismis shown in Figure 16.15b. Band-gap narrowing by carbon doping results invis photoabsorption, whereas carbon adsorbed on the surface can play twofunctions, that is, titania sensitization (not shown in the scheme) and as anelectron sink retarding recombination of charge carriers.

For doping–heterojunction nanostructure, the vanadium-doped titania mod-ified with 𝛽-In2S3 was proposed as visible-light-active photocatalyst [142]. Themechanism is similar to heterojunctions of two semiconductors capable of visabsorption. In brief, under visible-light irradiation, electrons are excited from VBto CB of both V—TiO2 and 𝛽-In2S3, and the band alignment of the heterojunc-tion allows an efficient electron transfer from CB of 𝛽-In2S3 into CB of V—TiO2.Moreover, it was proposed that V0 and electrons present in wide hole quencherwindow of V—TiO2 (comprises of split molecular orbitals in V—TiO2, the elec-tron trapping states and defects) can quench the holes from VB of 𝛽-In2S3.

The example of surface modification–heterojunctions allowing faster trans-fer of charge carriers has been presented for 𝛼-Fe2O3nanorod (core)/graphene(RGO)/BiV1−xMoxO4 (shell) [143]. Fe2O3/RGO (core) forms heterojunction withBiV1−xMoxO4 (shell). The energy bands of Fe2O3 and BiV1−xMoxO4 shift upwardand downward, respectively, along with diffusion of carriers until Fermi levelsreach equilibrium. Under irradiation, photogenerated holes are transferred fromVB of Fe2O3to VB of BiV1−xMoxO4, while photogenerated electrons from CBof BiV1−xMoxO4 to CB of Fe2O3. The reduced graphene oxide (RGO) interlayerbehaves as an electron conductor to enhance separation of charge carriers, dueto the band alignment and potential difference.

References 473

16.7 Summary

Although many ways of introduction of visible-light activity to semiconductorsexist, two mechanisms play the most significant role: (i) change of intrinsicproperties of photocatalysts by doping, which allow them for visible-lightabsorption, and (ii) interaction between semiconductor and modifier resultingin either charge transfer or energy transfer between them. Generally, dopingstrategy is responsible for reducing band-gap, while surface modificationcauses mainly an introduction of new energy bands. Other perspective waysare Z-scheme, interparticle electron transfer without electron mediator andsemiconductor–modifier composites, which absorb visible light, but do notchange intrinsic electronic properties of semiconductor. Quite uncommonstrategy, different from impurity doping, so-called “self-doping”, offers promisingvisible-light activation of photocatalyst without additional substances (dopants,modifiers, cocatalysts) contribution.

A design of efficient visible-light photocatalysts should fulfill general require-ments oriented for application: lower band-gap responsible for utilizing visiblelight, proper CB and VB positions depending on the driven redox reaction,minimization of electron–hole recombination, high stability of photocatalyst (nophotocorrosion) and low price of final photocatalytic material. A comparison ofdifferent visible-light photocatalysts is difficult and does not give the clear answerwhich materials are most efficient. Visible-light photocatalytic activities of suchdesigned semiconductors along with their stabilities are still not sufficient toperform significant commercial applications. Therefore, further comprehensiveresearch studies to make improvements in these fields and produce cheapand multitasking photocatalytic materials are necessary. In addition, dopedor modified visible-light-active photocatalysts could be dedicated to artificialphotosynthesis to recycle CO2 being a relevant topic of interest, addressing thecritically important issue of future sustainable society.

References

1 Hasegawa, G., Kanamori, K., Sugawara, Y., Ikuhara, Y., and Nakanishi, K.(2012) Flower-like surface modification of titania materials by lithiumhydroxide solution. J. Colloid Interface Sci., 374, 291–296.

2 Yan, X., Ohno, T., Nishijima, K., Abe, R., and Ohtani, B. (2006) Is methyleneblue an appropriate substrate for a photocatalytic activity test? A study withvisible-light responsive titania. Chem. Phys. Lett., 429, 606–610.

3 Ohtani, B. (2014) Revisiting the fundamental physical chemistry in heteroge-neous photocatalysis: its thermodynamics and kinetics. Phys. Chem. Chem.Phys., 16, 1788–1797.

4 Ribeiro, R.A.P., de Lazaro, S.R., and de Oliveira, C.R. (2016) Band-gap engi-neering for photocatalytic applications: anionic and cationic doping of TiO2anatase. Curr. Phys. Chem., 6, 22–27.



6 Stepanov, A.L. (2012) Applications of ion implantation for modification ofTiO2: a review. Rev. Adv. Mater. Sci., 30, 150–165.

7 Fromknecht, R., Khubeis, I., Massing, S., and Meyer, O. (1999) Ion implanta-tion in TiO2: damage production and recovery, lattice site location and elec-trical conductivity. Nucl. Instrum. Methods Phys. Res., Sect. B, 147, 191–201.

8 Anpo, M. and Takeuchi, M. (2001) Design and development ofsecond-generation titanium oxide photocatalysts to better our environ-ment – approaches in realizing the use of visible light. Int. J. Photoenergy, 3,89–94.

9 Yamashita, H. and Anpo, M. (2004) Application of an ion beam techniquefor the design of visible light-sensitive, highly efficient and highly selectivephotocatalysts: ion-implantation and ionized cluster beam methods. Catal.Surv. Asia, 8, 35–45.

10 Anpo, M. (2000) Use of visible light. Second-generation titaniumoxide photocatalysts prepared by the application of an advanced metalion-implantation method. Pure Appl. Chem., 9, 1787–1792.

11 Yen, C.C., Wang, D.Y., Shih, M.H., Chang, L.S., and Shih, H.C. (2010)A combined experimental and theoretical analysis of Fe-implanted TiO2modified by metal plasma ion implantation. Appl. Surf. Sci., 256, 6865–6870.

12 Zhu, J., Chen, F., Zhang, J., Chen, H., and Anpo, M. (2006) Fe3+—TiO2photocatalysts prepared by combining sol–gel method with hydrothermaltreatment and their characterization. J. Photochem. Photobiol., 180, 196–204.

13 Zhu, J., Zheng, W., He, B., Zhang, J., and Anpo, M. (2004) Characterizationof Fe—TiO2 photocatalysts synthesized by hydrothermal method and theirphotocatalytic reactivity for photodegradation of XRG dye diluted in water.J. Mol. Catal. A: Chem., 216, 35–43.

14 Tian, B., Li, C., Gu, F., Jiang, H., Hu, Y., and Zhang, J. (2009) Flame sprayedV-doped TiO2 nanoparticles with enhanced photocatalytic activity undervisible light irradiation. Chem. Eng. J., 151, 220–227.

15 Kuncewicz, J. and Ohtani, B. (2016) Rhodium-doped titania photocatalystswith two step bandgap excitation by visible light – influence of the dopantconcentration on photosensitization efficiency. RSC Adv., 6, 77201–77211.

16 Dong, F., Zhao, W., Wu, Z., and Guo, S. (2009) Band structure and visiblelight photocatalytic activity of multi-type nitrogen doped TiO2 nanoparticlesprepared by thermal decomposition. J. Hazard. Mater., 162, 763–770.

17 Yamanaka, K. and Morikawa, T. (2012) Charge-carrier dynamics innitrogen-doped TiO2 powder studied by femtosecond time-resolved diffusereflectance spectroscopy. J. Phys. Chem. C, 116, 1286–1292.

18 Samsudin, E.M. and Abd Hamid, S.B. (2017) Effect of band gap engineeringin anionic-doped TiO2 photocatalyst. Appl. Surf. Sci., 391, 326–336.

19 Liu, G., Yin, L.C., Wang, J., Niu, P., Zhen, C., Xie, Y., and Cheng, H.M.(2012) A red anatase TiO2 photocatalyst for solar energy conversion. EnergyEnviron. Sci., 5, 9603–9610.

References 475

20 Wang, W., Lu, C., Ni, Y., Su, M., Huang, W., and Xu, Z. (2012) Prepara-tion and characterization of visible-light-driven N—F—Ta tri-doped TiO2photocatalysts. Appl. Surf. Sci., 258, 8696–8703.

21 Shi, J.W., Zheng, J.T., Hu, Y., and Zhao, Y.C. (2007) Influence of Fe3+ andHo3+ co-doping on the photocatalytic activity of TiO2. Mater. Chem. Phys.,106, 247–249.

22 Xiong, L.B., Li, J.L., Yang, B., and Yu, Y. (2012) Ti3+ in the surface oftitanium dioxide: generation, properties and photocatalytic application.J. Nanomater., 831524.

23 Cronemeyer, D.C. (1959) Infrared absorption of reduced rutile TiO2 singlecrystals. Phys. Rev., 113, 1222.

24 Zuo, F., Wu, T., Zhang, Z., Borchardt, D., and Feng, P. (2010) Self-doped Ti3+

enhanced photocatalyst for hydrogen production under visible light. J. Am.Chem. Soc., 132, 11856–11857.

25 Justicia, I., Ordejon, P., Canto, G., Mozos, J.L., Fraxedas, J., Battiston, G.A.,Gerbasi, R., and Figueras, A. (2002) Designed self-doped titanium oxide thinfilms for efficient visible-light photocatalysis. Adv. Mater., 14, 1399–1402.

26 Nakano, T., Yazawa, S., Araki, S., Kogoshi, S., Katayama, N., Kudo, Y., andNakanishi, T. (2015) Experimental study of the visible-light photocatalyticactivity of oxygen-deficient TiO2 prepared with Ar/H2 plasma surface treat-ment. Jpn. J. Appl. Phys., 54, 01AE04.


28 Li, H., Chen, Z., Tsang, C.K., Li, Z., Ran, X., Lee, C., Nie, B., Zheng, L.,Hung, T., Lu, J., Pan, B., and Li, Y.Y. (2014) Electrochemical doping ofanatase TiO2 in organic electrolytes for high-performance supercapacitorsand photocatalysts. J. Mater. Chem. A, 2, 229–236.

29 Fan, C., Chen, C., Wang, J., Fu, X., Ren, Z., Qian, G., and Wang, Z. (2015)Black hydroxylated titanium dioxide prepared via ultrasonication withenhanced photocatalytic activity. Sci. Rep., 5, 11712.

30 Zhu, Q., Peng, Y., Lin, L., Fan, C.M., Gao, G.Q., Wang, R.X., and Xu, A.W.(2014) Stable blue TiO2−x nanoparticles for efficient visible light photocata-lysts. J. Mater. Chem. A, 2, 4429–4437.

31 Macyk, W. and Kisch, H. (2001) Photosensitization of crystalline and amor-phous titanium dioxide by platinum(IV) chloride surface complexes. Chem.Eur. J., 7, 1862–1867.

32 Macyk, W., Burgeth, G., and Kisch, H. (2003) Photoelectrochemical proper-ties of platinum(IV) chloride surface modified TiO2. Photochem. Photobiol.Sci., 2, 322–328.

33 Tada, H., Jin, Q., Nishijima, H., Yamamoto, H., Fujishima, M., Okuoka, S.,Hattori, T., Sumida, Y., and Kobayashi, H. (2011) Titanium(IV) dioxidesurface-modified with iron oxide as a visible light photocatalyst. Angew.Chem. Int. Ed., 50, 3501–3505.


34 Zang, L., Lange, C., Abraham, I., Storck, S., Maier, W.F., and Kisch, H.(1998) Amorphous microporous titania modified with platinum(IV) chlo-ride – a new type of hybrid photocatalyst for visible light detoxification.J. Phys. Chem. B, 102, 10765–10771.

35 Ohtani, B. and Nishimoto, S.-i. (1993) Effect of surface adsorption ofaliphatic alcohols and silver ion on the photocatalytic activity of TiO2suspended in aqueous solutions. J. Phys. Chem. B, 97, 920–926.

36 Kowalska, E., Remita, H., Colbeau-Justin, C., Hupka, J., and Belloni, J. (2008)Modification of titanium dioxide with platinum ions and clusters: applicationin photocatalysis. J. Phys. Chem. C, 112, 1124–1131.

37 Sclafani, A., Mozzanega, M.-N., and Herrmann, J.M. (1997) Influence of sil-ver deposits on the photocatalytic activity of titania. J. Catal., 168, 117–120.

38 Banmwenda, G.R., Tsubota, S., Nakamura, T., and Haruta, M. (1995) Pho-toassisted hydrogen production from water–ethanol solution: a comparisonof activities of Au—TiO2 and Pt—TiO2. J. Photochem. Photobiol., A, 89,177–189.

39 Ohtani, B., Kakimoto, M., Miyadzu, H., Nishimoto, S., and Kagiya, T. (1988)Effect of surface-adsorbed 2-propanol on the photocatalytic reduction ofsilver and/or nitrate ions in acidic TiO2 suspensions. J. Phys. Chem., 92,5773–5777.

40 Pichat, P., Herrmann, J.M., Disdier, J., Courbon, H., and Mozzanega, M.N.(1981) Photocatalytic hydrogen production from aliphatic alcohols overa bifunctional platinum on titanium dioxide catalyst. Nouv. J. Chim., 5,627–636.

41 Ohtani, B., Ogawa, Y., and Nishimoto, S.-i. (1997) Photocatalytic activity ofamorphous-anatase mixture of titanium(IV) oxide particles suspended inaqueous solutions. J. Phys. Chem. B, 101, 3746–3752.

42 Tian, Y. and Tatsuma, T. (2005) Mechanisms and applications ofplasmon-induced charge separation at TiO2 films loaded with gold nanopar-ticles. J. Am. Chem. Soc., 127, 7632–7637.

43 Kowalska, E., Abe, R., and Ohtani, B. (2009) Visible light-induced pho-tocatalytic reaction of gold-modified titanium(IV) oxide particles: actionspectrum analysis. Chem. Commun., 241–243.

44 Kominami, H., Tanaka, A., and Hashimoto, K. (2010) Mineralization oforganic acids in aqueous suspension of gold nanoparticles supported oncerium(IV) oxide powder under visible light irradiation. Chem. Commun.,46, 1287–1289.

45 Kominami, H., Tanaka, A., and Hashimoto, K. (2011) Gold nanoparticlessupported on cerium(IV) oxide powder for mineralization of organic acidsin aqueous suspensions under irradiation of visible light of 𝜆= 530 nm. Appl.Catal., A, 397, 121–126.

46 Thimsen, E., Le Formal, F., Gratzel, M., and Warren, S.C. (2011) Influenceof plasmonic Au nanoparticles on the photoactivity of Fe2O3 electrodes forwater splitting. Nano Lett., 11, 35–43.

47 Yu, H., Ming, H., Zhang, H., Li, H., Pan, K., Liu, Y., Wang, F., Gong, J., andKang, Z. (2012) Au/ZnO nanocomposite: facile fabrication and enhanced

References 477

photocatalytic activity for degradation of benzene. Mater. Chem. Phys., 137,113–117.

48 Lan, J.Y., Zhou, X.M., Liu, G., Yu, J.G., Zhang, J.C., Zhi, L.J., and Nie, G.J.(2011) Enhancing photocatalytic activity of one-dimensional KNbO3nanowires by Au nanoparticles under ultraviolet and visible-light. Nanoscale,3, 5161–5167.

49 Haes, A.J., Haynes, C.L., McFarland, A.D., Schatz, G.C., Van Duyne, R.P.,and Zou, S. (2005) Plasmonic materials for surface-enhanced sensing andspectroscopy. MRS Bull., 30, 368–375.

50 Halas, N. (2005) Playing with plasmons: tuning the optical resonant proper-ties of metallic nanoshells. MRS Bull., 30, 362–367.

51 Zijlstra, P., Chon, J.W.M., and Gu, M. (2009) Five-dimensional opticalrecording mediated by surface plasmons in gold nanorods. Nature, 459,410–413.

52 Ueno, K. and Misawa, H. (2013) Surface plasmon-enhanced photochemicalreactions. J. Photochem. Photobiol., C, 15, 31–52.

53 Verbruggen, S.W. (2015) TiO2 photocatalysis for the degradation of pollu-tants in gas phase: from morphological design to plasmonic enhancement.J. Photochem. Photobiol., C, 24, 64–82.

54 Sarina, S., Waclawik, E.R., and Zhu, H.Y. (2013) Photocatalysis on supportedgold and silver nanoparticles under ultraviolet and visible light irradiation.Green Chem., 15, 1814–1833.

55 Lou, Z.Z., Wang, Z.Y., Huang, B.B., and Dai, Y. (2014) Synthesis and activityof plasmonic photocatalysts. ChemCatChem, 6, 2456–2476.

56 Hou, W.B. and Cronin, S.B. (2013) A review of surface plasmonresonance-enhanced photocatalysis. Adv. Funct. Mater., 23, 1612–1619.

57 Subramanian, V., Wolf, E., and Kamat, P.V. (2001) Semiconductor-metalcomposite nanostructures. To what extent do metal nanoparticlesimprove the photocatalytic activity of TiO2 films? J. Phys. Chem. B, 105,11439–11446.

58 Rodriguez-Gonzalez, V., Zanella, R., del Angel, G., and Gomez, R. (2008)MTBE visible-light photocatalytic decomposition over Au/TiO2 andAu/TiO2—Al2O3 sol–gel prepared catalysts. J. Mol. Catal. A: Chem., 281,93–98.

59 Burgeth, G. and Kisch, H. (2002) Photocatalytic and photoelectrochemicalproperties of titania-chloroplatinate(IV). Coord. Chem. Rev., 230, 41–47.

60 Kowalska, E., Mahaney, O.O.P., Abe, R., and Ohtani, B. (2010)Visible-light-induced photocatalysis through surface plasmon excitationof gold on titania surfaces. Phys. Chem. Chem. Phys., 12, 2344–2355.

61 Wang, L., Zhou, X.M., Nguyen, N.T., and Schmuki, P. (2015)Plasmon-enhanced photoelectrochemical water splitting using Au nanoparti-cles decorated on hematite nanoflake arrays. ChemSusChem, 8, 618–622.

62 Furube, A., Du, L., Hara, K., Katoh, R., and Tachiya, M. (2007) Ultrafastplasmon-induced electron transfer from gold nanodots into TiO2 nanoparti-cles. J. Am. Chem. Soc., 129, 14852–14853.


63 Du, L., Furube, A., Hara, K., Katoh, R., and Tachiya, M. (2009) Plasmoninduced electron transfer at gold–TiO2 interface under femtosecond near-IRtwo photon excitation. Thin Solid Films, 158, 861–864.

64 Naya, S.-i., Teranishi, M., Isobe, T., and Tada, H. (2010) Lightwavelength-switchable photocatalytic reaction by gold nanoparticle-loadedtitanium(IV) dioxide. Chem. Commun., 46, 815–817.

65 Naya, S.-i., Inoue, A., and Tada, H. (2010) Self-assembly heterosupramolec-ular visible light photocatalyst consisting of gold nanoparticle-loadedtitanium(IV) dioxide and surfactant. J. Am. Chem. Soc., 132, 6292–6293.

66 Sakai, N., Fujiwara, Y., Arai, M., Yu, K., and Tatsuma, T. (2009) Electrode-position of gold nanoparticles on ITO: control of morphology and plasmonresonance-based absorption and scattering. J. Electroanal. Chem., 628, 7–15.

67 Sakai, N., Fujiwara, Y., Takahashi, Y., and Tatsuma, T. (2009)Plasmon-resonance-based generation of cathodic photocurrent at elec-trodeposited gold nanoparticles coated with TiO2 films. ChemPhysChem, 10,766–769.

68 Panayotov, D.A., DeSario, P.A., Pietron, J.J., Brintlinger, T.H., Szymczak, L.C.,Rolison, D.R., and Morris, J.R. (2013) Ultraviolet and visible photochem-istry of methanol at 3D mesoporous networks: TiO2 and Au—TiO2. J. Phys.Chem. C, 117, 15035–15049.

69 Brennan, L.J., Purcell-Milton, F., Salmeron, A.S., Zhang, H., Govorov, A.O.,Fedorov, A.V., and Gun’ko, Y.K. (2015) Hot plasmonic electrons for gener-ation of enhanced photocurrent in gold–TiO2 nanocomposites. NanoscaleRes. Lett., 10, 1–12.

70 Horiguchi, Y., Kanda, T., Torigoe, K., Sakai, H., and Abe, M. (2014) Prepara-tion of gold/silver/titania trilayered nanorods and their photocatalytic activi-ties. Langmuir, 30, 922–928.

71 Caretti, I., Keulemans, M., Verbruggen, S.W., Lenaerts, S., andVan Doorslaer, S. (2015) Light-induced processes in plasmonic gold/TiO2photocatalysts studied by electron paramagnetic resonance. Top. Catal., 58,776–782.

72 Priebe, J.B., Radnik, J., Lennox, A.J.J., Pohl, M.M., Karnahl, M.,Hollmann, D., Grabow, K., Bentrup, U., Junge, H., Beller, M., andBruckner, A. (2015) Solar hydrogen production by plasmonic Au—TiO2catalysts: impact of synthesis protocol and TiO2 phase on charge transferefficiency and H2 evolution rates. ACS Catal., 5, 2137–2148.

73 Tsukamoto, D., Shiraishi, Y., Sugano, Y., Ichikawa, S., Tanaka, S., andHirai, T. (2012) Gold nanoparticles located at the interface of anatase/rutileTiO2 particles as active plasmonic photocatalysts for aerobic oxidation.J. Am. Chem. Soc., 134, 6309–6315.

74 Long, R. and Prezhdo, O.V. (2014) Instantaneous generation ofcharge-separated state on TiO2 surface sensitized with plasmonic nanoparti-cles. J. Am. Chem. Soc., 136, 4343–4354.

75 Mukherjee, S., Libisch, F., Large, N., Neumann, O., Brown, L.V., Cheng, J.,Lassiter, J.B., Carter, E.A., Nordlander, P., and Halas, N.J. (2013) Hot elec-trons do the impossible: plasmon-induced dissociation of H2 on Au. NanoLett., 13, 240–247.

References 479

76 Hallett-Tapley, G.L., Silvero, M.J., Gonzalez-Bejar, M., Grenier, M.,Netto-Fereira, J.C., and Scaiano, J.C. (2011) Plasmon-mediated catalyticoxidation of sec-phenethyl and benzyl alcohols. J. Phys. Chem. C, 115,10784–10790.

77 Valenti, M., Dolat, D., Biskos, G., Schmidt-Ott, A., and Smith, W.A. (2015)Enhancement of the photoelectrochemical performance of CuWO4 thin filmsfor solar water splitting by plasmonic nanoparticle functionalization. J. Phys.Chem. C, 119, 2096–2104.

78 Liu, Z., Hou, W., Pavaskar, P., Aykol, M., and Cronin, S.B. (2011) Plas-mon resonance enhancement of photocatalytic water splitting under visibleillumination. Nano Lett., 11, 1111–1116.

79 Bouhadoun, S., Guillard, C., Dapozze, F., Singh, S., Amans, D., Boucle, J.,and Herlin-Boime, N. (2015) One step synthesis of N-doped and Au-loadedTiO2 nanoparticles by laser pyrolysis: application in photocatalysis. Appl.Catal., B, 174, 367–375.

80 Hou, W., Liu, Z., Pavaskar, P., Hsuan Hung, W., and Cronin, S.B. (2011)Plasmonic enhancement of photocatalytic decomposition of methyl orangeunder visible light. J. Catal., 277, 149–153.

81 Seh, Z.W., Liu, S.W., Low, M., Zhang, S.-Y., Liu, Z., Mlayah, A., andHan, M.-Y. (2012) Janus Au—TiO2 photocatalysts with strong localizationof plasmonic near fields for efficient visible-light hydrogen generation.Adv. Mater., 24, 2310–2314.

82 Cushing, S.K., Li, J.T., Meng, F.K., Senty, T.R., Suri, S., Zhi, M.J., Li, M.,Bristow, A.D., and Wu, N.Q. (2012) Photocatalytic activity enhanced by plas-monic resonant energy transfer from metal to semiconductor. J. Am. Chem.Soc., 134, 15033–15041.

83 Chen, X., Zhu, H.-Y., Zhao, J.-C., Zheng, Z.-F., and Gao, X.-P. (2008)Visible-light-driven oxidation of organic contaminants in air with goldnanoparticle catalysts on oxide supports. Angew. Chem. Int. Ed., 47,5353–5356.

84 Hou, W.B., Hung, W.H., Pavaskar, P., Goeppert, A., Aykol, M., andCronin, S.B. (2011) Photocatalytic conversion of CO2 to hydrocarbon fuelsvia plasmon-enhanced absorption and metallic interband transitions. ACSCatal., 1, 929–936.

85 Wang, C.J., Ranasingha, O., Natesakhawat, S., Ohodnicki, P.R., Andio, M.,Lewis, J.P., and Matranga, C. (2013) Visible light plasmonic heating ofAu—ZnO for the catalytic reduction of CO2. Nanoscale, 5, 6968–6974.

86 Mohamed, R.M. and Aazam, E.S. (2011) Characterization and catalyticproperties of nano-sized Au metal catalyst on titanium containing highmesoporous silica (Ti-HMS) synthesized by photo-assisted deposition andimpregnation methods. Int. J. Photoenergy, 2011, 7.

87 Trammell, S.A., Nita, R., Moore, M., Zabetakis, D., Chang, E., andKnight, D.A. (2012) Accelerating the initial rate of hydrolysis of methylparathion with laser excitation using monolayer protected 10 nm Aunanoparticles capped with a Cu(bpy) catalyst. Chem. Commun., 48,4121–4123.


88 Silva, C.G., Juarez, R., Marino, T., Molinari, R., and Garcia, H. (2011) Influ-ence of excitation wavelength (UV or visible light) on the photocatalyticactivity of titania containing gold nanoparticles for the generation of hydro-gen or oxygen from water. J. Am. Chem. Soc., 133, 595–602.

89 Son, M.S., Im, J.E., Wang, K.K., Oh, S.L., Kim, Y.R., and Yoo, K.H. (2010)Surface plasmon enhanced photoconductance and single electron effects inmesoporous titania nanofibers loaded with gold nanoparticles. Appl. Phys.Lett., 96, 023115.

90 Nishijima, Y., Ueno, K., Yokata, Y., Murakoshi, K., and Misawa, H. (2010)Plasmon-assisted photocurrent generation from visible to near-infraredwavelength using a Au-nanorods/TiO2 electrode. J. Phys. Chem. Lett., 1,2031–2036.

91 Leong, K.H., Chu, H.Y., Ibrahim, S., and Saravanan, P. (2015) Palladiumnanoparticles anchored to anatase TiO2 for enhanced surface plasmonresonance-stimulated, visible-light-driven photocatalytic activity. BeilsteinJ. Nanotechnol., 6, 428–437.

92 Zielinska-Jureka, A., Wei, Z., Wysocka, I., Szweda, P., and Kowalska, E.(2015) The effect of nanoparticles size on photocatalytic and antimicrobialproperties of Ag—Pt/TiO2 photocatalysts. Appl. Surf. Sci., 353, 317–325.

93 Liu, E.Z., Qi, L.L., Bian, J.J., Chen, Y.H., Hu, X.Y., Fan, J., Liu, H.C., Zhu, C.J.,and Wang, Q.P. (2015) A facile strategy to fabricate plasmonic Cu modifiedTiO2 nano-flower films for photocatalytic reduction of CO2 to methanol.Mater. Res. Bull., 68, 203–209.

94 Kouame, N.A., Alaoui, O.T., Herissan, A., Larios, E., Jose-Yacaman, M.,Etcheberry, A., Colbeau-Justin, C., and Remita, H. (2015) Visiblelight-induced photocatalytic activity of modified titanium(IV) oxide withzero-valent bismuth clusters. New J. Chem., 39, 2316–2322.

95 Wu, X.Y., Yin, S., Liu, B., Kobayashi, M., Kakihana, M., and Sato, T. (2014)A carbon modified NaTaO3 mesocrystal nanoparticle with excellentefficiency of visible light induced photocatalysis. J. Mater. Chem. A, 2,20832–20840.

96 O’Regan, B. and Graetzel, M. (1991) A low-cost, high-efficiency solarcell based on dye-sensitized colloidal titanium dioxide films. Nature(London, U. K.), 353, 737–740.

97 Wu, Y.Z. and Zhu, W.H. (2013) Organic sensitizers from D-pi-A toD-A-pi-A: effect of the internal electron-withdrawing units on molecularabsorption, energy levels and photovoltaic performances. Chem. Soc. Rev.,42, 2039–2058.

98 Planells, M., Pelleja, L., Clifford, J.N., Pastore, M., De Angelis, F., Lopez, N.,Marder, S.R., and Palomares, E. (2011) Energy levels, charge injection,charge recombination and dye regeneration dynamics for donor–acceptorpi-conjugated organic dyes in mesoscopic TiO2 sensitized solar cells. EnergyEnviron. Sci., 4, 1820–1829.

99 Odobel, F., Pellegrin, Y., and Warnan, J. (2013) Bio-inspired artificiallight-harvesting antennas for enhancement of solar energy capture indye-sensitized solar cells. Energy Environ. Sci., 6, 2041–2052.

References 481

100 Chou, C.S., Guo, M.G., Liu, K.H., and Chen, Y.S. (2012) Preparation ofTiO2 particles and their applications in the light scattering layer of adye-sensitized solar cell. Appl. Energy, 92, 224–233.

101 Kowalska, E., Yoshiiri, K., Wei, Z., Zheng, S., Kastl, E., Remita, H., Rau, S.,and Ohtani, B. (2015) Hybrid photocatalysts composed of titania modifiedwith plasmonic nanoparticles and ruthenium complexes for decompositionof organic compounds. Appl. Catal., B, 178, 133–143.

102 Abe, R., Shinmei, K., Koumura, N., Hara, K., and Ohtani, B. (2013)Visible-light-induced water splitting based on two-step photoexcitationbetween dye-sensitized layered niobate and tungsten oxide photocatalysts inthe presence of a triiodide/iodide shuttle redox mediator. J. Am. Chem. Soc.,135, 16872–16884.

103 Jiang, D., Xu, Y., Hou, B., Wu, D., and Sun, Y.H. (2007) Synthesis of visiblelight-activated TiO2 photocatalyst via surface organic modification. J. SolidState Chem., 180, 1787–1791.

104 Moongraksathum, B., Hsu, P.T., and Chen, Y.W. (2016) Photocatalytic activityof ascorbic acid-modified TiO2 sol prepared by the peroxo sol–gel method.J. Sol–Gel Sci. Technol., 78, 647–659.

105 Zou, W.W., Zhang, J.L., and Chen, F. (2010) Preparation, characterizationand application of TiO2 nanoparticles surface-modified by DDAT. Mater.Lett., 64, 1710–1712.

106 Mitoraj, D. and Kisch, H. (2010) On the mechanism of urea-induced titaniamodification. Chem. Eur. J., 16, 261–269.

107 Gopidas, K.R., Bohorquez, M., and Kamat, P.V. (1990) J. Phys. Chem., 94,6436.

108 Liu, D. and Kamat, P.V. (1993) Photoelectrochemical behavior of thinCdSe and coupled TiO2/CdSe semiconductor films. J. Phys. Chem., 97,10769–10773.

109 Abe, R., Takami, H., Murakami, N., and Ohtani, B. (2008) Pristine simpleoxides as visible light driven photocatalysts: highly efficient decomposition oforganic compounds over platinum-loaded tungsten oxide. J. Am. Chem. Soc.,130, 7780–7781.

110 Lalitha, K., Sadanandam, G., Kumari, V.D., Subrahmanyam, M., Sreedhar, B.,and Hebalkar, N.Y. (2010) Highly stabilized and finely dispersed Cu2O/TiO2:a promising visible sensitive photocatalyst for continuous production ofhydrogen from glycerol:water mixtures. J. Phys. Chem. C, 114, 22181–22189.

111 Amano, F., Nogami, K., and Ohtani, B. (2009) Visible light-responsivebismuth tungstate photocatalysts: effects of hierarchical architecture onphotocatalytic activity. J. Phys. Chem. C, 113, 1536–1542.

112 Zhang, J.Y., Xiao, F.X., Xiao, G.C., and Liu, B. (2016) Linker-assisted assem-bly of 1D TiO2 nanobelts/3D CdS nanospheres hybrid heterostructure asefficient visible light photocatalyst. Appl. Catal., A, 521, 50–56.

113 Wang, J. and Zhang, W.D. (2012) Modification of TiO2 nanorod arrays bygraphite-like C3N4 with high visible light photoelectrochemical activity.Electrochim. Acta, 71, 10–16.


114 Hitoki, G., Takata, T., Kondo, J.N., Hara, M., Kobayashi, H., and Domen, K.(2002) An oxynitride, TaON, as an efficient water oxidation photocata-lyst under visible light irradiation (lambda≤ 500 nm). Chem. Commun.,1698–1699.

115 Hitoki, G., Takata, T., Kondo, J.N., Hara, M., Kobayashi, H., and Domen, K.(2002) (Oxy)nitrides as new photocatalysts for water splitting under visiblelight irradiation. Electrochemistry, 70, 463–465.

116 Chun, W.J., Ishikawa, A., Fujisawa, H., Takata, T., Kondo, J.N., Hara, M.,Kawai, M., Matsumoto, Y., and Domen, K. (2003) Conduction and valenceband positions of Ta2O5, TaON, and Ta3N5 by UPS and electrochemicalmethods. J. Phys. Chem. B, 107, 1798–1803.

117 Hitoki, G., Ishikawa, A., Takata, T., Kondo, J.N., Hara, M., and Domen, K.(2002) Ta3N5 as a novel visible light-driven photocatalyst (lambda< 600 nm).Chem. Lett., 736–737.

118 Kasahara, A., Nukumizu, K., Hitoki, G., Takata, T., Kondo, J.N., Hara, M.,Kobayashi, H., and Domen, K. (2002) Photoreactions on LaTiO2N undervisible light irradiation. J. Phys. Chem. A, 106, 6750–6753.

119 Yamasita, D., Takata, T., Hara, M., Kondo, J.N., and Domen, K. (2004)Recent progress of visible-light-driven heterogeneous photocatalysts foroverall water splitting. Solid State Ionics, 172, 591–595.

120 Sato, J., Saito, N., Yamada, Y., Maeda, K., Takata, T., Kondo, J.N., Hara, M.,Kobayashi, H., Domen, K., and Inoue, Y. (2005) RuO2-loaded beta-Ge3N4 asa non-oxide photocatalyst for overall water splitting. J. Am. Chem. Soc., 127,4150–4151.

121 Maeda, K., Takata, T., Hara, M., Saito, N., Inoue, Y., Kobayashi, H.,and Domen, K. (2005) GaN:ZnO solid solution as a photocatalyst forvisible-light-driven overall water splitting. J. Am. Chem. Soc., 127,8286–8287.

122 Nukumizu, K., Nunoshige, J., Takata, T., Kondo, J.N., Hara, M.,Kobayashi, H., and Domen, K. (2003) TiNxOyFz as a stable photocatalystfor water oxidation in visible light (<570 nm). Chem. Lett., 32, 196–197.

123 Liu, Z., Zhao, Z.-G., and Miyauchi, M. (2009) Efficient visible light activeCaFe2O4/WO3 based composite photocatalysts: effect of interfacial modifica-tion. J. Phys. Chem. C, 113, 17132–17137.

124 Wang, Z., Hou, J.G., Jiao, S.Q., Huang, K., and Zhu, H.M. (2012) In situchemical reduction of the Ta3N5 quantum dots coupled TaON hollowspheres heterojunction photocatalyst for water oxidation. J. Mater. Chem.,22, 21972–21978.

125 Chen, S.S., Qi, Y., Hisatomi, T., Ding, Q., Asai, T., Li, Z., Ma, S.S.K.,Zhang, F.X., Domen, K., and Li, C. (2015) Efficient visible-light-drivenZ-scheme overall water splitting using a MgTa2O6−xNy/TaON heterostruc-ture photocatalyst for H2 evolution. Angew. Chem. Int. Ed., 54, 8498–8501.

126 Kim, E.S., Nishimura, N., Magesh, G., Kim, J.Y., Jang, J.W., Jun, H.,Kubota, J., Domen, K., and Lee, J.S. (2013) Fabrication of CaFe2O4/TaONheterojunction photoanode for photoelectrochemical water oxidation. J. Am.Chem. Soc., 135, 5375–5383.

References 483

127 Horiuchi, Y., Toyao, T., Takeuchi, M., Matsuoka, M., and Anpo, M. (2013)Recent advances in visible-light-responsive photocatalysts for hydrogenproduction and solar energy conversion – from semiconducting TiO2 toMOF/PCP photocatalysts. Phys. Chem. Chem. Phys., 15, 13243–13253.

128 Abe, R. (2010) Recent progress on photocatalytic and photoelectrochemicalwater splitting under visible light irradiation. J. Photochem. Photobiol., C, 11,179–209.

129 Sayama, K., Mukasa, K., Abe, R., Abe, Y., and Arakawa, H. (2001) Stoi-chiometric water splitting into H2 and O2 using a mixture of two differentphotocatalysts and an IO3−/I− shuttle redox mediator under visible lightirradiation. Chem. Commun., 2416–2417.

130 Higashi, M., Abe, R., Ishikawa, A., Takata, T., Ohtani, B., and Domen, K.(2008) Z-scheme overall water splitting on modified-TaON photocatalystsunder visible light (lambda< 500 nm). Chem. Lett., 37, 138–139.

131 Higashi, M., Abe, R., Teramura, K., Takata, T., Ohtani, B., and Domen, K.(2008) Two step water splitting into H2 and O2 under visible light byATaO(2)N (A=Ca, Sr, Ba) and WO3 with IO3−/I− shuttle redox mediator.Chem. Phys. Lett., 452, 120–123.

132 Maeda, K., Higashi, M., Lu, D.L., Abe, R., and Domen, K. (2010) Efficientnonsacrificial water splitting through two-step photoexcitation by visiblelight using a modified oxynitride as a hydrogen evolution photocatalyst.J. Am. Chem. Soc., 132, 5858–5868.

133 Tabata, M., Maeda, K., Higashi, M., Lu, D.L., Takata, T., Abe, R., andDomen, K. (2010) Modified Ta3N5 powder as a photocatalyst for O2 evo-lution in a two-step water splitting system with an iodate/iodide shuttleredox mediator under visible light. Langmuir, 26, 9161–9165.

134 Sasaki, Y., Iwase, A., Kato, H., and Kudo, A. (2008) The effect of co-catalystfor Z-scheme photocatalysis systems with an Fe3+/Fe2+ electron mediator onoverall water splitting under visible light irradiation. J. Catal., 259, 133–137.

135 Abe, R., Takata, T., Sugihara, H., and Domen, K. (2005) Photocatalytic over-all water splitting under visible light by TaON and WO3 with an IO3−/I−shuttle redox mediator. Chem. Commun., 3829–3831.

136 Kato, H., Hori, M., Konta, R., Shimodaira, Y., and Kudo, A. (2004) Con-struction of Z-scheme type heterogeneous photocatalysis systems for watersplitting into H2 and O2 under visible light irradiation. Chem. Lett., 33,1348–1349.

137 Sasaki, Y., Nemoto, H., Saito, K., and Kudo, A. (2009) Solar water splittingusing powdered photocatalysts driven by Z-schematic interparticle electrontransfer without an electron mediator. J. Phys. Chem. C, 113, 17536–17542.

138 Zheng, S.Z., Wang, K.L., Wei, Z.S., Yoshiiri, K., Braumuller, M., Rau, S.,Ohtani, B., and Kowalska, E. (2016) Mono- and dual-modified titania with aruthenium(II) complex and silver nanoparticles for photocatalytic degrada-tion of organic compounds. J. Adv. Oxid. Technol., 19, 208–217.

139 Zheng, S., Wei, Z., Yoshiiri, K., Braumuller, M., Ohtani, B., Rau, S., andKowalska, E. (2016) Titania modification with ruthenium(II) complex andgold nanoparticles for photocatalytic degradation of organic compounds.Photochem. Photobiol. Sci., 15, 69–79.


140 Cao, Y.Q., Yu, Y.L., Zhang, P., Zhang, L.L., He, T., and Cao, Y. (2013) Anenhanced visible-light photocatalytic activity of TiO2 by nitrogen andnickel–chlorine modification. Sep. Purif. Technol., 104, 256–262.

141 Zhang, Y., Zhang, P., Huo, Y.N., Zhang, D.Q., Li, G.S., and Li, H.X. (2012)Ethanol supercritical route for fabricating bimodal carbon modified meso-porous TiO2 with enhanced photocatalytic capability in degrading phenol.Appl. Catal., B, 115, 236–244.

142 Mumtaz, A., Mohamed, N.M., Mazhar, M., Ehsan, M.A., andSaheed, M.S.M. (2016) Core-shell vanadium modified titania@beta-In2S3hybrid nanorod arrays for superior interface stability and photochemicalactivity. ACS Appl. Mater. Interfaces, 8, 9037–9049.

143 Hou, Y., Zuo, F., Dagg, A., and Feng, P.Y. (2012) Visible light-drivenalpha-Fe2O3 nanorod/graphene/BiV1−xMoxO4 core/shell heterojunction arrayfor efficient photoelectrochemical water splitting. Nano Lett., 12, 6464–6473.

485

17

Roles of the Active Species Generated duringPhotocatalysisMats Jonsson

KTH Royal Institute of Technology, Applied Physical Chemistry, School of Chemical Science and Engineering,Teknikringen 30, SE-100 44 Stockholm, Sweden

17.1 Introduction

Ever since the landmark paper by Fujishima and Honda in 1972, heterogeneousphotocatalysis has attracted considerable attention in several different fields [1].One of the major forecasted applications of heterogeneous photocatalysis is waterpurification. In particular, heterogeneous photocatalysis using TiO2 has provento be successful in degrading organic pollutants that are difficult to degrade byother means [2]. One of the main advantages of this technology is the efficientdegradation of pollutants without any chemical additives. If driven by sunlight,the technology becomes truly environmentally sustainable. However, the tech-nology has some inherent limitations as well. One of them is that the majorityof the reactions contributing to the desired degradation of pollutants occur at orvery close to the surface of the photocatalyst. This makes it challenging to designreactors for handling large volumes of wastewater. It is essential to maximize thesurface area to solution volume ratio to reduce the residence time. At the sametime, fulfilling this criterion may result in a design where efficient light absorptionbecomes difficult.

The utilization of heterogeneous photocatalysis in water purification relieson the reactivity of the active species produced in the process. Hence, mech-anistic understanding is crucial for process optimization. The mechanisticunderstanding can be divided into two parts: (i) Mechanistic understandingof the fundamentals of heterogeneous photocatalysis in aqueous systems and(ii) Understanding the impact of the specific system on the overall mechanism.The latter implies understanding of how solutes and solution pH affects thechemistry induced by photocatalysis. The nature, identity, and yield of the activespecies are tightly connected to both parts. In this chapter, the potential role ofthe active species generated during heterogeneous photocatalysis of aqueoussystems will be discussed.


486 17 Roles of the Active Species Generated during Photocatalysis

17.2 Mechanism of Photocatalysis in TiO2/WaterSystems

The primary event in heterogeneous TiO2 photocatalysis is the photon-inducedcharge separation into a hole–electron pair (i). This process is followed by anumber of other processes such as charge recombination (ii) and eventually alsointerfacial charge transfer ((iii) and (iv)). The latter two are the key processes inthe chemistry induced by photocatalysis. Interfacial charge transfer can occurbetween the positive hole as well as the electron and species adsorbed to thesurface of the photocatalyst. Hence, the nature of the primary reactive specieswill depend on the surface affinity of solutes present in the aqueous phase. Theprimary processes are summarized below.

