FUEL CELL SCIENCE
Wiley Series on Electrocatalysis and Electrochemistry
Andrzej Wieckowski, Series Editor
Fuel Cell Catalysis: A Surface Science Approach, Marc T. M. Koper
Electrochemistry of Functional Supramolecular Systems, Margherita Venturi, Paola Ceroni,
and Alberto Credi
Catalysis in Electrochemistry: From Fundamentals to Strategies for Fuel Cell Development,
Elizabeth Santos and Wolfgang Schmickler
Fuel Cell Science: Theory, Fundamentals, and Biocatalysis, Andrzej Wieckowski and
Jens Nørskov
FUEL CELL SCIENCETHEORY, FUNDAMENTALS,AND BIOCATALYSIS
Edited by
Andrzej WieckowskiJens K. Nørskov
Copyright � 2010 by John Wiley & Sons, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey
Published simultaneously in Canada
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form
or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as
permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior
written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to
the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400,
fax 978-750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission
should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken,
NJ 07030, 201-748-6011, fax 201-748-6008, or online at http://www.wiley.com/go/permission.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts
in preparing this book, they make no representations or warranties with respect to the accuracy or
completeness of the contents of this book and specifically disclaim any implied warranties of
merchantability or fitness for a particular purpose. No warranty may be created or extended by sales
representatives or written sales materials. The advice and strategies contained herein may not be
suitable for your situation. You should consult with a professional where appropriate. Neither the
publisher nor author shall be liable for any loss of profit or any other commercial damages, including
but not limited to special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact
our Customer Care Department within the United States at 877-762-2974, outside the United States
at 317-572-3993 or fax 317- 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print
may not be available in electronic formats. For more information about Wiley products,
visit our web site at www.wiley.com.
Library of Congress Cataloging in-Publication Data:
Wieckowski, Andrzej
Fuel cell science : theory, fundamentals, and biocatalysis / edited by Andrzej Wieckowski
and Jens Nørskov.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-470-41029-5 (cloth)
Printed in the United States of America
10 9 8 7 6 5 4 3 2 1
CONTENTS
Foreword vii
Preface xi
Preface to the Wiley Series on Electrocatalysis and Electrochemistry xiii
Contributors xv
1. Hydrogen Reactions on Nanostructured Surfaces 1
Holger Wolfschmidt, Odysseas Paschos, and Ulrich Stimming
2. Comparison of Electrocatalysis and Bioelectrocatalysis
of Hydrogen and Oxygen Redox Reactions 71
Marc T. M. Koper and Hendrik A. Heering
3. Design of Palladium-Based Alloy Electrocatalysts
for Hydrogen Oxidation Reaction in Fuel Cells 111
Sung Jong Yoo and Yung-Eun Sung
4. Mechanism of an Enhanced Oxygen Reduction Reactionat Platinum-Based Electrocatalysts: Identification
and Quantification of Oxygen Species Adsorbed on
Electrodes by X-Ray Photoelectron Spectroscopy 147
Mitsuru Wakisaka, Hiroyuki Uchida, and Masahiro Watanabe
5. Biocathodes for Dioxygen Reduction in Biofuel Cells 169
Renata Bilewicz and Marcin Opallo
6. Platinum Monolayer Electrocatalysts: Improving
Structure and Activity 215
Kotaro Sasaki, Miomir B. Vukmirovic, Jia X. Wang, and Radoslav R. Adzic
7. The Importance of Enzymes: Benchmarks for Electrocatalysts 237
Fraser A. Armstrong
8. Approach to Microbial Fuel Cells and Their Applications 257
Juan Pablo Busalmen, Abraham Esteve-Nunez , and Juan Miguel Feliu
9. Half-Cell Investigations of Cathode Catalysts for PEM Fuel Cells:
From Model Systems to High-Surface-Area Catalysts 283
Matthias Arenz and Nenad M. Markovic
v
10. Nanoscale Phenomena in Catalyst Layers for PEM Fuel Cells:
From Fundamental Physics to Benign Design 317
Karen Chan, Ata Roudgar, Liya Wang, and Michael Eikerling
11. Fuel Cells with Neat Proton-Conducting Salt Electrolytes 371
Dominic Gervasio
12. Vibrational Spectroscopy for the Characterizationof PEM Fuel Cell Membrane Materials 395
Carol Korzeniewski
13. Ab Initio Electrochemical Properties of Electrode Surfaces 415
Ismaila Dabo, Yanli Li, Nic�ephore Bonnet, and Nicola Marzari
14. Electronic Structure and Reactivity of Transition Metal Complexes 433
Heather J. Kulik and Nicola Marzari
15. Quantitative Description of Electron Transfer Reactions 457
Patrick H.-L. Sit, Agostino Migliore, Michael L. Klein, and Nicola Marzari
16. Understanding Electrocatalysts for Low-Temperature Fuel Cells 489
Peter Ferrin, Manos Mavrikakis, Jan Rossmeisl, and Jens K. Nørskov
17. Operando XAS Techniques: Past, Present, and Future 511
Christina Roth and David E. Ramaker
18. Operando X-Ray Absorption Spectroscopy of Polymer
Electrolyte Fuel Cells 545
Eugene S. Smotkin and Carlo U. Segre
19. New Concepts in the Chemistry and Engineering
of Low-Temperature Fuel Cells 565
Fikile R. Brushett, Paul. J. A Kenis, and Andrzej Wieckowski
Index 611
vi CONTENTS
FOREWORD
IS THERE A COMMON “ACTIVITY YARDSTICK” THAT APPLIESTO ALL FUEL CELL ELECTROCATALYSTS?
Thinkingwhat should be themessage in the foreword to a book that covers extensively
awide frontier of fuel cell catalysis work, a tempting, albeit somewhat risky, idea kept
coming up in my mind: Is it possible to define a common “activity yardstick” that
applies to a large number of, if not to all, fuel cell electrocatalysts? Is it possible to
make such a generalization when considering the wide variety of catalytic surfaces
studied and practiced in low-temperature fuel cells?
