SELECTIVITY - KU Leuven
Transcript of SELECTIVITY - KU Leuven
Jia
n L
i P
OR
OU
S IO
N E
XC
HA
NG
E M
EM
BR
AN
ES
WIT
H IM
PR
OV
ED
MO
NO
VA
LE
NT
SE
LE
CT
IVIT
Y
Decem
ber 2
018
ARENBERG DOCTORAL SCHOOL
FACULTY OF ENGINEERING SCIENCE
FACULTY OF ENGINEERING SCIENCE DEPARTMENT OF CHEMICAL ENGINEERING
PROCESS ENGINEERING FOR SUSTAINABLE SYSTEMS Celestijnenlaan 200F BOX 2424 B-3001 HEVERLEE, BELGIUM
tel. + 32 485 632238 [email protected]
www.cit.kuleuven.be
POROUS ION EXCHANGE MEMBRANES WITH IMPROVED MONOVALENT SELECTIVITY
Jian Li
Dissertation presented in partial fulfilment of the requirements for the
degree of Doctor of Engineering Science (PhD): Chemical Engineering
December 2018
Supervisor: Prof. Bart Van der Bruggen
POROUS ION EXCHANGE MEMBRANES
WITH IMPROVED MONOVALENT
SELECTIVITY
Jian Li
Supervisor:
Prof. Bart Van der Bruggen
Members of the Examination Committee:
Prof. Jean Berlamont (Chairman)
Prof. Joos (Joseph) Vandewalle (deputy
chairman)
Prof. Guy Koeckelberghs
Prof. Luc Pinoy
Prof. Kitty Nijmeijer (Eindhoven University of
Technology)
Prof. Yang Zhang (Qingdao Institute of
Bioenergy & Bioprocess Technology,
Chinese Academy of Sciences)
Dissertation presented in
partial fulfilment of the
requirements for the degree
of Doctor of Engineering
Science (PhD):
Chemical Engineering
December 2018
© 2018 KU Leuven, Science, Engineering & Technology
Uitgegeven in eigen beheer, Jian Li, Celestijnenlaan 200F box 2424, B-3001 Leuven (Belgium)
Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd en/of openbaar gemaakt worden door middel van druk,
fotokopie, microfilm, elektronisch of op welke andere wijze ook zonder voorafgaande schriftelijke toestemming van de uitgever.
All rights reserved. No part of the publication may be reproduced in any form by print, photoprint, microfilm, electronic or any other
means without written permission from the publisher.
Acknowledgements
I
Acknowledgements
Our destiny offers not the cup of despair, but the chalice of opportunity. So let us
seize it, not in fear, but in gladness. ——R.M. Nixon
Thanks a lot for giving me this valuable chance to spend over three years in KU
Leuven to pursue my PhD degree. This is the most important and correct decision in
my life so far and I can still feel all the exciting moments. I vividly remember that
things began in September 2015, when Dr. Xin Li picked me at the Leuven railway
station. Thanks to all the wonderful people I met in Belgium. This is a special moment
to look back on the period just gone. The completion of my doctoral thesis benefits
from valuable supports from many people and I owe them a debt of gratitude.
Firstly, I would like to express my deepest gratitude and most sincere respect to my
promoter Prof. Bart Van der Bruggen for his excellent supervision, advices, research
guidance and numerous supports! This was a journey that I enjoyed very much, and I
will always be thankful for having had you as my promoter. You gave me a wonderful
opportunity to work at this world-class research group in membrane field. In addition
to the academic aspect, I appreciate you a lot for the way you are: work hard and keep
an open mind. I will follow your philosophy to support my future life. I feel very
lucky and proud to be one of your students. Let‘s never say goodbye, because it is just
a start, far from the end at the moment.
Secondly, I would like to thank my colleagues who help me a lot in ProcESS. First of
all, I would like to express my most sincere thanks to the seniors Jiuyang Lin, Ruixin
Zhang, Wenyuan Ye! I learned a lot from your personality on doing research and
optimistic attitude towards life. The biggest thanks to all my friends in ProcESS who
gave me a lot of assistances in my research and life in Leuven: Junyong Zhu, Shushan
Yuan, Jing Wang, Miaomiao Tian, Bin Liu, Xin Li. You are my strongest backing.
Although some of them were graduated, the assistance from all of you made me adapt
myself in our lab and Leuven. I hope you will have a nice future, realize your dreams
and fight for your belief! I also would like to thank my friends and colleagues in this
Acknowledgements
II
fantastic research group: Ruijun, Yan, Yi, Sofie, Ece, Ben, Carlos, Duc, Trang, Indah,
Fred, Mokgadi, Saeed and other members. Thank you for being such a united and
enterprising family. I cherish the international atmosphere we spend together. Bryant
McGill has his befitting quote for you – ―Cooperation is a higher moral principle than
competition.‖ We are trying our best to complement others rather than compete with
others. We are the best.
While I have the chance, I would like to thank Prof. Jean Berlamont from Department
of Civil Engineering and Prof. Joos (Joseph) Vandewalle from Department of
Electrical Engineering for being a chairman/deputy chairman of the jury for my PhD
defence. I also would like to express my utter appreciation to my assessors Prof. Guy
Koeckelberghs (Department of Chemistry, KU Leuven) and Prof. Luc Pinoy
(Sustainable Chemical Process Technology TC, Ghent and Aalst Technology
Campuses) for their kind and useful comments, remarks and engagement during my
entire doctoral years and correcting my thesis. Without your constructive suggestions
and help in my research during the past three years, I could not come to this part of
my PhD. I also express my sincere gratitude to Prof. Kitty Nijmeijer (Department of
Chemical Engineering and Chemistry, TU Eindhoven) and Prof. Prof. Yang Zhang
(Qingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of
Science) for agreeing to be my jury members.
I would like to thank for several special persons. Dr. Linfeng Li (Bettergy, US), you
broaden my international horizons and guided me how to be a qualified researcher.
Prof. Jiangnan Shen (Zhejiang University of Technology, China), I have gained a lot
since the first day working with you as a master student. It is hard to imagine my
research career without tremendous supports from you. You are also my introducer to
KU Leuven. Thank you so much for your assistance and time I spent with you. Prof.
Arcadio Sotto Diaz (Rey Juan Carlos University, Spain), I will never forget the
moment you come to my lab and bring me your encouraging words and valuable
suggestions.
Acknowledgements
III
Special thanks to Prof. Laurent Bazinet form Université Laval, Prof. Kang Li form
Imperial College London, Prof. Yong Wang and ShiPeng Sun from Nanjing Tech
University for many interesting discussions and suggestions on membrane technology
during conference.