Scheme

TiO2 + light → e− + h+ (17.1)e− + h+ → heat∕light (17.2)h+ + D → D⋅+ (17.3)e− + A → A⋅− (17.4)

The relative surface coverage will depend on the concentration and adsorptionequilibrium constant. It should be noted that the mechanism of interfacial chargetransfer has been extensively discussed over the years and the purpose of thischapter is not to repeat this discussion.

17.3 Active Species Generated at the Catalyst/WaterInterface

In general, both the electron and the positive hole display reactivity towardsolutes and solvents adsorbed to the surface. By scavenging or trapping theelectron, the lifetime of the positive hole is significantly prolonged since theprobability for hole–electron recombination is reduced. A good electronscavenger is a solute with high electron affinity. In aqueous systems, the mostcommonly occurring electron scavenger is molecular oxygen, which is essen-tially always present unless you actively try to remove it. In the majority ofpractical applications, we can always take for granted that oxygen is present. Inair-saturated water, the rate of oxidative photocatalytic degradation increasesby a factor of 5 compared to an oxygen-free solution. The initial productformed upon electron scavenging by molecular oxygen is the superoxide radicalanion, O2

∙−. In aqueous solution, the superoxide radical anion can undergo thefollowing reactions:

O2∙− + H+ → HO2

∙ (17.5)O2

∙− + O2∙− + 2H2O → O2 + H2O2 + 2OH− (17.6)

HO2∙ + HO2

∙ → O2 + H2O2 (17.7)

17.3 Active Species Generated at the Catalyst/Water Interface 487

The relative contribution from the different reactions will depend on the pH ofthe solution. In aqueous systems containing oxides, superoxide anions have beendetected at the surface of the oxide. However, this does not exclude that super-oxide anions or hydroperoxyl radicals are also present in the solution. In general,superoxide can act both as an oxidant and as a reductant. The reactivity is in gen-eral quite low, and this allows superoxide to be accumulated in the system and todiffuse significant distances from where it was originally formed. The protonatedform, HO2

∙, is considerably more reactive as oxidant, and it also undergoes dis-proportionation to form O2 and H2O2 much more rapidly than its deprotonatedcounterpart.

H2O2 formed in the disproportionation of HO2∙ and O2

∙− is also an efficientelectron scavenger. What makes H2O2 even more interesting as an electron scav-enger is its mode of reaction upon one-electron reduction. This is displayed inthe following reaction:

H2O2 + e− → HO∙ + HO− (17.8)

As can be seen, this reaction produces the strongly oxidizing hydroxyl radi-cal that will contribute to the oxidizing power of the system, provided oxidationis the desired process. The system will reach steady state with respect to H2O2when the rate of H2O2 formation is balanced by the rate of H2O2 consumption.Consumption primarily occurs through reduction by the electron as describedabove and through oxidation by the photocatalytically produced oxidant. As willbe discussed later, there are also other reactions contributing to the consumptionof H2O2.

The concentration of H2O2 produced from the disproportionation reactionof superoxide or the hydroperoxyl radical in aerated system is fairly low. Thesteady-state concentration of H2O2 in an air-saturated system is typically below1 μM [3]. To utilize the beneficial effects of H2O2 in a more efficient way, H2O2must be added to the system. By adding a concentration corresponding to theoxygen concentration in an air-saturated system, the rate of oxidative photocat-alytic degradation increases by a factor of 2 compared to the oxygen-containingsystem [3].

In aqueous systems, the positive hole is assumed to be reduced by H2O orOH− adsorbed to the surface. This leads to the formation of hydroxyl radicals.Whether hydroxyl radicals are produced in this process or not has been debatedfor a long time [4–17]. Arguments have been put forward claiming that it isthermodynamically impossible to produce the hydroxyl radical under these con-ditions. Yet, many studies have presented experimental evidence pointing towardthe hydroxyl radical.

Among the studies where evidence for the involvement of the hydroxyl radi-cal has been presented, we find studies where hydroxylated products have beenquantified. In principle, hydroxylated products can also be formed upon directoxidation followed by hydrolysis. However, the isomeric product distributionsformed in these two reactions usually differ to some extent. For this reason, com-plete product analysis can be used to verify the existence of the hydroxyl radical.

Goldstein et al. studied the hydroxylation of phenol under the followingconditions [18]: (i) 𝛾-radiolysis, (ii) reaction with SO4

∙−, and (iii) photocatalysis


in the presence of TiO2 colloid particles. The experiments were carried out inthe presence of cupric ions in order to facilitate oxidation of initially formedOH-adducts into stable products. The authors concluded that hydroxyla-tion of relatively low concentrations of phenol in the photocatalytic systemtakes place via hydroxyl radical transfer from the immobile surface oxidant{TiIV–O2–TiIV}–OH∙. At higher concentrations of phenol, adsorption becomessignificant and phenol is oxidized directly by positive holes. The authors clearlyemphasized that hydroxylation occurs via surface-bound hydroxyl radicals andnot by free hydroxyl radicals. Surface-bound hydroxyl radicals would requireless energy to be produced than free hydroxyl radicals. Hence, the photocatalyticformation of the hydroxyl radical may be thermodynamically favorable after all.It should be kept in mind that the reactivity of the adsorbed hydroxyl radical isexpected to be quite different from that of the free hydroxyl radical. The latteris expected to be considerably more reactive and thereby also less selective. Ina study by Lawless et al., the properties of the surface-bound hydroxyl radicalwere investigated experimentally [19]. It is interesting to note that these authorsestimate the reduction potential of the surface-bound hydroxyl radical to beabout 1.5 V versus SHE. This is approximately 400 mV lower than the reductionpotential of the free hydroxyl radical.

Recently, Liao and Reitberger used the strongly chemiluminescent3-hydroxyphthalic hydrazide formed upon hydroxylation of phthalic hydrazideto probe the existence of hydroxyl radicals in TiO2 photocatalysis [20]. On thebasis of their experimental results and a discussion based on previously publisheddata and theories, the authors argue for the formation of free hydroxyl radicals inthe system. The main argument being that 3-hydroxyphthalic hydrazide can onlybe formed in the reaction between free phthalic hydrazide and free hydroxylradicals. Under conditions where there is significant adsorption of phthalichydrazide to the TiO2 particles, very little chemiluminescence is observed.However, as pointed out by Goldstein et al., hydroxylation could very well occuras a result of the reaction between the surface-bound hydroxyl radical and theprobe molecule. Goldstein et al. report that direct oxidation by the positivehole occurs at higher phenol concentration which is analogous with conditionswhere there is significant adsorption. Hence, both studies report similar trendsbut draw different conclusions regarding the nature of the hydroxylating species.Liao and Reitberger also report on radiation chemical experiments performedin the presence of 100 ppm TiO2 particles compared to experimental results ofa system with no TiO2 particles. The two sets of experiments did not displayany difference in the observed chemiluminescence at a phthalic hydrazideconcentration of 0.5 mM. From this result, the authors concluded that freehydroxyl radicals are not scavenged by TiO2 particles in the system. It shouldbe noted that there is experimental evidence that hydroxyl radicals producedupon radiolysis of water can in fact be scavenged by oxide particles. This wasfirst demonstrated by Lawless et al. for TiO2 using pulse radiolysis [19]. Veryrecently, this was also demonstrated for TiO2, ZrO2, and Y2O3 using competitionkinetics [21]. However, the conditions under which these experiments wereperformed differ from those for the experiment by Liao and Reitberger. Themain difference is that the amount of oxide is significantly higher which probably

17.3 Active Species Generated at the Catalyst/Water Interface 489

explains why the scavenging of hydroxyl radicals is actually seen. The existenceof surface-bound hydroxyl radicals will be further discussed below.

Another approach to identify the reactive oxidant was to quantify the amountof H2O2 formed in TiO2 photocatalysis under different conditions [22]. As shownabove, H2O2 will be formed as a consequence of the reduction of molecular oxy-gen by the electron. However, in oxygen-free aqueous solutions, this reactionpath is not an option and we would not expect any H2O2 to be formed. In aseries of experiments, it was shown that H2O2 was formed also in N2-purgedsystems where the concentration of molecular oxygen is very low. Indeed, thepresence of H2O2 could always be argued due to traces of oxygen in the system.However, the steady-state levels of H2O2 did not differ sufficiently between theair-saturated and the N2-purged systems to support that explanation. To con-firm that the formation of H2O2 was indeed attributable to hydroxyl radicals,experiments were also performed with hydroxyl radical scavengers. Interestingly,it was shown that the formation of hydrogen peroxide decreased to undetectableamounts in the presence of a scavenger in N2-purged systems, while the oppo-site effect was observed in air-saturated systems. The latter is attributed to thefact that hydroxyl radicals consume hydrogen peroxide in general. Therefore, thehydroxyl radical scavenger has a protecting effect that results in increased con-centrations of hydrogen peroxide as long as there is another path for productionof H2O2. In aqueous TiO2 systems, the hydrogen peroxide is distributed betweenthe solution and the TiO2 surface. A significant fraction of the H2O2 is adsorbedto the surface.

It is worth noting that H2O2 adsorbs to most oxide surfaces. It has also beenknown for quite some time that H2O2 is catalytically decomposed on oxidesurfaces to produce H2O and O2. It was recently verified that the mechanismfor the catalytic decomposition of H2O2 involves the intermediate formationof surface-bound hydroxyl radicals [23]. The complete mechanism is depictedbelow.

H2O2 → 2OH∙ (17.9)OH∙ + H2O2 → H2O + HO2

∙ (17.10)HO2

∙ + HO2∙ → H2O2 + O2 (17.11)

This process occurs also on TiO2 although the efficiency of TiO2 as a catalyst forH2O2 decomposition is fairly low compared to many other oxides, such as ZrO2[24]. This is well in line with the observed photocatalytic formation of H2O2 inoxygen-free systems. If TiO2 was a better catalyst for H2O2 decomposition, H2O2would not be detected. In general, the catalysis of H2O2 decomposition on oxidesis solely attributed to the fact that the hydroxyl radical is much more stronglyadsorbed to the surface than is H2O2. Hence, the existence of surface-boundhydroxyl radicals is a prerequisite for the catalytic decomposition of H2O2 tooccur. The lower catalytic effect of TiO2 implies that the difference in adsorptionenergy between the hydroxyl radical and hydrogen peroxide is smaller thanfor other oxides. This, in turn, would indicate that the surface-bound hydroxylradicals on TiO2 are more reactive than surface-bound hydroxyl radicals onmany other oxides. This feature also contributes to the photocatalytic propertiesof TiO2.


The significantly lower affinity of the hydroxyl radicals to TiO2 compared tomany other oxides is also supported by systematic density functional theory(DFT) calculations on a number of oxide systems as well as by the radiationchemical experiments mentioned above [24]. The adsorption energy of thehydroxyl radical to metal oxides appears to be governed by the electronegativityof the metal atom [25]. However, it should be noted that, while the adsorptionof hydroxyl radicals to TiO2 is weaker than the adsorption to many other oxides,adsorption of the hydroxyl radical is still a highly exothermic process [24]. Thisis well in line with the proposed difference in reduction potential of the free andthe surface-bound species [19].

17.4 Oxidative Degradation of Solutes Present in theAqueous Phase

As the main reactions of interest take place at the photocatalyst surface, a veryimportant property of the solute to be degraded is its affinity to the photocatalystsurface. The rate of degradation will then depend on surface coverage of thesubstrate as well as on the chemical reactivity of the adsorbed substrate towardthe primary oxidizing species. This makes it a bit more complicated to comparedegradation of different substrates than if it was simply a question of comparingreactivity in solution. At a given substrate concentration, the rate of degra-dation could differ considerably due to differences in adsorption equilibriumconstant. The kinetics of solute degradation can therefore often be described asLangmuir–Hinshelwood kinetics [26].

r = k𝜃 (17.12)

where the surface coverage, 𝜃, is given by

𝜃 =KLHC

1 + KLHC(17.13)

Here KLH denotes the adsorption equilibrium constant and C is the equilibriumconcentration of the reactive solute. By combining these equations, we obtain

r = −dCdt

=kKLHC

1 + KLHC(17.14)

From this follows that the rate of degradation initially increases linearly withincreasing solute concentration. At the point where the surface becomes sat-urated with solute, the rate of degradation becomes independent of the soluteconcentration. Under these conditions, the maximum capacity of the photocat-alytic system is utilized. This is not the case if the solute absorbs the incident light.In this case, the efficiency will actually decrease with increasing solute concen-tration. It is important to emphasize that the photocatalytic efficiency should bequantified in terms of a quantum yield or the rate of degradation under a givenset of conditions.

The efficiency of a photocatalytic system is often determined using a chemicalprobe molecule. To be able to compare photocatalytic efficiencies determined

17.4 Oxidative Degradation of Solutes Present in the Aqueous Phase 491

using different probe molecules, it is essential to understand the mechanismof degradation. A probe that has been used quite extensively both to quan-tify surface-bound hydroxyl radicals produced upon catalytic decompositionof H2O2 on oxide surfaces and to determine photocatalytic efficiency isTris(hydroxymethyl)aminomethane [23, 27]. Upon hydrogen abstraction fromTris, formaldehyde is produced. This product is stable and can be readilyquantified. Tris is a well-known pH buffer which makes it possible also to controlthe pH of the system. Methanol can be used in the same way yielding the sameproduct, formaldehyde. The effects of pH and the presence of oxygen on thesetwo probe molecules have recently been quantified quite thoroughly [28].

It is evident that the presence of O2 facilitates the formation of formaldehydefrom both probes. The effect of O2 is larger for the methanol system than for theTris system. The yield of formaldehyde was also shown to increase with increasingpH in the Tris system. This should be accounted for when using Tris or methanolas probes.

Phenols are commonly used as probe molecules [29–33]. This type ofcompound is particularly relevant also as a commonly occurring pollutant inaqueous systems. In general, oxidation of a phenol or phenolate leads to the for-mation of the corresponding phenoxyl radical. The phenoxyl radical can undergoradical–radical combination with another phenoxyl radical. However, theprobability for this process to occur is relatively low unless the rate of phenoxylradical production is high. A competing process in oxygen-containing systems isthe reaction between the phenoxyl radical and the superoxide radical anion [34,35]. The latter being formed upon trapping of the electron from the conductionband. Phenoxyl radicals and superoxide can react in two ways. The first modeof reaction is radical–radical combination eventually leading to a ring-openingreaction and efficient consumption of the phenol. The second mode of reaction iselectron transfer to form molecular oxygen and phenol/phenolate. This reactioncounteracts the photocatalytic consumption of phenol. Hence, it is quite evidentthat the superoxide radical anion is a key reactant in phenol degradation. Thereactions involving the phenoxyl radical are depicted below [36]:

PhO− + ox → PhO∙ + red (17.15)2PhO∙ → (PhO)2 (17.16)PhO∙ + O2

∙− → (PhO)O2− (17.17)

PhO∙ + O2∙− → PhO− + O2 (17.18)

The relative rates of these reactions depend on the properties of the phenoxylradical. For phenoxyl radicals with a low standard reduction potential, corre-sponding to easily oxidized phenols, the combination reaction between the phe-noxyl radical and superoxide dominates over the electron-transfer reaction [34].Consequently, easily oxidized phenols are efficiently degraded by photocataly-sis. For phenoxyl radicals with relatively high standard reduction potential, cor-responding to phenols that are more difficult to oxidize, the electron-transferreaction gains considerably in importance. Hence, phenols that are more difficultto oxidize are also less efficiently degraded by photocatalysis [36]. The standardreduction potential of phenoxyl radicals is strongly connected to the substituent


pattern and can be described as a linear function of the substituent constant [37].This must be accounted for when using phenols as probes for photocatalysis aswell as when using heterogeneous photocatalysis for degradation of phenolic pol-lutants in aqueous systems.

Dyes are also often used as probe molecules and they are present as pollutantsin industrial wastewaters. When employing heterogeneous photocatalysis as amethod to clean up water from dyes, the dye concentration must be low enoughto allow illumination of the photocatalyst. Under such conditions, full photocat-alytic efficiency may not have been reached due to low surface coverage of thedye. This can also be an issue when using dyes as probes, that is, that the max-imum photocatalytic efficiency may not be possible to assess. One dye that hasbeen studied quite extensively is indigo carmine [38]. The degradation of indigocarmine was studied using gamma radiolysis and heterogeneous photocatalysis.In radiolysis of water, several of the reactive species formed are identical with theones formed upon heterogeneous photocatalysis of aqueous systems. The radia-tion chemical experiment showed that below pH= 4, the degradation of indigocarmine can be equally attributed to HO2

∙/O2∙− and the hydroxyl radical. Hence,

HO2∙/O2

∙− is expected to be a significant reactant also in photocatalytic degra-dation of indigo carmine in this pH range [38]. Above pH= 4, the degradation ofindigo carmine is dominated by the hydroxyl radical.

For many organic pollutants, the first step in photocatalytic degradation is theformation of a carbon-centered radical. This radical can react further with molec-ular oxygen and produce a peroxyl radical. The further fate of the peroxyl radicalis closely connected to the structure of the corresponding carbon-centeredradical from which it was formed. HO-, RO-, H2N-, and R2N-substituents in𝛼-position relative to the radical center will make the radical susceptible tooxidation. Upon reaction with molecular oxygen, the corresponding peroxylradical will have very short lifetime. Instead, molecular oxygen will oxidize theC-centered radical. Under certain conditions, peroxyl radicals formed couldinduce further hydrogen abstraction and thereby sustain a chain reaction. Thisrequires the presence of weak C—H bonds. Hence, molecular oxygen is a keycomponent in heterogeneous photocatalysis.

17.5 Impact of H2O2 on Oxidative Degradationof Solutes Present in the Aqueous Phase

As mentioned above, H2O2 can enhance the rate of photocatalytic degradationmore substantially than molecular oxygen due to the electron trapping followedby the formation of hydroxyl radicals [3]. Unlike the hydroxyl radicals producedupon electron transfer between the positive hole and adsorbed H2O/OH−, thehydroxyl radicals formed upon electron capture by H2O2 are not necessarily sur-face bound. They may therefore display a different reactivity toward other solutespresent in the system. It should be noted that degradation of phenol is not nec-essarily enhanced if H2O2 is replacing molecular oxygen given the crucial role ofsuperoxide in this process. Indeed, O2

∙−/HO2∙ can be formed upon oxidation of

17.6 The Role of Common Anions Present in the Aqueous Phase 493

H2O2, but it must be kept in mind that this is a process competing with the forma-tion of phenoxyl radicals and the overall degradation efficiency is not expectedto increase. In a relatively recent study of the impact of H2O2 on the photocat-alytic degradation of Tris to form formaldehyde, the H2O2-concentration depen-dence was elucidated [3]. The observed trend is quite interesting. At fairly lowconcentrations, there is a significant increase in the rate of photocatalytic degra-dation of Tris with increasing H2O2 concentration. After reaching a maximum,the rate of degradation decreases quite strongly with H2O2 concentration. Thisis followed by a weaker concentration dependence where the degradation ratedecreases even further with increasing H2O2 concentration. It is interesting tonote that the H2O2-concentration dependence does not appear to depend verymuch on the concentration of Tris (at least within the studied concentrationrange). H2O2 adsorption experiments in the presence and absence of Tris alsoshow that Tris does not compete with H2O2 for adsorption sites.

Similar trends have been observed for substances with relatively low affinityfor TiO2, such as the safira dye [39]. It should be noted that Tris has a very lowaffinity for TiO2. For substances with higher affinity for TiO2, such as Triclopyr,the trend is different [40]. In this case, there is a clear competition for adsorptionsites between the substrate molecule and H2O2. Consequently, the effect of H2O2becomes more predictable under these conditions.

17.6 The Role of Common Anions Present in theAqueous Phase

As discussed, solutes present in the aqueous phase can react with the primaryradical species and form secondary radical species. This is also the case for someinorganic anions. In a previous study, the impact of HPO4

2−, B4O72−, SO4

2−, Cl−,HCO3

−/CO32−, and Br− on the photocatalytic efficiency was investigated [38].

Radiation chemical experiments, where the hydroxyl radical is produced uponradiolysis of water, showed that the efficiency in degrading indigo carmine isreduced in the presence of HCO3

−/CO32− and Br− indicating the conversion of

the reactive hydroxyl radical into significantly less reactive radicals according tothe following reactions:

OH∙ + HCO3−∕CO3

2− → H2O∕OH− + CO3∙− (17.19)

OH∙ + Br− → OH− + Br∙ (17.20)Br∙ + Br− → Br2

∙− (17.21)

The other anions had no measurable effect on the efficiency which is com-pletely in line with the known (low) reactivity of these anions toward thehydroxyl radical. TiO2 photocatalysis was also studied under the same con-ditions. In these experiments, it was obvious that the efficiency in degradingindigo carmine increased in the presence of HCO3

−/CO32− and Br−. Again, the

other anions showed no significant effect on the degradation efficiency. Giventhe lower reactivity (and reduction potential) of the surface-bound hydroxylradical compared to the free hydroxyl radical, it is reasonable to assume that


there is essentially no reaction with the other anions. Furthermore, on thebasis of the proposed reduction potential of the surface-bound hydroxyl radicaland the well-known reduction potentials of the radicals formed above, thesereactions are close to thermoneutral. As stated above, the radicals formed uponreaction between the hydroxyl radical and HCO3

−/CO32− and Br−(CO3

∙− andBr2

∙−) are less reactive than the hydroxyl radical. However, lower reactivityimplies a longer lifetime. These radicals are also more stable with respect toradical–radical recombination. As a consequence, the radicals produced at thesurface will have time to diffuse away from the surface. This means that theaffected volume is significantly larger in the presence of these ions. Indeed, thiseffect can only be expected when the solute being degraded displays reactivitytoward the surface-bound hydroxyl radicals as well as the secondary radicals.It has also been proposed that the enhanced efficiency could be attributed to amore efficient direct charge transfer between the TiO2 surface and the anions inquestion.

17.7 Summary of Active Species Presentin Heterogeneous Photocatalysis in Water

On the basis of numerous experimental studies, it appears reasonable to con-clude that the active oxidizing species is the surface-bound hydroxyl radical.In more concentrated solutions, it appears to be more probable to have directelectron transfer from adsorbed solute to the positive hole. The surface-boundhydroxyl radical is expected to be less reactive than the free hydroxyl radicalsince the adsorption has a stabilizing effect. In air-saturated aqueous solutions,the electron formed upon photolysis will be scavenged by molecular oxygen.This results in the formation of superoxide, which, depending on pH, can beprotonated to yield the hydroperoxyl radical. Both the superoxide radical anionand the hydroperoxyl radical can undergo disproportionation to yield molecularoxygen and hydrogen peroxide. Hydrogen peroxide can also be produced inoxygen-free systems as a result of recombination of surface-bound hydroxylradicals. Hydrogen peroxide present in a photocatalytic system can have twoeffects. It can scavenge electrons from the photocatalyst and thereby enhancethe efficiency. In connection to the scavenging of the electron, a hydroxyl radicalis formed that enhances the efficiency even more. As mentioned above, thehydroxyl radicals formed in this process are not necessarily surface bound andmay therefore not display the same reactivity as the hydroxyl radicals producedfrom the positive hole. The other type of effect of H2O2 is oxidation of hydrogenperoxide in competition with oxidation of the substrate. This leads to formationof hydroperoxyl radicals that could be beneficial for the system depending on thesubstrate. Substrates forming C-centered radicals upon photocatalytic oxidationare reactive toward molecular oxygen. This usually leads to the formation ofperoxyl radicals. These radicals may react further with other substrate moleculesand thereby induce a chain reaction. However, the efficiency of such a reactionis strongly dependent on the structure of the substrate.

References 495

Table 17.1 One-electron reduction potentials ofactive species commonly present or generated uponheterogeneous photocatalysis in aqueous systems.

SpeciesReduction potential (Eo)/V versus SHE

OHsurf∙ About 1.5 [19]

OHfree∙ 1.9 [41]

O2∙− 1.00 (at pH 0) [41]

HO2∙ 0.79 [41]

RO2∙ About 0.8 [42]

Br2∙− 1.65 [41]

CO3∙− 1.6 [41]

O2 −0.33 [41]H2O2 0.8 (pH 7) [41]

In aqueous solutions containing HCO3−/CO3

2− or Br−, the primary photocat-alytic oxidant is scavenged by anions. This results in less reactive radicals thatwill be capable of diffusing longer distances from the photocatalyst surface. Theproperties of the active species discussed here are summarized in Table 17.1.

References


2 Frank, S.N. and Bard, A.J. (1977) Heterogeneous photocatalytic oxidation ofcyanide and sulfite in aqueous solutions at semiconductor powders. J. Phys.Chem., 81, 1484–1488.

3 Diesen, V. and Jonsson, M. (2013) Effects of O2 and H2O2 on TiO2photocatalytic efficiency quantified by formaldehyde formation fromtris(hydroxymethyl)aminomethane. J. Adv. Oxid. Technol., 16, 16–22.

4 Grela, M., Brusa, M., and Colussi, A. (1999) Efficiency of hot carrier trappingby outer-sphere redox probes at quantum dot interfaces. J. Phys. Chem. B,103, 6400–6402.

5 Tachikawa, T., Tojo, S., Fujitsuka, M., and Majima, T. (2004) Direct observa-tion of the cascade hole transfer giving free radical cations of trans-stilbenesduring TiO2 photocatalytic reactions. Chem. Phys. Lett., 392, 50–54.

6 Yu, Z. and Chuang, S.S. (2007) In situ IR study of adsorbed species and pho-togenerated electrons during photocatalytic oxidation of ethanol on TiO2.J. Catal., 246, 118–126.

7 Murakami, Y., Endo, K., Ohta, I., Nosaka, A.Y., and Nosaka, Y. (2007) CanOH radicals diffuse from the UV-irradiated photocatalytic TiO2-surfaces?Laser-induced-fluorescence study. J. Phys. Chem. C, 111, 11339–11346.


8 Byrne, J.A., Eggins, B.R., Dunlop, P.S., and Linquette-Mailley, S. (1998) Theeffect of hole acceptors on the photocurrent response of particulate titaniumdioxide. Analyst, 123, 2007–2012.

9 Ishibashi, K.–.I., Fujishima, A., Watanabe, T., and Hashimoto, K. (2000)Quantum yields of active oxidative species formed on TiO2 photocatalysts.J. Photochem. Photobiol. A, 134, 139–142.

10 Salvador, P. (2007) On the nature of photogenerated radical species active inthe oxidative degradation of dissolved pollutants with TiO2 aqueous suspen-sions: a revision in the light of the electronic structure of adsorbed water.J. Phys. Chem. C, 111, 17038–17043.

11 Ishibashi, K.–.I., Fujishima, A., Watanabe, T., and Hashimoto, K. (2000)Detection of active oxidative species in TiO2 photocatalysis using the fluores-cence technique. Electrochem. Commun., 2, 207–210.

12 Raja, P., Bozzi, A., Mansilla, H., and Kiwi, J. (2005) Evidence for superoxide-radical anion, singlet oxygen and OH-radical intervention during the degra-dation of the lignin compound (3-methoxy-4-hydroxyphenylmethylcarbinol).J. Photochem. Photobiol. A, 169, 271–278.

13 Mora-Seró, I., Villarreal, T.L., Bisquert, J., Pitarch, Á., Gómez, R., andSalvador, P. (2005) Photoelectrochemical behavior of nanostructured TiO2thin-film electrodes in contact with aqueous electrolytes containing dissolvedpollutants: a model for distinguishing between direct and indirect interfa-cial hole transfer from photocurrent measurements. J. Phys. Chem. B, 109,3371–3380.

14 Nosaka, Y., Koenuma, K., Ushida, K., and Kira, A. (1996) Acetic acid on illu-minated TiO2 powder studied by means of in situ electron spin resonancemeasurements. Langmuir, 12, 736–738.

15 Kormali, P., Triantis, T., Dimotikali, D., Hiskia, A., and Papaconstantinou, E.(2006) On the photooxidative behavior of TiO2 and PW12O40

3−: OH radicalsversus holes. Appl. Catal., B, 2006 (68), 139–146.

16 Nosaka, Y., Komori, S., Yawata, K., Hirakawa, T., and Nosaka, A.Y. (2003)Photocatalytic OH radical formation in TiO2 aqueous suspension studied byseveral detection methods. Phys. Chem. Chem. Phys., 5, 4731–4735.

17 Carneiro, J.T., Savenije, T.J., Moulijn, J.A., and Mul, G. (2009) Toward a phys-ically sound structure-activity relationship for TiO2-based photocatalysts.J. Phys. Chem. C, 114, 327–332.

18 Goldstein, S., Czapski, G., and Rabani, J. (1994) Oxidation of phenol byradiolytically generated ∙OH and chemically generated SO4

∙−. a distinctionbetween ∙OH transfer and hole oxidation in the photolysis of TiO2 colloidsolution. J. Phys. Chem., 98, 6586–6591.

19 Lawless, D., Serpone, N., and Meisel, D. (1991) Role of hydroxyl radicals andtrapped holes in photocatalysis. A pulse radiolysis study. J. Phys. Chem., 95,5166–5170.

20 Liao, H. and Reitberger, T. (2013) Generation of free OHaq radicals by blacklight illumination of degussa (evonik) P25 TiO2 aqueous suspensions.Catalysts, 3 (2), 418–443.

References 497

21 Lousada, C.M., Johansson, A.J., Brinck, T., and Jonsson, M. (2012) Mechanismof H2O2 decomposition on transition metal oxide surfaces. J. Phys. Chem. C,116, 9533–9543.

22 Diesen, V. and Jonsson, M. (2014) Formation of H2O2 in TiO2 photocatalysisof oxygenated and deoxygenated aqueous systems – a probe for photocatalyti-cally produced hydroxyl radicals. J. Phys. Chem. C, 118, 10083–10087.

23 Lousada, C.M. and Jonsson, M. (2010) Kinetics, mechanism, and activationenergy of H2O2 decomposition on the surface of ZrO2. J. Phys. Chem. C,114, 11202–11208.

24 Lousada, C.M., Johansson, A.J., Brinck, T., and Jonsson, M. (2013) Reactivityof metal oxide clusters with hydrogen peroxide and water – a DFT studyevaluating the performance of different exchange-correlation functionals.Phys. Chem. Chem. Phys., 15, 5539–5552.

25 Lousada, C.M., Brinck, T., and Jonsson, M. (2015) Application of reactivitydescriptors to the catalytic decomposition of hydrogen peroxide at oxidesurfaces. Comput. Theor. Chem., 1070, 108–116.

26 Fox, M.A. and Dulay, M.T. (1993) Heterogeneous photocatalysis. Chem. Rev.,93, 341–357.

27 Diesen, V. and Jonsson, M. (2012) Tris(hydroxymethyl)aminomethane asa probe in heterogeneous TiO2 photocatalysis. J. Adv. Oxid. Technol., 15,392–398.

28 Yang, M. and Jonsson, M. (2014) Evaluation of the O2 and pH effects onprobes for surface bound hydroxyl radicals. J. Phys. Chem. C, 118, 7971–7979.

29 Sivalingam, G., Priya, M.H., and Madras, G. (2004) Kinetics of photodegrada-tion of substituted phenols by solution combustion synthesized TiO2. Appl.Catal. B, 51, 67–76.

30 Ryu, J. and Choi, W. (2008) Substrate-specific photocatlytic activities of TiO2and multiactivity test for water treatment application. Environ. Sci. Technol.,42, 294–300.

31 Cheng, Y., Sun, H., Jin, W., and Xu, N. (2007) Photocatalytic degradationof 4-chlorophenol with combustion synthesized TiO2 under visible lightirradiation. Chem. Eng. J., 128, 127–133.

32 Orlov, A., Jefferson, D.A., Macleod, N., and Lambert, R.M. (2004) Photocat-alytic properties of TiO2 modified with gold nanoparticles in the degradationof 4-chlorophenol in aqueous solution. Catal. Lett., 92, 41–47.

33 Serpone, N., Sauve, G., Koch, R., Tahiri, H., Pichat, P., Piccinini, P.,Pelizzetti, E., and Hidaka, H. (1996) Standardization protocol of processefficiencies and activation parameters in heterogeneous photocatalysis:relative photonic efficiencies 𝜁 . J. Photochem. Photobiol. A, 94, 191–203.

34 Jonsson, M., Lind, J., Reitberger, T., Eriksen, T., and Merenyi, G. (1993)Free-radical combination reactions involving phenoxyl radicals. J. Phys. Chem.,97, 8229–8233.

35 Merenyi, G., Lind, J., and Jonsson, M. (1993) Autoxidation of closed-shellorganics – an outer-sphere electron-transfer. J. Am. Chem. Soc., 115,4945–4946.


36 Diesen, V. and Jonsson, M. (2014) Comment on the use of phenols as probesfor the kinetics of heterogeneous photocatalysis. Appl. Catal. B, 158-159,429–431.

37 Jonsson, M., Lind, J., Eriksen, T.E., and Merenyi, G. (1993) O–H bondstrenghts and one-electron reduction potentials of multisubstituted phe-nols and phenoxyl radicals – predictions using free-energy relationships.J. Chem. Soc. Perkin Trans. 1, 2, 1567–1568.

38 Liao, H., Stenman, D., and Jonsson, M. (2009) Study of Indigo carmine as rad-ical probe in photocatalysis. J. Photochem. Photobiol. A, 202, 86–91.

39 Sauer, T., Neto, G.C., Jose, H.J., and Moreira, R.F.P.M. (2002) Kinetics of pho-tocatalytic degradation of reactive dyes in a TiO2 slurry reactor. J. Photochem.Photobiol. A, 149, 147–154.

40 Poulios, I., Kositizi, M., and Kouras, A. (1998) Photocatalytic decomposi-tion of triclopyr over aqueous semiconductor suspensions. J. Photochem.Photobiol. A, 115, 175–183.

41 Wardman, P. (1989) Reduction potentials of One-electron couples involvingfree radicals in aqueous solution. J. Phys. Chem. Ref. Data, 18, 1637–1755.

42 Jonsson, M. (1996) Thermochemical properties of peroxides and peroxyl radi-cals. J. Phys. Chem., 1996 (100), 6814–6818.

499

18

Visible-Light-Active Photocatalysis: NanostructuredCatalyst Design, Mechanisms, and ApplicationsRamachandran Vasant Kumar and Michael Coto

University of Cambridge, Department of Materials Science & Metallurgy, 27 Charles Babbage Road,Cambridge CB3 0FS, UK

18.1 Introduction

The photocatalytic process is initiated by the generation of an electron–hole pairwithin a semiconductor upon absorption of a photon of sufficient energy (shownschematically in Figure 18.1) [1]. On the catalyst surface, it is considered that pho-togenerated electrons (e−) and holes (h+) react with dissolved oxygen and water,respectively, to give highly oxidizing radical species which can kill bacteria andoxidize organic and some inorganic compounds on, or proximate to, the photo-catalyst’s surface. Usually, the rate-limiting step in these reactions is the transferof electrons to O2 molecules to form superoxide radical anions (O−

2 ), which arean effective oxidizing agent for dyes and other less stable compounds.

The most extensively employed photocatalyst, commonly used for redox reac-tions, is titanium dioxide (TiO2). TiO2 is a highly stable, nontoxic, cost–effective,and abundant oxide semiconductor. However, its photoconversion efficiency islimited by high rates of exciton recombination, fast backward reaction speed ofredox species and redox products, and an inability to absorb light in the visi-ble spectrum due to its large bandgap energy. The most prevalent techniques foraddressing these problems are based upon doping and/or by the addition of noblemetal particles to the surface of TiO2. These metal particles, thanks to their lowFermi level, can act as local trapping sites and increase the lifetime of photogen-erated charges, while their surface plasmon resonance (SPR), induced by visiblelight, can locally enhance the light absorption and extend the activity of TiO2 intothe visible part of the solar spectrum [2].

18.2 Historical Background

Photocatalyzed oxidation of organic compounds under UV light on the surfaceof metal oxides such as TiO2 and ZnO was first reported almost six decadesago [3]. The excitement really began following a report in 1972 by Fujishimaand Honda [1] demonstrating the splitting of water into hydrogen and oxygen


500 18 Visible-Light-Active Photocatalysis

hν

O2 (ad)

H2O

Conduction band

Valence band

Eg

Chargeseparation

Recombination

h+

e–O2

HO

Figure 18.1 The Photocatalytic Mechanism. The photogenerated charge mechanism in TiO2consisting of electron and hole separation and recombination and reactions with adsorbedmolecules.

TiO2 n-type semiconductorphotoelectrode

Pt counter electrode

Electrolyte

O2

H2O

H2

H2O

(λ<415 nm)

VB

3.0 eV

CB

UV

e–

e–

h+

Bias

Figure 18.2 Water splitting using photocatalysis with oxygen evolution on irradiated TiO2 andhydrogen evolution on dark Pt [1].

on UV-irradiated TiO2 and Pt (in darkness), respectively, in a photoelectrolyticcell (see Figure 18.2). The promise of unlimited solar energy for generating fuelsspurred great interest. Two decades later the growing interest in nanotechnologyled directly to emphasis on synthesizing nanosized metal oxides for photocatal-ysis, thereby greatly enhancing the surface area, a topic of research that has nowgathered explosive research interests.

While TiO2 remains the most investigated photocatalyst material, other pho-toactive semiconductors such as ZnO, Fe2O3, WO3, CdSe, PbS, and GaAs and

18.3 Basic Concepts 501

double compounds such as titanates have also gained considerable attention [4].A paper published in 2001 by Asahi et al. [5] on TiO2 that has doped with nitrogen(N) offered a convincing path for research into extending the range of photocatal-ysis to longer wavelengths in the visible region. In this chapter, the primary focusis based upon nanosized TiO2 as the model system for elucidating the rapid devel-opment in the field of visible-light photocatalysis for harnessing photoassistedredox chemical reactions on the surface of the semiconducting photocatalysts.

18.3 Basic Concepts

Photocatalysis is a rapidly growing scientific field of research with a massivepotential for a wide range of industrial applications. Most of the attention isfocused on water disinfection, destruction of air-borne pollutants, decomposi-tion of water, synthesis renewable fuels, and generation of chemicals from simplefeedstocks. The basic concepts relevant to the above applications are derivedfrom the processes occurring on the semiconductor–fluid interfaces. The solidsemiconductor can be activated by photons and reactions under considerationare heterogeneous in nature taking place at the interface between two or morephases.

Photocatalysis refers to a process whereby the speed of redox reactions isincreased by the action of light on the semiconducting material which partakesin the process but is not consumed in the reaction. The key features of asemiconducting photocatalyst is its ability to absorb light, produce excited statesby virtue of interaction of light with its electronic structure, and transport theelectronic charges to facilitate surface reactions with interfacial chemical sub-strates leading to secondary chemical reactions and transportation of productsaway from the semiconducting surface to sustain the redox reactions.

The concept of insulators, conductors, and semiconductors is related to theenergy levels of electron in the materials. As shown in Figure 18.3, the upperlevel is called the conduction band (CB), while the lower level as the valence band(VB). In conductors, the CB is the high-energy level, and the VB is the low energylevel, and they overlap in metallic conductors whereas in semiconductors the VBand the CB are separated with a forbidden gap. The VB is far away from the CBin insulators. Due to these differences, conductors can always conduct electricity,while due to a large gap between VB and CB insulators cannot conduct electricityat all. Conduction of electricity in semiconductors can be induced thermally or bydoping or by exciting with photons. Electrons can be excited from the VB to theCB if a photon of energy equal or greater than the bandgap (Eg) is used to irradiatethe material. Once the condition is satisfied and electrons are transferred to theCB, then this material behaves as a conductor and thus can conduct electricity.Thus, for a semiconductor material to conduct, the following condition must besatisfied such that the frequency, 𝜈

𝜈 > Eg/

h (18.1)

where h is Plank’s constant, and Eg is bandgap energy.


(a) (c) (b)

Ef

E

Figure 18.3 Energy-level diagram of (a) conductor (b) insulator, and (c) semiconductor as afunction of bandgap width.

The energy-level diagrams are strictly valid for single crystals and bulk materi-als. For nanosized crystals, termination of the crystal structure at the free surfacehas a pronounced effect due to the very large surface area per unit volume ofthe material. Several possibilities arise, a very large concentration of dangling(incomplete bonds) may be present. Surface atoms may rearrange to relax theenergy penalty and thus rearrange quite distinctly from the underlying bulk. Stepsand kinks may be present and can readily adsorb or form strong chemical bondswith foreign atoms from the environment. The overall effect is to form localizedsurface electronic states with the bandgap thus profoundly affecting the elec-tronic and optical properties of the nanosized semiconductor. As redox reactionsare surface processes, nanostructuring can have major impacts on photocataly-sis. Surface space charge may arise due to spatial nonneutrality resulting in bandbending in the energy-level diagrams, with the band rising at the surface for an-type semiconductor and bending downward for a p-type semiconductor (seeFigure 18.4) [6].

When the surface of the semiconductor particle or crystal is irradiated withphotons of energies greater than the electronic bandgap of the material, electronsare excited from the VB to the CB, thus generating electron–hole pairs (e–h),most of these recombine rapidly dissipating the energy as heat (see Figure 18.1).Efficiency of photocatalysis depends on the extent to which the e–h pairs canbe separated to induce sufficiently favorable lifetimes of excitons, thus allowingopportunity for the redox reactions to take place at the solid–fluid interface.

The semiconductor photocatalysts used for harnessing redox reactions mustsatisfy several criteria. The main requirement is that the redox potential of thedesired reaction lies within the bandgap domain of the photocatalyst [1, 3–7].The effectiveness of the electron transfer reaction is related to the position of thesemiconductor’s CB and VB edges relative to the electrochemical redox poten-tials of the adsorbed substrate. For a desired electrodic reaction to occur, thepotential of the electron acceptor species should be located below the CB of the

18.3 Basic Concepts 503

Space chargeregion

Space chargeregion

EbbEbb

E-field E-field

Dipole Dipole

+ – +–

(a) (b)

Econduction

Evalence

Econduction

Evalence

Figure 18.4 (a) Band bending upward in a n-type nanosized semiconductor where electronsfrom close to the CB are transferred to the surface states; (b) band bending downward in ap-type semiconductor since electrons are transferred from surface states to acceptor levelsnear the valence band [6].

–2.0

ZrO2

KTaO3 SrTiO3 TiO2

–1.0

0

1.0

5.0

eV

3.4

eV

3.2

eV

3.0

eV

3.6

eV

ZnS

CdSe

GaP

SiC

Si

MoS2 H+/H2

O2/H2O

Fe2O3

1.7

eV

2.2

5 e

V

3.0

eV

1.1

eV

1.7

5 e

V

2.3

eV

2.8

eV

WO32.4

eV

CdS

2.0

3.0

4.0

V v

ers

us N

HE

(p

H0

)

Figure 18.5 Valence and conduction bands for a variety of semiconducting materials on apotential scale (V) versus the normal hydrogen electrode (NHE). Redox potentials for thewater-splitting half reactions are indicated by the dotted lines. In order for a semiconductor tobe an effective catalyst its conduction band energy should be higher than the H2 producingreaction potential and its valence band energy should be lower than the O2 producingreaction potential. (Reprinted with permission from Ref. [7]. Copyright 2011, Springer Nature.)

semiconductor (should be more positive), whereas the potential of the electrondonor species should be located above the VB (should be more negative).A schematic comparing the band energies of a variety of semiconducting mate-rials with the energies required to complete the reactions needed for hydrogengas production by splitting water is provided in Figure 18.5 [7].