When examining the relevant literature, it appears that themore recent searches for
active metal electrocatalysts and for active molecular electrocatalysts have had
somewhat different priorities. In the case of metal electrocatalysts, the focus has
been on tailoring the electronic properties of metal alloy surfaces to achieve an
optimized bond strength between the metal surface and the relevant adsorbed
intermediates [1]. Such studies have been supported by density functional theory
(DFT) calculations, yielding the energies of the bonds between catalytically active
surfaces and the likely reaction intermediates [2]. In all such studies, the assumption
has been that a complete description of the electrocatalytic process requires consid-
eration of a reactant molecule and a metal surface in contact with water, or aqueous
electrolyte. The electro element of electrocatalysis has been covered all along by
assuming that a change in the interfacial potential difference has an effect on, and only
on, the activation energy of any reaction step involving electron transfer. Accordingly,
the typical rate expression for an electrocatalytic process takes the formof a product of
a preexponential term and a two-component exponential term, with the rate depen-
dence on the electrode potentialE fully covered in the exponential term. For a cathodic
process within the so-called Tafel regime, the rate expression takes the following
general form
JðEÞ ¼ Fk0Acat* Cg
rexpf�DH*act=RTg expf�ðE�E� cell processÞ=bg ð1Þ
where F is the Faraday constant, k0 is a frequency factor, Acat* is the overall catalyst
surface area per unit electrode cross-sectional area, Cr is the concentration of the
reactantmolecule at the electrode surface, g is the reaction orderDHact* is the chemical
component of the activation energy, and b is the so-called Tafel slope. In the case of
molecular electrocatalysts, the more recent achievements in preparation of highly
vii
oxygen reduction reaction (ORR)-active, carbon-supported iron complexes [3],
resulted from efforts to maximize the overall surface density of effective redox
centers, N*. Lefevre et al. [3] showed that an effective center formed when the iron
complex was located on a specific, pretailored carbon surface site. The mechanism of
electrocatalytic processes taking place at such active surface sites is described in terms
of redox mediation, where the electrocatalytic activity at a potential E is expected to
involve a fraction of N*, Nactive (E), defined by
NactiveðEÞ ¼ N*f ðE�E�surface redoxÞ ð2Þ
For example, in the specific case of ORR catalyzed by a molecular iron complex, a
plausible mechanism of ORR at the active complex of iron, X–Fe(II), where X is a
surface anchor site and the iron complex is in its reduced form, has been described [3]
with a first step involving bonding of dioxygen to the active form of the iron complex,
X–Fe(II), assisted by electron transfer from the Fe(II) center:
O2 þX--FeðIIÞ ¼ X--FeðIIIÞ--O--Oþ e ð3aÞ
This step is followed by the completion of the 4e reduction process with regeneration
of the active form of the redox system, written in simplified form as follows:
X--FeðIIIÞ--O--Oþ eþ 3eþ 4Hþ ¼ X--FeðIIÞþ 2H2O ð3bÞ
This mechanism implies that only at a cathode potential sufficiently negative to
generate a significant population of the reduced form of the surface redox couple,
X–Fe(II), can the rate of the process in Equations (3a) and (3b) rise to a measurable
level. In an ideal case where the steady-state population of X–Fe(II) depends on
potential according to a simple Nernst equation, the number of active sites at an
electrode potential E will be given, for a cathodic process, as
NactiveðEÞ ¼ N*½1=ðZþ 1Þ� ð4Þ
whereZ ¼ expfðF=RTÞ ðE�E�surface redoxÞg. Inserting inEquation (1) this dependence
of active-site population on electrode potential, the rate expression will take the form
JðEÞ ¼ Fk0Acat* f ðE�E�
surface redoxÞCgr expf�DHact
* =RTg expf�ðE�E�cell processÞ=bg
ð5Þwhere in the simplest case, f ðE�E�
surface redoxÞ ¼ 1=ðZ þ 1Þ:The significant difference between Equations (5) and (1) is the appearance in (5) of
two sources of rate dependence on electrode potential, associated with two different
values ofE�. One is the dependence of the activation energy at an active surface site onthe overpotential, E�E�
cell process, and the second is a dependence of active-site
population on E�E�surface redox. The former appears in the exponential term of the
rate expression, whereas the latter appears in the preexponential term [4].
viii FOREWORD
The tacit assumption behind the use of the simpler expression [Eq. (1)] for
processes at metal surfaces is that availability of active sites on metal surfaces does
not depend on the electrode potential. This assumptionmisses, however, a key element
of electrocatalysis at metal surfaces [4]. For example, examination of the value of
E�M=M;ox for metal and metal alloy electrocatalysts of high ORR activity, reveals that
“ignition” of theORRprocess is tied consistentlywith the onset of cathodic generation
of some minimum surface population (e.g., 1%) of free metal sites on a surface that is
fully covered under open-circuit conditions by oxygen species that block ORR.
Recognizing that such change in surface composition is required for the onset of the
process, one can describe the ORR process at Pt in terms of surface redox mediation,
involving the Pt/PtOx surface redox system [4]. ORR ignition requires reduction of a
Pt surface oxide, or hydroxide species, for example, according to
2Pt�OHsurface þ 2Hþ þ 2e ¼ 2Ptsurface þ 2H2O ð6aÞ
followednextby reactionofO2atPt andwithmetal sites that becomeavailablebeyond
a threshold potential determined by E�M=MeOH according to
2Ptsurface þO2 þ 2eþ 2Hþ ¼ 2Pt�OHsurface ð6bÞ
Continuous repetition of (6a) þ (6b) sustains a steady-state rate of a 4e ORR process,
taking place at the active (metal) surface sites, with the active site continuously
regenerated beyond a threshold potential determined by E�M=MeOH.
Mediation by a surface redox system is apparently a common feature of a wide
variety of electrocatalysts, whether molecular or metallic, and this insight can lead
to an attempted definition of a “general key to active electrocatalysts.” From
Equation (5), an optimum value of E�surface redox will maximize the product of the
preexponential and exponential-terms at an electrode potential of technical interest,
that is, at a low overpotential-versus-E�cell process. Consequently,E
�surface redox must not
be too far from E�cell process, to electroactivate the mediating surface system and
thereby ignite the faradaic process at a low overpotential. However, too small a
difference between the two standard potentials will mean a small free-energy drive
for the reaction of the reactant molecule with the active form of the surface redox
system [e.g., reaction (6b)], because the standard free-energy change in that reaction
is FðE�cell process�E�
surface redoxÞ. The activation energy of a process like (6b) is
expected to be lower, the higher the ðE�cell process�E�
surface redoxÞ difference and,
conversely, very close proximity of the two standard potentials will likely result
in excessiveDHact* . We are looking, therefore, at a need to optimize the gap between
E�surface redox and E�
cell process, to satisfy the conflicting requirements of a low over-
potantial for electrode surface activation and a sufficient free-energy drive for the
reaction of the reactant molecule with the active surface site.