Furthermore, appreciations should be delivered to Herman Tollet, who took great
effort in helping me to construct the electro-membrane system. Besides, I would also
like to express my sincere gratitude to other important people: Christine, Hanne, and
Michele for your supporting on sample measurements. I am also deeply indebted to
Alena Vaes, Beatrice De Geest and Marie-Claude Deflem for their kindly
administrative help.
I gratefully acknowledge China Scholarship Council (CSC) of the Ministry of
Education, China, for funding me to study in KU Leuven. I benefit a lot from CSC
because it provides me a chance to change my thinking and broaden my horizon
through my stay in KU Leuven. A very special gratitude goes out to all down at
Research Fund for helping and providing the funding for the work.
Last but not least, many thanks also go to all my friends. Thank you very much for the
time spending with me together in KU Leuven. Your accompany and support make
my life in KU Leuven more colorful! You may not know what incredible thing you
did for me, I will cherish that for the rest of my life. Especially, I would like to extend
my gratitude to my girlfriend Yi Huang! Thank you for your accompany,
understanding and support. You are always standing by my side to help me through all
the tough time during my PhD. Moreover, as the only child in my family, I would like
to thank my parents and other family members.
Jian Li
December 2018
Abstract
IV
Abstract
Membranes have gained an important place in the field of chemical technology and
are being used increasingly in a broad range of applications. The key property that is
exploited in every application is the ability of a membrane to control the permeation
of a chemical species in contact with it. Electrodialysis is an electrochemical
separation process in which a gradient in electrical potential is used to separate ions
with charged, ion selective membranes. Considering the importance of ion exchange
membranes for contemporary developments, research efforts have been devoted to
developing novel IEMs or to modifying pristine ion exchange membranes for targeted
activities. Today, new membranes developed by regulating the ionic channels is a
novel direction of research.
In this thesis, ionic channels of the ion exchange membrane were regulated by several
methods concerning the membrane matrix and surface skin layer. From the point of
the membrane matrix, a dry-wet phase-inversion strategy by combining immersion
precipitation and dry-casting was used to control the membrane porosity with the
purpose of improving the physical and electrochemical properties of ion-exchange
membranes. Taking advantage of the porous structure, the desalination ratio reached
95%, and the current efficiency reached 100%. However, during the desalting
procedure, the driving force has two contributions, the electrical field and the
salinity-gradient. As a consequence, the porosity should be controlled to balance the
back diffusion from the concentrate to the diluate with membrane electrical resistance.
It was experimentally shown that a membrane prepared with a 1-h heating time has
more steric hindrance, which can decrease the diffusion of ions, so that a superior
desalination efficiency was obtained. On the other hand, a polyaniline grafted
ultrafiltration membrane was prepared with the purpose to separate monovalent ions
from multivalent ions. Similar with porous ion exchange membranes, transport of ions
by the electrical field was dominant at the beginning of the experiments, while
diffusion dialysis by the salinity gradient plays a larger role in ions transport at the
Abstract
V
end of the experiment. In single salt systems, the polyaniline skin layer can hinder the
transport of multivalent ions due to the electrostatic effect, while no obvious effect on
Na+ ions transport can be observed. In the binary system with Na
+ and Mg
2+ ions, the
value of permselectivity is almost doubled as the flux of Na+ was increased to 12.4×
10-8
mol·cm–2
·s–1
while the flux of Mg2+
was reduced to 3.1×10-8
mol·cm–2
·s–1
.
Furthermore, a facile strategy is reported for fabricating monovalent selective ion
exchange membrane based on the rapid co-deposition of biomimetic adhesive
polydopamine and poly(ethylene imine) by using CuSO4/H2O2 as a trigger. Through
this strategy, the surface properties and the permselectivity of the membranes can be
easily tailored by the addition of PEI and by tuning the PEI molecular weight. The
optimum membranes, with 4 h co-deposition of 60 mg PDA and 120 mg PEI,
permselectivity of SPES-PDA/PEI-2 was 2.5 times higher than that of the SPES
membrane. Especially, the flux of H+ was enhanced by the formation of acid-base
pairs. Remarkably, the PDA/PEI modified ion exchange membrane shows an
excellent operation stability for monovalent separation performance after immersion
in acid and alkaline solution for 7 days. Similarly, MIL(53)-Al with nanochannels was
introduced to the skin layer of the monovalent selective membranes through rapid
codeposition of PDA/PEI followed by a cross-linking reaction. The positive −NH2
allows to reject multivalent cations, while porous Mil(53)-Al can accelerate the
migration of Na+. A mass ratio of 0.2–0.4% (w/v) for Mil(53)-Al yielded a
permselectivity of about 0.3 and an ion flux of about 22.0 and 0.6 mol·cm–2
·s–1
for
Na+ and Mg
2+, respectively. At optimum conditions, the PDA-coated membrane
maintains a high monovalent selectivity with enhanced Na+ flux and an enhanced Na
+
and Mg2+
flux in single salt solutions was obtained. A similar material ZIF-8 was used
to replace MIL(53)-Al for a fabricating monovalent selective ion exchange membrane
via interfacial polymerization. Both Na+ and Mg
2+ exhibited a higher transport
efficiency after introducing the ZIF-8 nanoparticles in single salt solutions. When the
binary mixtures were applied, an enhanced monovalent selectivity and Na+ flux were
obtained.
Abstract
VI
In general, membranes with low resistance and/or selectivity for given ions are critical
in industrial processes. In this thesis, such porous membranes were approved to be
feasible to desalinate and separate monovalent ions. Porous ion exchange membrane
with porosity in the membrane matrix and skin layer by suitable membrane formation
techniques or nanoparticles incorporation can be an efficient way to reduce the
resistance and enhance the ion flux.
Samenvatting
VII
Samenvatting
Membranen hebben een belangrijke plaats verworven op het gebied van chemische
technologie en worden in toenemende mate gebruikt in een breder scala van
toepassingen. Een sleuteleigenschap die in elke toepassing wordt gebruikt, is het
vermogen van een membraan om een chemische soort selectief te transporteren.
Elektrodialyse is een elektrochemisch scheidingsproces waarbij een gradiënt in
elektrische potentiaal wordt gebruikt om ionen te scheiden met geladen, ion
selectieve membranen. Gezien het belang van ionenuitwisselingsmembranen voor
hedendaagse ontwikkelingen, zijn onderzoeksinspanningen gewijd aan het
ontwikkelen van nieuwe ionen uitwisselings membranen (IEMs) of aan het
modificeren van ionenuitwisselingsmembranen voor specifieke toepassingen. Het
reguleren van ionkanalen is hierbij een nieuwe richting van onderzoek.