The overall reaction is sum of two partial redox reactions such that the electronsconsumed in the reduction reaction is provided by the oxidation reaction. A dis-tinction should be made between those overall reactions that are spontaneous(i.e., ΔG∘

< 0) and those that are uphill (ΔG∘> 0). In either of the situations,

the total activation energy required is the sum of the activation energies for the


two half reactions, fully provided by the UV or visible light quanta in reactionsfully catalyzed by photons. In spontaneous reactions such as oxidation of organicpollutants, the photon quanta provide the energy to overcome the activationpotential energy barrier, whereas in uphill reactions such as splitting of waterto hydrogen and oxygen, not only the activation barriers have to be overcome,but photons must also supply the free energy to counter the positive free energyof the uphill reaction. Fujishima and Honda [1] applied an electrical bias to over-come the uphill reactions and utilized the photons energy to supply the activationpotential energy for the overall reaction. While the bandgap energy determinesthe optical absorption wavelength, the position of the highest point in the VBdetermines the oxidative power of the photocatalyst.

A compromise between bandgap and the free energy of the photogeneratedexcitons is important to consider. A large bandgap material such as ZrO2 (Eg >

5.0 eV) would require extremely energetic photons outside practical realms toderive its semiconducting properties in photoactivation. A moderately large valueof bandgap energy as for TiO2 (Eg > 3.0 eV) is achieved activation with UV lightand the excitons have high free energy available in comparison with relativelylower bandgap materials such as CdS (Eg > 2.6 eV). The chemical and the photo-chemical stability of larger bandgap energy materials is much greater than lowerbandgap materials. In any event, lower bandgap (Eg < 2.8 eV) materials are alsogreatly investigated as they are better adapted to the solar spectrum, offering thelarge percentage of visible spectra available from the sun.

A significant increase in efforts in extending the response to visible light usingnew and/or modified semiconductors with acceptable quantum efficiency shouldbe noted [8–12]. The main approaches aim to modify the electronic and/or theoptical properties of the semiconductors, and they consist of metallic hybridiza-tion or coatings, dye sensitization, doping with transition metals or nonmetallicelements, and the use of composite semiconductors.

18.4 Structure of TiO2

There are four commonly known polymorphs of TiO2 found in nature: anatase(tetragonal), brookite (orthorhombic), rutile (tetragonal), and TiO2 (B) (mon-oclinic) [7]. Of these four, only the anatase and rutile forms (and sometimesbrookite) are commonly used in experimental investigations.

The crystal structures of anatase, rutile, and brookite are shown in Figure 18.6.The titanium (Ti4+) ions in all the three forms are coordinated with six oxy-gen ions forming octahedra of TiO6. Due to oxygen deficiency, titanium dioxideis usually an n-type semiconductor, and some of the Ti4+ is converted to Ti3+

accompanied by vacant oxygen sites [4]. Rutile has a tetragonal structure withthe octahedral linked along the edges in chains along [001] direction, and thesechains are cross-linked by sharing of corners. Anatase also crystallizes in thetetragonal structure, while brookite takes the orthorhombic form. Differencesarise from the way the TiO6 are linked together as shown in Figure 18.6. In allforms of TiO2, the CB is comprised of 3d orbitals of Ti atoms while the VB ismade of mainly 2p orbitals of oxygen atoms.

18.4 Structure of TiO2 505

(a) (b) (c)

Figure 18.6 Crystal structures of TiO2 (a) anatase (b) rutile, and (c) brookite.

The rutile phase is stable at most temperatures and pressures up to 60 kbar andhas a bandgap of 3.02 eV. It is the thermodynamically stable phase and kineticallyfavored to form on thermal activation when anatase phase particles are in sizesgreater than 14 nm. In addition, once formed, the rutile particles were found togrow faster than their anatase counterparts [7].

However, rutile is generally regarded as a poorer photocatalyst, especially whencompared to the anatase phase of TiO2, which has higher electron mobility, lowerdensity, and a low dielectric constant. Anatase, which has a bandgap of 3.2 eV, hasa slightly larger distortion of the TiO6 octahedron in its tetragonal structure thanrutile. Its slightly higher Fermi level, lower capacity to absorb oxygen, and higherdegree of hydroxylation give anatase its increased photoreactivity [7, 11, 12].

The band structure of anatase is also advantageous for its use as a photocata-lyst. The VB of anatase TiO2 is comprised of the 2p orbital of oxygen hybridizedwith the 3d orbital of titanium, whereas the CB solely contains the 3d orbital oftitanium. Therefore, electrons that are excited into the CB are less likely to dropback down into the VB due to the difference in orbital structure, thereby makingelectron/hole recombination less likely. In addition, anatase TiO2 has inherentsurface band bending that forms spontaneously with a steeper potential gradientthan seen in rutile phase TiO2 (see Figure 18.7) [7]. This property leads to the fun-neling of photogenerated holes to the surface of the anatase particle, effectivelycreating spatial charge separation by trapping holes at the particle’s surface. Thissame funneling effect is not seen in the rutile phase, leading to increased bulkrecombination.

Although the TiO2 anatase phase is typically the phase investigated for pho-tocatalytic applications (and is the phase investigated in this work), some recentresearch has suggested that the combination of rutile and anatase phases has ahigher photocatalytic activity than either phase by itself. In addition, it has beenproposed that the combination of phases demonstrates an enhanced degree ofabsorption of visible light [7, 13].


En

erg

y CB

e–

e– e–

h+

h+ h+

VB

(a) (b)

CB

VB

Anatase Rutile

Figure 18.7 Surface band bending in the (a) anatase and (b) rutile phases of TiO2. The x-axissimply represents depth within the material with the surface of the material being indicatedby the vertical line at the right-hand side of each schematic. (Reprinted with permission fromRef. [7]. Copyright 2011, Springer Nature.)

18.5 Photocatalytic Reactions

The mechanism of reactions that can occur heterogeneously at the interface ofthe TiO2 and the fluid (water or humidified air) is well documented for over 2–3decades [14–16]. The excited electrons and holes can generate hydroxyl radicalsthrough chemical reactions as outlined in the following equations [14]:

TiO2 + h𝜗 → TiO2 + e−(CB) + h+(VB) (18.2)OH−(aq) + h+ → OH∙ (18.3)H2O(aq) + h+ → OH∙ + H+(aq) (18.4)O2 + e− → O∙−

2 (18.5)2O∙−

2 + H+ → HO∙2 (18.6)

The hydroxyl radical is even more oxidizing that all the well-known highly oxi-dizing substances such as ozone (see reduction potentials in Table 18.1). Eventhe superoxide radicals O∙−

2 generated by reduction of oxygen by the CB elec-trons are powerfully oxidative comparable to the oxidative power of hydrogen

Table 18.1 Standard electrochemical reductionpotentials of common oxidants.

OxidantReductionpotential (Eo)

∙OH (Hydroxyl radical) 2.8O3 (Ozone) 2.07H2O2 (Hydrogen peroxide) 1.77HClO (Hypochlorous acid) 1.49Cl2 (Chlorine) 1.36

18.6 Physical Architectures of TiO2 507

Anatase TiO2

Rutile TiO2

H2 O

O O

>Ti >Ti

Hole attack Trapped hole Release Original

Ti< Ti<

h+

H2 O

H2 O

O O OO⋅ OH

⋅OH

O O>Ti >Ti >Ti >Ti

Hole attack Hole attackForming Ti–OO–Ti Release

Ti< Ti< Ti< Ti<

h+ h+ h+

H2 O

O⋅ OH

⋅OH

OH

>Ti >TiTi< Ti<

Figure 18.8 Mechanism by which hydroxyl radicals are formed on the surface of anatase andrutile. (Reprinted with permission from Ref. [17]. Copyright 2014, The American ChemicalSociety.)

peroxide. Thus, the photocatalytic process unleashes radicals with superoxidativepower that can react with organic, microbes, and inorganic pollutants in waterand humid air to convert them into secondary products, ultimately at full oxida-tion to carbon dioxide and water. The mechanism by which hydroxyl radicals areformed on the surfaces of anatase and rutile is shown in Figure 18.8 [17].

18.6 Physical Architectures of TiO2

One of the key parameters investigated for the photocatalytic optimization ofTiO2 is the form of its physical structure and morphology. This section willexplore the different physical architectures which have been created, namelyfilms, nanotube arrays, and nanoparticles, as well as describe the methods usedto create them.

One type of TiO2 structure that has been widely researched is a mesoporousthin film. The increased surface area of the particles within the porous film as wellas the crystalline, structured spatial arrangement of the layers making up the filmcan help improve electron transfer and quantum efficiency [11, 18]. One of theeasiest methods to produce these thin films is sol–gel deposition, a techniquethat can produce highly ordered, crystalline films [17]. The sol–gel techniquehas several advantages over other methods, including chemical vapor deposition,plasma spraying, and anodization, since it does not require any special apparatus,can form uniform, multi-component films whose phase structure can be read-ily controlled, and produces porous, gel-like films with large specific areas [19].


The morphology of the resulting TiO2 porous films depends strongly on the typesof solvents and complexing agents used, and on the concentration of the template.The template material is often a polymer as it has been found that the use of non-ionic surfactants promotes the formation of film structures with exceptionallylarge surface areas.

Sputtering, while being a high cost process, can help engineer controlled poros-ity, particle size, thickness, orientation, and surface area on selected substrates[11]. Control of chemical composition, defect structures, in situ doping, and mor-phology control are also possible thus allowing opportunities for fundamentalinvestigations of thin films of photocatalytic materials. In short, both sputter-ing and sol–gel techniques have been proven capable of creating TiO2 thin filmswith high photocatalytic activity through the adjustment of the density, porosity,surface area, and crystallinity of the films.

Another physical structure that has been investigated is a TiO2 nanotube ornanowire array grown by anodic oxidation [13]. These arrays are structuredsuch that a high percentage of their total surface area is in contact with theelectrolyte, thus allowing for an increased amount of interaction between thephotogenerated holes (electrons) and the oxidizible (reducible) species in theelectrolyte. Although traditionally these arrays are grown on planar substrates,it is also possible to grow them on nonplanar geometries such as a half-pipe orfull-pipe. These geometries have the potential to limit optical loss by minimizingthe fraction of incident light that is reflected and maximizing the optical pathlength of light inside the array. They could also lead to enhancements in lightabsorption near the bandgap edge (375–400 nm). Initial experimentation intothis area is promising, with photoconversion efficiencies of 0.15% measuredfor arrays on full-pipe geometries, an improvement of 60% over the 0.094%efficiency measured for a flat geometry. Efficiencies of 0.126% were measuredfor the half-pipe geometry, a 34% increase over the flat geometry. The photo-conversion efficiencies and amount of reflected light observed in each of threedifferent geometries are compared in Figure 18.9.

Like TiO2 films, TiO2 nanotube arrays rely on high surface area and high crys-tallinity to achieve better photoconversion efficiencies. However, the fact that thegeometry of the nanotube array substrate can be altered to increase the amountof light absorbed offers an additional parameter through which photoconversionefficiencies can be maximized.

Finally, a large portion of the research on TiO2 has focused on its capabili-ties when it is in the form of nanoparticles. It has been established that TiO2is much more effective as a photocatalyst when in the form of nanoparticles thanwhen in the form of a bulk powder. Once the size of the TiO2 particle shrinksbelow 10 nm, it begins to behave quantum mechanically, leading to a shift in thebandgap edges which produces larger redox potentials. However, the free energyfor charge transfer remains unchanged, resulting in a larger net driving force forelectron transfer and a higher overall photoactivity [7].

The most common synthesis method to produce TiO2 nanoparticles isliquid phase processing, which uses a hydrothermal reaction to synthesize the

18.7 Visible-Light Photocatalysis 509

–1 –0.5

Potential (V vs Ag/AgCl)(a) (b)

Photo

conve

rsio

n e

ffic

iency (

%)

0 0.5

0.16

Foil0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

Half pipePipe

300

25

20

15

10

5

0350

Wavelength (nm)

% R

eflecta

nce

400 450 500

FoilHalf pipePipe

Figure 18.9 (a) Photocurrent efficiency as a function of applied potential for nanotube arraysanodized on a planar foil, a half-pipe of 3.75 mm diameter and a full-pipe of 3.75 mm diameterand aspect ratio 1.7. (b) Total reflectance measurements for the three geometries. (Reprintedwith permission from Ref. [13]. Copyright 2009, The American Chemical Society.)

nanoparticles. This method results in homogeneous products with a controlledstoichiometry. It is also capable of forming complex shapes and compositematerials. However, it is expensive and has long processing times. The reactionconditions, particularly the duration and temperature of the heat treatment,have a strong influence on the morphology, crystallinity, porosity, and surfacearea of the resulting structure. A study by Hou et al. illustrated that the mor-phology of Bi12TiO20 particles produced through a solution-phase hydrothermalprocess could be manipulated through variation of the reaction parameters,particularly the temperature, reagent concentration, and reaction time [10].Within a hydrothermal temperature range of 150–180 ∘C, the morphology wasprogressively transformed from nanosphere agglomerates to microflowers tonanowires to microspheres over time [10]. SEM images of some of the observedmorphologies can be seen in Figure 18.10. Similar opportunities exist for exploit-ing morphological changes in TiO2 to maximize surface area, photocatalyticefficiency, adsorption of substrates, post-filtration, and immobilization of thephotocatalytic material and optimization of reactor designs.

18.7 Visible-Light Photocatalysis

The physical structure of TiO2 plays a large role in determining its photocat-alytic activity with porous thin films, nanotube arrays, and nanoparticles emerg-ing as the favored architectures for producing high photoconversion efficiencies.In each of these structures, high crystallinity and high surface area are key param-eters for maximizing these efficiencies. However, additional processes, primarilyalterations to the chemical composition of the structure, must be considered toaddress the inherent shortcomings of TiO2 including high recombination rates,high backward reaction speeds, and a lack of absorption in the visible spectrum.


(a) (b)

(c)

20 μm 1 μm

1 μm 100 nm

(d)

Figure 18.10 SEM images of Bi12TiO20 structures (a,b) prepared at 150 ∘C and (c,d) prepared at180 ∘C. (a,c) overall product morphology; (b,d) enlarged image of the flower-like and nanowirestructures. (Reprinted with permission from Ref. [10]. Copyright 2010, The Royal Society ofChemistry.)

18.8 Ion Doping and Ion Implantation

One of the most widely investigated methods for extending the photocatalyticactivity of TiO2 into the visible spectrum is ion doping. Doping can be done witheither transition metal cations or with anions and is designed to decrease thebandgap or introduce intra-bandgap states that allow for an increased absorp-tion of visible light without altering the integrity of the TiO2 crystal structure [4].It is essential however that the ions are doped near the surface of the TiO2 parti-cles to ensure that trapped electrons and holes can be readily transferred to thesemiconductor–liquid interface where the photocatalytic reactions will occur. Ifthe ions are doped too deeply into the bulk, they tend to act as recombination cen-ters [9]. In addition, if the dopant concentration is too high, the depletion layermay become narrower than the penetration depth of the light, again leading toincreased recombination. In Figure 18.11, Cr doping has in fact led to reductionin photocatalysis rather than an increase [20].

In one study, where 21 different metal ion dopants were considered, Mo, Ru,Os, Re, V, and Rh ions were found to have a positive effect on the photocatalyticactivity of TiO2 whereas Co and Al ions were found to have a negative effect [9].Additional studies have noted increases in TiO2 photocatalytic activity followingdoping with Cu [21], Ce [22], and Fe [23] ions. The effectiveness of each dopantprimarily depends on their ability to trap and transfer either, or in some cases

18.8 Ion Doping and Ion Implantation 511

90%

Degra

dation

80

70

60

50

40

30

20

10

0

Dopant concentration

Vis

Irradiation time = 90 minUV–Vis

Ti 0.05 at.% Cr 0.5 at.% Cr 5 at.% Cr 20 at.% Cr 30 % at. Cr0.25 at.% Cr

Figure 18.11 Percentage of photodegradation of model pollutant BB (Basic Blue 41) as afunction of Cr3+ concentration. Dark bars: UV–Vis excitation; light bars: visible excitation.(Reprinted with permission from Ref. [20]. Copyright 2007, Elsevier.)

(Cu, Mn, Fe) both, electrons and holes [9]. The overall photocatalytic perfor-mance of the doped TiO2 was found to depend heavily on the dopant loadingmethod, particle size, and particle dispersion pattern [4]. Despite the positiveeffects of metal ion doping, there also appears to be an increase in thermal insta-bility and a decrease in carrier lifetimes associated with their presence [20].

An alternative to conventional metal ion doping by chemical methods is metalion implantation in which TiO2 is bombarded with high-energy transition metalions that become embedded in the lattice [4]. The resulting mixing of the Ti dorbital with the transition metal d orbital leads to bandgap narrowing and a shiftin the photoresponse of TiO2 into the visible region. Ion implantation experi-ments with V, Cr, Mn, Fe, and Ni ions have successfully promoted a redshift inthe absorption spectrum of TiO2, indicating the promise of this technique [4].

Another recent effort, which was first initiated in 2001 [5], is the doping of TiO2with anions such as N, C, F, and P. Although the larger ionic radii of anions com-pared to metallic ions likely result in an increase in strain on the TiO2 lattice,anions are less likely to form recombination centers than doped metal ions, indi-cating that they are also more effective at enhancing photocatalytic activity [5,11]. Nitrogen, the most widely studied of the anion dopants, is believed to workby mixing its own p states with the 2p orbital of O, thereby shifting the VB edgeupward and narrowing the overall bandgap of TiO2 [11] (see Figure 18.12). Theeffect of several different anions doped intoTiO2 [24] is depicted in Figure 18.13.However, there is some debate among the scientific community about the actualeffectiveness of anion doping, as the improvement in the absorption of visible


Normal TiO2 Doped TiO2

CB

Band gap

+ + + +VB

Doping

VB

Isolated energy levels

CB

Figure 18.12 Schematic mechanism of bandgap narrowing in anion doped TiO2.

B C

1.39

C 2p4

N

0.13

N 2p5

O F

0.82

1.19

F 2p6

Ti3+ 3d1

CB

VB

3.9

2.18

B 2p3

Substitutional

Figure 18.13 Schematic representations of B, C, N, and F doping in TiO2. (Reprinted withpermission from Ref. [24]. Copyright 2013, Elsevier.)

light is, at best, modest in many reports [13]. One hypothesis is that anion dopingsimply results in an increased number of oxygen vacancies rather than a narrow-ing of the bandgap. Further research is required to determine the exact nature ofthe effect which anion doping has on the band structure of TiO2. When TiO2 isdoped with hydrogen via a hydrogenation process, nanocrystals were shown tobe black containing paramagnetic Ti(III)ions and exhibited a higher visible-lightabsorption. The hydrogenated samples exhibited improvement in photocatalyticactivity under visible light (𝜆> 380 nm) for degrading methylene blue dye andgood response in term of photo current density. The band structure of the hydro-genated TiO2 is shown in Figure 18.14 [25].

18.9 Dye Sensitization 513

1.54 eV

2.18 eV

3.3 eV

Conductionband

Black TiO2 White TiO2

Valenceband

Figure 18.14 Electronic structure of black hydrogenated TiO2. (Reprinted with permissionfrom Ref. [25]. Copyright 2011, AAAS.)

18.9 Dye Sensitization

The addition of a chemisorbed or physisorbed dye to the photocatalyst surfacecan extend the range of wavelengths which activate the photocatalyst, a methodthat has been used extensively in the design of dye sensitised solar cells (DSSC) [4,26]. Upon excitation of the dye, which, depending on the dye, can occur undervisible light, a hole or electron is injected onto the surface of the photocatalyst(i.e., TiO2) which acts as a charge separator and site for the photocatalytic reac-tion [4]. Following the charge injection, the dye shifts into an oxidized state andrequires an electron sacrificial agent, such as iodide ions or EDTA, to regain itsfunctionality. An illustration of the full process is given in Figure 18.15 [27].

CB

Electroninjection Excited state

sensitizer: S*

Oxidized statesensitizer: S+

Electron mediator

Irradiatedby visiblelight

Sensitizer: SVB

H2

H2O

Figure 18.15 Mechanism of dye-sensitized photocatalytic hydrogen production on thesurface of a metal-oxide semiconductor (such as TiO2) under visible-light irradiation.(Reprinted with permission from Ref. [27].


18.10 Noble Metal Loading

A very prevalent technique for enhancing photocatalysis is shown to be by theaddition of noble metal particles to the surface of TiO2. These metal particles,thanks to their low Fermi level, can act as local trapping sites and increase thelifetime of photogenerated charges, while their SPR, induced by visible light (Vis),can locally enhance the light absorption and extend the activity of TiO2 into thevisible part of the solar spectrum.

Of all metal additions, silver is particularly promising due to its strong plas-monic behavior and relative affordability when compared to gold and platinum.Common methods of silver deposition onto TiO2 surfaces include photodeposi-tion, sputtering, impregnation, and chemical reduction of a silver salt in solution.Chemical reduction is particularly promising due to the method’s versatility andaffordability, while employment of suitable reducing agents in silver nanopar-ticle synthesis is also an effective tool in curbing the size of silver nanostruc-tures. Defects on a TiO2 surface, such as surface oxygen vacancies (V 0) and/orTi(3+) oxidation states, can also induce some visible-light photocatalysis due tothe formation of shallow trap states within the TiO2 bandgap. Thus, any processused to deposit Ag on TiO2 with a highly reducing chemical may lead to greatervisible-light photocatalysis through the reduction of surface Ti4+ to Ti3+ states,although presence of oxygen and its incorporation may reverse this process tosome extent.

The method was initially simply devised to prevent electron–hole recombi-nation. Because the Fermi level of noble metals is lower than that of TiO2, anyphotogenerated electrons will be transferred from the CB of TiO2 to the surfaceof the noble metal particles while the photogenerated holes will remain in theVB of TiO2 [2] (see Figure 18.16). This physical separation reduces the chances ofelectron–hole recombination, thus increasing the photocatalytic activity of TiO2.

Capture

Discharge

hν

e

h

e e e

Store

AgTiO2

Figure 18.16 Electron transfer mechanism in silver-loaded TiO2. (Reprinted with permissionfrom Ref. [2]. Copyright 2009, Elsevier.)

18.10 Noble Metal Loading 515

More recently however, noble metals have been found to improve the photo-catalytic activity of TiO2 in a second, possibly more significant way known asSPR. SPR occurs when the resonant frequency of incoming photons matchesthe natural vibration frequency of the material’s valence electrons, a vibrationthat arises from the attractive force of the positive nucleus and the repellingforces of surrounding electrons. This match in frequencies produces a collectiveoscillatory movement among the valence electrons that can produce spatiallynon-homogeneous oscillating electric fields as well as excite electrons to higherenergy states [28]. For many noble metals, including gold and silver, this matchin frequencies occurs at wavelengths exhibited by UV and visible light, the majorcomponents of sunlight.

Therefore, plasmonic noble metal nanostructures can influence the photocat-alytic properties of TiO2 in several ways. Any excited electrons formed on thesurface of the noble metal during the SPR process will have energies which are1.0–4.0 eV higher than the metal Fermi level [29]. These high-energy electronscan be transferred to the CB of TiO2 where they can be used in redox reac-tions at the semiconductor–metal–liquid interface (Figure 18.17) [29]. This bothincreases the number of electrons available for reactions and extends the pho-toactivity of TiO2 into the visible-light range. In addition, any strong electromag-netic fields produced by SPR can greatly increase photocatalytic activity as therate of electron–hole formation in a semiconductor scales with the square of theintensity of the local electric field [30]. In this case, electron–hole pair forma-tion will occur predominately in regions of the semiconductor’s surface that areclosest to the electric field, and therefore closest to the noble metal nanoparticles.This is advantageous for two reasons: (i) separation of the electron and hole tendsto occur rapidly at the surface due to the inherent surface band bending of theanatase TiO2 particle and (ii) the distance to the semiconductor–liquid interface,where reactions can occur, is short. Lastly, for larger plasmonic nanostructures,

Valence band

Conductionband

Agnanoparticle

Ag0 → Ag+

>450 nm

e–

Figure 18.17 Mechanism for light absorption of silver supported in TiO2. (Reprinted withpermission from Ref. [29], Copyright 2010, The American Chemical Society.)


SPR can scatter resonant photons, effectively increasing the average photon pathlength and providing more chances for electron–photon interaction within thecomposite metal–semiconductor photocatalyst [30].

It has been found that the properties of these noble metal–TiO2 compositesdepend strongly on the metal particle size, shape, dispersion, and composition.Interestingly, the resonant wavelength of the noble metal particles, and thereforethe position of the maximum SPR intensity, can be tuned by adjusting some ofthese properties. This could potentially provide a way to optimize the photoac-tivity of the metal–TiO2 composite for certain wavelengths of light. One of thesimplest ways to show this change in resonant frequency is by showing the changein the position of the extinction, or absorption, spectra for different types of par-ticles. Peaks in the extinction spectra indicate maximum photon absorption andtherefore maximum SPR excitation.

Our group has developed a rapid, one-pot chemical reduction method for theproduction of a highly black TiO2 Ag photocatalyst, without the need of a surfac-tant or complexing agent (see Figure 18.18) [12, 30]. It was found that the pho-tochemical properties of the resulting material are highly dependent on postre-duction annealing in air, which causes particle size and distribution changes,and alters the surface chemistry. The nanocomposite shows an increase of over800% in the rate of photocatalytic methylene blue dye degradation, comparedto commercial unmodified TiO2, under UV–Vis illumination. Unlike pure TiO2,the nanocomposite exhibits visible-light activation, with a corresponding drop inoptical reflectance from 100% to less than 10% [(Figure 18.19) [30]. The photocat-alytic properties were shown to be strongly enhanced by postreduction anneal-ing in air, which were observed to decrease, rather than coarsen, silver particlesize, and increase particle distribution (which ranges from subnanometers to

FormaldehydeH H

C

O

Ag+

Ag

N+

–O O–

O

Silver nitrate

Formaldehyde

H

H

CO

Ag+

Ag

TiO2Our

CatalystN+–O

O–

O

Silver nitrate

Figure 18.18 Modification of TiO2 nanoparticles with silver using silver nitrate andformaldehyde as a metal source and reducing agent, respectively, the resultant color changeof the powder after treatment is also shown on the right [12, 30].(http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/)

18.10 Noble Metal Loading 517

120

100

80

60

40

20

0300 400 500 600

Wavelength (nm)

Reflecta

nce (

%)

700 800

TiO2

TiO2-Ag (1 mol%)

TiO2-Ag (3 mol%)

TiO2-Ag (6 mol%)

TiO2-Ag (12 mol%)

Figure 18.19 Diffuse reflectance spectroscopy (DRS) of TiO2 and Ag–TiO2 at silver loadingsfrom 1–12 mol% following a 300 ∘C heat treatment for 30 min. (Reprinted with permissionfrom Ref. [30]. Copyright 2017, Elsevier.)

Plasmonic light harvesting antennae

Light

Metalnanoparticle

TiO2

Figure 18.20 Plasmonic light harvesting using core–shell metal-insulator nanoparticles.(Reprinted with permission from Ref. [26]. Copyright 2011, The American Chemical Society.)

approximately 10 nm). This, accompanied by a variation in the silver surface oxi-dation states, appears to dramatically affect the photocatalytic efficiency underboth UV and visible light. Such highly active photocatalysts (see Figure 18.20)could have wide ranging applications in water and air pollution remediation andsolar fuel production [26, 31].


O2

H2O

CB

CBCB

VBVB

VB

SnO2

ZnOTiO2

hν

hν

hνe–

e– e–

h+h+

h+

O2–•

OH•

Figure 18.21 Electron transfer mechanism in composite semiconductor. (Reprinted withpermission from Ref. [26]. Copyright 2011, The American Chemical Society.)

18.11 Coupled Semiconductors

Another way to extend the photocatalytic response of TiO2 into the visiblespectrum is semiconductor coupling. This method works on a similar principleas the dye sensitization technique except that instead of injecting electrons froma dye to a semiconductor, electrons are injected from one semiconductor toanother [32].

During semiconductor coupling, a large bandgap semiconductor whose con-duction band energy is more negative than the energy required for water splittingis paired with a small bandgap semiconductor that is capable of being excited byvisible light and has a conduction band energy which is more negative than thatof the large bandgap semiconductor [33]. Upon excitation of the small bandgapmaterial, electrons are injected into the large bandgap semiconductor, effectivelyincreasing the charge separation distance and decreasing the likelihood of recom-bination [34].

In this system, TiO2 frequently acts as the large bandgap semiconductor and ispaired with a material that has a smaller bandgap such as CdS [34], Bi2S3 [35],orWO3 [36]. All three of these semiconductors can absorb visible light as well astransfer charge quickly and efficiently. However, it has been reported that elec-tron sacrificial agents are required in systems where the small bandgap semicon-ductor is susceptible to photocorrosion, such as in the TiO2/CdS system. In acomposite system, three different semiconductors are coupled together as shownin Figure 18.21 [26].

18.12 Carbon–TiO2 Composites

Carbon has extensively been used in combination with TiO2 to enhance pho-tocatalytic rates and extend activity into the visible spectrum.TiO2 has been

18.12 Carbon–TiO2 Composites 519

successfully combined with many forms of carbon including: nanotubes [37],activated carbon (AC) [38], graphene [39], and fullerenes [40]. It is postulatedthat carbon may play several potential roles in photocatalysis, including actingas a photosensitizer for the absorption of visible-light photons, an electronsink for the promotion of photo generated charge separation and/or adsorptionagent, enhancing physical adsorption of organic species thereby promoting theiroxidation on or proximate to the photoactive surface. The exact role of carbonis still unclear, being highly dependent upon the chemical species, morphology,and the degree and nature of hybridization.

Although extensively explored for the last 15 years, activated carbon has onlybeen demonstrated to enhance photocatalysis under UV irradiation while provid-ing a high surface area support for immobilized TiO2 particles, thereby suggestingthis allotrope does not interact with TiO2 chemically to extend activity into thevisible region. On the other hand, carbon nanotubes (CNTs) with their tunablemorphology and advantageous electrical properties have been shown to enhancephotocatalysis rates as well as extend absorption into the visible region. Woanet al. proposed CNTs may act as sensitizers and efficient electron hole traps lead-ing to more efficient radical production under both Vis and UV light, as shown inFigure 18.22 [41]. A third possibility is also presented as the presence of Ti–O–Cbonds acting similarly to the effects of carbon doping in extending photo activityinto the visible spectrum.

Several synthesis methods of TiO2–CNT have been investigated, typicallyinvolving the deposition of small TiO2 particles onto CNTs. Methods include,but are not limited to, sol–gel coatings, hydrothermal synthesis, PVD and CVD,and even simple mixing, all of which have yielded enhanced photocatalysis.A comparative study by Yao et al. worked to elucidate the ideal arrangement ofCNT and TiO2 within a composite system to optimize the interphase contactbetween the two materials [42]. Deposition of small (5 nm) TiO2 nanoparticlesonto multi and single wall CNTs and coating of larger (100 nm) TiO2 withmulti and single walled CNTs were compared. It was found larger (100 nm)TiO2 coated with single wall CNT bundles gave optimal photocatalytic rates ofphenol degradation compared to other arrangements, an effect attributed to thepromotion of intimate interaction between the two materials favoring electronshuttling.

Graphene has become the center of a large research focus in recent yearsincluding in the field of photocatalysis. Indeed, graphene’s exceptional surfacearea and high theoretical electrical conductivity and mechanical strength makeit an ideal photocatalyst support. In situ production of photoreduced grapheneoxide (rGO)–TiO2 has been demonstrated by several researchers, showingenhanced rates of photocatalysis. The use of graphene oxide allows for scalable,cheap, and high yield production of rGO–TiO2 composites as GO functional-ized with carboxyl groups forms both stable suspensions and bonds with metaloxides. Subsequent UV illumination creates photo excited electrons on the TiO2surface which may directly interact with GO forming rGO with a subsequentcolor change from brown to black [43]. Lee et al. synthesized a graphenewrapped TiO2 composite via a hydrothermal route [44]. The resultant material


(a)

(b)

(c)

hν

– –

+ +

OH•

hν

–

+ +

OH•

O2

hν

–

– –

O2–

+

+ +

OH•

Figure 18.22 Photocatalytic mechanisms in TiO2–CNT composites (a) sensitization by carbon,(b) reduced recombination by a carbon actings as an electron–hole sink, and (c) presence ofintraband states by carbon doping. (Reprinted with permission from Ref. [41]. Copyright 2009,John Wiley & Sons.)

showed a redshift in light absorption and strong photocatalytic activation undervisible light compared to other graphene–TiO2 materials.

18.13 Alternatives to TiO2

Due to its many advantages, TiO2 has traditionally been the archetypal photo-catalyst. However, in recent years much research has been undertaken to dis-cover new photocatalytic materials that are both more efficient and inherentlyactive under visible light [10]. Indeed, synthesis of these new-generation photo-catalysts has expanded into a wide research field typically consisting of mixedmetal/nonmetal oxides [45], nitrides [46], and sulfides [47] – a selected few ofthese new compounds are discussed here.

18.14 Conclusions 521

Polymeric carbon nitride (C3N4) has been shown to be photocatalytic activeunder visible light for water [48]. Such metal-free polymer-like materials are boththermally and oxidation stable and can be made cheaply and are able to producehydrogen without the need for expensive noble metal cocatalysts. C3N4 has beenshown to behave like a conventional wide bandgap semiconductor, with a strongstep-like optical absorption edge at around 420 nm (although this can be subjectto change depending on the synthesis conditions) [49, 50].

Silver’s strong photoresponsive properties have been known for a longtime – exemplified by its traditional use in photography dating back to thenineteenth century. In recent years, several silver compounds have been usedfor highly active visible-light photocatalysts; common materials include: silverphosphates [51], bromides [52], and oxides [53]. Silver phosphate (Ag3PO4) dueto its highly positive valance band and excellent photocatalytic efficiency hasbeen extensively studied. Tauc plot analysis suggests the material has an indirectbandgap of 2.36 ev with an absorbance shoulder at around 530 nm. The atomicstructure consists of PO4 tetrahedra forming a body cantered cubic lattice withsix Ag+ ions distributed among 12 sites [54]. Silver phosphate has consistentlybeen shown to be a very effective photocatalyst in both the destruction of organiccontaminants and water splitting under visible light [45, 55]. Bi et al. synthesizedrhombic dodecahedral Ag3PO4 crystals, without the use of capping agents [56].It was found that rhombic dodecahedrons exhibit much higher photocatalyticactivity under visible light compared to conventional cubes or particles due topossessing more highly reactive {110} facets. However, despite its great promise,currently silver phosphates are limited by instability due to photo degradationand CB location preventing water oxidation without the use of electrical bias orelectron scavengers [57].

Bismuth-based nanomaterials such as bismuth oxyhalides [58], chalcoogenides[59], and vanadates [60] are considered as promising visible-light-active semicon-ductor photocatalysts; having been shown to be effective at both pollution reme-diation [61] and water splitting [62]. To increase the efficiency of bismuth-relatedphotocatalytic nanomaterials, much work has been carried out optimizing mor-phological and chemical parameters including doping with Er [63], S [64], andN [65] and forming hierarchical nanostructures assembled from nanofibers [66],spheres [67], and nanosheets [68]. In particular, layered bismuth oxyhalides haveshown interesting structures, suitable band positions, and high stability for effi-cient visible-light-driven photocatalysis [69].

18.14 Conclusions

TiO2, especially in the stable rutile phase, has been applied as a functional ceramicoxide in many applications such as in sensors, catalysis, paints, pigments, andcoatings over a long period. In particular, the photocatalytic property of TiO2 hasbeen of growing research interest in the last two decades. Many new applicationshave already been emerging in the field of solar cells, antibacterial agents, anti-stain, antifog, hydrophilic, disinfectant, dye degradation, pollutant destruction,and many other new applications. Search for improved visible-light photocatal-ysis is a rapidly growing research topic.


References

1 Fujishima, A. and Honda, K. (1972) TiO2 photoelectrochemistry and photo-catalysis. Nature, 238 (5358), 37–38.

2 Devi, L.G. and Kavitha, R. (2009) A review on plasmonic metal-TiO2 com-posite for generation, trapping, storing and dynamic vectorial transfer ofphotogenerated electrons across the Schottky junction in a photocatalyticsystem. Appl. Surf. Sci., 360, 601–622.

3 Markham, M.C. (1955) Photocatalytic properties of oxides. J. Chem. Educ., 32,540–543.

4 Gupta, S.M. and Tripathi, M. (2011) A review of TiO2 nanoparticles. Chin.Sci. Bull., 56, 1639–1657.

5 Asahi, R., Morikawa, T., Ohwaki, T., Aoki, K., and Taga, Y. (2001) Visible-lightphotocatalysis in nitrogen-doped titanium oxides. Science, 293 (5528),269–271.

6 Jones, P.M. and Dunn, S. (2007) Photo-reduction of silver salts on highlyheterogeneous lead zirconate titanate. Nanotechnology, 18, 185702.

7 Linic, S., Christopher, P., and Ingram, D.B. (2011) Plasmonic-metal nanos-tructures for efficient conversion of solar to chemical energy. Nat. Mater., 10,911–921.

8 Navarro Yerga, R.M., Alvarez Galván, M.C., del Valle, F., Villoria de la Mano,J.A., and Fierro, J.L. (2009) Water-splitting on semiconductor catalysts undervisible-light irradiation. ChemSusChem, 2, 471–485.

9 Bingham, S. and Daoud, W.A. (2011) Recent advances in making nano-sizedTiO2 visible-light active through rare-earth metal doping. J. Mater. Chem., 21,2041–2050.

10 Hou, J., Qu, Y., Krsmanovic, D., Ducati, C., Eder, D., and Kumar, R.V. (2010)Hierarchical assemblies of bismuth titanate complex architectures and theirvisible-light photocatalytic activities. J. Mater. Chem., 20, 2418–2423.

11 Jiang, W., Ullah, N., Divitini, G., Ducati, C., Kumar, R.V., Ding, Y., andBarber, Z.H. (2012) Vertically oriented TiO(x)N(y) nanopillar arrays withembedded Ag nanoparticles for visible-light photocatalysis. Langmuir,28 (12), 5427–5431.

12 Kumar, R.V., Ullah, N., Divitini, G., and Ducati, C. (2012) Photocatalyst com-posites. UK Patent GB 1217403.3, filed Sep. 28, 2012; PCT/EP2013/070301,filed Sep. 27, 2013; European Application: 13773667.4.

13 Shankar, K., Basham, J.I., Allam, N.K., Varghese, O.K., and Mor, G.K. (2009)Recent advances in the use of TiO2 nanotube and nanowire arrays for oxida-tive photoelectrochemistry. J. Phys. Chem. C, 113, 6327–6359.

14 Rincon, A.G., Pulgarin, C., Adler, N., and Peringer, P. (2001) Intercationbetween E. coli inactivation and DBP precursors – dihydroybenzene iso-mers – in the photocatalytic process of drinking water disinfection with TiO2.J. Photochem. Photobiol. A Chem., 139, 233–241.

15 Ollis, D.F. (1985) Environ. Sci. Technol., 19, 480–485.16 Serpone, N. (1995) Solar Energy Mater. Solar Cells, 38, 369–375.

References 523

17 Zhang, J. and Nosaka, Y. (2014) Mechanism of the OH radical generationin photocatalysis with TiO2 of different crystalline types. J. Phys. Chem. C,118 (20), 10824–10832.

18 Fumiaki, A., Yamakata, A., Nogami, K., Osawa, M., and Ohtani, B. (2008)Visible light responsive pristine metal oxide photocatalyst: enhancement ofactivity by crystallization under hydrothermal treatment. J. Am. Chem. Soc.Commun., 130 (52), 17650–17651.

19 Bu, S.J., Jin, Z.G., Liu, X.X., Yang, L.R., and Cheng, Z.J. (2005) Synthesis ofTiO2 porous thin films by polyethylene glycol templating and chemistry of theprocess. J. Eur. Ceram. Soc., 25, 673–679.

20 Bouras, P., Stathatos, E., and Lianos, P. (2007) Appl. Catal. B Environ., 73,51–57.

21 Yoong, L.S., Chong, F.K., and Dutta, B.K. (2009) Development ofcopper-doped TiO2 photocatalyst for hydrogen production under visiblelight. Energy, 34, 1652–1661.

22 Kozlova, E.A., Korobkina, T.P., Vorontsov, A.V., and Parmon, V.N. (2009)Enhancement of the O2 or H2 photoproduction rate in a Ce3+/Ce4+-TiO2system by the TiO2 surface and structure modification. Appl. Catal. A Gen.,367, 130–137.

23 Wang, W., Zhang, J., Chen, F., He, D., and Anpo, M. (2008) Preparation andphotocatalytic properties of Fe3+-doped Ag@TiO2 core–shell nanoparticles.J. Colloid Interface Sci., 323, 182–186.

24 Di Valentin, C. and Pacchioni, G. (2013) Fundamental understanding ofphotocatalytic conversions on semiconductor surfaces. Catal. Today, 206,12–18.

25 Chen, X., Liu, L., Yu, P.Y., and Mao, S.S. (2011) Increasing solar absorptionfor photocatalysis with black hydrogenated titanium dioxide nanocrystals.Science, 331 (6018), 746–750.

26 Brown, M.D., Suteewong, T., Kumar, R.S., D’Innocenzo, V., Petrozza, A.,Lee, M.M., Wiesner, U., and Snaith, H.J. (2011) Plasmonic dye-sensitizedsolar cells using core-shell metal-insulator nanoparticles. Nano Lett., 11 (2),438–445.

27 Tworski, E. (2012) Harnessing sunlight for redox reactions. MPhil thesis.University of Cambridge.

28 Ma, X.-C., Dai, Y., Yu, L., and Huang, B.-B. (2016) Energy transfer in plas-monic photocatalytic composites. Light Sci. Appl., 5, e16017.

29 Nolan, N.T., Seery, M.K., Hinder, S.J., Healy, L.F., and Pillai, S.C. (2010)J. Phys. Chem. C, 114, 13026–13034.

30 Coto, M., Ullah, N., Divitini, G., Ducati, C., and Kumar, R.V. (2017) Tuningthe properties of a black TiO2-Ag visible light photocatalyst produced by arapid one-pot chemical reduction. Mater. Today Chem., 4, 142–149.

31 Ni, M., Leung, M.K., Leung, D.Y., and Sumathy, K. (2007) A review andrecent developments in photocatalytic water-splitting using TiO2 for hydrogenproduction. Renew. Sustain. Energy Rev., 11, 401–425.

32 Shi, L., Li, C., Gu, H., and Fang, D. (2000) Morphology and properties ofultrafine SnO2–TiO2 coupled semiconductor particles. Mater. Chem. Phys.,62, 62–67.