On the basis of experimental results reported to date, the optimum value of
ðE�cell process�E�
surface redoxÞ for requiring electrocatalytic processes in low-temperature
fuel cells is in the range of 300–400mV. In the case of ORR at unalloyed Pt, for
example, (E�O2=H2O
�E�Pt=PtOx) is near 400mV and can be lowered further by about
FOREWORD ix
100mV by alloying [1], resulting in enhanced ORR activity. The rate enhancement
derived in this case from lowering of ðE�cell process�E�
surface redoxÞ indicates that the
beneficial effect of Pt alloying originates from lowering of the ignition overpotential,
resulting in an increase in the value of the preexponential term in Equation (5) at some
given cathode potential E. A metal surface where (E�O2=H2O
�E�M=MOx) is either
significantly smaller than 300mVor significantly higher than 400mVexhibits ORR
activity that is lower than that of Pt because it is associated with either high DHact* (in
the former case), or an excessive ignition overpotential (in the latter case). An
aggressive goal for the future would be to minimize further the difference between
the twoE� values. A surface redox systemwithE�surface redox removed less than 300mV
from E�cell process, while, at the same time, securing a low DHact
* for reaction of the
reactant molecule with the active surface site, will enable the onset of significant
current at overpotentials lower than those demonstrated to date. The reduction to
practice of suchadesirable surface function is highly challenging, however, becauseof
the low rates typically associated with processes driven by small changes in free
energy.
In summary, at a risk typical for all generalizations, a general rule for active fuel cell
electrocatalysts is proposed here, in the hope that it can provide a common ground for
the evaluation and development of new electrocatalysts. The rule is based on the
recognition that a wide variety of electrocatalytic processes, taking place at either
redox-functionalized or metal surfaces, are surface-redox-mediated, leading, in turn,
to the pursuit of an optimum value for (E�cell process�E�
surface redox) as the guideline for
maximizing the electrocatalytic activity. An optimized gap between these two
standard potentials best addresses the conflicting demands of a minimum over-
potential for surface activation and a high rate of the reaction between the reactant
molecule and the active surface site. Since themaximum rate is expected at an optimal
gap between the E� values, a plot of the rate of the electrocatalytic process versus
(E�cell process�E�
surface redox) will obviously take the famous form of a “volcano”;
however, this typical shape is now projected and explained in terms of a redox
mediation mechanism and the need to optimize the value of (E�cell process�E�
surface redox)
to achieve high rates at low overpotential. Enjoy the book!
S. GOTTESFELD
REFERENCES
1. H. A. Gasteiger and N. M. Markovi�c, Science 324(5923), 48–49 (2009).
2. J. Rossmeisel et al., inFuelCell Catalysis: A Surface ScienceApproach,M. T.M.Koper, ed.,
Wiley, Hoboken, NJ, 2009, pp. 57–93.
3. M. Lefevre et al., Science 324(5923), 71 (2009).
4. S. Gottesfeld, in Fuel Cell Catalysis: A Surface Science Approach, M. T. M. Koper, ed.,
Wiley, Hoboken, NJ, 2009, pp. 1–30.
x FOREWORD
PREFACE
The book covers some essential topics in the science of fuel cell electrocatalysis [1,2].
It shows an increase in importance of theory andmodeling, and the emerging newfield
of electrocatalysis science: bioelectrocatalysis. It shows a spectacular evolution of the
electrocatalysis concepts, froma simple statementof hydrogenevolution/oxidationon
platinum to reactions involving advanced nanoengineering and single-crystal sur-
faces, newmethods to study, and complicated chemical moieties (up to enzymes). It is
basically a materials/theory volume of chemical physics of fuel cell reactions,
including the electron transfer process and structure of the electric double layer, as
seen by a new generation of scientists, not necessarily electrochemists. It also shows
that operando measurements became possible because of the availability of synchro-
tron light. It forecasts thework for the future for the current and incominggenerationof
fuel cell scientists, namely, to use theory and understanding of the process involved
(see Chapter 19 and the Foreword), use the operando (advanced in situ) approach, and
expect theunexpected fromthe emergingnewfieldofbioelectrocatalysis.The future is
bright and exciting; the combination of the intellectual, high technology, and energy
issues makes us strong. We are looking forward.
AWacknowledges the splendid support by theNational ScienceFoundation and the
US Army Research Office toward his research in the preparation of this book.
J. NORSKOV
A. WIECKOWSKI
REFERENCES
1. S.-G. Sun, P.A. Christensen, and A. Wieckowski, eds., In-Situ Spectroscopic Studies of
Adsorption at the Electrode and Electrocatalysis, Elsevier, Amsterdam, 2007.
2. A. Wieckowski, E. Savinova, and C. Vayenas, eds., Catalysis and Electrocatalysis at
Nanoparticle Surfaces, Marcel Dekker, New York, 2003.
Note: Color versions of selected figures are available on ftp://ftp.wiley.com/
sci_tech_med/fuel_cell_catalysis.
xi
PREFACE TO THE WILEY SERIES ONELECTROCATALYSIS AND ELECTROCHEMISTRY
TheWiley Series on Electrocatalysis and Electrochemistry covers recent advances in
electrocatalysis and electrochemistry and depicts prospects for their contribution to
the industrial world. The series illustrates the transition of electrochemical sciences
from its beginnings in physical electrochemistry (covering mainly electron transfer
reactions, concepts of electrode potentials, and structure of the electrical double layer)
to a field in which electrochemical reactivity is shown as a unique aspect of
heterogeneous catalysis, is supported by high-level theory, connects to other areas
of science, and focuses on electrode surface structure, reaction environments, and
interfacial spectroscopy.