In deze thesis werden ionische kanalen van een ionenuitwisselingsmembraan
gereguleerd door verscheidene methoden met betrekking tot de membraanmatrix en
de oppervlaktelaag. Voor de membraanmatrix werd een droog-natte
fase-inversiestrategie toegepast, door het combineren van immersieprecipitatie met
droog gieten, om de porositeit van het membraan te reguleren met als doel om de
fysische en elektrochemische eigenschappen van het ionenuitwisselingsmembraan te
verbeteren. Door de voordelen van de poreuze structuur te benutten, bereikte de
efficiëntie van ontzilting 95% en de stroomefficiëntie 100%.
Tijdens het ontziltingsproces heeft de drijvende kracht echter twee bijdragen, namelijk
het elektrisch veld en de gradiënt in saliniteit. Daarom moet de porositeit worden
gecontroleerd om de omgekeerde diffusie van het concentraat naar het diluaat te
compenseren met de elektrische weerstand van het membraan. Experimenteel werd
aangetoond dat het membraan bereid met een verouderingsduur van 1 uur meer
sterische hinder vertoonde. Dit kan de diffusie van ionen verminderen, waardoor een
superieure ontziltingsefficiëntie kan worden verkregen.
Samenvatting
VIII
Anderzijds werd een met polyaniline geënt ultrafiltratiemembraan bereid voor het
scheiden van monovalente ionen en multivalente ionen. Vergelijkbaar met het poreus
ionenuitwisselingsmembraan was het transport van ionen door het elektrisch veld
dominant bij het begin van de experimenten, terwijl diffusiedialyse door de
saliniteitsgradiënt een grotere rol speelt bij het transport van ionen aan het einde van
het experiment. In systemen met slechts één zout kan de PANI-toplaag het transport
van multivalente ionen verhinderen vanwege het elektrostatische effect, terwijl er
geen duidelijk effect op het transport van Na+-ionen wordt waargenomen. In het
binaire systeem met Na+ en Mg
2+ ionen is de waarde van de permselectiviteit bijna
verdubbeld naarmate de flux van Na+
werd verhoogd tot 12.42×10-8
mol·cm–2
·s–1
,
terwijl de flux van Mg2+
werd gereduceerd tot 3.1×10-8
mol·cm–2
·s–1
.
Verder wordt een eenvoudige strategie gerapporteerd voor het fabriceren van een
monovalent ionenuitwisselingsmembraan op basis van de snelle co-depositie van
biomimetisch adhesief polydopamine en poly (ethyleenimine), door gebruik te maken
van CuSO4/H2O2 als trigger. Door deze strategie kunnen de
oppervlakte-eigenschappen en de permselectiviteit van de membranen gemakkelijk
worden aangepast door de toevoeging van PEI en door het PEI-molecuulgewicht in te
stellen. De optimale membranen, met 4 uur co-depositie van 60 mg PDA en 120 mg
PEI, vertoonden een permselectiviteit van SPES-PDA/PEI-2 die 2.5 keer hoger was
dan die van het SPES-membraan. Vooral de flux van H+ werd versterkt door de
zuur-baseparen in de synthese. Opmerkelijk is dat het door PDA/PEI gemodificeerde
ionenuitwisselingsmembraan een uitstekende operationele stabiliteit vertoont voor
monovalente scheidingsprestaties na onderdompeling in een zure en alkalische
oplossing gedurende 7 dagen. Evenzo werd MIL (53)-Al met nanokanalen
geïntroduceerd in de toplaag van de monovalente selectieve membranen door snelle
codepositie van PDA/PEI gevolgd door een verknopingsreactie.
De positieve -NH2 laat toe om multivalente kationen tegen te houden, terwijl poreus
Mil (53) -Al de migratie van Na+ kan versnellen. Een massaverhouding van 0,2-0,4%
Samenvatting
IX
(w/v) voor Mil (53) -Al leverde een permselectiviteit op van ongeveer 0,3 en een
ionenflux van ongeveer 22,0 en 0,6 mol·cm–2
·s–1
voor Na+ en Mg
2+, respectievelijk.
Bij een optimale conditie behoudt het PDA-gecoate membraan een hoge monovalente
selectiviteit met verbeterde Na+ en Mg
2+ flux in oplossingen met één zout. Een
vergelijkbaar materiaal, ZIF-8, werd gebruikt om Mil(53)-Al te vervangen voor het
vervaardigen van een monovalent ionenuitwisselingsmembraan via
grensvlakpolymerisatie. Zowel Na+ als Mg
2+ vertoonden een hogere
transportefficiëntie na introductie van de ZIF-8 nanodeeltjes in oplossingen met één
enkel zout. Wanneer binaire mengsels werden aangebracht, werden een verhoogde
monovalente selectiviteit en Na+ flux verkregen.
Samengevat zijn membranen met lage weerstand en/of goede selectiviteit voor
gegeven ionen van cruciaal belang in industriële processen. In deze thesis werd
aangetoond dat dergelijke poreuze membranen haalbaar zijn om te ontzilten en
monovalente ionen te scheiden. Het gebruik van een poreus
ionenuitwisselingsmembraan met porositeit in de membraanmatrix en toplaag kan
door geschikte technieken voor membraansynthese of incorporatie van nanodeeltjes
een efficiënte manier zijn om de weerstand te verminderen en transport van ionen te
verbeteren.