33 Serpone, N., Maruthamuthu, P., Pichat, P., Pelizzetti, E., and Hidaka, H.(1995) Exploiting the interparticle electron transfer process in the photocatal-ysed oxidation of phenol, 2-chlorophenol and pentachlorophenol: chemicalevidence for electron and hole transfer between coupled semiconductors.J. Photochem. Photobiol. A Chem., 85, 247–255.

34 Luo, J. et al. (2012) TiO2/(CdS, CdSe, CdSeS) nanorod heterostructures andphotoelectrochemical properties. J. Phys. Chem. C, 116, 11956–11963.

35 Peter, L.M., Wijayantha, K.G.U., Riley, D.J., and Waggett, J.P. (2003)Band-edge tuning in self-assembled layers of Bi2S3 nanoparticles used tophotosensitize nanocrystalline TiO2. J. Phys. Chem. B, 107, 8378–8381.

36 Papp, J., Soled, S., Dwight, K., and Wold, A. (1994) Surface acidity and photo-catalytic activity of TiO2, WO3/TiO2, and MoO3/TiO2 photocatalysts. Chem.Mater., 6, 496–500.

37 Yu, Y. et al. (2005) Enhancement of adsorption and photocatalytic activity ofTiO2 by using carbon nanotubes for the treatment of azo dye. Appl. Catal. BEnviron., 61, 1–11.

38 Ao, C. and Lee, S. (2004) Combination effect of activated carbon with TiO2for the photodegradation of binary pollutants at typical indoor air level.J. Photochem. Photobiol. A Chem., 161, 131–140.

39 Zhang, Y., Tang, Z.R., Fu, X., and Xu, Y.J. (2010) TiO2-graphene nanocompos-ites for gas-phase photocatalytic degradation of volatile aromatic pollutant: isTiO2-graphene truly different from other TiO2-carbon composite materials?ACS Nano, 4, 7303–7314.

40 Yang, M.-Q., Zhang, N., and Xu, Y.-J. (2013) Synthesis of fullerene-, carbonnanotube-, and graphene-TiO2 nanocomposite photocatalysts for selectiveoxidation: a comparative study. ACS Appl. Mater. Interfaces, 5.

41 Woan, K., Pyrgiotakis, G., and Sigmund, W. (2009) Photocatalyticcarbon-nanotube–TiO2 composites. Adv. Mater., 21, 2233–2239.

42 Yao, Y., Li, G., Ciston, S., Lueptow, R.M., and Gray, K.A. (2008) Photoreac-tive TiO2/carbon nanotube composites: synthesis and reactivity. Environ. Sci.Technol., 42, 4952–4957.

43 Williams, G., Seger, B., and Kamat, P.V. (2008) TiO2-graphene nanocompos-ites. UV-assisted photocatalytic reduction of graphene oxide. ACS Nano, 2,1487–1491.

44 Lee, J.S., You, K.H., and Park, C.B. (2012) Highly photoactive, low bandgapTiO2 nanoparticles wrapped by graphene. Adv. Mater., 24, 1084–1088.

45 Li, X.-Z. et al. (2014) Size-controlled synthesis of Ag3PO4 nanorods andtheir high-performance photocatalysis for dye degradation under visible-lightirradiation. Mater. Lett., 130, 97–100.

46 Gao, H. et al. (2013) Towards efficient solar hydrogen production byintercalated carbon nitride photocatalyst. Phys. Chem. Chem. Phys., 15,18077–18084.

47 Yang, H., Huang, C., Li, X., Shi, R., and Zhang, K. (2005) Luminescent andphotocatalytic properties of cadmium sulfide nanoparticles synthesized viamicrowave irradiation. Mater. Chem. Phys., 90, 155–158.

48 Wang, X. et al. (2009) A metal-free polymeric photocatalyst for hydrogenproduction from water under visible light. Nat. Mater., 8, 76–80.

References 525

49 Deifallah, M., McMillan, P.F., and Corà, F. (2008) Electronic and structuralproperties of two-dimensional carbon nitride graphenes. J. Phys. Chem. C,112, 5447–5453.

50 Yan, S.C., Li, Z.S., and Zou, Z.G. (2009) Photodegradation performance ofg-C3N4 fabricated by directly heating melamine. Langmuir, 25, 10397–10401.

51 Liu, J.-K., Luo, C.-X., Wang, J.-D., Yang, X.-H., and Zhong, X.-H. (2012) Con-trolled synthesis of silver phosphate crystals with high photocatalytic activityand bacteriostatic activity. CrystEngComm, 14, 8714–8721.

52 Kuai, L., Geng, B., Chen, X., Zhao, Y., and Luo, Y. (2010) Facile subsequentlylight-induced route to highly efficient and stable sunlight-driven Ag−AgBrplasmonic photocatalyst. Langmuir, 26, 18723–18727.

53 Wang, X., Li, S., Yu, H., Yu, J., and Liu, S. (2011) Ag2O as a new visible-lightphotocatalyst: self-stability and high photocatalytic activity. Chem. A Eur. J.,17, 7777–7780.

54 Yi, Z. et al. (2010) An orthophosphate semiconductor with photooxidationproperties under visible-light irradiation. Nat. Mater., 9, 559–564.

55 Martin, D.J., Umezawa, N., Chen, X., Ye, J., and Tang, J. (2013) Facet engi-neered Ag3PO4 for efficient water photooxidation. Energy Environ. Sci., 6,3380–3386.

56 Bi, Y., Ouyang, S., Umezawa, N., Cao, J., and Ye, J. (2011) Facet effect ofsingle-crystalline Ag3PO4 sub-microcrystals on photocatalytic properties.J. Am. Chem. Soc., 133, 6490–6492.

57 Martin, D.J. et al. (2015) Efficient visible driven photocatalyst, silver phos-phate: performance, understanding and perspective. Chem. Soc. Rev., 44,7808–7828.

58 Xiao, X. et al. (2012) Oxygen-rich bismuth oxyhalides: generalized one-potsynthesis, band structures and visible-light photocatalytic properties. J. Mater.Chem., 22, 22840–22843.

59 Weidong, H. et al. (2007) The photocatalytic properties of bismuth oxidefilms prepared through the sol–gel method. Thin Solid Films, 515, 5362–5365.

60 Kim, T.W. and Choi, K.-S. (2014) Nanoporous BiVO4 photoanodes withdual-layer oxygen evolution catalysts for solar water splitting. Science, 343,990–994.

61 Guo, Y. et al. (2010) Additive-free controllable fabrication of bismuth vana-dates and their photocatalytic activity toward dye degradation. Appl. Surf. Sci.,256, 2215–2222.

62 Ke, D., Peng, T., Ma, L., Cai, P., and Jiang, P. (2008) Photocatalytic water split-ting for O2 production under visible-light irradiation on BiVO4 nanoparticlesin different sacrificial reagent solutions. Appl. Catal. A Gen., 350, 111–117.

63 Zhou, T., Hu, J., and Li, J. (2011) Er3+ doped bismuth molybdate nanosheetswith exposed {010} facets and enhanced photocatalytic performance. Appl.Catal. B Environ., 110, 221–230.

64 Song, L., Zhang, S., Chen, C., Hu, X., and Wei, Q. (2011) Synthesis of sulphurdoped bismuth selenide photocatalysts by the solvothermal method and theirphotocatalytic activities under visible-light irradiation. Chem. Eng. J., 171,1454–1457.


65 Hou, J., Cao, R., Wang, Z., Jiao, S., and Zhu, H. (2012) Hierarchical nitrogendoped bismuth niobate architectures: controllable synthesis and excellentphotocatalytic activity. J. Hazard. Mater., 217, 177–186.

66 Wu, S. et al. (2010) Synthesis and photocatalytic properties of BiOClnanowire arrays. Mater. Lett., 64, 115–118.

67 Ai, Z., Ho, W., Lee, S., and Zhang, L. (2009) Efficient photocatalytic removalof NO in indoor air with hierarchical bismuth oxybromide nanoplate micro-spheres under visible light. Environ. Sci. Technol., 43, 4143–4150.

68 Guan, M. et al. (2013) Vacancy associates promoting solar-driven photocat-alytic activity of ultrathin bismuth oxychloride nanosheets. J. Am. Chem. Soc.,135, 10411–10417.

69 Li, J., Yu, Y., and Zhang, L. (2014) Bismuth oxyhalide nanomaterials: layeredstructures meet photocatalysis. Nanoscale, 6, 8473–8488.

527

Part V

Challenges and Perspectives of Visible Light ActivePhotocatalysis for Large Scale Applications

529

19

Quantum Dynamics Effects in PhotocatalysisAbdulrahiman Nijamudheen and Alexey V. Akimov

University at Buffalo, The State University of New York, Department of Chemistry, Buffalo, NY 14260-3000, USA

19.1 Introduction

Computational modeling is essential for advancing the vibrant field ofphotocatalysis. High-throughput computational screening accelerates theexperimental design of novel photocatalysts with improved energy conversionefficiencies [1–7]. Theoretical and computational investigations provide theatomic-level basis for understanding the photocatalysts’ solar energy conversionefficiencies, product quantum yields, and selectivity of the materials towardparticular types of reactions. Computations predict and rationalize manyphysicochemical and optoelectronic properties such as molecular and electronicstructure of materials, including their absorption spectra, excited-state energylevels, kinetics, and thermochemistry of the reactions catalyzed by the materialsunder certain conditions. Using modern excited-state dynamics approaches atquantum, semiclassical, or quantum-classical levels, the processes of nonadia-batic charge and energy transfer, electron–phonon interactions, tunneling, andenvironment-induced decoherence can be described in great detail.

Traditionally, the methodologies utilized in practical computational designof photocatalysts address various aspects of the catalytic process. The densityfunctional theory (DFT) and wavefunction (WF)-based methods are commonlyemployed to study static ground-state properties: structure and thermody-namic stability, defect formation and the ground-state reorganization energies,vibrational frequencies (phonons), ionization potentials, and electron affinities,as well as electronic energy levels within a single-particle orbital formulation.DFT has also found great use in investigations of the ground-state processes,including computing free energies and reaction pathways. Although chem-ical transformations are inherently nonadiabatic, the use of multireferenceground-state approaches accounts for these effects in the nominally staticcalculations. Properties more closely related to excited states, such as opticalabsorption and emission spectra, transport properties, and charge carriermobility, are studied within the time-dependent density functional theory(TD-DFT) or many-body wavefunction methods such as the GW approximation(GWA) [8] and the Bethe–Salpeter equation (SBE) [9, 10]. Such techniques


530 19 Quantum Dynamics Effects in Photocatalysis

are computationally demanding and have been restricted to relatively smallatomistic models. Nevertheless, a recent parallel implementation of the GWmethod by Govoni and Galli has been applied for studying the electronicproperties of solids and interfaces with ∼1500 electrons [11, 12], suggesting thatthe application of accurate many-body perturbation theories to photocatalyticprocesses in large-scale periodic systems may be feasible in the near future.Finally, the methodologies based on time-domain simulations of the dynamicsof excited electronic states and coupled electron-nuclear dynamics, knownbroadly as the nonadiabatic molecular dynamics (NA-MD), are indispensable forstudying the mechanisms of charge transfer in bulk materials or at the interfaces,photoinduced dissociation and electron or nuclear tunneling. The need forNA-MD approaches originates from the fact that the processes require at leasttwo electronic (or vibronic) states (e.g., donor and acceptor) that are coupled toeach other and to the environmental degrees of freedom.

On the one side, the conventional DFT and WF-based approaches to study-ing catalytic processes are well developed and broadly used. On the other side,the NA-MD methods and tool are being actively developed and applied to studyprocesses pertinent to solar energy conversion materials. The combination of thetwo, however, has not been well explored. Modeling photoinduced catalytic pro-cesses in a single simulation setup and within a single theoretical framework is achallenging task. It requires addressing questions at various levels:

a) What electronic structure method is suitable for such modeling?b) What dynamical methodology should be utilized to achieve the desired

accuracy?c) How the two approaches should be integrated with each other?d) How the timescales problem can be addressed to accelerate the dynamics?

In this chapter, we review the recent progress in addressing only one of theabove questions. Namely, the role of the adiabatic and nonadiabatic dynamicseffects in photocatalysis is investigated and the suitable approaches to incor-porate these effects are discussed. We further discuss the applications of theNA-MD methods in predicting and rationalizing the efficiency of the photocata-lysts and the underlying mechanisms of chemical reactions. Specifically, we focuson several actively studied photocatalytic processes: (i) methanol oxidation;(ii) water splitting, and (iii) carbon oxide redox reactions at semiconductor sur-faces. We refer the reader to several excellent reviews written on this topic. Anextensive account on the methods of NA-MD simulations, and computationaland experimental studies of the photocatalytic processes at semiconductorsurfaces has been written by Akimov et al. [13]. Recent developments on thesurface hopping based NA-MD methodologies have been summarized by Wanget al. [14]. Nyman [15] has reviewed methods for incorporating tunneling effectsin dynamics and rate calculations. Pham et al. have assessed the challenges in thecomputational modeling of heterogeneous interfaces for the solar water-splittingmechanisms [16]. Recent studies dedicated to the computational screening oftwo-dimensional materials for photocatalytic applications are reviewed by Singhet al. [17]. Wodtke has discussed the importance of nonadiabatic effects on metalsurfaces [18]. Others have reviewed the computational photocatalysis focusing

19.2 Computational Approaches to Model Adiabatic Processes in Photocatalysis 531

on organic semiconductors and nanomaterials [19, 20], organic conjugated mate-rials [21], quantum dots [22], and concerted electron-proton transfer [23, 24].

19.2 Computational Approaches to Model AdiabaticProcesses in Photocatalysis

The computational modeling of a catalytic cycle requires the knowledge of (i) thecorrect geometries of transition states; (ii) the relative energy of all intermediates,reactants, and products; and (iii) the activation free energies for all transitions.The geometry of all the transition states along the reaction path on the ground- orexcited-state adiabatic potential energy surfaces (PES) can be determined using anumber of sophisticated methods. The time-independent interpolation methodssuch as the nudged elastic band (NEB) [25], the climbing image nudged elasticband (CI-NEB) [26], and the growing string method (GSM) [27, 28] are widelyused to locate the transition states for solid-state systems and interfaces. Sub-sequently, local geometry optimizations and frequency analyses can verify thelocated transition states. Alternatively, local surface-walking algorithms [29–32]that use the information of the second derivative of the energies could replace theinterpolation methods. The walking algorithms are particularly useful when

a) a good initial guess for the actual transition state is available (e.g., from exper-imental studies or chemical intuition);

b) the frequency of the vibrational mode that corresponds to the reaction coor-dinate is not small;

c) the region around the saddle point is not shallow.

Synchronous transit methods, namely, linear synchronous transit (LST) andquadratic synchronous transit (QST) and transit-guided quasi-Newton (knownas QST3) methods, are popular local saddle point optimization methods [33–35].

The intrinsic reaction coordinate (IRC) calculations can confirm that the opti-mized transition state indeed connects the correct reactant and product geome-tries [36, 37]. Within the classical transition-state theories (TST), the knowledgeof the activation free energy obtained from the transition-state modeling givesan estimate of the reaction rates. If the reaction occurs adiabatically, one can usethe Arrhenius equation to calculate the rate, provided the pre-exponential factoris known.

Modeling catalytic cycles based on conventional TST and the adiabatic PESfails when (i) no knowledge of the elementary steps of the reactions exists; and (ii)the process occurs nonadiabatically. If the prior knowledge of most probable reac-tive steps is absent, a large number of possible reaction paths must be considered.Such paths can be supported by the chemical intuition requiring a lot of manualwork. In practice, the number of possible pathways that one can consider manu-ally is limited. Even when significant experimental evidences and chemical intu-ition are available, the computational studies can lead to wrong mechanisms [38].

On the contrary, the automated methods exist that aim to construct the mostexhaustive spaces of elementary reactions. The artificial-force-induced reaction


(AFIR) method developed by Maeda and Morokuma imposes an artificial forceto the reactants and the catalysts to find the PES through automated searches[39, 40]. The Zimmerman group has used a GSM termed Zstruct to automatethe search for adiabatic reaction mechanisms [41, 42]. The code AARON,devised by the Wheeler group is specifically useful for finding the mechanismsof organocatalysis [43]. The Martínez group has developed ab initio MD (AIMD)simulation tools to predict the outcomes of complex chemical reactions betweendifferent molecules as that could happen in an actual reaction vessel [44].Automated modeling of time-independent processes can be used to constructthe ground and excited-state adiabatic PESs separately. Born–Oppenheimermolecular dynamics (BOMD) and Car–Parrinello molecular dynamics (CPMD)methods, coupled with transition-state sampling algorithms, are useful tocalculate the timescales for the adiabatic reaction mechanisms [45–47]. Tosummarize, the theory of adiabatic processes relevant for photocatalysis is welldeveloped and advanced computational tools are available to study them.

19.3 Computational Approaches to ModelNonadiabatic Effects in Photocatalysis

A photocatalytic reaction is initiated by the photoexcitation of a solar energyharvester, in which an electron–hole pair is generated. The subsequent separa-tion of the charge carrier and their transfer to the reaction sites, including thoseat the catalysts or cocatalysts, stimulates the chemical reaction. The charge car-riers can lose their energy to a number of side reactions or undergo a recom-bination before they are transferred to the reaction site. The side processes aredetrimental to the overall performance of the material. The dynamics of chargeand energy transfer are strongly affected by the evolution of the system’s nucleiand the overall electronic wavefunction in terms of stationary electronic states.The Born–Oppenheimer approximation that assumes a domination of a singleelectronic state breaks down in many cases [48], including reactive processesand nonradiative electron relaxation from the excited states. For modeling suchprocesses, one needs a time-dependent treatment of nonequilibrium processes,which can be done within the framework of the NA-MD method, which accountsfor multiple coupled PES (Figure 19.1).

The electron dynamics of charge carrier generation, separation, transfer, andrecombination strongly affects and determines the photocatalytic efficiency.Experimental techniques such as pump-probe spectroscopy [49], transientgrating spectroscopy [50, 51], and ultrafast electron diffraction [52] can allprovide a quantitative measure of various processes involving change of elec-tronic states. Sum-frequency generation vibrational spectroscopy [53–56] canprovide vibronically resolved details of such processes. A broad spectrum ofmethods ranging from fully quantum to semiclassical and quantum-classical isavailable nowadays for modeling the nonadiabatic dynamics of charge carriers.Fully quantum treatment of electronic and nuclear degrees of freedom inboth the ground and excited states of a condensed matter system demands

19.3 Computational Approaches to Model Nonadiabatic Effects in Photocatalysis 533

En

erg

y

En

erg

y

Reaction coordinate(a) (b)

Adiabatic Nonadiabatic

Reaction coordinate

Figure 19.1 Schematic representation of (a) adiabatic and (b) nonadiabatic electrondynamics. The PESs that correspond to two different electronic states are shown.

large computational facilities. Therefore, quantum-classical and semiclassicalapproximations are utilized in practical simulations of the electron dynamicsin photocatalytic materials. Semiclassical and quantum-classical approaches totreat the dynamics of vibronic states use a direct product of the Hilbert space ofelectronic basis states (quantum) and the phase space of nuclear positions andmomenta (classical). Three types of popular computational techniques to studythe electron-nuclear coupling and electron dynamics are: (i) mean field (MF) orEhrenfest [57–59], (ii) trajectory surface hopping (TSH) [60, 61], and (iii) neglectof back reaction (NBR), often termed after a more general group of classicalpath approximations (CPA) [62]. The reader is referred to the recent reviewson NA-MD [63], MF theories [64], and TSH [14, 65] methodologies for moredetails. An important challenge quantum-classical approximations often face isthe incorrect description of quantum coherence. When decoherence effects arenot included, both the MF and surface hopping methods lead to faster rates forcarrier relaxation. In the past, many methods have been proposed to address theovercoherence problems in the calculations of nonadiabatic dynamics [66–71].

The electronic transitions in many photocatalytic systems may occur relativelyslowly – on the timescales of nanoseconds (ns) or microseconds (μs). Modelingsuch processes with stochastic approaches such as surface hopping method maybe unfeasible because many trajectories would be required to accurately samplethe rare transition. To circumvent this problem, our group has recently proposedand utilized an accelerated NA-MD method, X-NA-MD [72]. According to thetechnique, the electronic transitions are artificially accelerated via a uniform scal-ing of the NAC between all pairs of states. The relaxation timescales obtained fora range of NAC scaling coefficients (𝛼 > 1), are obtained using relatively smallnumber of short trajectories. Using the scaling law 𝜏

𝛼= 1

𝛼2A+B, the parameters A

and B are found and then used to predict the target timescale, 𝜏1, correspondingto the original NAC. The application of the X-NA-MD to the interfaces of sil-icene and germanane, and their methylated analogs with the TiO2 polymorphshave enabled us to access the electron–hole recombination timescales spanningfrom hundreds of femtoseconds to several dozen of nanoseconds [72].


Time-dependent wave packet (TDWP) approaches such as multiconfigu-rational time-dependent Hartree (MCTDH) [73, 74] and path-integral-basedring polymer molecular dynamics (RPMD) [75–78] are two popular quantumdynamics methods [15]. Although MCTDH provides an accurate descriptionof the evolution of nuclear wavefunction of a system, its use is limited to smallsystems due to very steep scaling of the computational costs with respect to thesystem’s size. Therefore, the use of the MCTDH method to study the details ofphotocatalytic processes in large and even moderately sized atomistic systemsis not practical at the present point. The RPMD is a classical-like quantumdynamics method that uses Kubo-transformed real-time correlation functionsto calculate the dynamical properties of a chemical system [75, 79]. In RPMD,the classical MD trajectories are run in an extended phase space where eachatom is represented by a ring polymer or a set of beads connected by harmonicsprings. This method accounts for the quantum mechanical zero-point energyand tunneling effects, and has been successfully applied to find the thermalrates for reactions in condensed phase systems [75]. Therefore, RPMD can beused to study the photocatalytic events where tunneling and zero-point energywill have critical roles in the rate of the reaction. The applications of RPMD insimulating the tunneling effects in reaction dynamics will be discussed further.One future challenge in this area is to develop RPMD methods that can accountfor the effects due to the electronic transitions. The quantum-classical andsemiclassical methods are computationally less expensive compared to thequantum dynamics approaches. Previous studies have proposed nonadiabatictransition-state theories (NA-TST) to calculate the rate of reactions that proceedthrough multiple PES [80–83]. The NA-TST is a statistical time-independentapproach where direct dynamic calculations are run as a function of the PEScalculated for the multiple electronic states of interest. The major advantagesof NA-TST are that it allows the calculation of the rates for interstate crossingand slow nonadiabatic transitions. The zero-point energy and tunneling can beincluded in the dynamics, although with significantly increased computationalcost. Recently, Sherman and Corcelli [84] have combined the Monte Carlotransition path sampling and fewest switches surface hopping methods to devisea nonadiabatic transition path sampling algorithm which can be used to findthe TS for a reaction in condensed phase while simultaneously consideringthe effects due to nonadiabatic electronic transitions. Although not tested inrealistic systems, this method is promising for finding the nonadiabatic TSs inphotocatalytic reactions.

The quantum-classical and semiclassical methods to perform the nonadiabaticdynamics are implemented in a number of open-source software: Newton-X [85],CPMD [86], JADE [87], SHARC [88, 89], Octopus [90], MCTDH package [91],PYXAID [92, 93], and Libra [94, 95]. Depending on the methods implemented,each of these programs has its own advantages and limitations. Newton-X can beused for on-the-fly TSH calculations with decoherence corrections in molecularsystems. The electronic structure calculations are driven by the external pack-ages, providing an access to various levels of theory, including the multicon-figurational treatment of electronic states. The JADE and CPMD packages relyon the TD-DFT and semiempirical methods coupled to the TSH simulations.

19.4 Quantum Tunneling in Adiabatic and Nonadiabatic Dynamics 535

SHARC is particularly suitable for TSH calculations in molecular systems andhas a nice feature of including spin-orbit coupling using spin-diabatic states. InOctopus, NA-MD simulations can be performed at the MF level through thedirect evolution of electronic density matrix in response to the time-dependentHamiltonian. The dynamics of nuclei are directly coupled to the evolution of elec-trons; however, the MF method is known for its inability to properly accountfor electron–phonon equilibration, which may significantly affect the computedthermal and kinetic properties. With the massive parallelization and support ofthe graphical processing unit (GPU) computations, the package may be a promis-ing tool for studying photocatalytic processes in realistic systems. The MCTDHpackage provides an access to quantum-mechanical description of nuclear wave-functions and their evolution, allowing one to study vibrational energy relaxationand scattering reactions in small molecules. The PYXAID program is especiallyuseful for the NA-MD simulations in condensed matter systems with hundredsof atoms. It provides access to the TSH and MF descriptions of electronic dynam-ics, but neglects the electron-nuclear back reaction, making it impossible to studyphotoinduced reactive processes directly. The recently developed Libra programenables such simulations by the inexpensive treatment of excited states usingthe so-called Δ-SCF approach. The built-in semiempirical Hamiltonians allowmodeling photoinduced dynamics in sufficiently large systems. Recently, Librahas been successfully interfaced to the Quantum Espresso [96] and GAMESS[97] packages, leading to the Libra-X [98] package, which enables more accuratetreatment of photoinduced nuclear dynamics, relying on the rigorous electronicstructure methods.

19.4 Quantum Tunneling in Adiabaticand Nonadiabatic Dynamics

Both electron and proton have small masses and large de Broglie wavelengths.Therefore, they are capable of tunneling through potential energy barriers – theprocesses that must be accounted for in modeling chemical transformations.Generally, electron tunneling contributes to the rates of electron transfer (ET)processes and therefore can facilitate certain redox processes. Electron tunnelingis also known as a through-space transition. It may be responsible for efficientlong-range ET and coherence preservation that leads to exceptional robustness ofthe excitation energy transfer in artificial and natural photosynthetic complexes.Depending on intrinsic properties of the system, tunneling may be suppressedin favor of an incoherent electron hopping regime, in which ET is realized via asequence of short-distance site-to-site hops. The prevalence of one mechanismover the other depends on the details of electronic structure of donor, bridge,and acceptor sites, as demonstrated for DNA [99, 100], olygopeptides [101,102], and proteins [103–105]. An important message from these observationsis that electron tunneling may be essential for efficient operation of complexphotocatalytic complexes, where a directed energy transfer over long distancesis necessary.


Adiabatic basis the entire system

Barrier, C

Diabatic basis of

Donor, A Acceptor, B

{|Ψi⟩}

{|ΨiD⟩}∼

{|ΨiA⟩}∼

Figure 19.2 Tunneling and superexchange as the consequences of representation choice.{|��D

i ⟩} and {|��Ai ⟩} are the eigenstates of the isolated donor (left) and acceptor (right),

respectively. These states are non-stationary from the point of view of the adiabatic states ofthe overall system A–C–B, {|𝜓 i⟩} (dotted line), but they can be regarded as the diabetic states.The time evolution of the projections ⟨��A

i |𝜓j⟩ describes the kinetics of tunneling.

At the fundamental level, tunneling may be related to the so-called superex-change effect. The latter arises when two states, A and B, are not directlycoupled to each other and are separated by a high-energy intermediate stateC (Figure 19.2). These states can also be regarded as the diabatic states repre-senting a donor, an acceptor, and a barrier, respectively. The transition betweenthe states A and B from the perspectives of a state hopping would requirean improbable transition from A to C, which would reduce the probability(and hence the rate) of the overall transition. At the same time, the coherentevolution of the wavefunction according to the time-dependent Schrödingerequation would result in a notably faster population to the state B. This resultcan be understood from the perspective of representation transformation. Theeigenstates of the overall system A–C–B, {|𝜓 i⟩}, are adiabatic (by definition) andare delocalized over the domain of all states (Figure 19.2, the dotted lines). Theinitially populated donor states, |��D

i ⟩, can be represented in the adiabatic basis,{|𝜓 i⟩}, of the states that are delocalized over the entire system. The evolution ofthe expansion coefficients will eventually lead to the increase of the projection ofthe adiabatic states of the entire system onto the diabatic states of the acceptor B,⟨��A

i |𝜓j⟩, mimicking tunneling. It should be emphasized that the process ofquantum mechanical tunneling is not fully differentiated from the standardnonadiabatic pathways for charge transfer. Within the exact quantum treatment,the distinction vanishes: quantum tunneling is already incorporated in thenature of the quantum transition. The distinction becomes more notable whenapproximate methodologies for NA-MD and tunneling are utilized.

Electronic tunneling can be accounted for within the NA-MD methods basedon the wave-packet propagation, path-integrals, and semiclassical theories.For instance, the RPMD have been extensively used by Ananth and Miller


[76, 106, 107], Cao and Voth [108–110], Ceriotti [111, 112], Markland [113–115],and others. There are good indications that tunneling (both electronic andnuclear) can be reasonably described within the trajectory-based schemes suchas quantized Hamiltonian dynamics (QHD) [116, 117] and entangled trajecto-ries [71, 118–120]. The approaches originally meant to treat the superexchangeproblem can well be applied to describe electron tunneling [71, 121, 122].Unfortunately, the latter methods have not been applied to atomistic systems sofar, although their classical-like nature may make them suitable for modelingtunneling in fully atomistic systems. The above discussion also indicates that awide range of tools for handling electronic tunneling under various situationshave been developed. The application of these tools and methods to photocat-alytic problems has not yet reached the routine stage and is yet a state of the art,waiting to be fully explored.

In the majority of chemical reactions, the neglect of atomic tunneling is welljustified and will not lead to any notable changes in the kinetics and reactionselectivity. However, in reactions where proton transfer is the rate-determiningstep, the inclusion of tunneling in the dynamics may become pivotal to obtainaccurate rates and the correct selectivity [123]. Proton tunneling will dominatethe kinetics when the reaction temperature is low, the barrier for the reaction issmall, and the barrier width is narrow. Proton tunneling increases the rates of ele-mentary processes in organic and organometallic transformations [123], enzymecatalysis [124], surface reactions [125–127], and interstellar chemistry [128]. Theunusually large H/D kinetic isotope effect (KIE) and a curved Arrhenius plot dueto the deviation from classical kinetic behavior are the two direct consequencesof tunneling in the rate of a reaction that can be verified from experiments.

Although rarely observed experimentally, tunneling can lead to productsthat are different from those predicted based on the activation free energy(ΔG‡) and the free energy (ΔG) of the reaction [129]. In a “tunneling-controlledreaction,” the tunneling product dominates over the kinetic and thermodynamicproducts. The rearrangement of methylhydroxycarbene (H3C—C—OH) in anAr matrix (at 11 K) is a classical example of a tunneling-controlled reaction(Figure 19.3) [130]. The hydroxyl hydrogen atom in the H3C—C—OH tunnelsthrough a barrier (TS1) of 28.0 kcal mol−1 to produce a [1, 2]-hydrogen shiftproduct, CH3CHO, whereas the TST predicts the formation of an unobservedproduct, vinyl alcohol (CH2CHOH), which has a significantly smaller activa-tion barrier (TS2, 22.6 kcal mol−1). Tunneling is the only reaction pathway herebecause the rate of the classical over-the-barrier reaction is extremely low at 11 Kto observe any product formation. Among the first approaches, a reaction-pathHamiltonian model, [131] the Wentzel–Kramers–Brillouin (WKB) approxima-tion, [132, 133] had been applied to study probabilities of proton tunneling. Thetunneling rate is calculated as a product of the WKB transmission coefficientand the classical rate, with which the reactant hits the barrier. This approach hasbeen used to calculate the tunneling lifetime in the reactions in small organicmolecules and metal surfaces [134–136]. The calculated tunneling lifetime(t1/2 = 71 min) for CH3CHO was in reasonable agreement with the experimentalone (t1/2 = 60 min). Multidimensional small curvature tunneling (SCT) [137]and large curvature tunneling (LCT) [138] approximations are usually applied to


H

TS1 TS2

H

‡‡

H

H

C O

O

HC O

H

H

H

H

C

H

H

H

H

C O

H

–39.8

22.6

28.0

0.0

ΔH(kcal

mol−1)

H

H

H

C OH

–50.7

H

H

Figure 19.3 PES for the rearrangement of H3C—C—OH in Ar matrix (11 K). H tunneling leadsto the experimentally observed product H3CCHO. ΔH indicates the relative enthalpy changes(in kilocalories per mole) along the reaction coordinate.

account for the corner-cutting tunneling effects on the adiabatic PES [139–142].The classical TST overestimates the rates because it does not consider therecrossing of the activated reactant species that reach the first-order saddlepoint. The variational TST (VTST) accounts for the recrossing effects, and isconsidered a suitable model to calculate the rates of chemical reactions from thePES when the tunneling corrections are included [143, 144]. Although less accu-rate than SCT and LCT, one-dimensional Eckart [145] tunneling correction canbe applied to find the rates in reasonably large systems [146, 147]. Alternatively,one can use the Feynman path-integrals-based methods, namely, instantontheory [148–151], ring-polymer molecular dynamics [78, 152], and centroiddensity method [153, 154], to account for tunneling in chemical reactions.

The proton transfer and the dissociation of O—H and C—H bonds are vitalprocesses during the photocatalytic water-splitting and photoinduced organicreactions. Therefore, the tunneling corrections should be added when simulat-ing the dynamics of these reactions, especially when the reaction temperature islow and the barrier is narrow. Tunneling can lead to subtle effects on the ratesof the rearrangement of reactive intermediates such as radicals and carbenes.Many photocatalytic reactions proceed through radical intermediates. Therefore,the tunneling corrections may be required to model the dynamics accurately.At cryogenic temperatures, the classical rate becomes negligible and tunnelingfrom the zero-point energy-level accounts for ∼100% of the reaction. In the limitof sufficiently large temperatures, the classical barrier crossing becomes dom-inant and the tunneling contribution diminishes. The photocatalytic reactionsare often performed at near room temperatures or higher (about 300 K). In thistemperature range, both thermally activated barrier crossing and tunneling fromthe excited vibrational energy levels contribute to determining the pathways ofphotocatalytic transformations.

Typically, tunneling due to atoms heavier than H can be neglected. However,the C atom tunneling has been shown to be important for kinetics of some


D

D

D

D

(a)

(b)

(c)

(d)

(e)

FF

Cl

Cl

N

MeS

H3C

CH3

CH3

CH3

CH3

CH4 +

+

CCH2C(CH3)2

+

CH3C(CH3)2

+C2H6

C+

N

MeS

(f)

(g)

(h)

CH2

CH2

Figure 19.4 Reactions that are dominated by C atom tunneling: (a) the automerization ofcyclobutadiene; (b) the Cope rearrangement of semibullvalene; (c,d) the ring opening ofcyclopropylcarbenyl radical and tetrahedryl-tetrahedrane; (e–g) the ring expansion of1-methylcyclobutylfluorocarbene, noradamantylchlorocarbene, and benzazirine; and (h) thedecomposition of C(CH3)5

+.

organic and organometallic reactions [155]. For instance, at the cryogenic tem-peratures, the ring opening of cyclopropylcarbenyl radical [140], the Coperearrangement of semibullvalene and barbaralane [156], the ring expan-sion of 1-methylcyclobutylfluorocarbene [157], benzazirine [158], andnoradamantylchlorocarbene [159], and the automerization of cyclobutadiene[160–162], all occur almost entirely due to C tunneling (∼100%) (Figure 19.4).Theoretical studies have shown the predominant role of C tunneling in thelow temperature decomposition of C(CH3)5

+ [163] and the rearrangement ofthe strained molecule, tetrahedryl-tetrahedrane to carbene [164]. For chemicalreactions that occur at the room temperature, the C atom tunneling from highervibrational state accelerates the rearrangement reactions and the reactionsinvolving radical/diradical intermediates [141, 165–171]. The tunneling correc-tions increase the rate by <50% in the temperature range used for the respectiveexperiments.

A number of semiclassical and quantum-classical NA-MD methods havebeen developed to incorporate tunneling in chemical dynamics. The Truhlargroup has developed a multistate approach that incorporates multidimensional


tunneling corrections via an army ants tunneling algorithm [172]. In the originalwork, the authors utilized the MF approximation to produce the nonadiabatictrajectories and the electronic wavefunction is assumed to be fully coherent inthe tunneling region whereas the decoherence is allowed in the classical regionsby the coherent switching with decay-of-mixing (CSDM) method [67, 173].Recent experiments have suggested that H tunneling contributes significantlyto the dynamics of dissociation of phenol (C6H5OH → C6H5O⋅ + H⋅) in its firstexcited state (S1) [174–177]. The photodissociation dynamics of phenol havebeen used as a model system for the theoretical studies of nonadiabatic tunnel-ing. Truhlar and coworkers have studied the details of this reaction by using acombination of four theoretical methods [178, 179]: (i) a direct diabatizationof electronic states by a fourfold method to obtain the diabatic PES [180, 181],(ii) anchor-points reactive potential (APRP) method to fit the diabatic PES as afunction of 33 nonredundant internal coordinates [182], (iii) CSDM method fortreating the coherence (tunneling region)–decoherence (classical regions) effectsof the electronic states, and (iv) army ants tunneling algorithm. The simulatedresults accurately reproduced the bimodal behavior of the kinetic energy spectraobserved in experiments for the photodissociation. This work emphasizes theimportance of electronic coherence and tunneling in correctly predicting thelow kinetic energy release found for the photodissociation of phenol.

By means of a two-dimensional model of coupled ab initio PESs, Xie et al.[183] have shown that the diabatic models fail to capture the tunneling near theavoided crossing region of two interacting electronic states. In another work,they used a three-dimensional wave-packet dynamics method to model the pho-todissociation dynamics of C6H5OH and C6H5OD in the S1 state as a function offull dimensional coupled diabatic electronic states. Their calculations inclusiveof nonadiabatic tunneling lead to a life time (2.29 ns) of C6H5OH in its lowestvibrational level (00), showing good agreements with the experimental values thatvaried from 1.2–2.4 ns [184–186]. As a clear indication of the tunneling effects,the rates of photodissociation are notably slowed down upon deuteriation, inC6H5OD.

Among other methodologies for studying nonadiabatic tunneling effects, theNakamura group has devised a one-dimensional analytical approach using modelsystems for incorporating the one-dimensional tunneling corrections in nonadi-abatic trajectories [83, 187]. The Marx group has adopted a combination of theTD-DFT and ab initio path integral (PI) formalism to incorporate the thermal andquantum effects in the photoabsorption spectra of lithium clusters at low tem-peratures [188]. The Hammes-Schiffer group has extensively studied the role ofquantum mechanical tunneling in the proton-coupled electron transfer (PCET)reactions [24, 189]. The nonadiabatic tunneling theories have evolved over theyears and are now capable of modeling electron and nuclear tunneling effects inthe photochemical reactions. However, these techniques are yet to be applied forthe simulation of systems beyond small molecules and model systems; especially,heterogeneous photocatalytic reactions that evolve through a large number ofelectronic, nuclear, and vibrational degrees of freedom.

19.5 The Mechanisms of Organic Reactions Catalyzed by Semiconductor Photocatalysts 541

19.5 The Mechanisms of Organic Reactions Catalyzedby Semiconductor Photocatalysts

Metal oxide semiconductors are widely explored as photocatalysts becausethey are thermally stable, inexpensive, abundant, and environmentally friendly[190–195]. Their bandgaps can be tuned by a suitable chemical modification toabsorb light in the UV–visible region (1–3 eV) [196–202]. The majority of themetal oxide semiconductors require UV–visible-light irradiation for activatingthe reactions. Generally, UV-light-induced reactions are nonselective comparedto the visible-light photocatalysis. Lang et al. have reviewed the scope of themetal oxide photocatalysts for selective organic transformations [192, 203]. Guoet al. have reviewed the recent mechanistic studies of the photochemical reac-tions catalyzed by TiO2 [204]. Here, we focus on recent computational studiesof nonadiabatic processes that are critical to understanding the photocatalyticmethanol oxidation, water splitting, and carbon oxide redox reactions.

19.5.1 Methanol Photooxidation on Semiconductor Surfaces

The photoinduced oxidation of methanol (CH3OH) to formaldehyde (CH2O) onrutile TiO2 (110) surface is a reaction that has been widely adopted as a modelfor the computational studies of heterogeneous photocatalysis. Many experimen-tal studies have provided directly certain details of the interfacial charge transferdynamics [205, 206] and the mechanisms of different stages of this reaction [204,207–210]. Therefore, a direct comparison of theoretical results with the exper-imental findings is possible. The elementary steps of the reaction occur on thetimescale ranging from sub-picoseconds to few picoseconds and a moderate size(<200 atoms) unit cell can be used to mimic the experimental reaction condi-tions. Thus, even sophisticated quantum-classical dynamics methods can be usedto model photooxidation of MeOH/TiO2 directly.

Previous studies have proposed two unique mechanistic pathways for thisreaction. According to the two-photon absorption mechanism [207, 211–216],a physisorbed MeOH molecule releases a proton of its OH group to nearbyO atoms of the TiO2 surface, even under ambient conditions. This reaction fur-nishes a ground-state methoxy (CH3O) species and a hydroxyl group adsorbedon the TiO2 surface. In a subsequent rate-limiting step, one of the C—H bondsdissociates upon a photon absorption to produce CH2O. The initial O—Hbond dissociation occurs under ambient thermal conditions, whereas a highertemperature (600 K) is required for the C—H bond dissociation. Using sumfrequency generation vibrational spectroscopy (SFG-VS), Feng et al. [217]. haveidentified that the binding mode of CH3O on TiO2 (mono- vs bidentate) stronglyinfluences the rates of the C—H bond photodissociation. The bidendate modehas been found less active or inert. This effect is not yet completely understood.One can argue that the observed effect could be due to stronger coupling of theadsorbate and substrate states in such configuration, enhancing dissipation ofelectronic excitation energy. Such a hypothesis is yet to be verified via directNA-MD simulations.


hν

or

H

(a)

(b)

(c)

H

H

O

C

HO

Ti

HH

H

O

C

HO

TiTi

hν

hν

HH

H

OO

O O

C

Ti

H H

H

HH H

OO

O

O

CH3

H

CH

H

O

O

TiTi

Ti Ti Ti

HH

H

OO

C

Ti

H H

H OO

C

Ti

H H

O

Ti

Δ

Figure 19.5 Possible two-photon mechanism for the photoinduced oxidation and C—Ccoupling reactions of CH3OH over TiO2. (a) thermal or photoinduced CH3O generation,(b) oxidation of CH3O to CH2O, and (c) C—C coupling of CH3O and CH2O producing HCO2CH3.

The presently available theoretical and experimental studies reveal that thereverse reaction of CH2O to CH3O is kinetically unfavorable, while the desorp-tion of CH2O from TiO2 surface is readily feasible under the ambient conditions[218, 219]. Mass spectrometry and scanning tunneling microscopy (STM) stud-ies by Friend and coworkers [209] revealed that the product CH2O can undergo afurther oxidation to form CHO. Subsequently, an ultrafast reaction of CHO withresidual CH3O led to the C—C coupling product, methyl formate (Figure 19.5)[208, 209]. The details of these processes have not been explored to the full extentyet. A comprehensive understanding of these processes would require reactivedynamics simulations that also account for electron–phonon energy transfer. Theconventional computational studies that utilize ab initio calculations to under-stand the static properties of ground- and excited-state electronic properties ofCH3OH/TiO2 and the mechanisms of some oxidative processes have been under-taken by several groups [212, 213, 220, 221].