The scope of this series ranges from electrocatalysis (practice, theory, relevance to
fuel cell science and technology) to electrochemical charge transfer reactions,
biocatalysis and photoelectrochemistry. While individual volumes may appear quite
diverse, the series promises up-to-date and synergistic reports on insights to further the
understanding of properties of electrified solid/liquid systems. Readers of the series
will also find strong reference to theoretical approaches for predicting electrocatalytic
reactivity by high-level theories such as DFT. Beyond the theoretical perspective,
further vehicles for growth are provided by the sound experimental background and
demonstration of the significance of such topics as energy storage, syntheses of
catalytic materials via rational design, nanometer-scale technologies, prospects in
electrosynthesis, new research instrumentation, surface modifications in basic re-
search on charge transfer, and related interfacial reactivity. In this context, one might
notice that new methods that are being developed for one specific field are readily
adapted for application in another.
Electrochemistry has benefited from numerous monographs and review articles
due to its applicability in the practical world. Electrocatalysis has also been the
subject of individual reviews and compilations. TheWiley Series on Electrocatalysis
and Electrochemistry hopes to address the current activity in both of these comple-
mentary fields by containing volumes that individually focus on topics of current and
potential interest and application. At the same time, the chapters intend to demon-
strate the connections of electrochemistry to areas in addition to chemistry and
physics, such as chemical engineering, quantum mechanics, chemical physics,
surface science, biochemistry, and biology, and thereby bring together a vast range
of literature that covers each topic. While the title of each volume informs of the
specific concentration chosen by the volume editors and chapter authors, the integral
outcome of the series aims is to offer a broad-based analysis of the total development
xiii
of the field. The progress of the series will provide a global definition of what
electrocatalysis and electrochemistry are concerned with now and how these fields
will evolve overtime. The purpose is twofold; to provide a modern reference for
graduate instruction and for active researchers in the two disciplines, and to document
that electrocatalysis and electrochemistry are dynamic fields that are ever-expanding
and ever-changing in their scientific profiles.
Creation of each volume required the editor involvement, vision, enthusiasm and
time. The Series Editor thanks all the individual volume editors who graciously
accepted the invitations. Special thanks are for Ms. Anita Lekhwani, the Series
AcquisitionsEditor,whoextended the invitation to theSeriesEditor and is awonderful
help in the assembling process of the Series.
ANDRZEJ WIECKOWSKI
Series Editor
xiv PREFACE TO THE WILEY SERIES ON ELECTROCATALYSIS AND ELECTROCHEMISTRY
CONTRIBUTORS
Radoslav R. Adzic, Brookhaven National Laboratory, Upton, NY 11973
Matthias Arenz, Department of Chemistry, University of Copenhagen,
Copenhagen, Denmark
Fraser A. Armstrong, Department of Chemistry, Oxford University, South Parks
Road, Oxford OX1 3QR, United Kingdom
Renata Bilewicz, Faculty of Chemistry, University of Warsaw, Pasteura 1, 02-093
Warsaw, Poland
Nic�ephore Bonnet, Department of Materials Science and Engineering, Massachu-
setts Institute of Technology, Cambridge, MA 02139
Fikile R. Brushett, Department of Chemical and Biomolecular Engineering,
University of Illinois at Urbana—Champaign, Urbana, IL 61801
Juan Pablo Busalmen, Laboratorio de Bioelectroquımica, INTEMA(CONICET),
UNMdP. Juan B. Justo 4302, B7608FDQ, Mar del Plata, Argentina
Karen Chan, Department of Chemistry, Simon Fraser University, Burnaby, British
Columbia, Canada
Ismaila Dabo, Universit�e Paris-Est, CERMICS, Projet Micmac ENPC-INRIA, 6-8
Avenue Blaise Pascal, 77455 Marne-la-Vall�ee Cedex 2, France
Michael Eikerling, Department of Chemistry, Simon Fraser University, Burnaby,
British Columbia, Canada
AbrahamEsteve-Nunez, DepartamentodeQuımicaAnalıtica e IngenierıaQuımica,
Universidad de Alcal�a, Madrid, Spain
JuanMiguel Feliu, Instituto de Electroquımica, Universidad de Alicante, Apartado
de Correos 99, 03080 Alicante, Spain
Peter Ferrin, Department of Chemical and Biological Engineering, University of
Wisconsin—Madison, Madison, WI 53706
Dominic Gervasio, Department of Chemical and Environmental Engineering,
University of Arizona, Tucson, AZ 85721
Hendrik A. Heering, Leiden Institute of Chemistry, Leiden University, PO Box
9502, 2300 RA Leiden, The Netherlands
xv
Paul J. A. Kenis, Department of Chemical and Biomolecular Engineering, Uni-
versity of Illinois at Urbana—Champaign, Urbana, IL 61801
Michael L. Klein, Institute for Computational Molecular Science, College of
Science and Technology, Temple University, Philadelphia, PA 19122
Marc T.M. Koper, Leiden Institute of Chemistry, Leiden University, PO Box 9502,
2300 RA Leiden, The Netherlands
Carol Korzeniewski, Department of Chemistry and Biochemistry, Texas Tech
University, Lubbock, TX 79409
Heather J.Kulik, Department ofMaterials Science andEngineering,Massachusetts
Institute of Technology, Cambridge, MA 02139
Yanli Li, Universit�e Paris-Est, CERMICS, Projet Micmac ENPC-INRIA, 6-8
Avenue Blaise Pascal, 77455 Marne-la-Vall�ee Cedex 2, France
Nenad M. Markovic, Materials Science Division Argonne National Laboratory,
Argonne, IL 60439
Nicola Marzari, Department of Materials Science and Engineering, Massachusetts
Institute of Technology, Cambridge, MA 02139
Manos Mavrikakis, Department of Chemical and Biological Engineering,
University of Wisconsin—Madison, Madison, WI 53706
Agostino Migliore, Center for Molecular Modeling, Department of Chemistry,
University of Pennsylvania, Philadelphia, PA 19104
JensK.Nørskov, Department ofPhysics,Center forAtomic-ScaleMaterialsDesign,
Technical University of Denmark, DK-2800, Lyngby, Denmark
Marcin Opallo, Institute of Physical Chemistry, Polish Academy of Sciences, ul.