List of Abbreviations
X
List of Abbreviations
AFM
ED
CED
BMED
RED
FTIR
IEC
NF
UF
PA
RO
SEM
EDAX
TFC
TFN
XPS
ZIF
MF
MOFs
AOPs
SGP
BPMs
PVC
IEMS
MIEMs
PANI
PPY
Atomic force microscopy
Electrodialysis
Conventional Electrodialysis
Bipolar Membrane Electrodialysis
Reverse Electrodialysis
Fourier-transform infrared spectroscopy
Ion exchange capacity
Nanofiltration
Ultrafiltration
Polyamide
Reverse osmosis
Scanning electron microscopy
Energy dispersive spectroscopy
Thin-film composite
Thin-film nanocomposite
X-ray photoelectron spectroscopy
Zeolitic imidazole framework
Microfiltration
Metal organic frameworks
advanced oxidization processes
Salinity gradient power
Bipolar ion exchange membranes
Polyvinyl chloride
Ion exchange membrane
Monovalent selective ion exchange
membrane
Polyaniline
Polypyrrole
List of Abbreviations
XI
PEI
IP
PDA
Ra
Rrms
Rm
MPD
TMC
Zreal
Zimag
I–V
Proton exchange membranes
Polyethyleneimine
Interfacial polymerization
Polydopamine
Average roughness
Root mean square roughness
Maximum vertical difference between the
highest and lowest points
m-phenylenediamine
Trimesoyl chloride
Real impedance
Imaginary impedance
Current–voltage
PEMs
Contents
XII
Contents
1. Introduction ....................................................................................................................... 1
1.1 Background ..................................................................................................................... 1
1.2 Ion exchange membranes ................................................................................................ 7
1.2.1 Preparation of ion exchange membranes ................................................................ 12
1.2.2 Monovalent selective ion exchange membrane ...................................................... 14
1.3 Porous ion exchange membranes .................................................................................. 18
1.4 Motivation and contents of the PhD thesis .................................................................... 20
2. Methods and materials ......................................................................................................... 22
2.1 Chemicals and methods ................................................................................................. 22
2.1.1 Chemicals ............................................................................................................... 22
2.1.2 Porous ion exchange membrane preparation .......................................................... 23
2.1.3 Preparation of monovalent selective ion exchange membrane based on
ultrafiltration membrane .................................................................................................. 24
2.1.4 Preparation of polydopamine/polyethyleneimine modified monovalent selective ion
exchange membrane ........................................................................................................ 25
2.1.5 Preparation of monovalent cation exchange membrane containing hydrophilic
MIL53-(Al) framework ................................................................................................... 27
2.1.6 Preparation of monovalent cation exchange membrane by interfacial
polymerization ................................................................................................................. 28
2.2 Membrane properties and characterization.................................................................... 29
2.2.1 Ion exchange capacity ............................................................................................ 29
2.2.2 Water uptake ........................................................................................................... 30
2.2.3 Water contact angle ................................................................................................ 30
2.2.4 Zeta potential of membrane surfaces ...................................................................... 31
2.2.5 Membrane electrical resistance .............................................................................. 31
Contents
XIII
2.2.6 Diffusion dialysis experiments ............................................................................... 34
2.2.7 Current-voltage and transport number measurements ............................................ 35
2.2.8 Electrodialysis experiments .................................................................................... 39
2.2.9 Structural stability of ion exchange membrane ...................................................... 42
2.2.10 Morphology and structure of membranes ............................................................. 42
2.2.11 Chemical structure and composition of membranes ............................................. 43
2.2.12 Water flux experiments ........................................................................................ 43
3. Cation exchange membranes with controlled porosity in electrodialysis application ..... 44
3.1 Introduction ................................................................................................................... 44
3.2 Results and discussion ................................................................................................... 45
3.2.1 SEM results and water flux .................................................................................... 45
3.2.2 IEC and water uptake ............................................................................................. 47
3.2.3 Contact angle measurements .................................................................................. 48
3.2.4 Membrane resistance and transport number ........................................................... 49
3.2.5 Diffusion dialysis.................................................................................................... 49
3.2.6 Electrodialysis experiments .................................................................................... 51
3.3 Conclusions ................................................................................................................... 54
4. Charge-assisted ultrafiltration membranes for monovalent ions separation in
electrodialysis .......................................................................................................................... 56
4.1 Introduction ................................................................................................................... 56
4.2 Results and discussion ................................................................................................... 58
4.2.1 SEM results ............................................................................................................ 58
4.2.2 FTIR results ............................................................................................................ 59
4.2.3 IEC, water uptake and contact angle ...................................................................... 60
4.2.4 Diffusion dialysis experiments ............................................................................... 62
4.2.5 Desalination parameters during ED: conductivity and pH ..................................... 63
Contents
XIV
4.2.6 Current efficiency ................................................................................................... 67
4.2.7 Monovalent selectivity measurements .................................................................... 69
4.3 Conclusions ................................................................................................................... 70
5. Mussel-inspired modification of ion exchange membrane for monovalent separation ... 71
5.1 Introduction ................................................................................................................... 71
5.2 Results and discussion ................................................................................................... 74
5.2.1 Chemical structure of the membrane surface ......................................................... 74
5.2.2 Morphologies of the membrane .............................................................................. 78
5.2.3 Zeta potential .......................................................................................................... 79
5.2.4 Water contact angle, ion exchange capacity and water uptake .............................. 80
5.2.5 Diffusion experiments ............................................................................................ 82
5.2.6 Electrochemical characterization of the monovalent selective ion exchange
membranes ....................................................................................................................... 83
5.2.7 Electrodialysis experiments .................................................................................... 85
5.2.8 Stability and effects of molecular weight of PEI .................................................... 86
5.3 Conclusions ................................................................................................................... 89
6. Mussel-inspired monovalent selective cation exchange membranes containing
hydrophilic MIL53(Al) framework for enhanced ion flux ...................................................... 90
6.1 Introduction ................................................................................................................... 90
6.2 Results and discussion ................................................................................................... 91
6.2.1 Surface morphology and chemical structure of the membrane .............................. 91
6.2.2 Contact angle, ion exchange capacity and water uptake ........................................ 94
6.2.3 Diffusion dialysis experiments ............................................................................... 96
6.2.4 Electrochemical properties of membranes ............................................................. 97
6.2.5 Electrodialysis experiments .................................................................................... 99
6.3 Conclusions ................................................................................................................. 102
Contents
XV
7. Thin-Film-Nanocomposite Cation ExchangeMembranes Containing Hydrophobic
Zeolitic Imidazolate Framework for Monovalent Selectivity ............................................... 103
7.1 Introduction ................................................................................................................. 103
7.2 Results and Discussion ................................................................................................ 105
7.2.1 Surface morphology and zeta potential ................................................................ 105
7.2.2 IEC and water uptake ........................................................................................... 109
7.2.3 Diffusion dialysis experiments ............................................................................. 110
7.2.4 Membrane resistance ............................................................................................ 111
7.2.5 Electrodialysis experiments .................................................................................. 112
7.2.6 Monovalent selectivity ......................................................................................... 114
7.3 Conclusions ................................................................................................................. 116
8. Conclusions and recommendations for further research ............................................... 117
8.1 General conclusions..................................................................................................... 117
8.2 Recommendations for further research........................................................................ 121
References ............................................................................................................................. 123
Curriculum Vitae ................................................................................................................... 141
Chapter 1
1
1. Introduction
1.1 Background
The increasing calls for environmental protection have stimulated to manage water
resources more holistically (Tzabiras et al., 2016). However, there is still a struggle to
achieve a sustainable, high quality and a sufficient amount of water supply. During the
Global Risks 2015 Report of the World Economic Forum, water shortage had been
identified as the most serious challenge for humanity in the next few decades (Liu et
al., 2017b). Already today, 50% of the world population is suffering from medium
water shortage while 10% are undergoing extreme water problems (Johnson et al.,
2016). Moreover, it is expected that the global population would grow by nearly 40%
in the next forty years (Pendergast and Hoek, 2011). The increasing demand for water
sources has posed a worldwide threat to water supply systems. More than seventy
percent of the Earth's surface is covered by water, but the available freshwater only
accounts for a tiny fraction of the earth‘s total water supply (Khawaji et al., 2008).
The available drinking water obtained from groundwater and lakes is limited because
much of it is too deep to access or cannot be exploited in a sustainable way.