Ground-state DFT calculations of Liu and coworkers [222] indicate that theC—H bond scission is the rate determining step in the methanol oxidation andmethyl formate formation on perfect and defect-containing rutile TiO2 (110)surfaces. The Luo group [223] described the energetics and the mechanisms forthe proton-coupled hole transfer reactions from small molecules such as H2O,CH3OH, HCO2H, and CH2O on the anatase TiO2 (101) surface. Recent, abinitio NA-MD calculations by Chu et al. rationalize the dynamics of forward andreverse hole transfer between TiO2 and CH3OH [224]. Their results suggested


that at 100 K, the reverse hole transfer from CH3OH to TiO2 is faster (∼10 fs)than the forward hole transfer (∼150 fs). Therefore, CH3OH is ineffective as ahole scavenger. Instead, the CH3O species may trap holes more effectively (15%).Their study indicates that the CH3O may be the active species involved in theoverall photooxidation to CH2O. Overall, the work demonstrates the great useof the NA-MD simulations in obtaining a deeper insight into the photocatalyticprocesses, by accounting also for the dynamical effects. Unfortunately, the lim-itations of the methodology utilized (neglect of electron-nuclear back reaction,neglect of electron–hole interaction and one-particle approximation) make itunable to provide on-the-fly excited-state nuclear dynamics. Therefore, thedetails of the photoinduced bond breaking remain inaccessible yet.

Several other authors have addressed the limitations of the method usedin the work of Chu et al., although not necessarily within the NA-MD sim-ulations context. For instance, Migani et al. had discussed the energy levelalignment of the methanol and methoxy species on TiO2 as computed bythe many-body quasiparticle techniques [221]. They suggested that only anadvanced self-consistent GW method was able to describe the experimentallymeasured spectra, whereas simpler theories based on the DFT description,even with hybrid functionals, as well as the non-self-consistent GW calcula-tions failed to reproduce the energy-level alignment. Together with this, theircalculations suggested that the experimentally measured optical absorptionspectra correspond to CH3OH molecule adsorbed on TiO2 rather than tothe CH3O species. Recently, a one-photon absorption mechanism has beensuggested [225], according to which the CH3OH molecule adsorbed on TiO2can be considered the main active species involved in the photooxidation toCH2O, in contrast to a two-photon photodissociation mechanism. The excitongenerated during the photoactivation of the system drives the conversion ofCH3OH to CH3O via an excitonic interfacial PCET mechanism. The studies ofMigani and coworkers point out to the importance of using high-level electronicstructure methods in describing energetics of the methanol and methoxy groupsat the interface. The use of such methods may lead to qualitative changes in thepredicted mechanism, although the calculations are admittedly expensive. It is,thus, highly desirable that efficient yet qualitatively accurate methodologies aredeveloped for describing the electronic structure of intermediates involved in allsteps of the photocatalytic process.

Together with the new developments and qualitatively new insights obtainedfrom the electronic structure calculations, new developments in describing cou-pled electron-nuclear dynamics have recently been reported. Kolesov et al. [226]have utilized the Δ-SCF approach to inexpensively describe electronic excitedstates. The methodology has been coupled to the Ehrenfest method to studymechanisms of the photooxidation of CH3O to CH2O on the rutile TiO2 (110)surface. TheΔ-SCF technique allows one to access the lowest singlet excited stateand accounts for electron–hole interactions. The approach directly addressesthe problem of electron-nuclear back reaction, extending the NA-MD dynamicsbeyond the CPA level. Recently, a similar technique has been successfully utilizedby one of us in modeling photoinduced isomerization of small molecules [94].However, unlike the work of Kolesov et al., the nonadiabatic transition has been


treated using the TSH approach, which has the advantage over the Ehrenfestmethod of being able to provide a proper electron-nuclear thermal distribution.The corresponding computational scheme is available in the open-source Libramethodology development library [94, 227]. Recently, we have extended theΔ-SCF-NA-MD approach to the DFT level via the open-source interface of theLibra package with the Quantum Espresso code [96].

In their work, Kolesov et al. performed NA-MD calculations for a numberof trajectories with the starting geometries of MeO/TiO2 randomly selectedfrom the ground-state distribution sampled by the BOMD. Starting with thefirst excited electronic state, all structures evolved into CH2O by C—H cleavagewithin the time scale of 200 fs. Although these initial insights are highly valuable,the known limitations of the Ehrenfest dynamics (e.g., thermal equilibrium andovercoherence) are yet to be addressed by using alternative quantum-classicalor fully quantum methods. In addition, such questions as the role of the initialphotoexcited state on the resulting products, the details of vibronic structureand resolution of electron–phonon couplings are yet to be addressed in thefuture studies. The investigation of the photoinduced reaction using NA-MDsimulations that include explicit excitonic effects is another potential route forfurther developments and studies.

19.5.2 Water-Splitting Reactions on Semiconductor Surfaces

Water splitting has been an active area of research for many experimentalistsand theorists for a long time [195, 228–236]. This is not surprising, since theprocess itself may have an enormous impact on the production of energy fuelsfrom renewable sources. Studies of the water-splitting reactions are important foradvancing our understanding of the photocatalytic materials and processes. Thestatic and dynamic properties of water/semiconductor interfaces and the mecha-nisms of water splitting have been studied extensively by using adiabatic theories.There are many reviews [16, 237–239] written on this topic and here we do notattempt to discuss them in detail. We only summarize some of the key aspects ofprevious computational studies.

1) It is not clear if water adsorbs associatively or dissociatively on many pho-tocatalyst surfaces. Therefore, both H2O and OH− adsorbed on surface havebeen proposed as reactive species, leading to distinct catalytic cycles.

2) DFT and hybrid DFT may fail to describe the excited-state electronicstructure of water/catalyst interface correctly. Therefore, the reliability ofDFT results should always be verified against the results from high-orderWFT and GW methods.

3) Theoretical studies usually neglect bulk water effects, or rely on implicitsolvent models and cluster models containing few explicit water molecules.These approaches may not necessarily represent the realistic system and canproduce erroneous results.

4) Cocatalysts play an important role in accelerating water-splitting reactions.But their participation in the dynamics and kinetics have not been seriouslyconsidered in the computational models.


5) Cluster models are widely used to replicate the semiconductor surfaces. Thereliability of these simplified models should always be validated by periodiccalculations.

6) Since the water decomposition involves protons, quantum effects of twotypes become important: (a) ET and nonadiabatic transitions; and (b) protontunneling/zero- point energy. The two processes are closely related such thatthey are not separated and may be treated within the PCET concept.

7) The dynamics of PCET are vital for water oxidation and are computationallystudied by using AIMD methods. From our discussions of tunneling in chem-ical reactions, it is evident that H tunneling should be accounted in dynamicsto fully rationalize PCET. H tunneling in PCET step of water oxidation cat-alyzed photosynthetic complexes, and other homogenous catalysts are welldocumented [240–243]. But for reactions occurring on periodic semiconduc-tor and heterostructured photocatalysts, further studies are required.

8) NA-MD simulations can unravel the recombination timescales and suggestmethods to suppress this process, and thereby increase the catalytic efficiency.Compared to many experimental studies [190, 244–247], only a small numberof computational investigations have addressed the nonadiabatic dynamicsof charge separation, transfer, and recombination during the photocatalyticwater-splitting reactions on semiconductors [248–251].

Here we discuss some of the theoretical studies that use NA-MD tech-niques in modeling the electron-nuclear dynamics during the photocatalyticwater-splitting reactions. The Furche group has studied photooxidation ofH2O over TiO2 nanoparticles using TD-DFT, NA-MD, and surface hoppingtechniques [252]. The most favorable mechanism consists of an electron-protontransfer from a physisorbed H2O to a localized hole on a protonated bridgingoxygen O (Obr) of TiO2. Explicit consideration of excitonic interaction ofelectron–hole pair was necessary to explain the mechanism. Ehrenfest NA-MDsimulations performed by assuming the physisorbed H2O as the active speciessuggest that the photooxidation of H2O is favored when certain Ti-interstitialdefects are present [253]. Several experimental [254–257] and theoretical studies[258–264] have demonstrated the use of GaN in photocatalytic water-splittingreactions, and the adiabatic dynamics and reaction mechanisms. But the nona-diabatic processes in these systems have only rarely been addressed. Akimovet al. have studied the dynamics of photogenerated holes and subsequent protontransfer reactions during the photocatalytic water splitting over the GaN (10–10)surface [265]. Their NA-MD calculations suggested that the hole generated atthe GaN/water interface relaxes to its lowest-energy excited state on a timescaleof 100 fs. The hole is localized on the surface N atoms during this process. Forthe water-splitting reaction, the hole should be localized on the O atom of thesurface hydroxyl. During the energy relaxation process, the hole transfers to theO atom. The short dwelling time of the hole (∼50 fs) on the surface hydroxylduring its relaxation kinetics explains the poor efficiency observed for watersplitting on pure GaN surface and the requirement of cocatalysts for enhancedcatalytic activities. Further AIMD calculations support the view that the faster—OH deprotonation accelerates the rate of water splitting, whereas the reverse


effects observed with —NH deprotonation [261]. NA-MD methods have provedsuccessful in studying certain details of water splitting on semiconductors.However, the dynamics of the complete catalytic cycle on various surfaces, theeffects of electron-nuclear coupling, excitonic effects, and tunneling effects areyet to be explored.

19.5.3 Carbon Oxide Redox Reactions on Semiconductor Surfaces

Similar to the water splitting reactions, CO oxidation and CO2 reductionare two fundamental chemical reactions with a wide range of environmentaland industrial applications [266, 267]. Theoretical studies have provided anatomic-level understanding of the mechanisms of these reactions [16, 268,269]. Here, we discuss some of the computational studies of the nonadiabaticprocesses in the reaction. The heterogeneous CO oxidation on a semiconductorphotocatalyst proceeds via the hole trapping by a surface-adsorbed O2

2− thatproduces physisorbed O2

− species. The reaction of O2− with CO furnishes

CO2 [270]. The Glezakou group has studied the charge transfer kinetics andnonadiabatic effects in the O2 adsorption, CO oxidation, and CO2 reduction onan oxygen vacancy (Ov) site of TiO2 rutile (110) surface [271]. Their kinetic studybased on charge-constrained DFT (cDFT) and Marcus rate theory explainsthat the thermal CO oxidation and CO2 reduction reactions over TiO2 are notfavored because of the weak electronic coupling between the initial and finalstates with localized and well-separated charges. But the reactions can occurunder photochemical conditions where the nonadiabatic coupling between theelectronic states becomes more important. Usually the CO oxidation reactionsare activated by the molecular oxygen adsorbed on the catalyst surface. Apartfrom directly participating in the stoichiometric elementary reaction, oxygenacts as an electron scavenger, thereby suppressing the electron–hole recombina-tion [272]. Theoretical studies predict that the absorption of molecular oxygenin the peroxide (bidendate, O2

2−) state is more stable than the superoxide (mon-odendate O2

−) state. The Selloni group has studied the dynamics, energetics, andthe nonadiabatic effects in the adsorption, desorption, and chemical reactions ofO2 on metal oxide photocatalysts using reactive AIMD calculations [272–275].Their studies establish that even though the peroxo (*O2

2−) species is morestable than the experimentally found *O2

−, the conversion from *O2− to *O2

2− isthermodynamically unfavorable due to an energy barrier of 0.3–0.4 eV.

Experiments have indicated that the turnover numbers of Ru-complex/Ta2O5-catalyzed photochemical CO2 reduction are two orders of magnitude higherwhen the PO3H2 group is used as a linker to connect Ru-complex on the Ta2O5surface, compared to the use of CO2H as a linker [276–278]. Contrary to thisfinding, the experimentally measured redox potentials suggest larger ET toreactant when COOH is used. A DFT computational study using Ru(di-X-bpy)(CO)2Cl2/Ta2O5 (bpy= bipyridyl ligand, X= anchor groups: PO3H2, COOH, andOH) as model systems suggests that the steric interactions and the localizationof acceptor states in the anchor groups determine the rate of ET to the reactioncenter and the catalytic efficiency [279]. The NA-MD simulations have beenapplied to rationalize distinct turnover numbers in Ru complexes attached to

19.6 Conclusions and Outlook 547

the substrate with various anchor group [280]. It was concluded that the surfacecoverage effects may dominate the otherwise favorable ET rates. The electronicstructure calculations and experiments indicate that the surface coverage will bereduced for systems with COOH because the COOH-surface bond can be easilydissociated, in contrast to the PO3H2 bond, which is significantly more stable.Another critical factor, the free energy change along the PES for the reactionhas not been addressed yet. All these studies point out to the importance of thecombined use of adiabatic electronic structure methods, adiabatic/nonadiabaticPES calculations, and the NA-MD calculations to fully rationalize the role ofanchor groups in photocatalysis.

19.6 Conclusions and Outlook

In this review, we have discussed recent developments in the computationalapproaches to study photocatalytic processes, involving charge transfer andexcitation-induced reactions. In particular, we have focused on the adiabaticand nonadiabatic dynamic processes as well as the tunneling effects. Thecomputational methods to study the ground- and excited-state adiabatic PESare well developed and are routinely applied to a variety of systems. The devel-opment of fully automated computational approaches to calculate thousands ofunique mechanistic pathways and predict the most favorable adiabatic reactionmechanisms without any prior knowledge from experiments is an active areaof research. The automated, reaction mechanism search methods are expectedto significantly improve the power of computational screening of advancedphotocatalysts for selective chemical transformations.

Quantum and quantum-classical dynamics methods for modeling the nona-diabatic processes, including in the photocatalysts, have advanced significantlyover the past couple of decades. The kinetics and mechanisms of charge carriergeneration, separation, and transfer can nowadays be calculated with a reason-able accuracy by using the MF or SH theories. The inclusion of decoherenceeffects is important for gaining a better accuracy of computed charge and energytransfer rates. However, many NA-MD implementations rely on one-particleapproximation of electronic states, and neglect electron–hole interaction andexcitonic effects. The development of new efficient methods accounting for theseeffects is needed. The comparative studies benchmarking these approximationsagainst rigorous many-body theories may help understand the validity scope ofthe used approximation and their potential impact on the accuracy of computedproperties. Some techniques for larger systems are limited by the neglect ofelectron-nuclear back reaction, making them inapplicable to modeling thephotodissociation processes. The fully quantum methodologies can capturethese effects, but are too expensive to be applied for realistic photocatalyticsystems. The development of the new efficient techniques that can be used tomodel photoinduced reactive processes, including bond breaking/formation aswell as structural reorganization, will be an important challenge of computa-tional photocatalysis to address in the coming years. Presently such efforts are


being undertaken in several groups, including our efforts on the Δ-SCF-NA-MD methodology, which relies on the time-domain DFT, TSH, and Δ-SCFtechniques.

The inclusion of electronic tunneling effects is critical for accurately modelingthe dynamics of long-range electron and excitation energy transfer. Electronictunneling has been accounted for in a number of wave-packet propagation,RPMD, and semiclassical NA-MD theories, and computational tools are avail-able to study them. The rates of PCET during the photocatalytic water splittingand C—H or N—H activation processes may be affected by the cooperativeaction of nonadiabatic electronic transitions and proton electron tunneling. Thenonadiabatic tunneling theories have shown the importance of electron andH tunneling in water splitting and photodissociation of small organic molecules,which may be an important factor to consider in photocatalytic processes. Heavyatom tunneling has been critical to explaining the rates of a number of chemicalreactions, but it is neglected in the computational modeling of photocatalyticreactions. With the advent of powerful computational methodologies in recentyears, we expect that the future studies will unravel the details of electron, H, andheavy atom tunneling effects in the photocatalytic reactions occurring on solids,surfaces, and interfaces. The inexpensive quantum-classical methodologies thatcapture tunneling and superexchange have been developed recently in variousresearch groups, but are yet to be applied to study the corresponding processes(electron or proton/heavy atom tunneling) in realistic systems. The choice ofthe method that delivers an optimal combination of accuracy and efficiencyin a broad range of realistic systems remains unclear, prompting for futurecomparative studies. It is also of interest to know the potential implications oftunneling on the selectivity of photocatalytic processes in particular applications,such as water oxidation/reduction.

We have demonstrated the successes and the limitations of existing NA-MDmethods in investigating the mechanisms of three sets of reactions, namely, (i)methanol oxidation, (ii) water splitting, and (iii) carbon oxide redox reactions.Computational studies have been successful in calculating the rates of charge andenergy transfer in the reactant/photocatalyst interfaces, which is vital for estimat-ing the efficiency of photocatalysts. These methods were successful in correctlydescribing the dynamics of certain elementary reactions, dominated by a singlePES. However, many important details of the mechanisms are yet to be under-stood. To name a few, during the methanol oxidation reaction, the preference forone-photon versus two-photon absorption mechanism, the role of initial pho-toexcited state in product distribution, the methyl formate formation dynamics,and the resolution of electron–phonon coupling are yet to be resolved. The tun-neling effects in PCET during semiconductor photocatalysis are rarely addressed.In regard to the CO photooxidation/CO2 photoreduction, we highlighted that thecombined use of adiabatic and nonadiabatic theories will be required to explainthe role of the anchor group in controlling the photocatalytic yields.

To conclude, a large host of semiclassical and quantum-classical methodsare presently available, but they are not routinely applied in modeling pho-tocatalytic processes. It is desirable that such methods become more widelyavailable and user-friendly for a broader community to use them. Certain

References 549

improvements of the existing techniques are also desirable, including the abilityto model photodissociation processes, to account for excitonic effects andelectron–phonon back reaction, and improved scaling characteristics of thecomputational expenses with respect to the system’s size. The quantum dynamicseffects discussed in this chapter play critical roles in quantifying the efficiencyand mechanisms of photocatalysis, but have only rarely been addressed byexperimental or theoretical studies. The future theoretical studies may utilizea combination of both adiabatic and nonadiabatic theories, and account forspecific nonclassical phenomena, whenever necessary. Novel methods arealso desirable to access modeling the nonadiabatic phenomena that occur atthe longer time scales (nanoseconds and above). The indicated challenges arebeing currently addressed in the open-source methodology development Librasoftware and the derived codes. It is, however, expected that the involvementof the research community working on the development and application of theabove-mentioned methodologies will be critical for the substantial progress inthe modeling of nonadiabatic and quantum nuclear effects in photocatalysis inthe long term.

References

1 Martsinovich, N. and Troisi, A. (2011) High-throughput computationalscreening of chromophores for dye-sensitized solar cells. J. Phys. Chem. C,115, 11781–11792.

2 Pyzer-Knapp, E.O., Suh, C., Gómez-Bombarelli, R., Aguilera-Iparraguirre, J.,and Aspuru-Guzik, A. (2015) What is high-throughput virtual screening?A perspective from organic materials discovery. Annu. Rev. Mater. Res., 45,195–216.

3 Gómez-Bombarelli, R., Aguilera-Iparraguirre, J., Hirzel, T.D., Duvenaud, D.,Maclaurin, D., Blood-Forsythe, M.A., Chae, H.S., Einzinger, M., Ha, D.-G.,Wu, T. et al. (2016) Design of efficient molecular organic light-emittingdiodes by a high-throughput virtual screening and experimental approach.Nat. Mater., 15, 1120–1127.

4 Hachmann, J., Olivares-Amaya, R., Atahan-Evrenk, S., Amador-Bedolla, C.,Sánchez-Carrera, R.S., Gold-Parker, A., Vogt, L., Brockway, A.M., andAspuru-Guzik, A. (2011) The Harvard clean energy project: large-scalecomputational screening and design of organic photovoltaics on the worldcommunity grid. J. Phys. Chem. Lett., 2, 2241–2251.

5 Setyawan, W. and Curtarolo, S. (2010) High-throughput electronic bandstructure calculations: challenges and tools. Comput. Mater. Sci., 49,299–312.

6 Carrete, J., Mingo, N., Wang, S., and Curtarolo, S. (2014) Nanograinedhalf-heusler semiconductors as advanced thermoelectrics: an ab initiohigh-throughput statistical study. Adv. Funct. Mater., 24, 7427–7432.

7 Curtarolo, S., Hart, G.L.W., Nardelli, M.B., Mingo, N., Sanvito, S., andLevy, O. (2013) The high-throughput highway to computational materialsdesign. Nat. Mater., 12, 191–201.


8 Aryasetiawan, F. and Gunnarsson, O. (1998) The GW method. Rep. Prog.Phys., 61, 237.

9 Salpeter, E.E. and Bethe, H.A. (1951) A relativistic equation for bound-stateproblems. Phys. Rev., 84, 1232–1242.

10 Bickers, N.E., Scalapino, D.J., and White, S.R. (1989) Conserving approxi-mations for strongly correlated electron systems: Bethe-Salpeter equationand dynamics for the two-dimensional Hubbard model. Phys. Rev. Lett., 62,961–964.

11 Govoni, M. and Galli, G. (2015) Large scale GW calculations. J. Chem.Theory Comput., 11, 2680–2696.

12 Scherpelz, P., Govoni, M., Hamada, I., and Galli, G. (2016) Implementationand validation of fully relativistic GW calculations: spin–orbit couplingin molecules, nanocrystals, and solids. J. Chem. Theory Comput., 12,3523–3544.

13 Akimov, A.V., Neukirch, A.J., and Prezhdo, O.V. (2013) Theoretical insightsinto photoinduced charge transfer and catalysis at oxide interfaces. Chem.Rev., 113, 4496–4565.

14 Wang, L., Akimov, A., and Prezhdo, O.V. (2016) Recent progress in surfacehopping: 2011–2015. J. Phys. Chem. Lett., 2100–2112.

15 Nyman, G. (2014) Computational methods of quantum reaction dynamics.Int. J. Quantum Chem., 114, 1183–1198.

16 Pham, T.A., Ping, Y., and Galli, G. (2017) Modelling heterogeneous interfacesfor solar water splitting. Nat. Mater. doi: 10.1038/nmat4803

17 Singh, A.K., Mathew, K., Zhuang, H.L., and Hennig, R.G. (2015) Computa-tional screening of 2D materials for photocatalysis. J. Phys. Chem. Lett., 6,1087–1098.

18 Wodtke, A.M. (2016) Electronically non-adiabatic influences in surfacechemistry and dynamics. Chem. Soc. Rev., 45, 3641–3657.

19 Kilina, S., Kilin, D., and Tretiak, S. (2015) Light-driven and phonon-assisteddynamics in organic and semiconductor nanostructures. Chem. Rev., 115,5929–5978.

20 Prezhdo, O.V., Duncan, W.R., and Prezhdo, V.V. (2008) Dynamics of the pho-toexcited electron at the chromophore–semiconductor interface. Acc. Chem.Res., 41, 339–348.

21 Nelson, T., Fernandez-Alberti, S., Roitberg, A.E., and Tretiak, S. (2014)Nonadiabatic excited-state molecular dynamics: modeling photophysics inorganic conjugated materials. Acc. Chem. Res., 47, 1155–1164.

22 Prezhdo, O.V. (2009) Photoinduced dynamics in semiconductor quantumdots: insights from time-domain ab initio studies. Acc. Chem. Res., 42,2005–2016.

23 Weinberg, D.R., Gagliardi, C.J., Hull, J.F., Murphy, C.F., Kent, C.A.,Westlake, B.C., Paul, A., Ess, D.H., McCafferty, D.G., and Meyer, T.J. (2012)Proton-coupled electron transfer. Chem. Rev., 112, 4016–4093.

24 Hammes-Schiffer, S. and Stuchebrukhov, A.A. (2010) Theory of coupledelectron and proton transfer reactions. Chem. Rev., 110, 6939–6960.

References 551

25 Henkelman, G. and Jónsson, H. (2000) Improved tangent estimate in thenudged elastic band method for finding minimum energy paths and saddlepoints. J. Chem. Phys., 113, 9978–9985.

26 Henkelman, G., Uberuaga, B.P., and Jónsson, H. (2000) A climbing imagenudged elastic band method for finding saddle points and minimum energypaths. J. Chem. Phys., 113, 9901–9904.

27 Peters, B., Heyden, A., Bell, A.T., and Chakraborty, A. (2004) A growingstring method for determining transition states: comparison to the nudgedelastic band and string methods. J. Chem. Phys., 120, 7877–7886.

28 Weinan, E., Ren, W., and Vanden-Eijnden, E. (2002) String method for thestudy of rare events. Phys. Rev. B, 66, 052301.

29 Simons, J., Joergensen, P., Taylor, H., and Ozment, J. (1983) Walking onpotential energy surfaces. J. Phys. Chem., 87, 2745–2753.

30 Jørgensen, P., Jensen, H.J.A., and Helgaker, T. (1988) A gradient extremalwalking algorithm. Theor. Chim. Acta, 73, 55–65.

31 Nichols, J., Taylor, H., Schmidt, P., and Simons, J. (1990) Walking on poten-tial energy surfaces. J. Chem. Phys., 92, 340–346.

32 Schlegel, H.B. (2011) Geometry optimization. Wiley Interdiscip. Rev. Comput.Mol. Sci., 1, 790–809.

33 Peng, C., Ayala, P.Y., Schlegel, H.B., and Frisch, M.J. (1996) Using redun-dant internal coordinates to optimize equilibrium geometries and transitionstates. J. Comput. Chem., 17, 49–56.

34 Peng, C. and Schlegel, H.B. (1993) Combining synchronous transit andquasi-Newton methods to find transition states. Isr. J. Chem., 33, 449–454.

35 Halgren, T.A. and Lipscomb, W.N. (1977) The synchronous-transit methodfor determining reaction pathways and locating molecular transition states.Chem. Phys. Lett., 49, 225–232.

36 Fukui, K. (1981) The path of chemical reactions – the IRC approach.Acc. Chem. Res., 14, 363–368.

37 Maeda, S., Harabuchi, Y., Ono, Y., Taketsugu, T., and Morokuma, K. (2015)Intrinsic reaction coordinate: calculation, bifurcation, and automated search.Int. J. Quantum Chem., 115, 258–269.

38 Plata, R.E. and Singleton, D.A. (2015) A case study of the mechanism ofalcohol-mediated Morita Baylis–Hillman reactions. The importance ofexperimental observations. J. Am. Chem. Soc., 137, 3811–3826.

39 Maeda, S. and Morokuma, K. (2011) Finding reaction pathways of typeA + B → X: toward systematic prediction of reaction mechanisms.J. Chem. Theory Comput., 7, 2335–2345.

40 Sameera, W.M.C., Maeda, S., and Morokuma, K. (2016) Computationalcatalysis using the artificial force induced reaction method. Acc. Chem. Res.,49, 763–773.

41 Zimmerman, P.M. (2015) Single-ended transition state finding with thegrowing string method. J. Comput. Chem., 36, 601–611.

42 Pendleton, I.M., Pérez-Temprano, M.H., Sanford, M.S., and Zimmerman,P.M. (2016) Experimental and computational assessment of reactivity andmechanism in C(sp3)–N bond-forming reductive elimination from palla-dium(IV). J. Am. Chem. Soc., 138, 6049–6060.


43 Doney, A.C., Rooks, B.J., Lu, T., and Wheeler, S.E. (2016) Design oforganocatalysts for asymmetric propargylations through computationalscreening. ACS Catal., 6, 7948–7955.

44 Wang, L.-P., Titov, A., McGibbon, R., Liu, F., Pande, V.S., and Martínez, T.J.(2014) Discovering chemistry with an ab initio nanoreactor. Nat. Chem., 6,1044–1048.

45 Schlegel, H.B. (2003) Exploring potential energy surfaces for chemicalreactions: an overview of some practical methods. J. Comput. Chem., 24,1514–1527.

46 Bolhuis, P.G., Chandler, D., Dellago, C., and Geissler, P.L. (2002) Transitionpath sampling: throwing ropes over rough mountain passes, in the dark.Annu. Rev. Phys. Chem., 53, 291–318.

47 Dellago, C., Bolhuis, P.G., Csajka, F.S., and Chandler, D. (1998) Transitionpath sampling and the calculation of rate constants. J. Chem. Phys., 108,1964–1977.

48 Butler, L.J. (1998) Chemical reaction dynamics beyond theBorn–Oppenheimer approximation. Annu. Rev. Phys. Chem., 49, 125–171.

49 Yan, Y.J., Fried, L.E., and Mukamel, S. (1989) Ultrafast pump-probe spec-troscopy: femtosecond dynamics in Liouville space. J. Phys. Chem., 93,8149–8162.

50 Brown, E.J., Zhang, Q., and Dantus, M. (1999) Femtosecond transient-gratingtechniques: population and coherence dynamics involving ground andexcited states. J. Chem. Phys., 110, 5772–5788.

51 Chen, K., Gallaher, J.K., Barker, A.J., and Hodgkiss, J.M. (2014) Transientgrating photoluminescence spectroscopy: an ultrafast method of gatingbroadband spectra. J. Phys. Chem. Lett., 5, 1732–1737.

52 Zewail, A.H. (2006) 4D Ultrafast electron diffraction, crystallography, andmicroscopy. Annu. Rev. Phys. Chem., 57, 65–103.

53 Wang, H.-F., Velarde, L., Gan, W., and Fu, L. (2015) Quantitativesum-frequency generation vibrational spectroscopy of molecular surfacesand interfaces: lineshape, polarization, and orientation. Annu. Rev. Phys.Chem., 66, 189–216.

54 Mifflin, A.L., Velarde, L., Ho, J., Psciuk, B.T., Negre, C.F.A., Ebben, C.J.,Upshur, M.A., Lu, Z., Strick, B.L., Thomson, R.J. et al. (2015) Accurateline shapes from sub-1 cm−1 resolution sum frequency generation vibra-tional spectroscopy of 𝛼-pinene at room temperature. J. Phys. Chem. A, 119,1292–1302.

55 Abraham, B., Nieto-Pescador, J., and Gundlach, L. (2016) Ultrafast relaxationdynamics of photoexcited zinc-porphyrin: electronic-vibrational coupling.J. Phys. Chem. Lett., 7, 3151–3156.

56 Nieto-Pescador, J., Abraham, B., Li, J., Batarseh, A., Bartynski, R.A.,Galoppini, E., and Gundlach, L. (2016) Heterogeneous electron-transferdynamics through dipole-bridge groups. J. Phys. Chem. C, 120, 48–55.

57 Li, X., Tully, J.C., Schlegel, H.B., and Frisch, M.J. (2005) Ab initio Ehrenfestdynamics. J. Chem. Phys., 123, 084106.

58 Parandekar, P.V. and Tully, J.C. (2005) Mixed quantum-classical equilibrium.J. Chem. Phys., 122, 094102.

References 553

59 Kapral, R. and Ciccotti, G. (1999) Mixed quantum-classical dynamics.J. Chem. Phys., 110, 8919–8929.

60 Tully, J.C. (1990) Molecular dynamics with electronic transitions.J. Chem. Phys., 93, 1061–1071.

61 Hammes-Schiffer, S. and Tully, J.C. (1994) Proton transfer in solution:molecular dynamics with quantum transitions. J. Chem. Phys., 101,4657–4667.

62 Augustin, S.D. and Rabitz, H. (1978) The classical path approximation intime-dependent quantum collision theory. J. Chem. Phys., 69, 4195.

63 Akimov, A.V. and Prezhdo, O.V. (2015) Theory of nonadiabatic elec-tron dynamics in nanomaterials, in Encyclopedia of Nanotechnology(ed. B. Bhushan), Springer, Netherlands, pp. 1–20.

64 Tully, J.C. (2012) Perspective: nonadiabatic dynamics theory. J. Chem. Phys.,137, 22A301.

65 Barbatti, M. (2011) Nonadiabatic dynamics with trajectory surface hoppingmethod. Wiley Interdiscip. Rev. Comput. Mol. Sci., 1, 620–633.

66 Schwartz, B.J., Bittner, E.R., Prezhdo, O.V., and Rossky, P.J. (1996) Quan-tum decoherence and the isotope effect in condensed phase nonadiabaticmolecular dynamics simulations. J. Chem. Phys., 104, 5942–5955.

67 Zhu, C., Nangia, S., Jasper, A.W., and Truhlar, D.G. (2004) Coherent switch-ing with decay of mixing: an improved treatment of electronic coherence fornon-Born–Oppenheimer trajectories. J. Chem. Phys., 121, 7658–7670.

68 Zhu, C. (2016) Restoring electronic coherence/decoherence for atrajectory-based nonadiabatic molecular dynamics. Sci. Rep., 6, 24198.

69 Jaeger, H.M., Fischer, S., and Prezhdo, O.V. (2012) Decoherence-induced sur-face hopping. J. Chem. Phys., 137, 22A545.

70 Akimov, A.V., Long, R., and Prezhdo, O.V. (2014) Coherence penalty func-tional: a simple method for adding decoherence in Ehrenfest dynamics.J. Chem. Phys., 140, 194107.

71 Akimov, A.V. and Prezhdo, O.V. (2014) Second-quantized surface hopping.Phys. Rev. Lett., 113, 153003.

72 Nijamudheen, A. and Akimov, A.V. (2017) Excited state dynamics in 2D het-erostructures: SiR/TiO2 and GeR/TiO2 (R = H, Me) as promising photocata-lysts. J. Phys. Chem. C. doi: 10.1021/acs.jpcc.7b00545

73 Beck, M.H., Jackle, A., Worth, G.A., and Meyer, H.-D. (2000) The multi-configuration time-dependent Hartree (MCTDH) method: a highly efficientalgorithm for propagating wavepackets. Phys. Rep., 324, 1–105.

74 Museth, K. and Billing, G.D. (1996) Generalization of the multiconfigura-tional time-dependent Hartree method to nonadiabatic systems. J. Chem.Phys., 105, 9191–9199.

75 Habershon, S., Manolopoulos, D.E., Markland, T.E., and III, T.F.M. (2013)Ring-polymer molecular dynamics: quantum effects in chemical dynam-ics from classical trajectories in an extended phase space. Annu. Rev. Phys.Chem., 64, 387–413.

76 Ananth, N. (2013) Mapping variable ring polymer molecular dynamics: apath-integral based method for nonadiabatic processes. J. Chem. Phys., 139,124102.


77 Duke, J.R. and Ananth, N. (2015) Simulating excited state dynamics in sys-tems with multiple avoided crossings using mapping variable ring polymermolecular dynamics. J. Phys. Chem. Lett., 6, 4219–4223.

78 Menzeleev, A.R., Bell, F., and Iii, T.F.M. (2014) Kinetically constrainedring-polymer molecular dynamics for non-adiabatic chemical reactions.J. Chem. Phys., 140, 064103.

79 Craig, I.R. and Manolopoulos, D.E. (2004) Quantum statistics and classicalmechanics: real time correlation functions from ring polymer moleculardynamics. J. Chem. Phys., 121, 3368–3373.

80 Hammes-Schiffer, S. and Tully, J.C. (1995) Nonadiabatic transition state the-ory and multiple potential energy surface molecular dynamics of infrequentevents. J. Chem. Phys., 103, 8528–8537.

81 Zhang, F., Ding, W.-J., and Fang, W.-H. (2006) Combined nonadiabatictransition-state theory and ab initio molecular dynamics study on selectivityof the 𝛼 and 𝛽 bond fissions in photodissociation of bromoacetyl chloride.J. Chem. Phys., 125, 184305.

82 Lykhin, A.O., Kaliakin, D.S., dePolo, G.E., Kuzubov, A.A., and Varganov, S.A.(2016) Nonadiabatic transition state theory: application to intersystem cross-ings in the active sites of metal-sulfur proteins. Int. J. Quantum Chem., 116,750–761.

83 Zhu, C. and Nakamura, H. (1994) Theory of nonadiabatic transition forgeneral two-state curve crossing problems. I. Nonadiabatic tunneling case.J. Chem. Phys., 101, 10630–10647.

84 Sherman, M.C. and Corcelli, S.A. (2016) Nonadiabatic transition path sam-pling. J. Chem. Phys., 145, 034110.

85 Barbatti, M., Granucci, G., Lischka, H., and Ruckenbauer, M (2006)Newton-X: a Package for Newtonian Dynamics Close to the Crossing Seam,www.univie.ac.at/newtonx (accessed 06 October 2017).

86 Doltsinis, N. and Marx, D. (2002) Nonadiabatic Car-Parrinello moleculardynamics. Phys. Rev. Lett., 88, 166402.

87 Du, L. and Lan, Z. (2015) An on-the-fly surface-hopping program JADE fornonadiabatic molecular dynamics of polyatomic systems: implementationand applications. J. Chem. Theory Comput., 11, 1360–1374.

88 Marquetand, P., Richter, M., González-Vázquez, J., Sola, I., and González, L.(2011) Nonadiabatic ab initio molecular dynamics including spin–orbitcoupling and laser fields. Faraday Discuss., 153, 261–273.

89 Richter, M., Marquetand, P., González-Vázquez, J., Sola, I., and González, L.(2011) SHARC: ab Initio molecular dynamics with surface hopping in theadiabatic representation including arbitrary couplings. J. Chem. TheoryComput., 7, 1253–1258.

90 Marques, M.A.L., Castro, A., Bertsch, G.F., and Rubio, A. (2003) Octopus:a first-principles tool for excited electron-ion dynamics. Comput. Phys.Commun., 151, 60–78.

91 Worth, G.A., Beck, M.H., Jackle, A., and Meyer, H.D. (2000) The MCTDHpackage, University of Heidelberg, Heidelberg, Germany.

References 555

92 Akimov, A.V. and Prezhdo, O.V. (2013) The PYXAID program fornon-adiabatic molecular dynamics in condensed matter systems. J. Chem.Theory Comput., 9, 4959–4972.

93 Akimov, A.V. and Prezhdo, O.V. (2014) Advanced capabilities of the PYXAIDprogram: integration schemes, decoherence effects, multiexcitonic states, andfield-matter interaction. J. Chem. Theory Comput., 10, 789–804.

94 Akimov, A.V. (2016) Libra: an open-source “methodology discovery” libraryfor quantum and classical dynamics simulations. J. Comput. Chem., 37,1626–1649.

95 Akimov, A.V. (2016) Nonadiabatic molecular dynamics with tight-bindingfragment molecular orbitals. J. Chem. Theory Comput., 12, 5719–5736.

96 Gianozzi, P., Baroni, S., Bonini, N., Calandra, M., Car, R., Cavazzoni, C.,Ceresoli, D., Chiarotti, G.L., Cococcioni, M., Dabo, I. et al. (2009) QUAN-TUM ESPRESSO: a modular and open-source software project for quantumsimulations of materials. J. Phys. Condens. Matter, 21, 395502.

97 Schmidt, M.W., Baldridge, K.K., Boatz, J.A., Elbert, S.T., Gordon, M.S.,Jensen, J.H., Koseki, S., Matsunaga, N., Nguyen, K.A., Su, S. et al. (1993)General atomic and molecular electronic structure system. J. Comput.Chem., 14, 1347–1363.

98 Alexvakimov/LibraX: Libra-driven interfaces with Quantum Espresso andGAMESS programs (2017).

99 Genereux, J.C. and Barton, J.K. (2010) Mechanisms for DNA charge trans-port. Chem. Rev., 110, 1642–1662.

100 Venkatramani, R., Keinan, S., Balaeff, A., and Beratan, D.N. (2011) Nucleicacid charge transfer: black, white and gray. Coord. Chem. Rev., 255,635–648.

101 Shin, Y.K., Newton, M.D., and Isied, S.S. (2003) Distance dependence ofelectron transfer across peptides with different secondary structures: therole of peptide energetics and electronic coupling. J. Am. Chem. Soc., 125,3722–3732.

102 Malak, R.A., Gao, Z., Wishart, J.F., and Isied, S.S. (2004) Long-range electrontransfer across peptide bridges: the transition from electron superexchangeto hopping. J. Am. Chem. Soc., 126, 13888–13889.

103 Hartings, M.R., Kurnikov, I.V., Dunn, A.R., Winkler, J.R., Gray, H.B., andRatner, M.A. (2010) Electron tunneling through sensitizer wires bound toproteins. Coord. Chem. Rev., 254, 248–253.

104 Gray, H.B. and Winkler, J.R. (2005) Long-range electron transfer. Proc. Natl.Acad. Sci. U.S.A., 102, 3534–3539.

105 Gray, H.B. and Winkler, J.R. (2009) Electron flow through proteins. Chem.Phys. Lett., 483, 1–9.

106 Ananth, N. and Miller, T.F. (2010) Exact quantum statistics for electronicallynonadiabatic systems using continuous path variables. J. Chem. Phys., 133,234103.

107 Menzeleev, A.R., Ananth, N., and Miller, T.F. (2011) Direct simulation ofelectron transfer using ring polymer molecular dynamics: comparison withsemiclassical instanton theory and exact quantum methods. J. Chem. Phys.,135, 074106.


108 Cao, J. and Martyna, G.J. (1996) Adiabatic path integral molecular dynamicsmethods. II. Algorithms. J. Chem. Phys., 104, 2028–2035.

109 Cao, J. and Voth, G.A. (1994) The formulation of quantum statisticalmechanics based on the Feynman path centroid density. V. Quantum instan-taneous normal mode theory of liquids. J. Chem. Phys., 101, 6184.

110 Hernandez, R., Cao, J., and Voth, G.A. (1995) On the Feynman path cen-troid density as a phase space distribution in quantum statistical mechanics.J. Chem. Phys., 103, 5018.

111 Ceriotti, M., More, J., and Manolopoulos, D.E. (2014) i-PI: a Python interfacefor ab initio path integral molecular dynamics simulations. Comput. Phys.Commun., 185, 1019–1026.

112 Rossi, M., Ceriotti, M., and Manolopoulos, D.E. (2014) How to remove thespurious resonances from ring polymer molecular dynamics. J. Chem. Phys.,140, 234116.

113 Kelly, A., Brackbill, N., and Markland, T.E. (2015) Accurate nonadiabaticquantum dynamics on the cheap: making the most of mean field theory withmaster equations. J. Chem. Phys., 142, 094110.

114 Marsalek, O. and Markland, T.E. (2016) Ab initio molecular dynamics withnuclear quantum effects at classical cost: ring polymer contraction fordensity functional theory. J. Chem. Phys., 144, 054112.

115 Pfalzgraff, W.C., Kelly, A., and Markland, T.E. (2015) Nonadiabatic dynam-ics in atomistic environments: harnessing quantum-classical theory withgeneralized quantum master equations. J. Phys. Chem. Lett., 6, 4743–4748.

116 Prezhdo, O.V. (2006) Quantized Hamilton dynamics. Theor. Chem. Acc., 116,206–218.

117 Prezhdo, O.V. (2002) Classical mapping for second-order quantized Hamilto-nian dynamics. J. Chem. Phys., 117, 2995–3002.

118 Donoso, A. and Martens, C. (2001) Quantum tunneling using entangledclassical trajectories. Phys. Rev. Lett., 87, 223202.

119 Wang, L., Martens, C.C., and Zheng, Y. (2012) Entangled trajectory molec-ular dynamics in multidimensional systems: two-dimensional quantumtunneling through the Eckart barrier. J. Chem. Phys., 137, 034113.

120 Roman, E. and Martens, C.C. (2007) Independent trajectory implementa-tion of the semiclassical Liouville method: application to multidimensionalreaction dynamics. J. Phys. Chem. A, 111, 10256–10262.

121 Wang, L., Trivedi, D., and Prezhdo, O.V. (2014) Global flux surface hoppingapproach for mixed quantum-classical dynamics. J. Chem. Theory Comput.,10, 3598–3605.