Kasprzaka 44/52, 01-224 Warsaw, Poland
Odysseas Paschos, Department of Physics, Technische Universit€at M€unchen,James Franck Strasse 1, D-85748, Garching, Germany
David E. Ramaker, Chemistry Department, George Washington University,
Washington, DC 20052
Jan Rossmeisl, Department of Physics, Center for Atomic-Scale Materials Design,
Technical University of Denmark, DK-2800, Lyngby, Denmark
Christina Roth, Institute for Materials Science, Technische Universit€at, Darmstadt,
Germany
Ata Roudgar, Department of Chemistry, Simon Fraser University, Burnaby, British
Columbia, Canada
Kotaro Sasaki, Brookhaven National Laboratory, Upton, NY 11973
xvi CONTRIBUTORS
Carlo U. Segre, Physics Division, Illinois Institute of Technology, 3101 S. Dearborn
St., Chicago, IL 60616
Patrick H.-L. Sit, Center for Molecular Modeling, Department of Chemistry,
University of Pennsylvania, Philadelphia, PA 19104 and Institute for
Computational Molecular Science, College of Science and Technology,Temple
University, Philadelphia, PA 19122
Eugene S. Smotkin, Department of Chemistry and Chemical Biology, 417 Hurtig
Hall, Northeastern University, Boston, MA 02116
Ulrich Stimming, Department of Physics, Technische Universit€at M€unchen, James
Franck Strasse 1, D-85748 Garching, Germany and ZAE Bayern Division 1,
Walther Meissner Strasse 6, D-85748 Garching, Germany
Yung-Eun Sung, World Class University Program of Chemical Convergence for
Energy and Environment, School of Chemical and Biological Engineering, Seoul
National University, Seoul 151-744, Korea
Hiroyuki Uchida, Clean Energy Research Center, University of Yamanashi,
4 Takeda, Kofu 400-8510, Japan
Miomir B. Vukmirovic, Brookhaven National Laboratory, Upton, NY 11973
Mitsuru Wakisaka, Fuel Cell Nanomaterials Center, University of Yamanashi,
4 Takeda, Kofu 400-8510, Japan
Jia X. Wang, Brookhaven National Laboratory, Upton, NY 11973
Liya Wang, Department of Chemistry, Simon Fraser University, Burnaby, British
Columbia, Canada
Masahiro Watanabe, Fuel Cell Nanomaterials Center, University of Yamanashi,
4 Takeda, Kofu 400-8510, Japan
Andrzej Wieckowski, Department of Chemistry, University of Illinois at Urbana—
Champaign, Urbana, IL 61801
Holger Wolfschmidt, Department of Physics, Technische Universit€at M€unchen,James Franck Strasse 1, D-85748 Garching, Germany
Sung Jong Yoo, Fuel Cell Center, Korea Institute of Science and Technology, Seoul
136-791, Korea
CONTRIBUTORS xvii
CHAPTER 1
Hydrogen Reactionson Nanostructured Surfaces
HOLGER WOLFSCHMIDT and ODYSSEAS PASCHOS
Department of Physics, Technische Universit€at M€unchen, Garching, Germany
ULRICH STIMMING
Department of Physics, Technische Universit€at M€unchen and ZAE Bayern Division 1,Garching, Germany
Hydrogen catalysis is an important scientific field since hydrogen reactions (e.g.,
hydrogen evolution and hydrogen oxidation) play a key role in electrochemical
devices such as fuel cells and electrolyzers. The latter devices have the potential to
provide clean and sustainable energy with high efficiencies. This chapter reviews
hydrogen catalysis in detail. Details on hydrogen reaction studies from theoretical and
experimental perspectives are presented. The former usually complement the results
from experimental studies and are used to strengthen them. Various systems that have
been explored throughout the years are reviewed. These include model surfaces as
well as applied systems. Model catalyst systems comprise Pt and Pd nanoislands
deposited on planar surfaces of inert supports, high-quality single-crystal materials,
or single nanoparticles created with scanning tunneling microscopy tips. Applied
systems consist of metallic nanoparticles deposited on high-surface-area carbon
supports. Theory versus experiment, and model versus applied systems are reviewed
in detail, and useful insights for hydrogen reactions in these systems are demonstrated
1.1 INTRODUCTION
Whereas the nineteenth century was the stage of the steam engine and the twentieth
centurywas the stageof the internal-combustion engine, it is likely that the twenty-first
century will be the stage of the fuel cell. Fuel cells have captured the interest of people
Fuel Cell Science: Theory, Fundamentals, and Biocatalysis,Edited by Andrzej Wieckowski and Jens K. NørskovCopyright � 2010 John Wiley & Sons, Inc.
1
around theworld as one of the next great energy alternative. They are nowon the verge
of being introduced commercially, revolutionizing the present method of power
production. Fuel cells can use hydrogen as fuel and oxygen or air as oxidant, offering
the prospect of supplying the world with clean, sustainable electrical power, heat,
and water.
This chapter focusesonhydrogen reactions such as thehydrogenoxidation reaction
(HOR) and the hydrogen evolution reaction (HER). These reactions are of utmost
importance in developing and improving fuel cell devices. The discussion here is
directed principally toward hydrogen electrocatalysis from an experimental as well as
theoretical perspective. Starting with an overview on the fundamentals of hydrogen
reactions in Section 1.2, studies on single crystals as well-defined and high-quality
surfaces are reviewed. An introduction to theoretical work calculating important
fundamentals for hydrogen catalysis regarding material properties is then discussed.
As predicted by theory, the behavior of nanostructured and bimetallic surfaces differs
from that of bulk material. Similar findings supporting the theoretical predictions are
shown for large nanostructured surfaces as well as single particles. The section
concludes with a short overview of carbon-based catalysts.
The fundamentals of hydrogen reactions are reviewed in Section 1.2. Starting from
thegeneral reversible hydrogen reaction, the different reaction pathways suggested by
Volmer, Heyrowsky, and Tafel are introduced. Because of the importance of the
hydrogen adsorption mechanism and the important findings with new experimental
techniques, a short overview of results obtained since the late 1990s is given. An
introduction to the correlation between catalytic behavior and the catalyst material
significance of this correlation, completes this section using experimental and
theoretical calculations, with a conclusion regarding the long-range.
Single crystals and well-defined surfaces play a very important role in surface
science. Many scientific contributions are available that study these well-defined
surfaces. Section 1.3 introduces the electrochemical behavior toward hydrogen
reactions on Pt, Au, and Pd surfaces. The quality of single crystals rapidly increased
in the 1990s, resulting in newand different insights. Because of the importance of Pt as
a catalyst, themain part of this section focuses on this element. The dependence of the
crystallographic orientation toward adsorption as well as electrocatalytic activity is
discussed. An introduction to Pd as a catalyst material with the property to absorb
hydrogen and Au as an inert support material is the last topic in that section.