Furthermore, severe ecosystem damage caused water depletion at a striking rate
across the world (Lattemann and Höpner, 2008). Oceans, containing the most
abundant water resources on the earth, can provide an inexhaustible, continuous and
high-quality water supply without damaging the original freshwater ecosystems
(Elimelech and Phillip, 2011). Thus, developing advanced water treatment
technologies for desalination is imperative.
Several technologies have been developed for sustainable water purification, such as
adsorption, flocculation, distillation, air flotation and advanced oxidization processes
(AOPs) (Zhang et al., 2016c). However, most of the technologies mentioned above
are often energetically, chemically and operationally intensive, and thus require
Chapter 1
2
considerable infusion of capital, engineering expertise and infrastructure, all of which
precludes their use in much of the world (Shannon et al., 2008). Even in highly
industrialized countries, the cost and time needed to develop state-of-the-art
conventional water and wastewater treatment facilities make it arduous to address all
the problems mentioned above. Furthermore, intensive chemical treatments and
residuals resulting from treatment can be other factors that limit industrial
applications. Hence, reducing chemical treatment via developing more effective,
low-cost, robust methods to supply clean water are of paramount importance.
Fortunately, there is much more that science and technology can do to mitigate
environmental impact, and to increase efficiency. Membrane technologies like reverse
osmosis (RO), nanofiltration (NF), ultrafiltration (UF), microfiltration (MF), and
electrodialysis (ED) processes are emerging as effective methods to realize the
removal of contaminants and serving as important technologies in water supply for
different purposes (Hu and Mi, 2013). A semipermeable membrane is required for all
these processes, which can permeate some components with specific
physical/chemical properties while rejecting the others (Hou, 2016; Liu et al., 2016).
Some features of membrane processes include: no chemical additives, no phase
change, low energy consumption, operation simplicity and easy to scale up (Kang and
Cao, 2014).
UF and MF are low pressure filtration membranes which have been extensively used
in particle and natural organic materials removal. The pore size of MF generally spans
from 0.1 µm to 10 µm while the pore size of UF ranges from 0.01 µm to 0.1 µm
(Figure 1.1) (Fiksdal and Leiknes, 2006). RO, which can reject all solutes, is a crucial
technology in desalination. NF is a separation process which has caused widespread
attention. The pore size of a NF membrane is between that of an RO and UF
membrane, which could be potentially applied in separating dissolved organic matter
and divalent ions. Different from traditional pressure driven membranes, rejection
mechanisms for NF membranes should take electrostatic repulsion into consideration.
Chapter 1
3
Nevertheless, the development of membrane technology is largely limited by the
membrane materials, and membrane fouling can be another serious problem (Guo et
al., 2012)
Fig. 1.1 Classification of membranes according to pore size (Mikhaylin and Bazinet, 2016; Xu
and Zhang, 2016)
Electrodialysis (ED), an electric driven process, has been commercially applied for
diverse purposes (Sadrzadeh and Mohammadi, 2008). ED can be mainly operated in
three ways, 1. Conventional Electrodialysis (CED); 2. Bipolar Membrane
Electrodialysis (BMED); 3. Reverse Electrodialysis (RED). A typical CED setup
normally contains alternately arranged cation and anion exchange membranes
between two electrodes. Spacers are incorporated with the gaskets to prevent the
contact of cation and anion exchange membranes. While a direct current is employed
as driving force, cations and anions transport toward the cathode and anode to form a
concentrate and a diluate compartment, respectively. A schematic diagram of an ED
cell is presented in Fig. 1.2. Xu et al. reviewed the industrial applications of CED, this
is summarized in Table 1.1 (Xu, 2005). During ED operation, parameters such as the
desalination ratio, water recovery efficiency, current efficiency, energy consumption
and operating cost should be taken into consideration. It can be concluded from Table
1.1 that the industrialization of CED has been limited due to the high costs. The main
reason is attributed to the high operating cost and membrane cost. In addition, the
Chapter 1
4
issues concerning membrane fouling also lead to high energy consumptions.
Membrane fouling is typically induced by the adsorption of organic matter or the
precipitation of metallic cations. Over the past few years, significant progress in ED
has been made. More promising results were expected by combining ED with other
technologies. By introducing IEMs with extraordinary antifouling and desalination
properties, the disadvantages mentioned above have been resolved to a greater extent.
Fig. 1.2 Schematic diagram of electrodialysis experimental setup (Khan et al., 2016)
Chapter 1
5
Tab
le 1
.1 I
ndust
rial
appli
cati
on o
f co
nven
tional
ele
ctro
dia
lysi
s (E
D)
Lim
itat
ion
s
Co
nce
ntr
atio
n o
f fe
ed a
nd
cost
s
Pro
du
ct w
ater
qu
alit
y a
nd c
ost
s
Mem
bra
ne
pro
per
ties
an
d c
ost
s
Pro
du
ct w
ater
qu
alit
y a
nd c
ost
s
Mem
bra
ne
sele
ctiv
ity
an
d c
ost
s
Co
sts
Co
sts
Sta
tus
of
app
lica
tio
n
Co
mm
erci
al
Co
mm
erci
al
Co
mm
erci
al
Co
mm
erci
al
Com
mer
cial
or
pil
ot
ph
ase
Co
mm
erci
al
Pil
ot
ph
ase
Sta
ck a
nd p
roce
ss d
esig
n
Shee
t fl
ow
, to
rtuous
pat
h s
tack
,
rever
se p
ola
rity
Shee
t fl
ow
, to
rtuous
pat
h s
tack
,
rever
se p
ola
rity
Shee
t fl
ow
sta
ck,
unid
irec
tional
Shee
t fl
ow
, to
rtuous
pat
h s
tack
,
rever
se p
ola
rity
Shee
t fl
ow
, to
rtuous
pat
h s
tack
,
unid
irec
tional
Shee
t fl
ow
, unid
irec
tional
Shee
t fl
ow
, unid
irec
tional
Indust
rial
appli
cati
on
s
Bra
ckis
h w
ater
des
alin
atio
n
Boil
er f
eed w
ater
pro
duct
ion
Was
te a
nd p
roce
ss w
ater
trea
tmen
t
Ult
ra p
ure
wat
er p
roduct
ion
Dem
iner
aliz
atio
n o
f fo
od
pro
duct
s
Tab
le s
alt
pro
duct
ion
Conce
ntr
atio
n o
f re
ver
se o
smosi
s
bri
ne
BMED is a technology that can combine resource recycling, energy conversion, and
environmental protection. Bipolar ion exchange membranes (BPMs) can provide a
Chapter 1
6
continuous H+ and OH
- supply by splitting water molecules between the cation and
anion exchange layer with high efficiency, without the need of introducing any other
salt ions. The most common application of BMED technology is the production of
inorganic acids and bases from the corresponding salts. A number of applications have
been operated at pilot scale, such as the production of mineral acids and bases from
salts, or the generation of acid and base from RO desalination concentrates. Some of
them can even be found in commercial plants, like recycling of HF and HNO3 from
steel picking solutions. Furthermore, the separation of organic acid salts and
acidification of various streams have also been explored. The feasibility has been
proved concerning the conversion of gluconic acid (Alvarez et al., 1997), propionic
acid (Boyaval et al., 1993), lactic acid (Lee et al., 1998), acetic acid (Zhang et al.,
2011), malic acid (Liu et al., 2014a), vitamin C (Yu et al., 2002), formic acid (Ferrer
et al., 2006), lactobionic acid (Gutiérrez et al., 2013), salicylic acid (Liu et al., 2015),
citric acid (Tongwen and Weihua, 2002), and amino acid (Eliseeva et al., 2001).