122 Akimov, A.V., Trivedi, D., Wang, L., and Prezhdo, O.V. (2015) Analysis of thetrajectory surface hopping method from the Markov state model perspective.J. Phys. Soc. Jpn., 84, 094002.

123 Meisner, J. and Kästner, J. (2016) Atom tunneling in chemistry. Angew.Chem. Int. Ed., 55, 5400–5413.

124 Cha, Y., Murray, C.J., and Klinman, J.P. (1989) Hydrogen tunneling inenzyme reactions. Science, 243, 1325–1330.

References 557

125 Hama, T., Ueta, H., Kouchi, A., Watanabe, N., and Tachikawa, H. (2014)Quantum tunneling hydrogenation of solid benzene and its control viasurface structure. J. Phys. Chem. Lett., 5, 3843–3848.

126 Hama, T., Ueta, H., Kouchi, A., and Watanabe, N. (2015) Quantum tunnelingobserved without its characteristic large kinetic isotope effects. Proc. Natl.Acad. Sci., 112, 7438–7443.

127 Wonchoba, S.E., Hu, W.-P., and Truhlar, D.G. (1995) Surface diffusion of Hon Ni(100): interpretation of the transition temperature. Phys. Rev. B, 51,9985–10002.

128 Hiraoka, K., Sato, T., and Takayama, T. (2001) Tunneling reactions in inter-stellar ices. Science, 292, 869–870.

129 Ley, D., Gerbig, D., and Schreiner, P.R. (2012) Tunnelling control of chem-ical reactions – the organic chemist’s perspective. Org. Biomol. Chem., 10,3781–3790.

130 Schreiner, P.R., Reisenauer, H.P., Ley, D., Gerbig, D., Wu, C.-H., andAllen, W.D. (2011) Methylhydroxycarbene: tunneling control of a chemicalreaction. Science, 332, 1300–1303.

131 Miller, W.H., Handy, N.C., and Adams, J.E. (1980) Reaction path Hamilto-nian for polyatomic molecules. J. Chem. Phys., 72, 99–112.

132 Garrett, B.C. and Truhlar, D.G. (1984) WKB approximation for thereaction-path Hamiltonian: application to variational transition state the-ory, vibrationally adiabatic excited-state barrier heights, and resonancecalculations. J. Chem. Phys., 81, 309–317.

133 Dittrich, W. and Reuter, M. (2016) The WKB approximation, in Classicaland Quantum Dynamics, Springer International Publishing, pp. 299–310.

134 Schreiner, P.R., Reisenauer, H.P., Pickard Iv, F.C., Simmonett, A.C.,Allen, W.D., Mátyus, E., and Császár, A.G. (2008) Capture of hydroxymethy-lene and its fast disappearance through tunnelling. Nature, 453, 906–909.

135 Carrington, T. Jr.,, Hubbard, L.M., Schaefer, H.F. III,, and Miller, W.H. (1984)Vinylidene: potential energy surface and unimolecular reaction dynamics.J. Chem. Phys., 80, 4347–4354.

136 Mastromatteo, M. and Jackson, B. (2013) The dissociative chemisorption ofmethane on Ni(100) and Ni(111): classical and quantum studies based onthe reaction path Hamiltonian. J. Chem. Phys., 139, 194701.

137 Skodje, R.T., Truhlar, D.G., and Garrett, B.C. (1981) A generalsmall-curvature approximation for transition-state-theory transmissioncoefficients. J. Phys. Chem., 85, 3019–3023.

138 Garrett, B.C., Truhlar, D.G., Wagner, A.F., and Dunning, T.H. Jr., (1983)Variational transition state theory and tunneling for a heavy–light–heavyreaction using an ab initio potential energy surface. 37Cl+H(D) 35Cl→H(D)37Cl+35Cl. J. Chem. Phys., 78, 4400–4413.

139 Wu, A., Mader, E.A., Datta, A., Hrovat, D.A., Borden, W.T., and Mayer, J.M.(2009) Nitroxyl radical plus hydroxylamine pseudo self-exchange reactions:tunneling in hydrogen atom transfer. J. Am. Chem. Soc., 131, 11985–11997.

140 Gonzalez-James, O.M., Zhang, X., Datta, A., Hrovat, D.A., Borden, W.T., andSingleton, D.A. (2010) Experimental evidence for heavy-atom tunneling in


the ring-opening of cyclopropylcarbinyl radical from intramolecular 12C/13Ckinetic isotope effects. J. Am. Chem. Soc., 132, 12548–12549.

141 Nijamudheen, A., Karmakar, S., and Datta, A. (2014) Understanding themechanisms of unusually fast H–H, C–H, and C–C bond reductive elimina-tions from gold(III) complexes. Chem. Eur. J., 20, 14650–14658.

142 Bhattacharjee, R., Nijamudheen, A., Karmakar, S., and Datta, A. (2016)Strain control: reversible H2 activation and H2/D2 exchange in Pt complexes.Inorg. Chem., 55, 3023–3029.

143 Truhlar, D.G. and Garrett, B.C. (1980) Variational transition-state theory.Acc. Chem. Res., 13, 440–448.

144 Truhlar, D.G. and Garrett, B.C. (1984) Variational transition state theory.Annu. Rev. Phys. Chem., 35, 159–189.

145 Eckart, C. (1930) The penetration of a potential barrier by electrons. Phys.Rev., 35, 1303–1309.

146 Mandal, D., Ramanan, R., Usharani, D., Janardanan, D., Wang, B., andShaik, S. (2015) How does tunneling contribute to counterintuitiveH-abstraction reactivity of nonheme Fe(IV)O oxidants with alkanes? J. Am.Chem. Soc., 137, 722–733.

147 Mandal, D. and Shaik, S. (2016) Interplay of tunneling, two-state reactiv-ity, and Bell–Evans–Polanyi effects in C–H activation by nonheme Fe(IV)Ooxidants. J. Am. Chem. Soc., 138, 2094–2097.

148 Langer, J.S. (1969) Statistical theory of the decay of metastable states. Ann.Phys., 54, 258–275.

149 Miller, W.H. (1975) Semiclassical limit of quantum mechanical transitionstate theory for nonseparable systems. J. Chem. Phys., 62, 1899–1906.

150 Mil’nikov, G.V. and Nakamura, H. (2001) Practical implementation of theinstanton theory for the ground-state tunneling splitting. J. Chem. Phys., 115,6881–6897.

151 Althorpe, S.C. (2011) On the equivalence of two commonly used forms ofsemiclassical instanton theory. J. Chem. Phys., 134, 114104.

152 Pérez de Tudela, R., Aoiz, F.J., Suleimanov, Y.V., and Manolopoulos, D.E.(2012) Chemical reaction rates from ring polymer molecular dynamics: zeropoint energy conservation in Mu+H2 → MuH+H. J. Phys. Chem. Lett., 3,493–497.

153 Voth, G.A., Chandler, D., and Miller, W.H. (1989) Rigorous formulationof quantum transition state theory and its dynamical corrections. J. Chem.Phys., 91, 7749.

154 Voth, G.A. (1993) Feynman path integral formulation of quantum mechani-cal transition-state theory. J. Phys. Chem., 97, 8365–8377.

155 Borden, W.T. (2016) Reactions that involve tunneling by carbon and the rolethat calculations have played in their study. Wiley Interdiscip. Rev. Comput.Mol. Sci., 6, 20–46.

156 Zhang, X., Hrovat, D.A., and Borden, W.T. (2010) Calculations predict thatcarbon tunneling allows the degenerate Cope rearrangement of semibullva-lene to occur rapidly at cryogenic temperatures. Org. Lett., 12, 2798–2801.

References 559

157 Zuev, P.S., Sheridan, R.S., Albu, T.V., Truhlar, D.G., Hrovat, D.A., andBorden, W.T. (2003) Carbon tunneling from a single quantum state. Science,299, 867–870.

158 Inui, H., Sawada, K., Oishi, S., Ushida, K., and McMahon, R.J. (2013) Arylnitrene rearrangements: spectroscopic observation of a benzazirine and itsring expansion to a ketenimine by heavy-atom tunneling. J. Am. Chem. Soc.,135, 10246–10249.

159 Kozuch, S., Zhang, X., Hrovat, D.A., and Borden, W.T. (2013) Calculationson tunneling in the reactions of noradamantyl carbenes. J. Am. Chem. Soc.,135, 17274–17277.

160 Carpenter, B.K. (1983) Heavy-atom tunneling as the dominant pathway in asolution-phase reaction? Bond shift in antiaromatic annulenes. J. Am. Chem.Soc., 105, 1700–1701.

161 Redington, R.L. (1998) State-specific vibrational anharmonicities in cyclobu-tadiene and evidence for fast automerization by 12C4H4. J. Chem. Phys., 109,10781–10794.

162 Huang, M.J. and Wolfsberg, M. (1984) Tunneling in the automerization ofcyclobutadiene. J. Am. Chem. Soc., 106, 4039–4040.

163 Kozuch, S. (2015) On the tunneling instability of a hypercoordinated carbo-cation. Phys. Chem. Chem. Phys., 17, 16688–16691.

164 Kozuch, S. (2014) A quantum mechanical “jack in the box”: rapid rearrange-ment of a tetrahedryl-tetrahedrane via heavy atom tunneling. Org. Lett., 16,4102–4105.

165 Ertelt, M., Hrovat, D.A., Borden, W.T., and Sander, W. (2014) Heavy-atomtunneling in the ring opening of a strained cyclopropene at very low temper-atures. Chem. Eur. J., 20, 4713–4720.

166 Karmakar, S. and Datta, A. (2016) Role of heavy atom tunneling inMyers–Saito cyclization of cyclic enyne-cumulene systems. J. Phys. Chem.B, 120, 945–950.

167 Karmakar, S. and Datta, A. (2015) Metal free azide–alkyne click Reac-tion: role of substituents and heavy atom tunneling. J. Phys. Chem. B, 119,11540–11547.

168 Gerbig, D., Ley, D., and Schreiner, P.R. (2011) Light- and heavy-atom tun-neling in rearrangement reactions of cyclopropylcarbenes. Org. Lett., 13,3526–3529.

169 Greer, E.M., Cosgriff, C.V., and Doubleday, C. (2013) Computationalevidence for heavy-atom tunneling in the Bergman cyclization of a10-membered-ring enediyne. J. Am. Chem. Soc., 135, 10194–10197.

170 Kozuch, S. (2014) Heavy atom tunneling in the automerization of pentaleneand other antiaromatic systems. RSC Adv., 4, 21650–21656.

171 Weck, P.F. and Balakrishnan, N. (2005) Heavy atom tunneling in chemicalreactions: study of H+LiF collisions. J. Chem. Phys., 122, 234310.

172 Zheng, J., Xu, X., Meana-Pañeda, R., and Truhlar, D.G. (2014) Army antstunneling for classical simulations. Chem. Sci., 5, 2091.

173 Jasper, A.W., Zhu, C., Nangia, S., and Truhlar, D.G. (2004) Introductorylecture: nonadiabatic effects in chemical dynamics. Faraday Discuss., 127,1–22.


174 Tseng, C.-M., Lee, Y.T., Lin, M.-F., Ni, C.-K., Liu, S.-Y., Lee, Y.-P., Xu, Z.F.,and Lin, M.C. (2007) Photodissociation dynamics of phenol. J. Phys. Chem.A, 111, 9463–9470.

175 Ashfold, M.N.R., Devine, A.L., Dixon, R.N., King, G.A., Nix, M.G.D., andOliver, T.A.A. (2008) Exploring nuclear motion through conical intersectionsin the UV photodissociation of phenols and thiophenol. Proc. Natl. Acad.Sci., 105, 12701–12706.

176 Lee, J., Ryu, S., Ku, J.S., and Kim, S.K. (2001) Charge transfer photodissocia-tion of phenol on Ag(111). J. Chem. Phys., 115, 10518–10524.

177 Ashfold, M.N.R., Cronin, B., Devine, A.L., Dixon, R.N., and Nix, M.G.D.(2006) The role of 𝜋𝜎* excited states in the photodissociation of heteroaro-matic molecules. Science, 312, 1637–1640.

178 Xu, X., Zheng, J., Yang, K.R., and Truhlar, D.G. (2014) Photodissociationdynamics of phenol: multistate trajectory simulations including tunneling.J. Am. Chem. Soc., 136, 16378–16386.

179 Yang, K.R., Xu, X., Zheng, J., and Truhlar, D.G. (2014) Full-dimensionalpotentials and state couplings and multidimensional tunneling calculationsfor the photodissociation of phenol. Chem. Sci., 5, 4661–4680.

180 Yang, K.R., Xu, X., and Truhlar, D.G. (2013) Direct diabatization of elec-tronic states by the fourfold-way: including dynamical correlation bymulti-configuration quasidegenerate perturbation theory with completeactive space self-consistent-field diabatic molecular orbitals. Chem. Phys.Lett., 573, 84–89.

181 Nakamura, H. and Truhlar, D.G. (2002) Direct diabatization of electronicstates by the fourfold way. II. Dynamical correlation and rearrangementprocesses. J. Chem. Phys., 117, 5576–5593.

182 Yang, K.R., Xu, X., and Truhlar, D.G. (2014) Anchor points reactive potentialfor bond-breaking reactions. J. Chem. Theory Comput., 10, 924–933.

183 Xie, C., Yarkony, D.R., and Guo, H. (2017) Nonadiabatic tunneling via con-ical intersections and the role of the geometric phase. Phys. Rev. A, 95,022104.

184 Ratzer, C., Küpper, J., Spangenberg, D., and Schmitt, M. (2002) The structureof phenol in the S1-state determined by high resolution UV-spectroscopy.Chem. Phys., 283, 153–169.

185 Pino, G.A., Oldani, A.N., Marceca, E., Fujii, M., Ishiuchi, S.-I., Miyazaki, M.,Broquier, M., Dedonder, C., and Jouvet, C. (2010) Excited state hydrogentransfer dynamics in substituted phenols and their complexes with ammonia:𝜋𝜋

*–𝜋𝜎* energy gap propensity and ortho-substitution effect. J. Chem. Phys.,133, 124313.

186 Roberts, G.M., Chatterley, A.S., Young, J.D., and Stavros, V.G. (2012) Directobservation of hydrogen tunneling dynamics in photoexcited phenol. J. Phys.Chem. Lett., 3, 348–352.

187 Zhu, C., Teranishi, Y., and Nakamura, H. (2001) Nonadiabatic transitionsdue to curve crossings: complete solutions of the landau-zener-stueckelbergproblems and their applications, in Advances in Chemical Physics (eds I.Prigogine and S.A. Rice), John Wiley & Sons, Inc, pp. 127–233.

References 561

188 Della Sala, F., Rousseau, R., Görling, A., and Marx, D. (2004) Quantum andthermal fluctuation effects on the photoabsorption spectra of clusters. Phys.Rev. Lett., 92, 183401.

189 Hammes-Schiffer, S. (2015) Proton-coupled electron transfer: movingtogether and charging forward. J. Am. Chem. Soc., 137, 8860–8871.


191 Ibhadon, A.O. and Fitzpatrick, P. (2013) Heterogeneous photocatalysis:recent advances and applications. Catalysts, 3, 189–218.

192 Lang, X., Chen, X., and Zhao, J. (2013) Heterogeneous visible light photo-catalysis for selective organic transformations. Chem. Soc. Rev., 43, 473–486.

193 Hisatomi, T., Kubota, J., and Domen, K. (2014) Recent advances in semicon-ductors for photocatalytic and photoelectrochemical water splitting. Chem.Soc. Rev., 43, 7520–7535.

194 Hoffmann, M.R., Martin, S.T., Choi, W., and Bahnemann, D.W. (1995) Envi-ronmental applications of semiconductor photocatalysis. Chem. Rev., 95,69–96.



197 Yin, W.-J., Wei, S.-H., Al-Jassim, M.M., and Yan, Y. (2011)Double-hole-mediated coupling of dopants and its impact on band gapengineering in TiO2. Phys. Rev. Lett., 106, 066801.

198 Yin, W.-J., Tang, H., Wei, S.-H., Al-Jassim, M.M., Turner, J., and Yan, Y.(2010) Band structure engineering of semiconductors for enhanced photo-electrochemical water splitting: the case of TiO2. Phys. Rev. B, 82, 045106.

199 Mayer, M.A., Yu, K.M., Speaks, D.T., Denlinger, J.D., Reichertz, L.A.,Beeman, J.W., Haller, E.E., and Walukiewicz, W. (2012) Band gap engineer-ing of oxide photoelectrodes: characterization of ZnO1-xSx. J. Phys. Chem. C,116, 15281–15289.

200 Khan, S.U.M. (2002) Efficient photochemical water splitting by a chemicallymodified n-TiO2. Science, 297, 2243–2245.

201 Zhao, L., Chen, X., Wang, X., Zhang, Y., Wei, W., Sun, Y., Antonietti, M.,and Titirici, M.-M. (2010) One-step solvothermal synthesis of acarbon@TiO2 dyade structure effectively promoting visible-light photo-catalysis. Adv. Mater., 22, 3317–3321.


203 Lang, X., Ma, W., Chen, C., Ji, H., and Zhao, J. (2014) Selective aerobicoxidation mediated by TiO2 photocatalysis. Acc. Chem. Res., 47, 355–363.

204 Guo, Q., Zhou, C., Ma, Z., Ren, Z., Fan, H., and Yang, X. (2016) Elementaryphotocatalytic chemistry on TiO2 surfaces. Chem. Soc. Rev., 45, 3701–3730.


205 Tamaki, Y., Furube, A., Murai, M., Hara, K., Katoh, R., and Tachiya, M.(2006) Direct observation of reactive trapped holes in TiO2 undergoing pho-tocatalytic oxidation of adsorbed alcohols: evaluation of the reaction ratesand yields. J. Am. Chem. Soc., 128, 416–417.

206 Dimitrijevic, N.M., Shkrob, I.A., Gosztola, D.J., and Rajh, T. (2012) Dynamicsof interfacial charge transfer to formic acid, formaldehyde, and methanolon the surface of TiO2 nanoparticles and its role in methane production.J. Phys. Chem. C, 116, 878–885.

207 Zhou, C., Ren, Z., Tan, S., Ma, Z., Mao, X., Dai, D., Fan, H., Yang, X.,LaRue, J., Cooper, R. et al. (2010) Site-specific photocatalytic splitting ofmethanol on TiO2(110). Chem. Sci., 1, 575–580.

208 Yuan, Q., Wu, Z., Jin, Y., Xu, L., Xiong, F., Ma, Y., and Huang, W. (2013)Photocatalytic cross-coupling of methanol and formaldehyde on a rutileTiO2 (110) surface. J. Am. Chem. Soc., 135, 5212–5219.

209 Phillips, K.R., Jensen, S.C., Baron, M., Li, S.-C., and Friend, C.M. (2013)Sequential photo-oxidation of methanol to methyl formate on TiO2 (110).J. Am. Chem. Soc., 135, 574–577.

210 Mao, X., Wei, D., Wang, Z., Jin, X., Hao, Q., Ren, Z., Dai, D., Ma, Z.,Zhou, C., and Yang, X. (2015) Recombination of formaldehyde and hydrogenatoms on TiO2 (110). J. Phys. Chem. C, 119, 1170–1174.

211 Henderson, M.A., Otero-Tapia, S., and Castro, M.E. (1999) The chem-istry of methanol on the TiO2 (110) surface: the influence of vacancies andcoadsorbed species. Faraday Discuss., 114, 313–329.

212 Guo, Q., Xu, C., Ren, Z., Yang, W., Ma, Z., Dai, D., Fan, H., Minton, T.K.,and Yang, X. (2012) Stepwise photocatalytic dissociation of methanol andwater on TiO2 (110). J. Am. Chem. Soc., 134, 13366–13373.

213 Liu, S., Liu, A., Wen, B., Zhang, R., Zhou, C., Liu, L.-M., and Ren, Z. (2015)Coverage dependence of methanol dissociation on TiO2 (110). J. Phys. Chem.Lett., 6, 3327–3334.

214 Li, B., Zhao, J., Onda, K., Jordan, K.D., Yang, J., and Petek, H. (2006) Ultra-fast interfacial proton-coupled electron transfer. Science, 311, 1436–1440.

215 Shen, M. and Henderson, M.A. (2011) Identification of the active speciesin photochemical hole scavenging reactions of methanol on TiO2. J. Phys.Chem. Lett., 2, 2707–2710.

216 Wang, Z., Hao, Q., Mao, X., Zhou, C., Dai, D., and Yang, X. (2016) Photo-catalytic chemistry of methanol on rutile TiO2 (011)-(2 × 1). Phys. Chem.Chem. Phys., 18, 10224–10231.

217 Feng, R., Liu, A., Liu, S., Shi, J., Zhang, R., and Ren, Z. (2015) In situ studieson the dissociation and photocatalytic reactions of CH3 OH on TiO2 thinfilm by sum frequency generation vibrational spectroscopy. J. Phys. Chem. C,119, 9798–9804.

218 Feng, H., Liu, L., Dong, S., Cui, X., Zhao, J., and Wang, B. (2016) Dynamicprocesses of formaldehyde at terminal Ti sites on the rutile TiO2 (110)surface. J. Phys. Chem. C, 120, 24287–24293.

219 Yu, X., Zhang, Z., Yang, C., Bebensee, F., Heissler, S., Nefedov, A., Tang, M.,Ge, Q., Chen, L., Kay, B.D. et al. (2016) Interaction of formaldehyde with the

References 563

rutile TiO2 (110) surface: a combined experimental and theoretical study.J. Phys. Chem. C, 120, 12626–12636.

220 Bates, S.P., Gillan, M.J., and Kresse, G. (1998) Adsorption of methanol onTiO2 (110): a first-principles investigation. J. Phys. Chem. B, 102, 2017–2026.

221 Migani, A., Mowbray, D.J., Iacomino, A., Zhao, J., Petek, H., and Rubio, A.(2013) Level alignment of a prototypical photocatalytic system: methanol onTiO2 (110). J. Am. Chem. Soc., 135, 11429–11432.

222 Lang, X., Wen, B., Zhou, C., Ren, Z., and Liu, L.-M. (2014) First-principlesstudy of methanol oxidation into methyl formate on rutile TiO2 (110).J. Phys. Chem. C, 118, 19859–19868.

223 Ji, Y., Wang, B., and Luo, Y. (2014) A comparative theoretical study ofproton-coupled hole transfer for H2O and small organic molecules (CH3OH,HCOOH, H2CO) on the anatase TiO2 (101) Surface. J. Phys. Chem. C, 118,21457–21462.

224 Chu, W., Saidi, W.A., Zheng, Q., Xie, Y., Lan, Z., Prezhdo, O.V., Petek, H.,and Zhao, J. (2016) Ultrafast dynamics of photongenerated holes at aCH3OH/TiO2 rutile Interface. J. Am. Chem. Soc., 138, 13740–13749.

225 Migani, A. and Blancafort, L. (2016) Excitonic interfacial proton-coupledelectron transfer mechanism in the photocatalytic oxidation of methanol toformaldehyde on TiO2 (110). J. Am. Chem. Soc., 138, 16165–16173.

226 Kolesov, G., Vinichenko, D., Tritsaris, G.A., Friend, C.M., and Kaxiras, E.(2015) Anatomy of the photochemical reaction: excited-state dynamicsreveals the C–H acidity mechanism of methoxy photo-oxidation on titania.J. Phys. Chem. Lett., 6, 1624–1627.

227 Akimov, A.V. (2016) Libra, https://github.com/Quantum-Dynamics-Hub.228 Seitz, L.C., Chen, Z., Forman, A.J., Pinaud, B.A., Benck, J.D., and Jaramillo,

T.F. (2014) Modeling practical performance limits of photoelectrochemicalwater splitting based on the current state of materials research. Chem-SusChem, 7, 1372–1385.

229 Tachibana, Y., Vayssieres, L., and Durrant, J.R. (2012) Artificial photosynthe-sis for solar water-splitting. Nat. Photonics, 6, 511–518.

230 Haussener, S., Xiang, C., Spurgeon, J.M., Ardo, S., Lewis, N.S., andWeber, A.Z. (2012) Modeling, simulation, and design criteria for photo-electrochemical water-splitting systems. Energy Environ. Sci., 5, 9922–9935.



233 Chen, W.-F., Sasaki, K., Ma, C., Frenkel, A.I., Marinkovic, N.,Muckerman, J.T., Zhu, Y., and Adzic, R.R. (2012) Hydrogen-evolutioncatalysts based on non-noble metal nickel–molybdenum nitride nanosheets.Angew. Chem. Int. Ed., 51, 6131–6135.

234 McKone, J.R., Lewis, N.S., and Gray, H.B. (2014) Will solar-drivenwater-splitting devices see the light of day? Chem. Mater., 26, 407–414.

235 Gray, H.B. (2009) Powering the planet with solar fuel. Nat. Chem., 1, 7.


236 Lewis, N.S. and Nocera, D.G. (2006) Powering the planet: chemical chal-lenges in solar energy utilization. Proc. Natl. Acad. Sci., 103, 15729–15735.

237 Sun, C., Liu, L.-M., Selloni, A., Lu, G.Q.(.M.)., and Smith, S.C. (2010)Titania-water interactions: a review of theoretical studies. J. Mater. Chem.,20, 10319–10334.

238 Vittadini, A., Casarin, M., and Selloni, A. (2007) Chemistry of and onTiO2-anatase surfaces by DFT calculations: a partial review. Theor. Chem.Acc., 117, 663–671.

239 Liu, L.-M., Crawford, P., and Hu, P. (2009) The interaction between adsorbedOH and O2 on TiO2 surfaces. Prog. Surf. Sci., 84, 155–176.

240 Hammes-Schiffer, S. (2009) Theory of proton-coupled electron transfer inenergy conversion processes. Acc. Chem. Res., 42, 1881–1889.

241 Hammes-Schiffer, S. (2001) Theoretical perspectives on proton-coupled elec-tron transfer reactions. Acc. Chem. Res., 34, 273–281.

242 Gagliardi, C.J., Vannucci, A.K., Concepcion, J.J., Chen, Z., and Meyer, T.J.(2012) The role of proton coupled electron transfer in water oxidation.Energy Environ. Sci., 5, 7704–7717.

243 Goyal, P. and Hammes-Schiffer, S. (2017) Tuning the ultrafast dynamicsof photoinduced proton-coupled electron transfer in energy conversionprocesses. ACS Energy Lett., 2, 512–519.

244 Cowan, A.J. and Durrant, J.R. (2013) Long-lived charge separated states innanostructured semiconductor photoelectrodes for the production of solarfuels. Chem. Soc. Rev., 42, 2281–2293.

245 Onda, K., Li, B., Zhao, J., Jordan, K.D., Yang, J., and Petek, H. (2005) Wetelectrons at the H2O/TiO2(110) surface. Science, 308, 1154–1158.

246 Tang, J., Durrant, J.R., and Klug, D.R. (2008) Mechanism of photocatalyticwater splitting in TiO2. Reaction of water with photoholes, importance ofcharge carrier dynamics, and evidence for four-hole chemistry. J. Am. Chem.Soc., 130, 13885–13891.

247 Kubacka, A., Fernández-García, M., and Colón, G. (2012) Advancednanoarchitectures for solar photocatalytic applications. Chem. Rev., 112,1555–1614.

248 Zhang, Y., Qiu, C., and Kilin, D.S. (2014) Electron dynamics in charged wetTiO2 anatase (001) surface functionalised by ruthenium ions. Mol. Phys.,112, 441–452.

249 Huang, S., Balasanthiran, C., Tretiak, S., Hoefelmeyer, J.D., Kilina, S.V., andKilin, D.S. (2016) Dynamics of charge at water-to-semiconductor interface:case study of wet [0 0 1] anatase TiO2 nanowire. Chem. Phys., 481, 184–190.

250 Huang, S., Inerbaev, T.M., and Kilin, D.S. (2014) Excited state dynamics ofRu10 cluster interfacing anatase TiO2 (101) surface and liquid water. J. Phys.Chem. Lett., 5, 2823–2829.

251 Sapp, W., Koodali, R., and Kilin, D. (2016) Charge transfer mechanismin titanium-doped microporous silica for photocatalytic water-splittingapplications. Catalysts, 6, 34.

252 Muuronen, M., Parker, S.M., Berardo, E., Le, A., Zwijnenburg, M.A., andFurche, F. (2016) Mechanism of photocatalytic water oxidation on smallTiO2 nanoparticles. Chem. Sci. doi: 10.1039/c6sc04378j

References 565

253 Tritsaris, G.A., Vinichenko, D., Kolesov, G., Friend, C.M., and Kaxiras, E.(2014) Dynamics of the photogenerated hole at the rutile TiO2 (110)/waterinterface: a nonadiabatic simulation study. J. Phys. Chem. C, 118,27393–27401.

254 Maeda, K., Teramura, K., Lu, D., Takata, T., Saito, N., Inoue, Y., andDomen, K. (2006) Photocatalyst releasing hydrogen from water. Nature,440, 295.

255 Ohno, T., Bai, L., Hisatomi, T., Maeda, K., and Domen, K. (2012) Photocat-alytic water splitting using modified GaN:ZnO solid solution under visiblelight: long-time operation and regeneration of activity. J. Am. Chem. Soc.,134, 8254–8259.

256 Ryu, S.-W., Zhang, Y., Leung, B., Yerino, C., and Han, J. (2012) Improvedphotoelectrochemical water splitting efficiency of nanoporous GaN photoan-ode. Semicond. Sci. Technol., 27, 015014.

257 Hayashi, T., Deura, M., and Ohkawa, K. (2012) High stability and efficiencyof GaN photocatalyst for hydrogen generation from water. Jpn. J. Appl. Phys.,51, 112601.

258 Chen, P.-T., Sun, C.-L., and Hayashi, M. (2010) First-principles calculationsof hydrogen generation due to water splitting on polar GaN surfaces. J. Phys.Chem. C, 114, 18228–18232.

259 Shen, X., Allen, P.B., Hybertsen, M.S., and Muckerman, J.T. (2009) Wateradsorption on the GaN (101 0) nonpolar surface. J. Phys. Chem. C, 113,3365–3368.

260 Wang, J., Pedroza, L.S., Poissier, A., and Fernández-Serra, M.V. (2012) Waterdissociation at the GaN (101 0) surface: Structure, dynamics and surfaceacidity. J. Phys. Chem. C, 116, 14382–14389.

261 Ertem, M.Z., Kharche, N., Batista, V.S., Hybertsen, M.S., Tully, J.C., andMuckerman, J.T. (2015) Photoinduced water oxidation at the aqueousGaN (101 0) interface: deprotonation kinetics of the first proton-coupledelectron-transfer step. ACS Catal., 5, 2317–2323.

262 Zhang, X. and Ptasinska, S. (2016) Electronic and chemical structure of theH2O/GaN(0001) interface under ambient conditions. Sci. Rep., 6, 24848.

263 Wood, B.C., Schwegler, E., Choi, W.I., and Ogitsu, T. (2013) Hydrogen-bonddynamics of water at the interface with InP/GaP(001) and the implicationsfor photoelectrochemistry. J. Am. Chem. Soc., 135, 15774–15783.

264 Shen, X., Small, Y.A., Wang, J., Allen, P.B., Fernandez-Serra, M.V.,Hybertsen, M.S., and Muckerman, J.T. (2010) Photocatalytic water oxidationat the GaN (101 0)−water interface. J. Phys. Chem. C, 114, 13695–13704.

265 Akimov, A.V., Muckerman, J.T., and Prezhdo, O.V. (2013) Nonadiabaticdynamics of positive charge during photocatalytic water splitting onGaN(10–10) surface: charge localization governs splitting efficiency. J. Am.Chem. Soc., 135, 8682–8691.

266 Hwang, S., Lee, M.C., and Choi, W. (2003) Highly enhanced photocatalyticoxidation of CO on titania deposited with Pt nanoparticles: kinetics andmechanism. Appl. Catal. B Environ., 46, 49–63.

267 Habisreutinger, S.N., Schmidt-Mende, L., and Stolarczyk, J.K. (2013) Pho-tocatalytic reduction of CO2 on TiO2 and other semiconductors. Angew.Chem. Int. Ed., 52, 7372–7408.


268 Ip, C.M. and Troisi, A. (2016) A computational study of the competing reac-tion mechanisms of the photo-catalytic reduction of CO2 on anatase (101).Phys. Chem. Chem. Phys., 18, 25010–25021.

269 Ji, Y. and Luo, Y. (2016) Theoretical study on the mechanism of photore-duction of CO2 to CH4 on the anatase TiO2 (101) surface. ACS Catal., 6,2018–2025.

270 Kweon, K.E., Manogaran, D., and Hwang, G.S. (2014) Synergetic role of pho-togenerated electrons and holes in the oxidation of CO to CO2 on reducedTiO2 (110): a first-principles study. ACS Catal., 4, 4051–4056.

271 Yoon, Y., Wang, Y.-G., Rousseau, R., and Glezakou, V.-A. (2015) Impact ofnonadiabatic charge transfer on the rate of redox chemistry of carbon oxideson rutile TiO2 (110) surface. ACS Catal., 5, 1764–1771.

272 Li, Y.-F., Aschauer, U., Chen, J., and Selloni, A. (2014) Adsorption andreactions of O2 on anatase TiO2. Acc. Chem. Res., 47, 3361–3368.

273 Setvín, M., Aschauer, U., Scheiber, P., Li, Y.-F., Hou, W., Schmid, M.,Selloni, A., and Diebold, U. (2013) Reaction of O2 with subsurface oxygenvacancies on TiO2 anatase (101). Science, 341, 988–991.

274 Li, Y.-F. and Selloni, A. (2016) Pathway of photocatalytic oxygen evolutionon aqueous TiO2 anatase and insights into the different activities of anataseand rutile. ACS Catal., 6, 4769–4774.

275 Setvin, M., Aschauer, U., Hulva, J., Simschitz, T., Daniel, B., Schmid, M.,Selloni, A., and Diebold, U. (2016) Following the reduction of oxygen onTiO2 anatase (101) step by step. J. Am. Chem. Soc., 138, 9565–9571.

276 Sato, S., Arai, T., Morikawa, T., Uemura, K., Suzuki, T.M., Tanaka, H., andKajino, T. (2011) Selective CO2 conversion to formate conjugated with H2Ooxidation utilizing semiconductor/complex hybrid photocatalysts. J. Am.Chem. Soc., 133, 15240–15243.

277 Sato, S., Morikawa, T., Saeki, S., Kajino, T., and Motohiro, T. (2010)Visible-light-induced selective CO2 reduction utilizing a ruthenium com-plex electrocatalyst linked to a p-type nitrogen-doped Ta2O5 semiconductor.Angew. Chem. Int. Ed., 49, 5101–5105.

278 Suzuki, T.M., Tanaka, H., Morikawa, T., Iwaki, M., Sato, S., Saeki, S.,Inoue, M., Kajino, T., and Motohiro, T. (2011) Direct assembly synthe-sis of metal complex–semiconductor hybrid photocatalysts anchored byphosphonate for highly efficient CO2 reduction. Chem. Commun., 47, 8673.

279 Akimov, A.V., Jinnouchi, R., Asahi, R., and Prezhdo, O.V. (2014) Theoret-ical insights into the impact of Ru catalyst anchors on the efficiency ofphotocatalytic CO2 reduction on Ta2O5. J. Phys. Chem. B, 119, 7186–7197.

280 Akimov, A.V., Asahi, R., Jinnouchi, R., and Prezhdo, O.V. (2015) What makesthe photocatalytic CO2 reduction on N-doped Ta2O5 efficient: insights fromnonadiabatic molecular dynamics. J. Am. Chem. Soc., 137, 11517–11525.

567

20

An Overview of Solar Photocatalytic Reactor Designs andTheir Broader Impact on the EnvironmentJustin D. Glover1, Adam C. Hartley1, Reid A. Windmiller2, Naoma S. Nelsen2, andJoel E. Boyd1

1Erskine College Department of Chemistry, The Erskine Center for Environmental Stewardship, 2 WashingtonStreet, Due West, SC 29639, USA2Erskine College Department of Biology, The Erskine Center for Environmental Stewardship, Due West, SC29639, USA

20.1 Introduction

Since research began into the photocatalytic usage of metal oxide semiconductors,countless published manuscripts have mentioned the “promise” provided bythese photocatalytic materials for future commercial applications. The fact thatthese large-scale applications have been a long time in the making is a testamentto both the magnitude and the multitude of the barriers to commercializationthat were present at the outset. The raw volume of work that has been done inthis area since the 1970s has helped us understand most of these barriers, andindeed, to overcome many of them. The prior chapters of this book are filled withaccounts of innovations in materials, methods, and applications that have madeit possible for large-scale progress to become a reality. This chapter is focusedon the larger-scale applications in the published literature and benefits greatlyfrom some outstanding reviews on this topic [1–6]. Interestingly, the volumeof material published on a particular photocatalytic issue or application is notnecessarily indicative of readiness for commercial application. In fact, despite thepreponderance of aqueous phase laboratory-based publications in photocatal-ysis, there are actually more commercial applications of gas-phase work at thepresent time [1]. Although water treatment is the focus of this chapter, gas-phaseapplications will be briefly discussed as well. In order for photocatalysis to makeits way from the laboratory scale to the industrial scale, the following issuesmust be addressed, and the extent to which they have already been addressed isindicated by the type of large-scale applications published to date:

• Materials: Efficient use of illumination (both visible and UV) and the amountof the photocatalyst itself is essential. For aqueous-phase applications, the deci-sion of whether to use the photocatalyst in a slurry, or whether to deposit iton a support material within the reactor has important implications for workoutside of the laboratory.


568 20 Solar Photocatalytic Reactor Designs and Their Broader Impact on the Environment

• Reactor design: Many different types of reactor designs have been used inlaboratory studies, but a relatively small number have been implemented on alarge scale. This is due to the nontrivial challenges in colocating the irradiance,the photocatalyst, and the contaminant. A successful reactor design is onethat accomplishes this task in a cost- and space-efficient manner within theconstraints of the intended site-specific application.

• Retention: An artifact of the materials, deposition strategy, and reactordesign, the retention of the nanoscale photocatalyst is of utmost importance.Loss of photocatalyst diminishes the long-term effectiveness of the reactorsystem while releasing into the environment nanoscale materials can inturn become environmental hazards of their own. The final section of thischapter reviews the current state of the art in the environmental fate andtoxicology of nanoscale titanium dioxide, the most commonly used metaloxide photocatalyst.

20.2 Materials

TiO2 (titania) is the most commonly used photocatalytic material due to itslow cost, high availability, photo and chemical stability, as well as its moderatebandgap (3.2 eV for anatase) allowing it to utilize UV light (𝜆< 385 nm) [2, 7, 8].Most of this chapter will deal with titania photocatalysts, although other pho-tocatalysts have been successfully used as well. Titania can be implemented forphotocatalysis in a variety of morphologies including nanoparticles, nanotubes,nanorods, sheets, and various interconnected architectures [9]. Each formprovides its own potential advantages, with the benchmark material being thecommercially prepared P-25 nanoparticles from Evonik (formerly Degussa) [10].Solar illumination is a highly desirable and green option to provide excitationfor the photocatalyst, and yet UV is a relatively small portion of the overall solarspectrum. Thus, the modification of titania to utilize visible light is an importantstep in facilitating large-scale solar photocatalytic applications. Furthermore,the toxicity and cost of the modified photocatalyst is an important factor forlarge-scale installations. Two techniques for visible-light modification thatmeet these requirements are Fe-modification and N-doping. Modifying titaniawith transition metal cations such as Fe3+ has been shown to shift the titaniabandgap, allowing the utilization of visible light for excitation. The results ofthese studies have been mixed, but in general, the improvement in visible-lightactivity is at least somewhat offset by a loss of UV-induced photocatalytic activity[7]. N-doped titania also has the ability to shift the titania bandgap. N atomscan fill interstitial voids in the titania lattice, or they can replace O atoms oreven Ti atoms within the titania structure. N-doped titania shows visible-lightabsorbance in between 400 and 600 nm, and has limited visible-light activity inthat range as well [7]. The metal modification of titania with noble metals suchas Pt and Pd has shown great promise in many photocatalytic reports, but itis unlikely that the expense and risk of release of these metals into the treatedwater would ever allow such metals to be used in a large-scale solar applicationof titania photocatalysis for water purification.

20.4 Deposited Photocatalysts 569

20.3 Slurry-Style Photocatalysis

The direct addition of the nanoscale photocatalyst to the contaminated waterto form a slurry-style reactor has several advantages. In this format, the pho-tocatalyst has the maximum exposed surface area in contact with the solution.Additionally, solution transport issues (migration of reactants and products toand away from the activated surface) are optimized in a slurry reactor. Disad-vantageously, however, the light penetration depth below the slurry surface isminimal, resulting in the need for relatively thin fluid layers in order to achievesolar activation. In addition to the absorbance and scattering of the photocatalystitself, absorbance and scattering of light from the aqueous contaminants them-selves can reduce the penetration depth of the illumination and reduce the lightavailable for activation of the photocatalyst. Arguably the greatest drawback to aslurry-style application is that the nanoscale photocatalyst must be removed fromthe water after the photocatalytic process is complete. The photocatalyst can beseparated by agglomeration, sedimentation, and microfiltration [3, 11]. This needfor postuse photocatalyst removal dictates that the reactor system be designedfor batch purification rather than continuous flow, and the settling time can takehours [3]. For some applications, this limitation is not problematic, whereas inothers a continuous flow approach is optimal. The perceived drawbacks to reac-tors using nanoscale slurries have certainly not prevented very creative medium-and even large-scale applications involving titania slurries, which benefit fromthe enhanced photocatalytic activity of dispersed titania.

20.4 Deposited Photocatalysts

Deposition of the photocatalyst on a support material inherently decreasesthe available surface area of the photocatalyst, and a concomitant reduction inphotocatalytic efficiency is nearly always observed. The secondary challenge ofthe postuse removal of the titania is avoided when the photocatalyst is firmlybound to the support material [11]. Many different supports have been usedincluding glass, quartz, sand, polymers, vermiculite, concrete, and activatedcarbon [1, 11–19]. The ideal support material would be low cost, mechanicallydurable, UV stable, and highly resistant to photocatalytic oxidation [20]. Becausethe titania is in direct contact with the support material, the resistance of thesupport material to photocatalytic attack is directly related to the longevityof the photocatalytic composite material. If the titania destroys the supportmaterial interface, it can figuratively eat its way off of the surface and release intothe solution in much the same way that latex paint gets “chalky” from long-termsolar exposure of the titania pigment within the paint [1]. The pH of the aqueousmatrix varies widely across various applications, and thus the stability of thetitania-support interface in acid, neutral, and basic conditions is also important[11]. In addition to UV, oxidative, and pH stability, a composite photocatalyticmaterial must also be stable with respect to potential degradation byproductsof the intended target species. For example, 2-propanol has been shown tobe photocatalytically converted into acetone [21]. Many polymeric support


materials, including poly(methyl methacrylate), could be adversely affected bythe acetone in solution, resulting in the release of the titania from the polymericsurface [14, 22]. A further consideration is the temperature range over which thephotocatalytic composite would be used. Depending on the geographic location,season, and reactor design, some solar applications can reach temperatures ashigh as 60–70 ∘C in reactors even without solar concentrating reflectors [3].Thermal degradation of the support, or even thermal expansion and contractionacross the temperature range of use could lead to cracking of the materialor separation of the titania from the support surface. The titania lost due toany of these failure modes is not likely to be recovered, since a supportedphotocatalytic reactor design typically does not contain any of the mechanismsfor titania recovery used in a slurry reactor system. If such recovery techniquesare necessary, there is no remaining reason to use a deposited titania systemat all. Although bulk titania is known to be nontoxic, the rapidly burgeoningfield of nanotoxicology has yet to provide a definitive answer regarding theimpact of titania nanoparticles on human health and the environment [2, 7, 14].The UV stability of a support material is important for a solar application, butUV transparency is also a highly desirable property for some reactor designs.A UV-transparent support material allows for the possibility of illuminationthrough the support material, precluding the need for the light to traverse theaqueous solution, facilitating many applicable reactor configurations.