Besides experimental work, numerous theoretical calculations for hydrogen
catalysis have been performed. Computational methods such as density functional
theory (DFT) and Monte Carlo simulations are powerful tools in surface science and
catalysis. Theoretical as well as experimental work has been combined in several
scientific publications and complement each other well. The first principles of
theoretical techniques and theoretical results are shown in Section 1.4. As a main
topic, the adsorption behavior of hydrogen is considered and the d-band model is
introduced. Calculations regarding the hydrogen oxidation reaction and the influence
of different reactions pathways are also shown. Theoretical calculations of metals on
thinfilms and supported onvarious foreignmetals are reviewed and are correlatedwith
experimental findings.
2 HYDROGEN REACTIONS ON NANOSTRUCTURED SURFACES
The chemical behavior ofmetal nanoparticles often differs from that of bulkmetal.
Different effects such as particle size, interparticle distance, and support effects have
to be considered in this nanometer-scale regime. Since Pd and Pt are important
materials in catalysis, much work was done in the last few decades describing the
abovementioned effects. In particular, multilayers, monolayers, and submonolayers
of Pd and Pt onto foreign metal supports have shown unexpected behavior. Pd
on Au(111) regarding several electrochemical properties introduces this section.
Different types of adsorption, absorption, and desorption behavior as well as
electrocatalytic activity toward hydrogen reactions are shown and discussed. The
deposition of Pd onother supports and the influence of hydrogen reactions hindered by
adsorbing foreign adsorbates as well as investigations of Pt overlayers onAu(111) are
also discussed. A summary and detailed discussion in Section 1.5 also includes
theoretical aspects.
As mentioned above, planar surfaces are thoroughly investigated and serve as
widely accepted reference systems with high-quality, reproducible results. For local
investigation of small structures, new approaches and setups have to be designed
and applied. For this purpose, the electrochemical scanning tunneling microscope
(EC-STM) has been modified by several groups in order to create small nanoparticles
and nanoparticle arrays and also to investigate corrosion, deposition, dissolution, and
reactivity. Due to the tunneling effect, high resolution is achievable and thus leads to a
precise techniquewith atomic resolution. TheSTM tip can be used in different ways in
the electrochemical environment to create and investigate local reactivity of nanos-
tructures. Experimental and theoretical results are compared and are shown to
complement each other. Specifically, the activity of a single Pd particle is shown.
Adiscussion of the experimental results of the stability of Pd particles deposited onAu
(111) and their reactivity towardHER follows.A summary completes Section 1.6with
a comparison between results obtained at extended Pd nanostructured Au(111)
surfaces and single Pd particles.
Section 1.7 presents an overview of studies performed on carbon-based systems.
Since carbon has high electrical conductivity, is relatively inexpensive to use, and is
highly available, it has been the favored support material for many years. Of the many
scientific contributions, only a few can be presented here regarding the mechanism of
HER and HOR using metallic nanoparticles with carbon-based supports. The reac-
tivity of these catalysts for hydrogen reactions and CO oxidation is also of major
interest. These catalyst systems include glassy carbon, carbon nanofibers, Vulcan,
and carbon black for support for metallic nanoparticles, and the more highly oriented
and defined pyrolytic graphite (HOPG) are also presented and discussed.
1.2 FUNDAMENTALS OF HYDROGEN REACTIONS
1.2.1 Hydrogen Catalysis
Over the years a number of studies have been performed in order to investigate
the characteristics of hydrogen-related reactions. The general reversible reaction is
as follows:
FUNDAMENTALS OF HYDROGEN REACTIONS 3
Hþ þ e� $ 12H2 ð1:1Þ
Its standard potential is set to 0V. In the case of proton discharge to formmolecular
hydrogen the reaction is called a hydrogen evolution reaction (HER), while the
reverse pathway describes the hydrogen oxidation reaction (HOR). However, for the
reaction to proceed at sufficient rate, it needs to be catalyzed on an electrode surface.
Possible catalyst candidates include various metals such as Pt, Pd, and Ru, as well as
enzymes with active centers. Much research focused on finding parameters that
influence the activity of materials toward hydrogen electrocatalysis. Even though
much progress has been made on this matter, it is still not clearly known how
various properties influence the catalytic activity. More details will be given later
in this section.
Today it is generally accepted that hydrogenevolutiononPtoccursvia twodifferent
pathways consisting of every two reaction steps:
Discharge reaction of a proton to form an adsorbed hydrogen atom, known as the
Volmer reaction [1]:
PtþHþ þ e� !H�Pt ð1:2ÞCombination of two adsorbed hydrogen atoms to form molecular hydrogen,
known as the Tafel reaction [2]:
2ðH�PtÞ!H2 þ 2Pt ð1:3ÞCombination of an adsorbed hydrogen atom with a proton and an electron to form
molecular hydrogen, known as the Heyrovsky reaction [3]:
H�PtþHþ þ e� !H2 þ Pt ð1:4Þ
Two different pathways can occur; the first one is described as a combination of
reactions (1.2) and (1.3), known as the Volmer–Tafel mechanism. With this mecha-
nism, protons from the solution are discharged on the catalyst surface, forming
adsorbed hydrogen atoms. Then, two adjacent adsorbed hydrogen atoms combine to
form molecular hydrogen. The second mechanism, known as the Volmer–Heyrovsky
mechanism, can be described by combining reactions (1.2) and (1.4). A proton from
electrolyte solution is discharged on the catalyst surface to forman adsorbed hydrogen
atom. This step is followed by combination with another proton and electron to form
molecular hydrogen.