Due to the concerns arising from energy problems, developing renewable and
sustainable energy conversion and production technologies is crucial. Salinity
gradient power (SGP) can be harvested from mixing water streams of different
salinity. Theoretically, approximately 0.8 kWh is obtainable when 1 m3 of fresh water
flows into the sea, which translates into nearly 2 TW of SGP on the basis of the total
freshwater flow of the major rivers worldwide (Mei and Tang, 2018). RED is an
emerging membrane based technology that captures electricity from controlled
mixing of two water streams of different salinities. The configuration of the setup is
similar to a continuous ED setup, except that a suitable redox couple (the Fe2+
/Fe3+
redox couple or the [Fe(CN)6]4–
/Fe(CN)6]3–
redox couple) is applied. As shown in Fig.
1.3, freshwater channels are between two salt water channels, which are separated by
a CEM on one side and an AEM on the other. For a sodium chloride solution, sodium
ions permeate through the cation exchange membrane in the direction of the cathode,
and chloride ions permeate through the anion exchange membrane in the direction of
the anode. Electroneutrality of the electrode rinse solution is maintained via oxidation
Chapter 1
7
and reduction at the anode surface and cathode surface, respectively. The electrical
current generated is captured directly by an external load.
Fig. 1.3 The schematic representation of a RED system (Vermaas et al., 2014)
1.2 Ion exchange membranes
As discussed above, ED is expected to solve issues related to energy and environment
problems. It is generally accepted that the key of an electrodialysis setup is the ion
exchange membranes. The ion exchange membranes are composed of substrates,
functionalized groups, and movable counter-ions. According to the charge of
functional groups, the ion exchange membrane can be a cation exchange membrane
(CEM), or an anion exchange membrane (AEM). Sulfonic acid, phosphoric acid and
carboxylic acid groups are the most common functional moieties for CEMs, while
guanidinium cations, imidazole cations, and quaternary ammonium cations are the
moieties anchored inside the AEMs matrix. According to the connection of charged
groups to the matrix or their chemical structure, ion exchange membranes can be
further classified into homogeneous and heterogeneous membranes. For
heterogeneous membranes, the charged groups are chemically bonded to or physically
Chapter 1
8
mixed with the membrane matrix. Table 1.2 lists some commercially available
homogeneous and heterogeneous ion exchange membrane membranes. In general, the
thickness of the commercial membrane is around 0.2 mm with a resistance smaller
than 5 Ω·cm2.
Chapter 1
9
Table 1.2 The parameters of commercially available ion exchange membranes
Compamy Membrane type Thickness
(mm)
IEC
(meq/g)
Water
uptake
(%)
Resistance
(Ω·cm2)
Permselectivity
(%)
Shandong
Tianwei
Membrane
Technology CO.,
LTD., China
TWBPI (BPMs)
TWEDAI (AEMs)
TWEDCI (CEMs)
0.18-0.23
0.13-0.16
0.10-0.13
-
-
-
20-30
30-40
20-30
-
≤4
≤4
-
-
-
Tingrun
Membrane Tech,
China
TRJBM (BPMs)
TRJCM (CEMs)
TRJAM (CEMs)
0.16-0.23
0.16-0.23
0.16-0.23
-
1.7-2.0
1.5-1.8
33-40
33-38
22-24
-
2.5-5.5
5.0-8.3
90-95
95-99
90-95
PCA
-Polymerchemie
Altmeier GmbH,
Germany
PC-SK (CEMs)
PC-SA (AEMs)
0.13
0.09-0.13
1
1.3
-
-
0.75-3
1-1.5
96
93
Dupont Co., Inc.
USA
Nafion® 117 (CEMs)
Nafion®
112 (PEMs)
0.183
0.089
-
0.9
-
16
-
1.5
-
97
Fuma-Tech
GmbH, Germany
FKS (CEMs)
FAS (AEMs)
0.09-0.11
0.1-0.12
0.9
1.1
-
-
2-4
2-4
-
-
ASTOM
Corporation,
Japan
CSE (CEMs)
ASE (AEMs)
0.16
0.15
-
-
-
-
1.8
2.6
-
-
Asahi Kasei
Chemicals
Corporation,
Japan
Selemion CMV-SK
(CEMs) 0.13-0.15 2.4 25 2.0-3.5 95
Selemion AMV-SA
(AEMs) 0.11-0.15 1.9 19 1.5-3.0 92
Selemion ASV-SA
(AEMs) 0.11-0.15 - - 2.3-3.5 -
Chapter 1
10
From a historical point of view, the development of IEMs began in the eighteenth
century based on the conception of ―osmosis‖. In 1889, Maigrot and Sabates used a
non-selective separator in an electrochemical process to desalination of a sugar syrup
solution (Paidar et al., 2016). After that, Ostwald began to study the properties of
semipermeable membranes and found that a membrane could reject any electrolyte as
long as the membrane is impermeable either for its cation or its anion. This
phenomenon was confirmed by Donnan. A mathematical equation named ―Donnan
exclusion potential‖ was developed to describe the“membrane potential‖ at the
boundary layer (Donnan, 1911). The term ―electrodialysis‖ was used around 1900,
which was much earlier than the elaboration of the ―Donnan exclusion potential‖.
However, the basic research related to ion exchange membranes lagged behind. Only
after the 1930s, the interest in industrial applications prompted the development of
new methods for synthesis of ion exchange membranes. At that time, anion and
cation-selective membranes were assembled together in one apparatus to form parallel
solution compartments. However, industrial applications were significantly impeded
at that time due to the high electric resistance. With the maturation of membrane
fabrication technology, commercial applications for demineralizing and concentrating
electrolyte solutions were realized in the 1950s. In 1970s, a sulfonated
polytetrafluorethylene based ion exchange membrane with excellent stability was
developed by Dupont as Nafion®. The evolution step resulting in an upsurge in the
application in the chlor-alkali production industry (Grot, 1973). Moreover, bipolar
membranes were created in 1976 by Chlanda et al.. By combining a cation exchange
layer with an anion exchange layer, the application domain of electrodialysis with
bipolar membrane was largely amplified (Chlanda et al., 1978). Nowadays, enhanced
properties of ion exchange membranes have been realized with higher selectivity,
lower electrical resistance and improved mechanical stability. New ion exchange
membrane based technologies such as capacitive deionization, continuous
electrodeionization or diffusion dialysis etc. have gained great interest in water and
waste water treatment, food/drug, chemical process industry as well as biotechnology.