20.5 Applications

There are many obstacles to overcome in taking photocatalysis from the labora-tory scale to the industrial scale. The successes in this area are testaments to greatengineering and materials science ingenuity. As mentioned previously, a substan-tial fraction of the commercial applications of titania photocatalysis have been inthe gas phase, and this section will begin with a few such examples.

20.5.1 Gas Phase and Self-Cleaning Applications

One of the earliest commercial applications of solar photocatalysis is theintegration of titania in cementitious materials for the degradation of atmo-spheric hazards such as NOx and SOx in urban environments [23]. The basicpH of a cement surface provides a strong binding affinity for acidic gas-phasecontaminants such as these, providing a synergistic effect for titania-cementcomposite materials [1, 24]. Furthermore, the addition of photocatalytic titaniato roadway surfaces allows for the photocatalytic degradation of the NOxfrom automobiles very close to the source itself where concentrations aregreatest. An illuminated titania film possesses “superhydrophilic” properties,which causes an illumination-dependent wettability that provides notable anddurable self-cleaning properties [25]. Large structures including AT&T Stadiumin Arlington Texas (home of the Dallas Cowboys of the National FootballLeague) utilize titania-impregnated architectural textiles as self-cleaning roofingmaterials [1]. The presence of titania on travertine (an architecturally important

20.5 Applications 571

limestone) has been shown to provide self-cleaning properties as well as NOxdegradation capability [26]. Another commercialized gas-phase application oftitania photocatalysis is in ethylene oxidation in produce storage and transportapplications [1, 8]. Negishi and Sano reported a vertical photocatalytic towerreactor designed for the destruction of volatile organic compounds (VOCs)produced in small to medium-sized urban industrial sites where space limi-tations require a minimal spatial footprint [27]. Monteiro et al. developed apilot-scale reactor that utilized solar illumination by day and artificial illumi-nation by night for the continual photocatalytic oxidation of n-decane in thegas phase [28]. Gas-phase and self-cleaning applications are significantly aheadof water-cleaning applications in terms of commercialization [1], but a greatdeal of progress has been made in the area of photocatalytic water purificationusing solar illumination, and that will be the main focus of the next section ofthis chapter.

20.5.2 Water Purification Applications

There is a broad diversity within the laboratory-scale photocatalytic literatureregarding the reactor design and implementation. For large solar applications,however, most published reactors fall into three main categories, each of whichmay involve titania in a slurry or deposited within the reactor system. Each ofthese reactor types will be described below, with specific application examples,and a brief description of their relative advantages and disadvantages. These char-acteristics are well evaluated by criteria recently spelled out in a review articleby Malato et al. They listed five properties that should be present in a reactorconfiguration for photocatalytic water treatment [5]:

i) Low costii) Ability to use both direct and indirect solar illumination

iii) Low temperature increase of the wateriv) High photonic efficiency with low e−/h+ pair recombinationv) Turbulent water flow to maximize mass transfer.

These five properties provide an excellent reference in evaluating the threemain reactor types described below.

20.5.3 Inclined Plate Collectors

The simplest type of photocatalytic reactor system for solar water purificationis the inclined plate reactor (IPR). In this system, the contaminated water flowsover a plate ideally inclined for normal incidence of the solar radiation. Thetitania is either deposited on this inclined plate or present with the water as aslurry. The inclined plate can be planar, corrugated, or stepped. The corrugatedand stepped surface increases the residence time of the water on the illuminatedsurface and maximizes flow turbulence for enhanced reaction rates [4, 29].IPRs do not concentrate the solar radiation and are able to use both direct anddiffuse illumination, and are thus typically not equipped with solar-trackingcapability. Due to their lack of solar concentration, IPR systems thus require a


relatively large footprint in order to obtain a functional amount of photocatalyticfunction [3]. Furthermore, in order to maintain desirable water film thicknesseson the reactor surface, the flow rates are severely limited. Higher flow ratescan help ameliorate mass transfer limitations, but in the absence of turbulentflow, higher flow rates can also produce lamellar flow problems where only thesolution in immediate contact with the inclined surface experiences substantialinteraction with the photocatalyst [4]. Without covering the reactor surface, theevaporation rate of both water and the contaminants can be extreme. Coveringthe surface inevitably results in decreased illumination intensity on the titaniasurface [1, 4]. The greatest advantage of an IPR lies in its engineering simplicity,which minimizes installation costs and maintenance hassles. Although IPRshave since fallen out of favor due to these drawbacks, a classic IPR applicationelegantly captures the simple functionality of this approach. Pichat et al. useda simple yet effective photocatalytic reactor system involving a 1 m2 inclinedcorrugated steel panel coated with titania for the photocatalytic degradation ofpesticide-laden equipment rinse-water from a commercial vineyard [30]. Thisapparatus was powered by a photovoltaic panel driving the aquarium-type pumpthat circulated the water over the corrugated surface from the reservoir filledwith 80 l of contaminated water. In a similar refrain to other IPR reports, thesize of this IPR system proved to be insufficient, and the authors estimated thattripling the size of the system would likely be necessary to treat the pesticiderinse-water from this 0.15 km2 vineyard [4, 30].

20.5.4 Parabolic Trough Concentrator

In a parabolic trough concentrator (PTC), the photocatalyst is placed along withthe contaminated water within an “absorber tube” located at the focus of theparabolic trough. The origins of this photocatalytic reactor design lie in solarthermal applications such as solar hot water [3–5]. Since the focus of a parabolicreflector is only illuminated when the light is incident along the “vertical” axisof the parabolic reflector, a PTC system requires a solar tracking mechanism toensure that direct illumination is maintained throughout the day. This also meansthat indirect diffuse illumination, which can be an enormous fraction of the totallight available on even a sunny day, is not usable by a PTC reactor [1, 3]. PTCreactors are, however, very effective at concentrating direct solar illuminationwhen properly aligned, and concentration factors of 5–35 suns are commonfor photocatalytic water purification, [1, 4] while 20–150 suns are common inreactors designed for solar photochemical synthesis reactions [6]. These highconcentration factors can help minimize the amount of photocatalyst necessaryin the system to achieve effective yields, thus lowering the cost and overall sizeof the system [1, 4]. The treated water within a PTC reactor is enclosed withinthe absorber tubes, and is thus not prone to evaporation losses of the wateror volatile contaminants [1]. The large concentration factors of PTC systemscan cause increased e−/h+ pair recombination due to excessive excitation andelevated temperatures, and cooling systems are often used to control reactortemperature [3, 29]. The first PTC reactor was installed at Sandia National Labsin Albuquerque, New Mexico, USA, in 1989. This reactor, a repurposed solar


thermal collector with solar tracking, had a total surface area of 465 m2, andhad a solar concentration factor of 50 suns. The reactor was designed to removetrichloroethylene, trichloroethane, and heavy metals using suspended titaniaphotocatalyst [1, 4, 31, 32]. The fundamental success of this reactor systemprovided a solid foundation for the years of large-scale photocatalytic waterpurification experiments that have followed.

20.5.5 Compound Parabolic Concentrator Reactor

Compound parabolic concentrators (CPCs) were also originally designed forsolar thermal applications without the need for solar tracking, and have thegreatest capacity to fulfill the five desirable traits for photocatalytic reactorsystems listed previously [1, 5]. These reactors are similar to PTC in that theyuse a concentrating reflector and a central absorber tube containing both thewater and the photocatalyst. The reflector is formed from the intersection oftwo truncated parabolas just below the absorber tube. The dual-parabolic,omega-like shape of the CPC reflector system means that the incident radiationis not truly focused on the absorber tube, but is rather directed to illuminate theentire circumference of the tube – the top surface directly illuminated, whilethe back side is illuminated from the reflected light. The light need not enterthe CPC reflector directly from above, and diffuse light is readily utilized. In acommon CPC configuration that maximizes the utilization of diffuse light, theabsorber tube circumference matches the span of the reflector opening, resultingin a concentration factor of 1. In that case, a CPC functions as a hybrid betweena PTC and ITR, and benefits from many of the best attributes of both. Since thedirectionality of the incident light is no longer important, the reflectivity require-ments on the CPC reflector is much lower than the imaging-quality reflectorsdesirable for a PTC [3]. This also means that a CPC system does not requiresolar tracking functionality, lowering the cost and complexity of the system [1,4, 5, 29]. The irradiance within a CPC absorber tube is much more uniform thanin a PTC, and thus the light is more efficiently utilized. CPCs are reportedly30–200% more effective than comparably sized PTC reactor systems [4]. Thewater temperature is much lower in a CPC than in a PTC, further allowing fora wider choice of absorber tube materials [3]. Pilot-scale systems ranging from3 to 150 m2 have been reported in the literature [1]. Common failure modes ofCPC reactors include optical degradation of absorber tube materials with time,plugging, and fouling of the photocatalyst-support material in a packed-tubereactor when slurries are not used, which limits the long-term reuse of the sys-tem [29]. Colina-Marquez et al. utilized a recirculating solar CPC reactor systemfor the degradation of estrogens with a system containing an optimal suspendedtitania loading of 0.4 g l−1 within 29 mm glass tubes having a total length of12 m. This system provided 56.7% degradation of 40 l of 3 ppm commercial 17-βestradiol and nomegestrol acetate in a tap water matrix after 40 min of solarillumination, with a total organic carbon reduction of 31% [33]. Estrogens andother endocrine disrupting compounds are an important emerging contaminantfor which photocatalysis is a potential solution [34]. Andronic et al. recentlyreported the use of a CPC reactor along with titania compared to a titania-fly ash


composite [35]. The titania-fly ash composite materials facilitated the postuseseparability of the photocatalyst. These materials were used in a solar pilot-scaleplant equipped with a CPC reactor at the Plataforma Solar de Almeria. Phenolwas present in a 35 l solution into which the photocatalyst was added at 0.2 g l−1.After 150 min of solar exposure, the phenol removal efficiency was 66%, withtitania, and just under 35% with the titania-fly ash composite. The titania-flyash composites had lower photocatalytic activity in a compromise with theirimproved postuse removal. This example clearly indicates both the utility ofthe CPC reactor system, as well as the trade-offs that are necessary betweenactivity and postuse titania removal. The lower degradation rate of the supportedphotocatalyst can only be weighed against the potential costs of the release ofnanoscale titania into the environment.

20.5.6 The Environmental Impact of Nanoscale Titania

The use of engineered nanoparticles in large-scale water purification systemsraises concerns related to the release of nanoparticles into the environment andthe potential toxicity associated with environmental exposure. Predicting risksassociated with nanoparticle exposure, however, has proved challenging becauseof difficulties in accurately estimating the levels of exposure and toxicity [36]. Theunique physicochemical properties of nanoparticles and their interactions withcomplex biological and environmental matrices ultimately results in alterationsto the nanoparticles that are temporally and spatially dynamic [37]. As a result,many of the standard testing methods used to quantify the concentration andtoxicity of soluble chemicals do not apply to nanoparticles [36, 38].

Recently, researchers in the field of nanotoxicology have worked to identifyknowledge gaps in the ecological risk assessment of engineered nanoparticles andto define research priorities that will provide reliable data for use in establishingregulations for the sustainable development of nanomaterials [36, 39–41]. Theprimary research goals identified by these panels seek to address current limita-tions in the ability to (i) detect and quantify engineered nanoparticles in environ-mental and biological matrices, (ii) predict the fate of engineered nanoparticles inthe environment, (iii) assess the hazards associated with environmental exposure,and (iv) develop quantitative risk assessment models to determine the probableoutcome of exposure under specified conditions [36].

20.5.7 Detecting and Quantifying Nanoparticles

Analysis of exposure levels during toxicity testing requires quantitation of thetest material in the environment and/or the test organism [40]. For nanopar-ticles, a challenge is presented because both the chemical and physical statesaffect bioavailability and toxicity, and both need to be considered when selectingthe most appropriate techniques for risk assessment [36]. Further, the physico-chemical state changes over time as nanoparticles undergo transformations in theenvironment. Therefore, the exposure concentration is not constant over time[40]. Additionally, there is a high occurrence of artifacts when exposure con-centrations are measured because the physicochemical state of nanoparticles is


dependent on their environment and may be altered by the process of samplecollection and analysis [38]. Many analytical techniques developed for colloidshave been adapted for use with nanoparticles, and a review outlining the currenttechniques and applications for quantifying nanoparticle concentrations, as wellas the limitations in the technology, is available [38].

In complex environmental samples, quantitation of nanoparticle concentra-tion becomes more challenging. Very few studies provide analytical data for theconcentration of engineered nanoparticles in the natural environment, and mostrisk assessments are based on predicted environmental concentrations, ratherthan actual values [42]. The ability to easily and accurately quantify nanoparticlesin complex environmental matrices is limited by their low concentrations (pre-dicted to be at parts per billion (ppb) or sub-ppb levels [43]) and the difficulty indistinguishing between engineered nanoparticles and similar naturally occurringspecies [44]. In order to improve risk assessment, it will be necessary to continuedeveloping techniques to accurately quantify nanoparticles in complex matri-ces at environmentally relevant concentrations, particularly in soil and sedimentsamples where nanoparticles tend to accumulate [40]. For metal-based nanopar-ticles, there is also a need for techniques to differentiate between nanoparticles,agglomerates, and dissolution products [39].

Monitoring the uptake of nanoparticles by organisms and tracking their fateand distribution within the tissues is also critical for risk assessment. The tech-niques used for analysis of assimilation and distribution of nanoparticles in bio-logical samples must be able to differentiate between the various transformationstates of the nanoparticle, and a review of current imaging and analytical meth-ods is available [45]. Recently, the use of stable isotope tracers has proved to bea promising approach in monitoring the fate of nanoparticles in biological sam-ples [44]. In one study, researchers were able to detect the bioaccumulation of47TiO2 nanoparticles in the gut of zebra mussels (Dreissena polymorpha) exposedat environmental concentrations, despite the high background of natural Ti foundin the tissues [46].

20.5.8 Transformation of Nanoparticles in the Environment

Most research examining the toxicity of engineered nanoparticles has relied onthe use of pristine, or as manufactured, nanoparticles [47]. However, much ofthe nanoparticle waste that enters the environment is in the form of modifiednanoparticles embedded in manufactured products or released from these prod-ucts [36, 47]. The physical, chemical, and biological properties exhibited by thesemodified nanoparticles differ from those of pristine nanoparticles, and thereforethe manufacturing history of the nanoparticles will play a role in determining theenvironmental fate and toxicity of these molecules [40, 47].

Following release, nanoparticles in the environment undergo further chemicaland physical transformations that affect their activity and bioavailability, effec-tively creating a heterogeneous population of nanoparticle species that vary intheir environmental behavior [37, 47–50]. Several excellent reviews summarizingthe possible environmental transformations of nanoparticles are available [37,47–51]. Transformations of nanoparticles in the aquatic environment, including


dissolution, homoagglomeration, heteroagglomeration, deagglomeration, andsurface coating, are dynamic processes that affect the localization of nanoparti-cles in environmental compartments [37, 50]. These transformation processesare dependent on the physicochemical characteristics of the nanoparticle(composition, size, surface area and morphology, surface charge potential, andcrystal structure) as well as the environment (salinity, pH, water hardness,and the presence of natural organic materials) [40]. In order to predict theenvironmental behavior of nanoparticles, it is therefore necessary to considerthe properties of both the nanoparticle and the aqueous environment.

Dissolution of nanoparticles releases soluble ions or molecules that are activein the water column [47]. Dissolution reduces the persistence of the nanoparticlein the environment but has the potential to increase toxicity by releasing toxicions (such as Ag+, Cd2+, Zn2+, and Cu2+) [37]. The solubility of nanoparticlesand the rate of dissolution are dependent on the physicochemical propertiesof the nanoparticle and the aqueous environment, as described above [52].Surface modification, such as the use of capping agents or surfactants, can alterthe dissolution behavior of nanoparticles [52]. The extent of dissolution alsosignificantly affects cellular uptake pathways, with free ion transport occurringthrough ion channels or ion transporters and uptake of nanoparticles throughendocytic pathways [52].

Aggregation and agglomeration of nanoparticles can lead to precipitation,reducing nanoparticle concentrations in the water column and increasing expo-sure concentrations in the sediment [50]. Because of this, sediments and benthicorganisms are expected to be the main sinks for nanoparticles in aquatic envi-ronments [48]. However, much of the research assessing nanoparticle toxicity inaquatic systems focuses on pelagic organisms that live in surface waters [36]. Theeffects of natural organic matter (NOM) on nanoparticle stability in the watercolumn are complex, making it difficult to predict the fate of the nanoparticles[37]. Aggregation/dissolution and agglomeration/de-agglomeration are dynamicprocesses that affect the bioavailability and activity of nanoparticles in the watercolumn and the sediment. Because these transformation processes are often notat equilibrium, real-time kinetic measurements will be required to determineexposure concentration in the different environmental compartments [37].

20.5.9 Toxicity of Nanoparticles

Exposure to nanoparticles can cause damage to cells and tissues through severalrelated mechanisms. First, dissolution processes can cause the release of toxicions, such as Ag+ and Zn2+, resulting in chemical toxicity [52]. Nanoparticles,because of their unique physicochemical properties, can also induce physicalstress in cells, resulting in the production of reactive oxygen species (ROS) bythe cell and leading to oxidative stress [37]. Some nanoparticles attach to thesurface of the cell and disrupt membrane function [48]. Photocatalytic nanopar-ticles can also cause damage to cells through the production of ROS [37]. ROSdisrupt cell membranes through lipid peroxidation and cause damage to DNAand proteins [48]. In response to the presence of ROS, cells produce superoxidedismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx) [37]. These

References 577

enzymes play a critical role in inactivating ROS, and are often used as biomarkersin toxicity assays [37].

Numerous studies have been performed to assess the toxicity of nanoparti-cles in the laboratory, and several excellent reviews summarizing these studiesare available [37, 49, 50, 53–56]. Recent analysis of the literature has identifiedcommon limitations in many of these studies and provided recommendations toimprove the quality and reliability of the research. These include testing nanopar-ticle activity in complex media, using environmentally relevant concentrationsof nanoparticles, using extended exposure times, standardizing test methods,including appropriate controls, and using microcosm/mesocosm studies to rep-resent natural systems [36, 40].

A shift in the approach toward analyzing nanoparticle toxicity has beengenerated in response to these recommendations. Standardized test methods fortoxicity assays are being developed [39]. Also, experiments are being designed tobetter reflect natural conditions in terms of assessing environmentally relevantnanoparticle concentrations [46, 57], using of complex test media [58–62],conducting mesocosm studies [63], and assessing toxicity in benthic systems[61, 64–68]. Additionally, a number of studies are beginning to address thequestion of how interactions between nanoparticles in the environment affecttheir activity [61, 69–72].

20.6 Conclusion

Large-scale photocatalytic water purification is not only “promising,” but is acurrent reality. The barriers to widespread industrial adoption are fewer thanever. Advances in nanotoxicology are providing an improving basis for evaluatingthe potential environmental hazard posed by nanoscale photocatalysts in theenvironment. Adequate knowledge of those hazards is essential to a cost-benefitanalysis involving photocatalytic water purification applications. Sustainablephotocatalytic water purification systems of the future will need photocatalystswith high activity, and reactor designs that make efficient use of both lightand photocatalyst. An emphasis on photocatalyst retention will enhance thelong-term durability of these reactor systems and avoid environmental exposureto nanoscale photocatalysts. The material and reactor design platform is in placefor large-scale photocatalytic water purification applications that can add animportant capability to the incumbent methods in water purification.

References

1 Spasiano, D., Marotta, R., Malato, S., Fernandez-Ibañez, P., and Di Somma,I. (2015) Solar photocatalysis: materials, reactors, some commercial, andpre-industrialized applications. A comprehensive approach. Appl. Catal. B:Environ., 170–171, 90–123.

2 Gulyas, H. (2014) Solar heterogeneous photocatalytic oxidation for water andwastewater treatment: problems and challenges. J. Adv. Chem. Eng., 4, 1–11.


3 Malato, S., Fernández-Ibáñez, P., Maldonado, M.I., Oller, I., and Polo-López,M.I. (2013) Solar Photocatalytic Pilot Plants: Commercially Available Reac-tors Photocatalysis and Water Purification, Wiley-VCH Verlag GmbH & Co.KGaA, pp. 377–397.

4 Braham, R.J. and Harris, A.T. (2009) Review of major design and scale-upconsiderations for solar photocatalytic reactors. Ind. Eng. Chem. Res., 48,8890–8905.

5 Malato, S., Maldonado, M.I., Fernández-Ibáñez, P., Oller, I., Polo, I., andSánchez-Moreno, R. (2016) Decontamination and disinfection of water bysolar photocatalysis: the pilot plants of the Plataforma Solar de Almeria.Mater. Sci. Semicond. Process., 42 Part 1, 15–23.

6 Oelgemöller, M. (2016) Solar photochemical synthesis: from the beginnings oforganic photochemistry to the solar manufacturing of commodity chemicals.Chem. Rev., 116, 9664–9682.

7 Daghrir, R., Drogui, P., and Robert, D. (2013) Modified TiO2 for environmen-tal photocatalytic applications: a review. Ind. Eng. Chem. Res., 52, 3581–3599.

8 Keller, N., Ducamp, M.-N., Robert, D., and Keller, V. (2013) Ethylene removaland fresh product storage: a challenge at the frontiers of chemistry. Towardan approach by photocatalytic oxidation. Chem. Rev., 113, 5029–5070.

9 Nakata, K. and Fujishima, A. (2012) TiO2 photocatalysis: design and applica-tion. J. Photochem. Photobiol., C: Photochem. Rev., 13, 169–189.

10 Malato, S., Blanco, J., Richter, C., and Maldonado, M.I. (2000) Optimizationof pre-industrial solar photocatalytic mineralization of commercial pesti-cides: application to pesticide container recycling. Appl. Catal. B: Environ, 25,31–38.

11 Malato, S., Blanco, J., Vidal, A., and Richter, C. (2002) Photocatalysis withsolar energy at a pilot-plant scale: an overview. Appl. Catal. B: Environ., 37,1–15.

12 Hiroyuki, U., Shigeyoshi, I., and Hiroshi, Y. (1993) Photocatalytic decomposi-tion of propyzamide using TiO2 supported on activated carbon. Chem. Lett.,22, 1995–1998.

13 Loetscher, L.H., Carey, J.M., Skiles, S.L., Carey, V.M., and Boyd, J.E. (2009)Titania−acrylic coil reactor for photocatalytic water purification and steriliza-tion. Ind. Eng. Chem. Res., 48, 4697–4702.

14 Stewart, B.D., Andrews, L.G., Pelletier, B.S., Daly, C.A. Jr., and Boyd, J.E.(2015) Porous PMMA-titania composites: a step towards more sustainablephotocatalysis. J. Water Process Eng., 8, 179–185.

15 Mansilla, H.D., Bravo, C., Ferreyra, R., Litter, M.I., Jardim, W.F., Lizama,C., Freer, J., and Fernández, J. (2006) Photocatalytic EDTA degradation onsuspended and immobilized TiO2. J. Photochem. Photobiol. A: Chem, 181,188–194.

16 Jin, L. and Dai, B. (2012) TiO2 activation using acid-treated vermiculite as asupport: characteristics and photoreactivity. Appl. Surf. Sci., 258, 3386–3392.

17 Machado, L.C.R., Torchia, C.B., and Lago, R.M. (2006) Floating photocata-lysts based on TiO2 supported on high surface area exfoliated vermiculite forwater decontamination. Catal. Commun., 7, 538–541.

References 579

18 Wang, L., Wang, X., Cui, S., Fan, X., Zu, B., and Wang, C. (2013) TiO2supported on silica nanolayers derived from vermiculite for efficient pho-tocatalysis. Catal. Today, 216, 95–103.

19 Ling, C.M., Mohamed, A.R., and Bhatia, S. (2004) Performance of photocat-alytic reactors using immobilized TiO2 film for the degradation of phenol andmethylene blue dye present in water stream. Chemosphere, 57, 547–554.

20 Carlson, P.J., Pretzer, L.A., and Boyd, J.E. (2007) Solvent deposition of tita-nium dioxide on acrylic for photocatalytic application. Ind. Eng. Chem. Res.,46, 7970–7976.

21 Yamashita, H., Nose, H., Kuwahara, Y., Nishida, Y., Yuan, S., and Mori, K.(2008) TiO2 photocatalyst loaded on hydrophobic Si3N4 support for efficientdegradation of organics diluted in water. Appl. Catal., A350, 164–168.

22 Cantarella, M., Sanz, R., Buccheri, M.A., Ruffino, F., Rappazzo, G., Scalese, S.,Impellizzeri, G., Romano, L., and Privitera, V. (2016) Immobilization of nano-materials in PMMA composites for photocatalytic removal of dyes, phenolsand bacteria from water. J. Photochem. Photobiol. A: Chem., 321, 1–11.

23 Janus M., Zajac K. (2016) Concretes with photocatalytic activity. In: High Per-formance Concrete Technology and Applications. S. Yilmaz (Ed.), InTech, DOI:10.5772/64779

24 Cassar, L. (2004) Photocatalysis of cementitious materials: clean buildings andclean air. MRS Bull., 29, 328–331.


26 Quagliarini, E., Bondioli, F., Goffredo, G.B., Cordoni, C., and Munafò, P.(2012) Self-cleaning and de-polluting stone surfaces: TiO2 nanoparticles forlimestone. Constr. Build. Mater., 37, 51–57.

27 Negishi, N. and Sano, T. (2014) Photocatalytic solar tower reactor for theelimination of a low concentration of VOCs. Molecules, 19, 16624–16639.

28 Monteiro, R.A.R., Rodrigues-Silva, C., Lopes, F.V.S., Silva, A.M.T., Boaventura,R.A.R., and Vilar, V.J.P. (2015) Evaluation of a solar/UV annular pilot scalereactor for 24 h continuous photocatalytic oxidation of n-decane. Chem. Eng.J., 280, 409–416.

29 Abdel-Maksoud, Y., Imam, E., and Ramadan, A. (2016) TiO2 solar pho-tocatalytic reactor systems: selection of reactor design for scale-up andcommercialization—analytical review. Catalysts, 6, 138.

30 Pichat, P., Vannier, S., Dussaud, J., and Rubis, J.-P. (2004) Field solar photocat-alytic purification of pesticides-containing rinse waters from tractor cisternsused for grapevine treatment. Solar Energy, 77, 533–542.

31 Anderson, J.V., Link, H., Bohn, M., and Gupta, B. (1991) Development of solardetoxification technology in the USA – an introduction. Sol. Energy Mater.,24, 538–549.

32 Pacheco, J.E., Prairie, M.R., and Yellowhorse, L. (1993) Photocatalytic destruc-tion of chlorinated solvents in water with solar energy. J. Solar Energy Eng.,115, 123–129.

33 Colina-Marquez, J., Machuca-Martinez, F., and Li, P.G. (2015) Modeling thephotocatalytic mineralization in water of commercial formulation of estrogens


17-β Estradiol (E2) and nomegestrol acetate in contraceptive pills in a solarpowered compound parabolic collector. Molecules, 20, 13354–13373.

34 Das, R. (2014) Application photocatalysis for treatment of industrial wastewater—a short review. Open Access Library J., 1, 1–17.

35 Andronic, L., Isac, L., Miralles-Cuevas, S., Visa, M., Oller, I., Duta, A., andMalato, S. (2016) Pilot-plant evaluation of TiO2 and TiO2-based hybridphotocatalysts for solar treatment of polluted water. J. Hazard Mater., 320,469–478.

36 Klaine, S.J., Koelmans, A.A., Horne, N., Carley, S., Handy, R.D., Kapustka, L.,Nowack, B., and von der Kammer, F. (2012) Paradigms to assess the environ-mental impact of manufactured nanomaterials. Environ. Toxicol. Chem., 31,3–14.

37 Vale, G., Mehennaoui, K., Cambier, S., Libralato, G., Jomini, S., andDomingos, R.F. (2016) Manufactured nanoparticles in the aquaticenvironment-biochemical responses on freshwater organisms: a criticaloverview. Aquat. Toxicol., 170, 162–174.

38 von der Kammer, F., Ferguson, P.L., Holden, P.A., Masion, A., Rogers, K.R.,Klaine, S.J., Koelmans, A.A., Horne, N., and Unrine, J.M. (2012) Analysis ofengineered nanomaterials in complex matrices (environment and biota): gen-eral considerations and conceptual case studies. Environ. Toxicol. Chem., 31,32–49.

39 Petersen, E.J., Diamond, S.A., Kennedy, A.J., Goss, G.G., Ho, K., Lead,J., Hanna, S.K., Hartmann, N.B., Hund-Rinke, K., Mader, B., Manier, N.,Pandard, P., Salinas, E.R., and Sayre, P. (2015) Adapting OECD aquatic toxic-ity tests for use with manufactured nanomaterials: key issues and consensusrecommendations. Environ. Sci. Technol., 49, 9532–9547.

40 Selck, H., Handy, R.D., Fernandes, T.F., Klaine, S.J., and Petersen, E.J. (2016)Nanomaterials in the aquatic environment: a European union-United Statesperspective on the status of ecotoxicity testing, research priorities, and chal-lenges ahead. Environ. Toxicol. Chem., 35, 1055–1067.

41 Wang, J., Gerlach, J.D., Savage, N., and Cobb, G.P. (2013) Necessity andapproach to integrated nanomaterial legislation and governance. Sci. TotalEnviron., 442, 56–62.

42 Gottschalk, F., Sun, T., and Nowack, B. (2013) Environmental concentra-tions of engineered nanomaterials: review of modeling and analytical studies.Environ. Pollut., 181, 287–300.

43 Gottschalk, F., Sonderer, T., Scholz, R.W., and Nowack, B. (2010) Possibilitiesand limitations of modeling environmental exposure to engineered nanoma-terials by probabilistic material flow analysis. Environ. Toxicol. Chem., 29,1036–1048.

44 Part, F., Zecha, G., Causon, T., Sinner, E.K., and Huber-Humer, M. (2015)Current limitations and challenges in nanowaste detection, characterisationand monitoring. Waste Management (New York, N.Y.), 43, 407–420.

45 Schultz, C., Powell, K., Crossley, A., Jurkschat, K., Kille, P., Morgan, A.J.,Read, D., Tyne, W., Lahive, E., Svendsen, C., and Spurgeon, D.J. (2015) Ana-lytical approaches to support current understanding of exposure, uptake and

References 581

distributions of engineered nanoparticles by aquatic and terrestrial organisms.Ecotoxicology (London, England), 24, 239–261.

46 Bourgeault, A., Cousin, C., Geertsen, V., Cassier-Chauvat, C., Chauvat, F.,Durupthy, O., Chaneac, C., and Spalla, O. (2015) The challenge of studyingTiO2 nanoparticle bioaccumulation at environmental concentrations: crucialuse of a stable isotope tracer. Environ. Sci. Technol., 49, 2451–2459.

47 Nowack, B., Ranville, J.F., Diamond, S., Gallego-Urrea, J.A., Metcalfe, C., Rose,J., Horne, N., Koelmans, A.A., and Klaine, S.J. (2012) Potential scenarios fornanomaterial release and subsequent alteration in the environment. Environ.Toxicol. Chem., 31, 50–59.

48 Klaine, S.J., Alvarez, P.J., Batley, G.E., Fernandes, T.F., Handy, R.D., Lyon,D.Y., Mahendra, S., McLaughlin, M.J., and Lead, J.R. (2008) Nanomaterials inthe environment: behavior, fate, bioavailability, and effects. Environ. Toxicol.Chem., 27, 1825–1851.

49 Schaumann, G.E., Philippe, A., Bundschuh, M., Metreveli, G., Klitzke, S.,Rakcheev, D., Grun, A., Kumahor, S.K., Kuhn, M., Baumann, T., Lang, F.,Manz, W., Schulz, R., and Vogel, H.J. (2015) Understanding the fate and bio-logical effects of Ag- and TiO(2)-nanoparticles in the environment: the questfor advanced analytics and interdisciplinary concepts. Sci. Total Environ., 535,3–19.

50 Thwala, M., Klaine, S.J., and Musee, N. (2016) Interactions of metal-basedengineered nanoparticles with aquatic higher plants: a review of the state ofcurrent knowledge. Environ. Toxicol. Chem., 35, 1677–1694.

51 Maurer-Jones, M.A., Gunsolus, I.L., Murphy, C.J., and Haynes, C.L. (2013)Toxicity of engineered nanoparticles in the environment. Anal. Chem., 85,3036–3049.

52 Misra, S.K., Dybowska, A., Berhanu, D., Luoma, S.N., and Valsami-Jones,E. (2012) The complexity of nanoparticle dissolution and its importance innanotoxicological studies. Sci. Total Environ., 438, 225–232.

53 Canesi, L. and Corsi, I. (2016) Effects of nanomaterials on marine inverte-brates. Sci. Total Environ., 565, 933–940.

54 Jovanovic, B. (2015) Review of titanium dioxide nanoparticle phototoxicity:developing a phototoxicity ratio to correct the endpoint values of toxicitytests. Environ. Toxicol. Chem., 34, 1070–1077.

55 Nam, D.-H., Lee, B.-c., Eom, I.-c., Kim, P., and Yeo, M.-K. (2014) Uptake andbioaccumulation of titanium- and silver-nanoparticles in aquatic ecosystems.Molec. Cell. Toxicol., 10, 9–17.

56 Semenzin, E., Lanzellotto, E., Hristozov, D., Critto, A., Zabeo, A., Giubilato,E., and Marcomini, A. (2015) Species sensitivity weighted distribution for eco-logical risk assessment of engineered nanomaterials: the n-TiO2 case study.Environ. Toxicol. Chem., 34, 2644–2659.

57 Miranda, R.R., Damaso da Silveira, A.L., de Jesus, I.P., Grotzner, S.R., Voigt,C.L., Campos, S.X., Garcia, J.R., Randi, M.A., Ribeiro, C.A., and Filipak,N.F. (2016) Effects of realistic concentrations of TiO(2) and ZnO nanopar-ticles in Prochilodus lineatus juvenile fish. Environ. Sci. Pollut. Res. Int., 23,5179–5188.


58 Cupi, D., Hartmann, N.B., and Baun, A. (2015) The influence of naturalorganic matter and aging on suspension stability in guideline toxicity testingof silver, zinc oxide, and titanium dioxide nanoparticles with Daphnia magna.Environ. Toxicol. Chem., 34, 497–506.

59 Fan, W., Peng, R., Li, X., Ren, J., Liu, T., and Wang, X. (2016) Effect of tita-nium dioxide nanoparticles on copper toxicity to Daphnia magna in water:role of organic matter. Water Res., 105, 129–137.

60 Li, S., Ma, H., Wallis, L.K., Etterson, M.A., Riley, B., Hoff, D.J., and Diamond,S.A. (2016) Impact of natural organic matter on particle behavior and photo-toxicity of titanium dioxide nanoparticles. Sci. Total Environ., 542, 324–333.

61 Ma, T., Wang, M., Gong, S., and Tian, B. (2017) Impacts of sediment organicmatter content and ph on ecotoxicity of coexposure of TiO2 nanoparti-cles and cadmium to freshwater snails Bellamya aeruginosa. Arch. Environ.Contam. Toxicol., 72, 153–165.

62 Wormington, A.M., Coral, J., Alloy, M.M., Damare, C.L., Mansfield, C.M.,Klaine, S.J., Bisesi, J.H., and Roberts, A.P. (2016) Effect of natural organic mat-ter on the photo-induced toxicity of titanium dioxide nanoparticles. Environ.Toxicol. Chem., 36, 1661–1666.

63 Kulacki, K.J., Cardinale, B.J., Keller, A.A., Bier, R., and Dickson, H. (2012)How do stream organisms respond to, and influence, the concentration oftitanium dioxide nanoparticles? A mesocosm study with algae and herbivores.Environ. Toxicol. Chem., 31, 2414–2422.

64 Girardello, F., Custodio Leite, C., Vianna Villela, I., da Silva Machado, M.,Luiz Mendes Juchem, A., Roesch-Ely, M., Neves Fernandes, A., Salvador,M., and Antonio Pegas Henriques, J. (2016) Titanium dioxide nanoparti-cles induce genotoxicity but not mutagenicity in golden mussel Limnopernafortunei. Aquat. Toxicol., 170, 223–228.

65 Jovanovic, B. and Guzman, H.M. (2014) Effects of titanium dioxide (TiO2)nanoparticles on caribbean reef-building coral (Montastraea faveolata).Environ. Toxicol. Chem., 33, 1346–1353.

66 Li, S., Wallis, L.K., Ma, H., and Diamond, S.A. (2014) Phototoxicity of TiO(2)nanoparticles to a freshwater benthic amphipod: are benthic systems at risk?Sci. Total Environ., 466–467, 800–808.

67 Rosenfeldt, R.R., Seitz, F., Zubrod, J.P., Feckler, A., Merkel, T., Luderwald,S., Bundschuh, R., Schulz, R., and Bundschuh, M. (2015) Does the pres-ence of titanium dioxide nanoparticles reduce copper toxicity? A factorialapproach with the benthic amphipod Gammarus fossarum. Aquat. Toxicol.,165, 154–159.

68 Wallis, L.K., Diamond, S.A., Ma, H., Hoff, D.J., Al-Abed, S.R., and Li, S.(2014) Chronic TiO(2) nanoparticle exposure to a benthic organism, Hyalellaazteca: impact of solar UV radiation and material surface coatings on toxicity.Sci. Total Environ., 499, 356–362.

69 Boran, H., Boyle, D., Altinok, I., Patsiou, D., and Henry, T.B. (2016) AqueousHg(2+) associates with TiO2 nanoparticles according to particle size, changesparticle agglomeration, and becomes less bioavailable to zebrafish. Aquat.Toxicol., 174, 242–246.

References 583

70 Canesi, L., Frenzilli, G., Balbi, T., Bernardeschi, M., Ciacci, C., Corsolini,S., Della Torre, C., Fabbri, R., Faleri, C., Focardi, S., Guidi, P., Kocan, A.,Marcomini, A., Mariottini, M., Nigro, M., Pozo-Gallardo, K., Rocco, L.,Scarcelli, V., Smerilli, A., and Corsi, I. (2014) Interactive effects of n-TiO2and 2,3,7,8-TCDD on the marine bivalve Mytilus galloprovincialis. Aquat.Toxicol., 153, 53–65.

71 Fang, Q., Shi, Q., Guo, Y., Hua, J., Wang, X., and Zhou, B. (2016) Enhancedbioconcentration of bisphenol A in the presence of Nano-TiO2 can leadto adverse reproductive outcomes in zebrafish. Environ. Sci. Technol., 50,1005–1013.

72 Mansouri, B., Maleki, A., Johari, S.A., Shahmoradi, B., Mohammadi, E.,Shahsavari, S., and Davari, B. (2016) Copper bioaccumulation and depura-tion in common carp (Cyprinus carpio) following co-exposure to TiO2 andCuO nanoparticles. Arch. Environ. Contam. Toxicol., 71, 541–552.

585

21

Conclusions and Future WorkSrabanti Ghosh

CSIR - Central Glass and Ceramic Research Institute, Fuel Cell and Battery Division, 196, Raja S.C. Mullick Road,Kolkata 700032, India

In the twenty-first century, the key challenge is the development of alternativeenergy sources for ever-escalating energy demand due to diminishing fossil fuels,global warming issues, and other environmental problems such as contaminatedgroundwater and toxic air contaminants. Catalysis is one of the superlativeapproaches toward a solution for solar energy conversion and environmentalremediation. Design of nanostructures and hybridization with specific activematerials has emerged as an interesting platform for light harvesting andvisible-light-driven photocatalysis. Notably, artificial systems are being devel-oped with the aim of mimicking natural photosynthesis and directly harvestingand converting solar energy into renewable energy through water splitting. Thekey idea of visible-light-active photocatalysis is that the materials should possessenhanced light absorption in the visible region, promote the charge separationand transportation, and enhance the redox catalytic activity. Very impressiveresults have been obtained in the last few years with the realization of light-drivendye degradation, hydrogen generation, CO2 reduction, and organic transfor-mations by using nanostructures catalytic materials. The book focuses on theprinciples of visible-light-induced photocatalytic activities followed by some keyexamples of oxide-based materials, carbon materials, and so on. Chapters 2–18include a wide range of structural modification and crystal growth processesof catalysts leading to composites, heterostructures, including semiconductor/semiconductor, polymer/semiconductor, and multiheteronanostructures inorder to improve the performance by enhancing the light absorption andefficient charge separation and stability against photocorrosion. Various pho-tochemical processes like solar water splitting, hydrogen generation reaction,photoreduction of CO2, organic transformation, dye degradation, and photo-catalytic detoxification of organic pollutants have been discussed. Moreover,a deeper understanding of the principles underlying electronic behavior ofsize and shape tunable catalytic nanostructures and the evaluation of theireffectiveness as well as perspectives in solar light harvesting has been included.

In the past few years, much progress has been made in the design and synthesisof materials based on semiconductor metal oxides, polymeric carbon nitride, and


586 21 Conclusions and Future Work

plasmonic photocatalysts to achieve efficient light harvesting capacity. Despitesignificant efforts, it has not been possible to design a single material that hassufficient efficiency, stability, and low cost. To integrate the desired character-istics into a single component, heterogeneous photocatalysts are designed withmultiple functional components which could combine the advantages of differ-ent components to overcome the drawbacks of single component photocatalysts.Oxide-based semiconductors, for example, TiO2, are marked as the benchmarkphotoresponsive materials but are limited with wide bandgap and their low effi-ciency in charge separation. Therefore, a remarkable effort has been made tosensitize oxide-based semiconductors in the visible spectral range via dopingor surface tuning. A series of novel catalysts such as black hydrogenated TiO2,plasmonic photocatalysts, and other metal oxides/oxynitride with high photocat-alytic activity under solar light have been developed. Moreover, synergistic andcooperative interactions among different functionalities in nanoheterojunctionor nanohybrids open an alternative strategy for designing molecular materialsfor photocatalysis and artificial photosynthesis. Compared to single-phase pho-tocatalysts, the Z-scheme systems are composed of two or more semiconductorswhich promote the spatial charge separation and reduction and oxidation reac-tion happen at two different reaction sites.