Hydrogen oxidation reaction on Pt can be described in a similar way using the
reaction pathways in reverse order:
Dissociation of molecular hydrogen into one adsorbed hydrogen atom and
immediate discharge of the other atom into proton and electron, similar to
the Heyrovsky reaction:
H2 þ Pt!H�PtþHþ þ e� ð1:5Þ
4 HYDROGEN REACTIONS ON NANOSTRUCTURED SURFACES
Adsorption of molecular hydrogen on the catalyst surface in the form of two
hydrogen atoms, similar to the Tafel reaction:
H2 þ 2Pt! 2ðH�PtÞ ð1:6Þ
Discharge of an adsorbed hydrogen atom to proton and electron, similar to the
Volmer reaction:
H�Pt! PtþHþ þ e� ð1:7Þ
Similar to hydrogen evolution, the hydrogen oxidation reaction can follow two
different pathways. The first mechanism is a combination of reactions (1.5) and (1.7).
A hydrogen molecule is positively charged (H2 !Hþ2 ), and immediately one of its
atoms is discharged into proton and electron,while the other is adsorbed on the surface
of the catalyst. Then the adsorbed hydrogen atom is discharged into proton and
electron. The second one is a combination of reactions (1.6) and (1.7). With this
mechanismmolecular hydrogen is adsorbed on the catalyst surface in the form of two
hydrogen atoms, followed by discharge of the atoms into proton and electron.
1.2.2 Hydrogen Adsorption Mechanism and Experimental Setups
Aswas shown above, in both hydrogen reactions (oxidation and evolution) the step of
forming a hydrogen adsorbate on the catalyst surface exists in both pathways.
Research was performed in order to study the mechanism of hydrogen adsorption
onPt single crystals. Pt is one of themostwidely studied catalysts because of its ability
to catalyze hydrogen reactions with small overpotentials. Initial studies focused on
determining the heat of adsorption of hydrogen on Pt(111) single crystals. An
interesting review was published by Markovic and Ross [4], who showed the values
for the heat of hydrogen adsorption reported in early years to be inconsistent.
Christmann and Ertl [5] reported in 1976 that the value for Pt(111) is approximately
equal to 50–60 kJ/mol. However, McGabe and Schmidt [6] in 1977 and Salmeron
et al. [7] in 1979 reported higher values, between 70 and 90 kJ/mol. Later, it was found
that these corresponded to adsorption of hydrogen on defect sites. Until relatively
recently it was accepted that hydrogen tends to adsorb on highly coordinated sites,
which for the case of Pt(111) would be the threefold hollow sites. These would lead,
though, to a very high coverage of two hydrogen atoms per Pt; therefore, in order to
ensure agreement with experimental values, it was accepted that hydrogen occupies
the threefold next-nearest-neighbor sites (for details, see Section 1.3). Olsen et al. [8],
performing DFT calculations, showed that hydrogen tends to occupy the top sites.
Nevertheless, in all cases the values reported were close to each other.
Depending on the overpotential, the adsorbed hydrogen atom on the catalyst
surface is referred to as under- or overpotential deposited hydrogen (Hupd or Hopd).
Hupd refers to hydrogen atoms adsorbed at potentials positive of the reversible
hydrogen electrode (RHE) potential, while Hopd occurs at potentials negative of the
RHEpotential. The state ofHupd andHopd depends also on the pHof the electrolyte and
FUNDAMENTALS OF HYDROGEN REACTIONS 5
is generated from either protons or water molecules [4] and can be described by the
following reactions:
PtþH3Oþ þ e� ! Pt�Hupd þH2O ðpH � 7Þ ð1:8Þ
PtþH2Oþ e�Pt�Hupd þOH� ðpH � 7Þ ð1:9Þ
There are several possible reasonswhy reported experimental values sometimes do
not agree. The first would be the quality of the single crystal. Crystals having defect
sites or impurities adsorbed on their surfaces act differently toward electrochemical
reactions. Also, it has been shown that different single crystal faces of Pt have different
electrocatalytic rates. Markovic and Ross [4] showed that the activity increases in the
order of (111) < (100) < (110) (Fig. 1.1).Barber et al. [11] showed a slightly different result (Fig. 1.2), where the activity for
HER/HOR increases in the order of (100) < (111) < (551) < (110).However, despite these small differences, it is clearly shown that the activity is
strongly affected by the orientation of the Pt single crystal.
The experimental technique also plays a determining role on the results obtained
mainly with the appearance of a limiting current density above certain overpotentials
for the hydrogen reactions. Especially for the case of HORon Pt, the exchange current
density is high in acidic solutions, but simultaneously, because of the low solubility of
hydrogen, the limiting diffusion current is low.Quaino et al. [12] showed that by using
Levich–Koutecky analysis the j(g) dependence for HOR cannot always be obtained
accurately and may be underestimated. Bagotzky and Osetrova [13] were the first
to propose an alternative experimental setup that had the potential to solve many
issues related to the investigation ofHOR. Their setup consisted of Ptmicroelectrodes
2
0
-2
-4
-6-0.2 0.0
i[mA
cm-2
]
0.2 0.4
(a) 0.05 H2SO4, 274 K
0.6 0.8
Pt(110) Pt(100) Pt(111)
FIGURE 1.1 Polarization curves for HER and HOR on Pt(hkl) in 0.1MHClO4 at sweep rate
20mV/s [4].
6 HYDROGEN REACTIONS ON NANOSTRUCTURED SURFACES
(�A� and �B�) embedded in fused-glass tubes with two different polished surfaces.
Because of the small thickness of the electrode, therewas an enhanced mass transport
of hydrogen. Therefore, values for limiting diffusion current that were one order of
magnitude higher than those obtained from RDE setups could be reached. However,
the results obtained were affected by the roughness of the microelectrode surface,
which can be clearly seen in Figure 1.3.
0.2
25i / m
A c
m-2
(geo
m)
50
75
0.4ϕr / v
1
4
3
2
0.6 0.8 1.0
FIGURE 1.3 Dependence of hydrogen ionization current on potential on microelectrodes
with different roughness valuesA andB in 0.5MH2SO4: (1)A, (2)B and in 1MKOH: (3)A,
(4) B [13].
-0.20
-0.15
-0.10
-0.05
0.0010-3 10-2 10-1 100
(100)SI (111) (511) (110)
current-density / A cm-2
E v
s. R
HE
/ V
101 3232323232
FIGURE1.2 DerivedTafel plots, less the diffusion effect, for the Pt (100)SI, (511), (111), and
(110) faces as marked on the plot [11].