Chapter 1
11
The development history of ion exchange membranes is schematically shown in Fig.
1.4. The related ion exchange membrane based processes and their applications are
summarized in Table 1.3. However, all of today's available electro-membrane
processes and components used in these processes still have technical and commercial
limitations. Despite substantial ongoing developments, it is of critical importance to
explore new methods to improve products and processes.
Table 1.3 Electrodialysis and related processes and their applications (Strathmann, 2010)
Ion exchange membrane based processes Technical applications
Electrodialysis Water desalination and salt pre-concentration
Diffusion dialysis Acid and base recovery from industrial waste waters
Donnan dialysis Water softening, and exchange of ions
Bipolar membrane electrodialysis Production of acids and bases from corresponding
salts
Electrodeionization Production of ultra pure water
Capacitive deionization Water desalination and water softening
Reverse electrodialysis Electrodialytic energy generation
Fig. 1.4 Time line visualization of ion exchange membrane development and their related
processes (Xu, 2005)
Chapter 1
12
1.2.1 Preparation of ion exchange membranes
As mentioned above, according to their structure and preparation procedure, IEMs
can be divided into two major categories: homogeneous and heterogeneous. In
homogeneous IEMs, the fixed charge groups are evenly distributed over the entire
membrane matrix while heterogeneous membranes have distinct macroscopic
uncharged polymer domains of ion exchange resins in the membrane matrix. The
specific properties of ion exchange membranes are all related to the presence of these
charged groups. The distinct of amount, type and distribution of ion exchange groups
determine the most important membrane properties.
Ionic groups can be introduced to the homogeneous ion exchange membrane by three
methods: 1. Polymerization or polycondensation of monomers; at least one of them
must contain a moiety that either is or can be made anionic or cationic, respectively; 2.
Introduction of anionic or cationic moieties into a preformed solid film; 3.
Introduction of anionic or cationic moieties into a polymer, such as polysulfone,
followed by the dissolving of the polymer and casting it into a film.
Fluorinated materials are one of the most widely used materials to fabricate IEMs.
Fluorinated membranes have an extreme chemical and thermal stability in
applications in the chlor-alkali industry and in fuel cell applications. A typical
example for successful exploration of fluorocarbon based ion exchange membranes is
the product developed by DuPont with trade name ‗‗Nafion‘‘. However, the
applications of perfluorinated membranes with high equivalent weights were limited
in fuel cells due to their high cost and fuel loss. Membranes with low equivalent
weights were synthesized by Dow Chemical Company as Dow epoxy in 1998 (Souzy
et al., 2004). This polymeric structure can be described as a Teflon-like backbone with
a side chain attached via an ether group. Styrene-divinylbenzene based IEMs are
Chapter 1
13
involved in many industrial ED applications. Such membranes are synthesized by
polymerization of styrene and divinylbenzene, followed by a sulfonation and
amination reaction after chloridization. Cation exchange membranes are synthesized
by a sulfonation reaction using chlorosulfonic acid or concentrated sulphuric acid
while anion-exchange membranes are obtained by amination after chloromethylation.
The reaction equations for the membrane fabrication are shown in Fig. 1.5.
Fig. 1.5 The preparation of styrene-divinylbenzene based ion-exchange membranes (Nagarale et
al., 2006)
Engineering plastics such as polysulfone and polyethersulfone have been widely used
as a base polymer for ultrafiltration and gas separation because of their excellent
HC CH2
HC CH2
CHH2C
n
SO3H
SO3H
n
a) BPO; 60 ℃
b) RT; H2SO4
n
CH2Cl
CH2Cl
n
BPO; 60 ℃
CH3CH2OCH2Cl
SnCl4; 45 ℃; 4 h
n
n
CH2
NH3C
CH3H3C
CH2
NH3C
CH3H3C
Chapter 1
14
workability and mechanical strength. To date, most of the IEMs consist of polymeric
backbones prepared by functionalization of engineering plastics. However, the
variation of the main chain type and side chain type significantly affects the overall
performance of IEMs. Regarding the chemical structure, IEMs are similar with ion
exchange resins because both of them are bearing functional groups. The major
distinction is primarily related to the mechanical requirements of the membrane
process. Consequently, heterogeneous ion exchange membranes can be fabricated by
mechanical incorporation of powered ion-exchange resin into sheets of rubber,
polyvinyl chloride (PVC), acrylonitrile copolymers or some other extrudable or
moldable matrix (Gohil et al., 2004). By choosing suitable reinforcing materials,
heterogeneous ion exchange membranes can be obtained with controlled
electrochemical properties and mechanical strength. The present fabrication processes
include: a. calendering ion-exchange particles into an inert plastic film; b. dry
moulding of inert film forming polymers and ion exchange particles and then milling
the mould stock; c. resin particles can be dispersed in a solution containing a film
forming binder and then the solvent is evaporated to give ion-exchange membrane.
Comparatively, homogeneous membranes have good electrochemical properties
whereas heterogeneous membranes rather have a very good mechanical strength.
1.2.2 Monovalent selective ion exchange membrane
The separation of monovalent ions from multivalent ions in aqueous solution is a
recurrent problem in industrial applications where water salinity is important. A
typical example is seawater purification, where Ca2+
and Mg2+
should be removed to
prevent scaling. Thus, the optimal drinking water production method for a hard source
water requires the rejection of divalent ions while the monovalent ions should
permeate through the membranes. Further examples can be found for e.g.,
electroplating wastewaters containing heavy metals, which should be selectively
removed. Traditional methods to treat these wastewaters include precipitation,
Chapter 1
15
chelating and chlorination. However, excessive sludge production and slow
precipitation progress with insufficient metal removal limit the industrial application.
Electrodialysis with monovalent selective ion exchange membrane (MIEMs) provides
a method to treat these wastewaters, by separating monovalent and multivalent ions.
Currently, MIEMs are explored in a wide range of applications including seawater
desalination, acid recovery and the removal of specific ions.
The mechanisms to separate a specific ion were proposed by Sata in 1994 (Sata,
1994). The permselectivity between ions is governed by several factors, such as
hydrated radii, affinities with IEMs and migration rates in the membrane phase.