Dye-sensitized heterogeneous photocatalytic processes are potential in manyaspects, but the stability of the functional dyes under aerobic conditions can bean important concern. For example, dye-modified TiO2 photocatalysts rapidlylose their activity with repeated usage or gradually deactivated even in the darkand stability in aqueous solution which is a critical requirement for most appli-cations of visible-light-driven photocatalysts using water as solvent. Concerningthe stability and regeneration issue of catalyst, semiconductor-glass nanocom-posites are used as novel photocatalyst to improve the stability and reusabilityof photocatalyst and get uninterrupted H2 production from photocatalytic H2Sas well as water splitting. Another important issue regarding the light-drivenconversion of greenhouse gas CO2 to value-added products such as formicacid, formaldehyde, methanol and methane, and mitigating CO2 pollution hasbeen focused recently. Various heterogeneous photocatalysts, metal oxides,sulfides, phosphides, and oxynitrides have been tested but all heterogeneoussemiconductor catalysts display inadequate catalytic activity and selectivity forphotocatalytic CO2 reduction. Further advancement in the area requires robustphotocatalysts that competitor inorganic complexes and semiconductor oxidesin terms of light absorption can thus exploit their high and varied reactivityfor activating chemical processes in the literature. In particular, carbon-basedmaterials such as graphene-based derivatives, polymeric graphitic carbon, andconjugated polymer have been considered as new generation visible light activephotocatalysts for H2 generation and CO2 reduction. These novel materials couldbe useful for other different fields, such as in solar cells, self-cleaning surface, andother photoresponsive materials and devices. Further structure modification,morphology control, and bandgap engineering are suggested to enhance thecharge transfer efficiency and improve the photocatalytic performance in orderto use these polymeric materials as efficient solar conversion materials. Pho-tocatalytic water splitting directly converts solar energy into storable chemical

21 Conclusions and Future Work 587

energy and can be used in large-scale solar energy utilization and clean energyproduction. Interestingly, conjugated polymer-based photocatalysts are of sig-nificant interest in the field of photocatalytic water splitting as the observed smallbandgaps lead to activity in the visible range. Recently, polymeric graphitic car-bon nitride (commonly known as g-C3N4), metal organic frameworks (MOFs),covalent organic frameworks (COFs), and conjugated microporous polymers(CMPs) constitute new generation photocatalysts which demonstrate promisein visible-light-driven photoreactions such as pollutant degradation and H2generation. In this regard, graphene-supported semiconductor photocatalystsor CMPs have been found to exhibit significant photocatalytic activities due toenhanced adsorption ability, extended absorption range of light, formation ofheterojunctions, and improved hydrophilicity and dispersity. This developmenthas laid an excellent foundation for future work in this area. Notably, theefficiency of visible-light-induced selective organic transformations could beamplified by the combination of heterogeneous photocatalysts with molecularphotocatalysts, which may provide understanding of the photoinduced interfa-cial electron transfer processes. Particularly, metal-free organic photocatalystsoffer novel pathway to achieve environmentally benign organic transformationsand are expanded to utilize solar energy in organic synthesis, chemical industry,and pharmaceutical industry. Deeper understanding of the mechanism of thephotoredox processes revealed development of robust and efficient photoredoxcatalysts which are essential requirement for future directions of visible-lightphotoredox catalysis.

Upon excitation with suitable wavelength of light, reactive oxygen species(ROS) such as O2

∙−, ∙OH, H2O2, and ∙HO2 are generated from semiconductorcatalysts, which are able to degrade organic pollutants. Additionally, fundamen-tal mechanisms of heterogeneous photocatalysis, including thermodynamic andkinetics requirements, role of oxidizing species are systematically summarized.The computational modeling of photocatalysis based on reactive dynamicand nonadiabatic molecular dynamic methodologies has been discussed tounderstand quantum effects on various stages of photocatalysis and to measureenergy and charge carrier transfer timescales, reaction selectivity, carrier recom-bination, energy relaxation pathways, and solar energy conversion efficiency inphotocatalytic systems. The improvement in new technologies needs collabo-ration with a strong theoretical background for a better understanding of thephotocatalysis mechanism, in order to come up with a low-cost and environmen-tally friendly photochemical process. Furthermore, the design and fabrication ofvarious reactors based on larger-scale field and pilot-scale studies are consideredfor the purification of water with nanoscale metal oxide photocatalysts undersolar illumination. However, one of the major obstacles is the environmentalimpact of nanomaterials that are released from the photocatalytic reactor systemand limit the large-scale solar photocatalytic applications for water purification.

It may be concluded that hybrid nanostructures containing diverse function-alities and active materials will be assembled together to harvest solar light andwater splitting or carbon dioxide reduction with energy input from sunlight.Designing heterostructures with the appropriate morphological orientation andband position of the semiconductor and the polymer or another semiconductor

588 21 Conclusions and Future Work

unit is still challenging. Moreover, various useful experimental techniques, par-ticularly ultrafast spectroscopy have been developed to study the photogeneratedcharge transfer properties of semiconductor which reveals charge separation isthe key factors influencing photocatalytic performance. In spite of considerableprogress, there are many challenges and opportunities to rationally design highlyefficient visible-light-active photocatalysts toward various applications. Theamalgamation of the interface, composite, and tunable structure (morphologyand surface area control) may show potential for designing the cheap and highlyefficient hybrid photocatalysts as the next-generation photocatalytic systems indifferent application fields. In this regard, future research efforts should focuson the underlying mechanisms for enhanced photocatalytic activity of photocat-alysts through combing the theoretical calculation and experimental evidence.In this regard, various in situ techniques such as Raman, STM, and synchrotronradiation techniques are highly desirable to obtain a true picture of surfacesand/or interfaces in action, and theoretical materials design and photocatalyticprocess simulations should take into account the surface–interface reactionsof photogenerated charge carriers. However, great challenges need to beresolved before sunlight can be used as a viable source of energy and substantialbreakthroughs can be visualized in the search of a green and renewable energyin future.

589

Index

aab initio molecular dynamics (AIMD)

simulation tools 295, 532ab initio nonadiabatic molecular

dynamics 13action spectrum (AS) 142, 150–152,

459active species

generation at catalyst/water interface486–490

in heterogeneous photocatalysis 494acyl imidazoles

with benzyl bromides 89enantioselective alkylation of 89, 90photoinduced enantioselective

alkylation of 89adiabatic dynamics

quantum tunneling in 535–540and reaction mechanisms 545

adiabatic processin photocatalysis 531–532theory of 532

advanced oxidation processes (AOPs)27

aerobic indole C-3 formylation reaction402–404

aerobic oxidation 12, 463Ag3PO4-glass nanocomposite

179–183air-saturated system

H2O2 in 487opposite effect 489

alcohols to carbonyl derivativesoxidation 413–414

Al-doped zinc oxide (AZO) 200

alizarin red (AR) 80, 266, 403alkyl halides

activation of 77–91C–C bond forming reactions 78photoreduction of 104

alkynesarylphosphine oxides with 3971,2-diketones synthesis from

399–400anatase

band structure 505crystal structure 504–506substitutional doping of 451TiO2 505

anatase-rich titania 451anchor-points reactive potential (APRP)

method 540anions

atomic p levels 454doping with 425hypophosphite (H2PO2

–) 428role of 493–494

apparent quantum efficiency (AQE)10, 141, 142, 151

apparent quantum yield (AQY) 9, 202,238, 241, 347, 369, 431

arenesazolylations of 105electronic effect of substituents 414perfluoroarylation of 396–397polyhalogenated 109

artificial-force-induced reaction (AFIR)531–532


590 Index

arylamidines oxidative annulation405–406

arylboronic acids, oxidativehydroxylation 409–410

aryl halidesactivation 91–108carbon–halogen bond 108–109C–H arylation reactions 99photoredox catalysts and visible-light

92atom transfer radical addition (ATRA)

reactions 84Au–Cu nanoparticles 146–148

bband alignments 193–195, 205, 215,

255, 291, 293, 432, 433, 472band bending

semiconductors 503surface 506

band gap 5, 6, 54, 168, 196energy 4, 504engineering 174, 285–291heterojunctions 468–470modification for visible-light

sensitization 452semiconductor 255

benzimidazole synthesis 411–413benzodithiophene (BDT) 240benzo[b]phosphole oxides synthesis

397–398benzothiazole synthesis 411–413benzyl bromide 88, 89Bethe–Salpeter equation (SBE) 529bimetallic nanoparticles (NPs) 130,

131, 146–155bimodal carbon modification effect

472binding energy 7, 302, 351, 455Bi2S3–glass nanocomposite 178BiVO4 195–196

carbon-based materials 199TiO2 199WO3 197ZnO 197–199

black TiO2core-shell formation 121

designing 118–122enhanced catalytic properties of 120high thermal stability 123nanomaterials 120, 121nanotubes 120photocatalyst 122, 123synthesis strategies for 119

Bohr radius 177, 178, 231, 373Born–Oppenheimer approximation

532Born–Oppenheimer molecular

dynamics (BOMD) 532, 5442-bromoacetophenone 77, 80bulk modification 450

cC3-acylation of indoles 404cadmium sulfide (CdS) 428

CdS-glass nanocomposite 174–178nanospheres 469

calcium chloride (CaCl2) 169cadmium selenide 9capping agents 170, 171carbon–bromine bonds 85, 86, 95, 104carbon–carbon bond formation

reactions 80, 89carbon dioxide (CO2) 283, 421

emissions 53energy crisis and water splitting 14harvest solar light and water splitting

587mineralization of bacteria cells 138photocatalytic conversion of 31photoreduction into CH4 429reduction 30–44

carbon–fluorine bonds 108carbon–halogen bonds

activation of 76, 80, 91arylated products 109photoredox catalytic activation

75–110carbon modified metal oxides 33–34carbon nanotubes (CNTs) 33, 425,

519carbon neutral cycle 6carbon nitrides 332, 334, 336,

348–349, 430–434

Index 591

carbon oxide redox reactions 546–547carbon-TiO2 composites 518–520Car–Parrinello molecular dynamics

(CPMD) method 532cascade radical cyclizations 80catalyst/water interface 486–490C atom tunneling 538, 539CB edge (CBE) 196, 197C–C coupling reaction 10, 542cetyltrimethyl ammonium bromide

(CTAB) 170, 426CFO/WO3 composites 469chalcogenide photocatalysts 170charge carrier 3

of BiVO4 197charge transfer processes/chemical

reactions 264dynamics 13, 130, 156mobility of 7, 150, 156fast recombination of 142formation 5kinetics in heterojunction structure

209, 215in photocatalyst 129photogeneration of 28recombination 134separation 6transfer 6

charge transfermechanisms 437resistance 212

chemical reduction methods 119chemical vapor deposition (CVD) 119,

172chemiluminescence 488C–H functionalization 75–110,

407–408Chini clusters 133Claus process 166clean and renewable energy 329C:N ratio 335CO2 capture, storage (CCS) 28codoping 455–456COFs, see covalent organic frameworks

(COFs)

coherent switching withdecay-of-mixing (CSDM)method 540

colorless pollutants 265–268compound parabolic collectors (CPCs)

310compound parabolic concentrators

(CPCs) 41, 311, 312, 573–574conducting polymers

applications of 233–245hard templates 232heterostructures 242–245nanocomposites 231nanoscale 231nanostructured materials 231–233organic semiconductor 228–230for photocatalytic water splitting

237–242soft templates 232–233synthesis of 231, 233template-free method 233

conduction band (CB) 8, 10, 27, 54,167, 191, 193, 452, 501, 503

conduction band edge (CBE) 285, 289conduction band minimum (CBM)

330, 368, 451conjugated microporous polymers

(CMPs) 10, 238, 351conjugated polymer nanostructures

(CPNs) 6conjugated polymers, see conducting

polymersconsecutive photoinduced electron

transfer (conPET) processes92, 109

constant phase element (CPE) 212constrained DFT (cDFT) 546contact angle 304contact potential difference (CPD) 213CoP/CdS hybrid catalyst 10core-shell metal-insulator nanoparticles

517core/shell nanocrystals 36core(gold)-shell(Ag) structure 463coupled reactions 386coupled semiconductors 518

592 Index

covalent organic frameworks (COFs)10, 331, 350–351, 435

covalent triazine frameworks (CTFs)351, 436

CPCs, see compound parabolicconcentrators (CPCs)

Creutz–Brunschwig–Sutin model 260cross-dehydrogenative

couplingreactions 407of tertiary amines 406–407

Cu2O-based junctions 204–207cyclic voltammetry (CV) method 234

dDDAT molecules 467decarboxylative/decarbonylative

C3-acylation, of indoles404–405

decoherence effects 533, 547degradation

organic pollutants 485oxidative 490–492oxidative photocatalytic 486photocatalytic 492

degree of transformation 263density functional theory (DFT) 13,

287, 332, 372, 529deposited photocatalysts 569–570deposition precipitation method with

urea (DPU) 146detoxification

graphene-TiO2 composites for299–303

metal doped photocatalysts for296–299

photocatalysis applications of303–312

diazo compounds 406–4072,4-dichlorophenol under visible-light

irradiation 453dielectric constant 139diffuse reflectance spectroscopy (DRS)

146, 150, 1551,2-diketones synthesis, alkynes

399–4001, 4-diphenylbutadiyne (DPB) monomer

234

direct band gap 179direct C–H arylation of heteroarenes

398, 399direct photoexcitation 773,6-di(pyridin-2-yl)-1,2,4,5-tetrazine

411–414donor −𝜋 bridge–acceptor (D−𝜋−A)

structures 261doped metal oxides 34–35doping 450, 451, 510

codoping 455–456metal ion 451–453of metal oxides 451nonmetal ion 453–455self-doping 450, 457–458

doping-heterojunction nanostructure472

dual catalytic cycle 78dye, see also organic dyes

aggregation 260photodegradation 263sensitization 513

dye modified TiO2 photocatalysts269–270, 586

dye pollutantsmineralization of 264self-sensitized degradation of

262–265dye radical cations 263, 266dye-SC electronic coupling 261dye sensitized

degradation process 264heterogeneous photocatalytic process

586mechanism of 513mesoporous CN 347photocatalytic hydrogen production

513dye sensitized solar cells (DSSCs) 466

development of 262electron transfer in 465energy conversion 260–262photoelectrochemical 257TiO2 sensitizers for 262

dye sensitization 513dye-TiO2 photocatalytic system 265

Index 593

eelectrochemical anodization method

308electrochemical impedance

spectroscopy (EIS) 202,211–213

electrochemical reductionnegative 29of oxidants 506process 120

electromagnetic energy 130electromagnetic waves 5electron and hole localisation 293–296electron diffraction (ED) pattern 177electron donor-acceptor (EDA) 398electron/hole

acceptors 4separation 14

electron–hole pair 5, 9, 10, 28, 63, 66,136, 152, 153, 193, 212, 213, 227,229, 284, 295, 299–301, 342, 366,422, 436, 449, 455, 456, 461, 462,499, 502, 515, 532, 545

electron–hole recombination 4, 122,133, 138, 139, 146, 206, 208–210,296, 307, 473, 514, 533, 546

electronic paramagnetic resonance(EPR) 266

electronic transitions 174, 457conduction band or from valence

band 287HOMO to LUMO 467nonadiabatic 534, 548in photocatalytic systems 533RPMD methods 534uniform scaling of 533from VB of titania 471

electronic tunneling 536, 537, 548electron injection process 257electron-nuclear back reaction 535,

543, 547electron photogenerated 464electron scavenger 61, 65, 137, 149,

210, 238, 298, 486, 487, 521, 546electron spin resonance (ESR)

spectroscopy 121, 259electron transfer (ET) 460, 461, 463

in composite semiconductor 518in DSSCs 465feasibility of 271reactions 257processes 28, 193, 366

electron trappingenergy evaluation 296NPs 149Pd-based nanoparticles 154process 268, 269rutile (100) surface 295in TiO2 nanocrystals 295

electrospinning 232, 233endergonic reaction 369energy conversion efficiency 192, 204,

238, 349, 587energy-dispersive X-ray (EDX) mapping

146energy-dispersive X-ray spectroscopy

(EDS) 147, 150energy gap 54, 234, 261, 285, 287, 289,

293, 385, 437energy relaxation pathways 13, 587energy transfer 144, 243, 257, 394,

459–464, 473, 529, 532, 535, 542,547, 548

enhanced NIR absorption 119Eosin Y (EY) 80

photocatalysis by 396–401redox potentials of 397

ethylene glycol 120, 426

ffast dye regeneration 270Fe2O3-based junctions 199–200Fe2O3/OEC 207–209Fermi level energy (EF) 29, 55–57, 130,

139, 193, 194, 197, 204, 210, 230,376, 465, 472, 499, 505, 514, 515

Fisher–Tropsch synthesis 54flat-band potential 57, 208, 213, 470fluorescein 80, 400Förster resonance energy transfer

(FRET) 243Fourier transform infra-red (FTIR)

spectroscopy 121

594 Index

ggallium nitride (GaN) 37gas phase and self-cleaning applications

570–571gel combustion method 119Gibbs free energy 62, 166, 167, 191,

366glass–bandgap 174glass transition temperature 174glassy photocatalysts

glass by melt-quench technique172–174

glasses preparation 172–174semiconductor–glass

nanocomposites 171–172gold nanoparticles 130, 132, 138–144graphene

and CNT based nanocomposites345

nitrogen doping in 382TiO2 composite via hydrothermal

route 519two-dimensional graphene sheets

170graphene/semiconductor

nanocomposites (GSNs) 33graphical processing unit (GPU)

computations 535graphitic carbon nitrides (g-C3N4) 8,

10, 238, 331, 332, 334halogens doping 344hard templating 337–339metal doping 341metal oxides/g-CN nanocomposites

344, 345nanosheets 340nitrogen doping 342nonmetal doping 342–344oxygen doping 342phosphorus doping 343photocatalyticwater splitting

331–349precursor-derived 334–336soft templating 339–340sulfur doping 342, 343template-free 340–341templating methods 336

growing string method (GSM) 531,532

GW approximation (GWA) 529

hHaber–Bosch cycle 54halogens doping 344Hamiltonian model 537Hammes-Schiffer group 540heat exchanger 310heavy atom tunneling 548heptazine-based microporous polymers

(HMPs) 349, 351heteroarenes

applications 91C–H arylation of 398–399Eosin Y 105, 107

heteroatom doping 341–344heterogeneous photocatalysis 485

active species in 494–495degradation of phenolic pollutants

492fundamental mechanisms of 587TiO2 255, 283, 485

heterojunctions 468band alignment in 193band-gap 468–470charge carrier kinetics in 209–215Co–Pi/BiVO4/ZnO 198excitation of one component

468–469photoanodes 196semiconductors 436–437

H2 evolution reaction (HER) 10, 62,65, 202–206, 346, 349, 375

high angle annular dark field scanningtransmission electronmicroscopy (HAADF–STEM)146

high concentrating/high temperaturesystems 42–43

high-energy radiation 131high intensity LED-based photoreactor

266high-resolution transmission electron

microscopy (HRTEM) 121,146, 148, 150, 153, 179, 181, 243

Index 595

highest occupied molecular orbital(HOMO) 54, 261

HOMO–LUMO 334, 380homolytic aromatic substitution (HAS)

reaction 81, 85hot electrons 156, 460–462Hu’s reagent 86hybrid nanostructures 471–472hydrogenation, TiO2 118, 287, 289,

512hydrogen dissociation 463hydrogen peroxide (H2O2)

in air-saturated system 487decomposition 489formation 257, 489oxidative degradation of solutes

492–493production 487

hydrogen plasma 118, 119, 121, 123hydrogen sulfide (H2S) splitting

designing assembly 168–170fundamentals of 166–168photocatalyst and reagent system

169–170photocatalyst, role of 167–168standardization 168–169thermodynamics of 166–167

hydroperoxyl radical 27, 404, 487, 494hydroxylation

of arylboronic acids 240, 409–410phenol 487

hydroxyl radicals 492, 493, 506, 507photocatalytic formation 488surface bound 494

3-hydroxyphthalic hydrazide 488

iimidazoheterocycles, thiocyanation of

401iminium cation 80impedance (Z) 211inclined plate collector (IPC) 39,

571–572inclined plate reactor (IPR) 571indirect bandgap 67, 179, 521indium-based oxides 427indoles

aerobic indole C-3 formylationreaction 402–404

coupling of bromopyrroloindolinewith 87

decarboxylative/decarbonylativeC3-acylation of 404–405

inorganic semiconductors 229, 230,245

metal oxides 424–428oxynitrides 429–430sulfides 428–429

intensity-modulated photocurrentspectroscopy (IMPS) 213

intensity-modulated photovoltagespectroscopy (IMVS) 213

iondoping 510implantation 452, 510–513

IPR, see inclined plate reactor (IPR)

kKelvin Probe 213kinetic isotope effect (KIE) 537Kubelka–Munk function 335

lLangmuir–Hinshelwood kinetics 490large curvature tunneling (LCT) 537ligand-to-metal charge transfer (LMCT)

467light absorption 13, 14, 33, 34, 67, 117,

129, 153, 154, 168, 182, 195, 198,200, 239, 260, 287, 291, 301, 340,341, 347, 366, 462, 463, 469, 485,499, 508, 514, 515, 520, 585, 586

light-harvesting nanoheterojunction(LHNH) 242–245

light harvesting units 366–369, 371,378, 431

light-induced electron transfer 453light irradiation sensitizer 368local electromagnetic field (LEMF)

462localized surface plasmon (LSP) 130localized surface plasmon resonance

(LSPR) 5, 136, 138–140,142–144, 156, 459–462

596 Index

lowest unoccupied molecular orbital(LUMO) 54, 203, 245, 258, 261,334, 465, 467, 469

lyotropic liquid crystal (LC) 233

mMarcus rate theory 546medium concentrating/medium

temperature system 40–42melt-quench technique 172–174metal chalcogenides 10, 36–37metal complexes

dyes 270, 460photocatalytic reduction of 132

metal-enhanced fluorescence (MEF)136

metal-free organic photocatalysts 587metal ion

doping 451–453implantation 452

metal nanoparticlesby photodeposition method

130–132by radiolysis method 130–132

metal organic frameworks (MOFs)352, 434–435, 587

metal oxides 424–428, 449band-gap structure 449doping of 451photocatalysts 14semiconductors 567

metal oxides/g-CN nanocomposites344–345

metal surface modification 458–467methanol photooxidation 541–544methylene blue (MB) 122, 301

degradation of 10, 34heteropolyaromatic 31photocatalysis by 409–410photophysical properties of 402

methyl orange (MO) 31, 35, 122, 146,234, 236, 266

microwave irradiation 118, 119MOFs, see metal organic frameworks

(MOFs)multiconfigurational time dependent

Hartree (MCTDH) 534

multielectron transfer processes 256,269

multiwalled carbon nanotubes(MWCNT) 299, 345

nNaBH4 119nanoarchitectures 14, 351, 466nanoclusters 139, 145, 148–150, 155,

289, 295, 464nano-Einstein (nano-ein) 133nanofibers 6, 120, 123, 231–234, 236,

243, 244, 521nanometer scale 3nanoparticles 29

Au–Cu 146–148Au–CuO 148–150of controlled size and shape 131core–shell metal-insulator 517detecting and quantifying 574–575Ni and Au 150–152TiO2 with Ag 136–138TiO2 with Au 138–144TiO2 with Bi clusters 144–145TiO2 with bimetallic 146–155TiO2 with Pd 135–136toxicity 576–577transformation of 575–576

nanorod array electrodes 197nanoscale 5, 14, 228, 231, 245, 295,

568, 569, 574, 577, 587nanoseconds 209, 257, 270, 533, 549nanosheets 10, 340, 378nanosized

crystals 502photocatalysts 171

nanospheres (NSPs) 6, 10, 32, 337,469, 509

nanostructuresconducting polymers 227–246designing black TiO2 118–122effect of size 369–370g-CN photocatalysts 336hybrid 471–472

nanotubes (NTs) 7, 10, 118–121, 123,149, 150, 199, 231–233, 307, 308,

Index 597

345, 430, 457, 466, 507–509, 519,568

nanowire hybrid nanostructures 10NaTaO3, surface modification 464natural organic matter (NOM) 576N–B–TiO2 (red anatase TiO2) 455negative electrochemical reduction 29Nernst’s equation 60N-ethoxy-2-methylpyridinium (EMP)

397, 398nitrates

organic 306photocatalytic reduction of 135

nitrogen doping 342, 454N ,N-diisopropylethylamine (DIPEA)

78, 94noble metals (NM) 11, 459

cocatalysts 235, 521loading 514–518NPs 129, 130, 136, 156, 256, 268on semiconductor 267

nonadiabatic dynamicsof charge carriers 532effects in photocatalysis 530quantum tunneling in 535–540

nonadiabatic effectson metal surfaces 530in O2 adsorption 546in photocatalysis 532–535

nonadiabatic molecular dynamics(NA-MD) 530

nonadiabatic transition state theories(NA-TST) 534

nonadiabatic tunnelingeffects 540theories 540, 548

non concentrating reactor (NCC)39–40

nonconcentrating solar collectors(NCCs) 310

nonmetalsion doping 453–455surface modification 464

nonradiative transition 394normal hydrogen electrode (NHE) 61,

167, 191, 423, 503

nuclear magnetic resonance (NMR)121

nudged elastic band (NEB) 531Nyquist plot 211, 212

oOhm’s Law 211one-electron reduction potentials 495one-step photocatalytic process

371–374optical antenna effect 5optical cut-off wavelength 174optical electron transfer (OET) 260,

467organic compounds

colorless and color 464–467mineralization of 33oxidative decomposition of 460, 468photocatalytic oxidation of 454surface modification 464

organic dyeselectrochemical processes in 394for organic synthesis 393–415pollutants 262photophysical processes in 394photophysical properties of 393visible-light photoredox catalysts

394, 396, 414organic pollutants 492

conducting polymer nanostructuresfor 233–237

solar degradation of 38–44organic reactions

catalyzed by semiconductorphotocatalysts 541–547

photoinduced 538organic semiconductors

carbon nitride and composite430–434

conducting polymers 228–230covalent organic frameworks (COFs)

435donor–acceptor junction 245metal organic frameworks (MOFs)

434–435for photocatalytic water splitting

331–332

598 Index

organic transformationsnanostructures catalytic materials

585photocatalysis for 11–12

organocatalytic cycle 79, 80oxidation

of alcohols to carbonyl derivatives413–414

reaction 503oxidative annulation, of arylamidines

405–406oxidative coupling, of primary amines

414oxidative cross-coupling, of thiols

408–409oxidative degradation, of solutes

490–492oxidative hydroxylation, of arylboronic

acids 409–410oxidative photocatalytic degradation

486oxide-based semiconductors 3, 4, 586oxidized photosensitizer 466oxygen doping 342oxygen evolution catalysts (OECs)

195, 197, 207, 375oxygen evolution reaction (OER) 8, 61,

195, 377oxynitrides 6, 8, 11, 35, 36, 195, 379,

429–430, 437, 438, 469, 471, 586

pparabolic trough collector 40, 41, 309parabolic trough concentrator (PTC)

572–573Pd nanoparticles 135–136perfluoroarylation, of arenes 105, 107,

108, 396–397, 415perovskite 35–36, 44, 65peroxyl radical 492perylenediimide (PDI) 80, 92, 239,

240, 435phenacyl bromides 77–79, 89phenol

degradation of 134hydroxylation 487photodegradation 297

phenothiazine dyes 414phenoxyl radicals 491, 49310-phenylphenothiazine (PTH) 103phosphorus doping 343photoanodes 9, 58, 65–68, 165, 194,

196, 197, 200, 201, 206–208, 210,260, 269

photo-Arbuzov reactions 103photocatalysis

computational approaches to modeladiabatic processes in 531–532

computational approaches to modelnonadiabatic effects in532–535

computational modeling 529, 587on C-modified NaTaO3 464by 3,6-di(pyridin-2-yl)-1,2,4,5-

tetrazine 411efficiency of 502by Eosin Y 396–401heterogeneous 485mechanism of 284, 450, 500by methylene blue 409, 410by phenothiazine dyes 414plasmon-assisted 459plasmonic 459by Rose Bengal (RB) 402slurry-style 569in TiO2/water systems 486

photocatalystsdegradation 8, 122, 134, 135, 145,

227, 236, 244, 262, 264, 453, 486,487, 492, 493, 570, 572

deposited 569–570mono and bimetallic nanoparticles

155–156organic reactions catalyzed by

semiconductor 541–547plasmonic 463semiconductor 502solar photochemical splitting of water

369TiO2 255–271

photocatalytic CO2 reductionactivity by MnCo2O4 427advantages 421inorganic semiconductors 424–430

Index 599

NH2-MIL-101(Fe) 435organic semiconductors 430, 435principles 422–424Ru/CN composite 432semiconductor heterojunctions

436–437photocatalytic degradation

of 2,4-dichlorophenol undervisible-light irradiation 453

photocatalytic detoxification 284, 585photocatalytic hydrogen generation

61, 122, 123, 152, 380, 382photocatalytic oxidation

of organic compounds 454processes 235reaction mechanism of 343terbutylazine 265

photocatalytic reactions 5, 12, 13, 54,60, 65, 136, 183, 415, 469, 470,506–507, 510, 513, 532, 534, 538,540, 548

photocatalytic reductionof CO2 5, 6, 14, 422–425, 428–431,

433, 437, 438of Cr(VI) 7, 268in GO 300of metal complexes 132of nitrate 135

photocatalytic systemefficiency of 490electronic transitions in 533

photocatalytic water splittingconducting polymers 237, 240crystalline carbon nitrides 347–348development of 54–56effect of light 58–62electrolysis and photoelectrolysis

63–65organic semiconductors for

331–332process 329semiconductors, electrochemistry of

56–58soft templating, g-CN 339–340solar photocatalysts for 65–66structural modification with organic

groups 345–347

sulfur doping 342, 343sunlight conversion and storage

62–63template-free 340–341templating methods, g-CN 336visible-light absorbing metal oxides

66–67photocathodes 65, 68, 204–206, 210photochemical process 12, 39, 191,

585, 587photoconductivity 130, 132, 151photoconversion efficiency 65, 242,

499, 508, 509photocorrosion 30, 32, 36, 54, 60, 65,

149, 168, 170, 174, 175, 179, 183,192, 201, 203, 207, 330, 331

photocurrent efficiency 509photodegradation 4, 31, 33, 35, 36,

122, 130, 146, 147, 234, 236, 245,262–266, 297, 299–301, 511

photodissociation 463, 540, 541, 543,547–549

photoelectrochemical (PEC) 9, 54, 66,192, 215, 365

electrode 123water reduction 65

photoelectrochemical impedancespectroscopy (PEIS) 213

photoelectrodes 9, 62, 63, 65, 66, 68,200, 207, 213, 214, 500

photogenerated charge mechanism,(TiO2) 500

photogenerated electrons 5, 7–9, 28,29, 34, 59, 62, 65, 66, 132, 156,183, 202, 206, 214, 229, 237, 284,285, 299, 300, 334, 384, 386, 424,433, 449, 453, 455, 459, 464, 469,471, 472, 499, 514

photoinduced charge transfer process243, 457

photoinduced electron transfer (PET)92, 258, 260, 393, 396

photoinduced oxidation 231, 541, 542photoluminescence spectrometry (PL)

146, 155, 211photooxidation, methanol 541–544photoreactors 43, 309

600 Index

photoreactors (contd.)for solar degradation 38–44temperature 169

photoredox catalysis, see visible-light,photoredox catalysis

photoredox catalyticactivation of carbon–halogen bonds

75–110approach 76

photoredox transformations 80, 91,395

photosensitization 269, 270, 453photosensitizers 9, 132, 261, 262, 266,

269, 368, 382, 402, 403, 409, 460,465, 466, 519

photosynthesis, Z-scheme of 255photosystem II (PSII) 193, 376photovoltaic cells 53phthalocyanines (Pc) 8, 259, 261, 267,

270, 347plasma laser ablation techniques 118plasmon-assisted catalysis 462plasmon-assisted photocatalysis 459plasmon-based photocatalysts 6plasmon energy transfer (PRET) 461plasmonic heating 459, 462, 463plasmonic light harvesting 517plasmonic nanostructures 4, 515plasmonic photocatalysis 5, 156, 459,

462, 464plasmonic photosensitizer 460platinized SrTiO3 65p–n heterojunctions 6, 255poly(diphenylbutadiyne) (PDPB) 6,

234, 243poly(3, 4 ethylene dioxythiophene)

(PEDOT) 6, 228, 233, 235, 347poly(3-hexylthiophene) (P3HT) 233,

237, 243polychromatic and monochromatic light

150polyethylene glycol (PEG) 426polymer-based composites 4polymeric graphitic carbon nitride 4,

12, 331, 431, 587polyvinylpyrrolidone (PVP) 170

porous conjugated polymers (PCPs)239, 350

potential energy surfaces (PES) 531powder photocatalysts 38, 170–171powder X-ray diffraction (PXRD) 121primary amines, oxidative coupling

414PROPHIS reactor 41, 309, 310proton-coupled electron transfer

(PCET) reactions 540proton transfer 531, 537, 538, 545proton tunneling 537, 545PTC, see parabolic trough concentrator

(PTC)pulse radiolysis 488pyrrole heterocycle 81, 85, 99PYXAID program 534, 535

qQDs, see quantum dots (QDs)quantized Hamiltonian dynamics

(QHD) 537quantum-classical dynamics methods

547quantum dots (QDs) 10, 171, 176, 262,

470quantum effects 13quantum tunneling

in adiabatic dynamics 535–540in nonadiabatic dynamics 535–540

quantum yields 3, 9, 10, 30, 34, 110,129, 141, 191, 202, 203, 207, 227,238, 241, 331, 347, 350, 369, 402,428, 431, 432, 449, 490, 529

rradiative transition 394radical trifluoromethylation 410–411Raman spectroscopy 121Randle circuit 212reactive dioxygen species 402reactive oxygen species (ROS) 27, 257,

262, 283, 451, 576, 587receiver–reactor tubes 309reduced graphene oxide (rGO) 199,

299, 303, 472, 519resonant energy transfer (RET) 462

Index 601

rhodamine B (RB) 8, 35, 36, 80, 134,135, 145, 206, 237, 245, 262–264,400, 405, 407

rhodamine 6G (Rh-6G) 94–96,99–103, 296, 298

ring polymer molecular dynamics(RPMD) 534, 536, 538

Rose Bengal (RB) 8, 259, 400, 402–404Ru(bpy)3

2+ 77, 79, 80, 82, 85, 368, 369,382, 409, 410

ruthenium-based dyes 368rutile 30, 31, 35, 65, 117–119, 123, 129,

133, 139, 152, 269, 295–297, 301,308, 453, 458, 504–507, 521,541–543, 546

sSakata–Hashimoto–Hiramoto model

260scanning transmission electron

microscopy (STEM) 155Schottky barrier 139, 194, 269S–C junctions 194self-cleaning

glasses, principle of 305materials 303–307self-doping 450, 457–458semiconductor liquid junction (SCLJ)

193semiconductor photocatalysts 4

bandgap of 27electron–hole pair in 422organic reactions catalyzed by

541–547photoexcitation of 28

semiconductors 449band bending 503coupled 518doping 450electron transfer mechanism in

composite 518heterojunctions 436, 450inorganic 424–430organic 430oxides-based 586photocatalysts 502photo absorption 132

semiconductor surfacescarbon oxide redox reactions on

546–547methanol photooxidation on

541–544water splitting reactions on

544–546sensitization dye 513SHARC 534, 535silica doped TiO2 296silver nanoparticles 130, 132, 136–138,

514silver phosphate 10, 179, 521single electron transfer (SET) 75–78,

80, 82, 84–86, 89, 91, 94, 99, 395slurry-style photocatalysis 569small curvature tunneling (SCT) 537sol–gel

method 9process 172, 452technique 119

solar degradation, photoreactors for38–44

solar emission spectrum 44solar energy

chemical energy 269fuels 365, 500conversion 3, 4, 11, 62, 63, 65, 68,

227, 238, 349, 352, 366, 435, 449,470, 529, 530, 585, 587

harvesting 231overall efficiency (𝜂c) of 366renewable energy 585

solar illumination 568, 571–573, 587solar irradiation 10, 38, 191, 206, 299,

306, 365, 372solar light induced photocatalysis 4,

228, 246solar photocatalytic

applications 309, 568, 587CPC plant 311reactor/collector 38reactor designs 13, 567–577

solar photoreactors 29, 30, 38, 41, 312solar reactors

parameters of 43–44photocatalytic 40

602 Index

solar reactors (contd.)thermal efficiency of 39

solar spectrum 4, 43, 61, 65, 67, 117,129, 191, 227, 271, 284, 287–289,313, 330, 449, 499, 504, 514, 568

solar-to-chemical energy conversion365

solar-to-hydrogen (STH)conversion 123, 192, 196efficient 61

solar tracking devices 310SOLARDETOX 311SOLARIS reactor 309, 310solvothermal-assisted method 122solvothermal process 433standard hydrogen electrode (SHE)

291, 345stoichiometric ratio 335, 371, 372substitutional doping 285, 451sulphides 331sulfur doping 342–343superoxide radical anion 400,

402–404, 407, 486, 491, 494, 499surface band bending 505, 506, 515surface bound hydroxyl radicals 488,

489, 491, 493, 494surface-enhanced Raman scattering

(SERS) 136surface modification 450, 458

heterojunctions 472metals 458NaTaO3 464nonmetals 464organic compounds 464

surface-modified semiconductor 450surface photovoltage (SPV)

measurement 13spectroscopy 199, 211, 213–215

surface plasmonic effect 54surface plasmon resonance (SPR) 12,

143, 182, 499, 515

ttemplate-free method 233, 340terminal oxidant 402tertiary amines𝛼-functionalization of 408

cross-dehydrogenative coupling of406

tetrakis(hydroxymethyl) phosphoniumchloride (THPC) 146

tetramethylethylenediamine (TMEDA)402

thermodynamic energy 63thermodynamic favourability of

reactions 423–424thermodynamic, H2S splitting

166–167thiocyanation of imidazoheterocycles

401thiols, oxidative cross-coupling

408–409time-dependent density functional

theory (TD-DFT) 13, 529time dependent wave packet (TDWP)

approach 534time-resolved microwave conductivity

(TRMC) method 13, 130,132–138, 145, 146, 150, 266, 464

titania, see also titanium dioxide (TiO2)band structure of 472modification, urea-induced 467with monometallic nanoparticles

133–138with Pt clusters 133–135sensitization 458

titaniananobelts (TNBs) 469titanium butoxide 119, 120titanium dioxide (TiO2) 8–11, 499, 568

alternatives to 520–521anatase stability and photocatalytic

activity 296anion doped 512antibacterial action of 307band gap of 454bactericidal 307–308Bi clusters 144, 145bimetallic nanoparticles 146–155carbon-TiO2 composites 518–520catalytic effect 489co-doping 290coupled semiconductors 518coupling of 255crystal structure 505

Index 603

from DFT simulations 287doping and surface modification of

285–291DSSCs, sensitizers for 262dye adsorption on 263dye-modified 256–259dye sensitization 513electronic structure 513electron transfer (ET) mechanism

268, 514energy gap 293energy levels alignment 293gold nanoparticles 138–144high temperature stable anatase

296ion doping and ion implantation

510light absorption of silver supported in

515in limitation 129materials for energy conversion

260–262metal-free organic dye 267mono-and bimetallic nanoparticles

155, 156nanocluster 289nanomaterials 118, 119nanoparticles 516nanowire system 291nobel metals loading 514–518particles 488Pd nanoparticles 135, 136photoanode 269photocatalyst system 312photocatalytic activity of 129, 456photocatalytic mechanisms in

TiO2-CNT composites 520photocatalytic reactions 506–507photo-generated charge mechanism

500physical architectures 507–509physical structure 509rare earth metal ions, effect of

297–299silver nanoparticles 136–138structure 504transition metals, effect of 296–297

undoped metal oxides different from30–33

valence band (VB) of 453for visible-light assisted degradation

265–268visible-light photocatalysis 509water systems, photocatalysis in 486

Togni reagents 86total organic carbon (TOC) 237, 573total sun energy 3, 372toxicity, of nanoparticles 576trajectory surface hopping (TSH) 533transient absorption spectroscopy

(TAS) 7, 206, 209–211, 460transition state theories (TST) 531,

534transmission electron microscopy

(TEM) 177, 243, 343triazine and heptazine based organic

polymers 349–350tri-doped titania 455triethylamine (TEA) 105, 237, 240,

396, 428trifluoromethylation

of aromatic heterocycles 84, 85radical 410–411

Tris 491, 493tunneling-controlled reaction 537turnover frequency (TOF) 369, 376turnover number (TON)/turnover

frequency (TOF) 369two-step photoexcitation process

374–376

uultrasonication 118, 120ultrathin Bi2WO6 nanoplates 425ultraviolet (UV)

irradiation 3, 129, 134, 135, 139,191, 203, 236, 258, 451

light 330photocatalysis 284radiation 165

undoped TiO2 30–32urea-induced titania modification

467

604 Index

vvalence band (VB) 8, 10, 27, 54, 121,

142, 165, 167, 191, 193, 235, 256,257, 283, 287, 289, 291, 293, 329,332, 342–344, 346, 423, 451, 453,456, 501, 503

valence band maximum (VBM) 330,342, 366, 368, 380, 451–454

vanadium 67, 298, 472variational TST (VTST) 538VB edge (VBE) 196visible-light

absorption in metal oxides 66active photocatalytic materials 4active photocatalytic water splitting

329, 352alcohols, induced oxidation of 11assisted degradation of colorless

pollutants 258band gap modifications for 452benzene, induced hydroxylation of

12cross-dehydrogenative coupling 406degradation processes 227development of 11by dye photosensitization 269electron donating compounds 266environmental protection,

photocatalysis for 4–8Eosin Y 105general mechanism 396harvesting capability 4harvesting efficiencies 6heterojunction for 4high surface area anatase 11illumination 199induced organic synthesis 12induced photocatalysts 6induced photocatalytic activity 5, 7irradiation 9, 10, 61, 91, 139, 237,

258mechanistic studies of 13–14organic dye as 396, 414photocatalyst/OEC junctions 207,

209photoactivity of small metal clusters

133

photoredox catalysis 393range 8responsive junctions 195–207of TiO2 materials 4wavelengths of 94

wwastewater

detoxification 228, 308–312pollutants 5, 8, 262purification 4treatment 29, 38, 303, 310, 311

water oxidationand CBE 291–293oxygen 376–380valence 291–293

water oxidizing complex (WOC) 376water purification

applications 571heterogeneous photocatalysis 485photocatalysts for 33

water reductionC3N4 380–382multicomponent heterostructures

383–386semiconductors 382–383

water splitting 8–11band gap 284fundamentals of 366–367one-step photocatalytic process

371–373photocatalysis with oxygen evolution

500reactions on semiconductor surfaces

544–546thermodynamic requirements of

368Z-scheme photocatalysis/two-step

photoexcitation process374–376

wavefunction (WF)-based methods529

Wentzel–Kramers–Brillouin (WKB)approximation 537

WO3-based junctions 201–202WO3/OEC 208, 209

Index 605

xX-ray fluorescence spectrometry (XRF)

155X-ray photoelectron spectroscopy (XPS)

121, 146, 150, 266, 335X-ray powder diffraction (XRD) 153,

155, 175, 334

zzero-point energy 13, 534, 538, 545zinc gallium oxynitride (ZnGaNO)

nanorods 429Z-scheme photocatalysis 11, 374, 376,

470–471

Visible-Light-Active Photocatalysis: Nanostructured Catalyst Design, Mechanisms, and Applications

Documents

Transcript of Visible-Light-Active Photocatalysis: Nanostructured Catalyst Design, Mechanisms, and Applications