FUNDAMENTALS OF HYDROGEN REACTIONS 7
Two electrodes with different roughness values, differentiated by the degree of
polishing,were used to study hydrogen oxidation in both acidic and alkaline solutions,
and as can be seen, the results are different for the two electrodes. Quaino et al. [14]
used a similar setup to studyhydrogenoxidation. Theywere able to demonstrate that at
low overpotentials the Tafel–Volmer route dominates the kinetics of HOR. At high
overpotentials the Tafel–Volmer effect diminishes while the Volmer–Heyrovsky
mechanism becomes dominant.
Also, traces of impurities that can be present in unpurified solutions can compete
with the reactions under certain conditions, especially at low current densities [15],
resulting in misleading interpretation of the results.
1.2.3 Correlation between Activity toward Hydrogen Reactionsand Physicochemical Properties of Catalyst Material
In the early years research was focused on finding a relationship between the activity
toward hydrogen evolution and oxidation and a property of the catalyst. Conway and
Bockris [16] reported a correlation between the exchange current density j0 and the
electronic workfunction f.Workfunction is defined as the energy with which electrons
near the Fermi level are bound to the material. According to their study [16], the
relationship between j0 and f arises from the dependence of heat of adsorption on f.
Additionally, they showed that the bond strength between adsorbed hydrogen and
metal calculated from Pauling�s equation was smaller than the one obtained from
experiments. Theyalso, as canbeen seen inFigure1.4, usingvalues from the literature,
demonstrated that for various metals (e.g., Ta, Mo, W, Cu, Ni, Fe, Rh, Pd, Pt) the
logarithm of j0 increases as the heat of adsorption of H decreases, while an opposite
trend is observed for Hg, Cd, Pb, and Tl.
For HER, it was also shown (Fig. 1.5) that the logarithm of j0 increases as the d
character of the material increases.
The latterwas explained by the fact that as the d character increases,more electrons
have paired spins and hence require more energy to extract, them causing DH of
adsorbed hydrogen atoms to decrease.
Parsons [17] studied the relationship between exchange current density and
the ability of the electrode to adsorb atomic hydrogen in terms of the standard free
energy DGH. His theoretical studies showed that log j0 reaches a maximum when
DGH� 0. Even though he mentions a disagreement between experimental and
theoretical results (a similar disagreement is also mentioned by Trasatti [18] for the
heat of adsorption, the observed trend should still be valid. Metals that adsorb
hydrogen weakly (DGH has positive values), such as Hg, Zn, and Sn, have low
exchange current densities. Metals such as Pt that adsorb moderately hydrogen have
high values of j0 andmetals that adsorb hydrogen strongly, such asMo, Ta, andW, also
have low j0 values.
It was shown that there is dependence between the exchange current density for
hydrogen reactions and the workfunction of the catalyst material. However, work-
function values were usually used by electrochemists as obtained from physical
experiments. These values were usually measured using nonelectrochemical inputs
8 HYDROGEN REACTIONS ON NANOSTRUCTURED SURFACES
such as adsorption of gas hydrogen on metals and without taking into consideration
the chemical environment surrounding the catalyst. Trasatti [18] published an
interesting review on several aspects in order to obtain more accurate data regarding
the correlation of hydrogen reactions to physicochemical properties of materials.
He argued that the sign of the charge of electrode surface is usually ignored. If the
exchange current density j0 is plotted versus the workfunction (Fig. 1.6), then two
fairly parallel lines can be obtained.
One line consists of data from transition metals and sp metals with positively
charged surfaces,while the other includes data from spmetalswith negatively charged
surfaces. It is also noteworthy that the lines are approximately 0.4 eVapart from each
other. Trasatti explained the division of materials in these two groups in terms of
orientation ofwatermolecules on the catalyst surface. As is shown in Figure 1.7, if the
surface is positively charged, then water will be positioned with an oxygen atom
toward the metal, whereas in the case of a negatively charges catalyst surface, an
opposite orientation is expected.
3.5
−3
−4
−5
−6
−7
−8
−9
−10
−11
−12
−13 xPb
Hg
Nbx
TI
xAIxCd
MoXxCd
CuFex
w
xAu Au
FeNi
Ag
Pt (high c.d.)Rh
Pd
Pt (low c.d.)
4.0 4.5φ IN ELECTRON VOLTS
LOG
10 I 0
(A
mp
Cm
−2)
5.0 5.5
FIGURE 1.4 Linear dependence of log10 of the exchange current (i0) of HER on the
electronic workfunction f. Values of log10 i0 are taken from the literature [16].
FUNDAMENTALS OF HYDROGEN REACTIONS 9
Although the plot shows a clear difference between transition and sp metals, it
includes no information regarding themechanismof reaction. This information can be
factored in only if the exchange current density is plotted versus the heat of adsorption
of hydrogen on themetal surface.Asmentioned previously, the rate andmechanismof
HER depends on the bond strength between themetal and the hydrogen atom (M�H).
Parsons reported that it should pass through amaximum, and a similar volcano-shaped
curve was reported by Krishtalik and Delahay [19], as shown in Figure 1.8.
As shown inFigure 1.8, Pt is on the top of thevolcano curvewhere the Pt�Hbond is
neither too strong nor too weak. The general trend observed in the volcano curve is
that for several metals, as the bonding energy of hydrogen to the metal increases,
the activity also increases, reaching a maximum. Then an opposite trend is
observed, where log j0 decreases as the bonding strength of hydrogen to the metal
increases.
A similar study was done by Nørskov et al. [20]. Density functional theory (DFT)
calculations demonstrated a volcano-type behavior of hydrogen chemisorption
energies with respect to exchange current density for hydrogen evolution (Fig. 1.9).
Platinum was again found to be a better catalyst than other metals for HER
primarily because hydrogen evolution reaction on Pt is thermoneutral at the equilib-
rium potential. The findings of this work can be used to predict behavior of other
bimetallic systems for HER as well as HOR. The analysis was reported as a new
method to obtain H adsorption free energies and understand trends for different
systems that are of electrochemical interest.
-8
-7To
Fe
Ni
W
Mo
Pt (high c.d.)
Pt (low c.d.)
PdRh
-6
-5
-4
-3
-236 38 40
% d – CHARACTER42
LOG
10 I 0
(A
mp
Cm
-2)
44 46 48 50
FIGURE 1.5 Log10 i0 for HER as a function of percent d character of the metal [16].
10 HYDROGEN REACTIONS ON NANOSTRUCTURED SURFACES
Top Related