Strategies used for regulating such factors are as follows. The addiction of a
crosslinked skin layer on membrane surface is deemed to be an effective way to
obtain MIEMs. By introducing a dense skin layer, the migration of ions with different
hydrated radii occurs on the basis of a size sieving effect. Hou et al. developed a
highly selective MIEM by forming a polyamide selective layer on the membrane skin
layer. The degree of crosslinking was evaluated based on the surface elemental
composition and the results indicated that membrane with a higher cross-linked
structure lead to an improved mono/divalent cations permselectivity (Hou et al., 2018).
Apart from the widely reported covalent crosslinking, acid-base cross-linking and
crystallization are new emerging technologies to improve the membrane surface
density. An annealing treatment strategy was proposed by Ge et al. (Ge et al., 2014) to
enhance the density of polyvinyl alcohol (PVA)-based CEMs by tuning the structural
crystallinity, which can improve the pore-size sieving effect for ions. Due to the
semi-crystalline nature of PVA, the crystallinity of membranes can be controlled by
adjusting the annealing temperature. The increase of the annealing temperature
dramatically decreases the water uptake. In addition, the ion exchange capacity (IEC)
was maintained, indicating that the membrane intrinsic properties are highly
crystallinity dependent. However, in some cases, crosslinking can decrease the
difference in electrostatic repulsion between the cationic layer and cations with
different valence and weaken the monovalent selectivity of the modified membrane. A
Chapter 1
16
modification method using covalent immobilization and self-crosslinking of chitosan
layer on CEM surface was devised by Wang et al. (Wang et al., 2013). The Zn2+
and
Mg2+
leakage of the membrane samples crosslinked by formaldehyde were increased
by 23% and 69%, respectively, compared to those of membranes only grafted with a
chitosan layer. The reason might be that –NH2 groups of the chitosan layer were
consumed by the crosslinking reaction and the generated secondary amine and tertiary
amine only embraced relatively a weak positive charge due to the solvent effect.
Nevertheless, an increase of the electrical resistance is inevitable for most modified
membranes as the migration of ions is mitigated. To solve this problem, introducing
conducting polymers like polyaniline (PANI) or polypyrrole (PPY) can be an
alternative method to both reject the multivalent ions and reduce the membrane
resistance. Kumar et al. chemically modified organic-inorganic hybrid CEMs by in
situ polymerization of aniline in acidic medium using FeCl3 as an oxidizing agent
(Kumar et al., 2013). The thin layer of PANI grafted on the membrane surface and in
the membrane matrix was thermally stable and dense. The Na+ transport number
across the membranes remained unchanged, whereas the transport number of Zn2+
and
Al3+
decreased after modification with PANI. Sivaraman et al. also reported a method
to modify CEMs by electrochemical deposition (Sivaraman et al., 2007). By this
method of modification, the amount of PANI can be controlled easily. Furthermore,
the presence of a conducting polymer PPY layer on the surface of IEMs resulted in a
slower electro-migration of bivalent ions (Ca2+
, Mg2+
and Cu2+
) in comparison to
monovalent ions, which can serve as an alternative material to PANI for modification
of the IEMs (Gohil et al., 2006a).
Deposition of an oppositely charged layer relative to the charge of bulk IEMs also
enables the preparation of MIEMs. For membranes obtained by this method,
multivalent ions with the same charge as the membrane surface are rejected due to the
relatively large electrostatic repulsion, while monovalent ions are still able to pass
through the skin layer. Polyethyleneimine (PEI), a commercially available cationic
polyamine, is one of the most effective and frequently used polyelectrolytes to
Chapter 1
17
fabricate MIEMs. The combination of PEI with IEMs was realized by adsorption or
electro-deposition (Amara and Kerdjoudj, 2003; Guesmi et al., 2012). However, the
monovalent cation permselectivity of the membranes gradually deteriorates during
electrodialysis because the PEI was desorbed from the membrane surface. Thus,
fixation of the cationic polyelectrolyte on the membrane surface by covalent bonding
has been actively studied. One of the effective methods is to fix PEI on the membrane
surface by acid–amide bonding (Takata et al., 2000). After introducing sulfonyl
chloride groups into the membrane matrix, thin PEI layers were formed on the
membrane surfaces by sulfonyl-amide bonding.
Specific interactions between the ion exchange groups and the mobile ions can be
utilized to regulate the membrane selectivity. By choosing the ionic functional groups
and IEM matrices, selective separation can be achieved. As ionic functional groups
are anchored inside the membrane matrix, the mobility of the multivalent cation is
low in the membrane phase compared to monovalent ions. As a consequence, the
permselectivity of specific ions through the membrane may be changed by altering the
interaction between specific ions, ion exchange groups and the membrane matrix
(Sata, 1994).
Recently, research has proved that the membrane hydrophilicity also impact the
selectivity of the membrane. By decreasing the hydrophilicity, the permselectivity of
MIEMs increases (Gohil et al., 2006b). To improve the monovalent selectivity of
IEMs, the methods mentioned above can also be combined. Although IEMs are
industrially applied in numerous fields, the comprehensive properties of MIEMs still
lag behind to meet the requirements of applications on an industrial scale. More
research attention should be focused on releasing the limitations related to a high
electrical resistance, low permselectivity and insufficient ion flux.
Chapter 1
18
1.3 Porous ion exchange membranes
Advances in membrane technology are related to the development of
high-performance membranes with high permselectivity, low electrical resistance,
high mechanical strength, and high chemical stability. In general, a homogeneous or
heterogeneous ion-exchange membrane is a dense membrane that defines the
transport of ions. Although many strategies have been developed to accelerate ion
transport, these strategies were not economically competitive or resulted in an inferior
electrochemical performance (Malik et al., 2016). Recently, UF membranes were used
in the electrodialysis process to facilitate the migration of ions according to their
charge and molecular size (Bazinet and Moalic, 2011b). Chitosan oligomers are
widely used in biotechnology, pharmaceutical and health food industry because of
their bioactivity (Yamada et al., 2005). During the industrial production, the final
product is a mixture of molecules of different molecular weights and contains
minerals. Thus, fractionation of such mixtures is crucial (Horowitz et al., 1957). Aider
et al. suggested that electrodialysis with ultrafiltration method could be used as a
powerful process for the separation of bioactive chitosan oligomers of interest from
complex feed solutions under mild conditions, and other applications in food,
bio-pharmaceutical, and nutraceutical industries (Aider et al., 2008, 2009). However,
for the application of ultrafiltration membranes in electrodialysis, the separation of
smaller ionic species still remains a challenge, because of the relatively large pore
sizes of these membranes.
Inspired by the methods of performing electrodialysis with ultrafiltration membranes
and preparing pore-filled membranes, porous membranes prepared from charged
polymers with functional groups represent another option. Such membranes with
porous structure might facilitate the migration of ions by reducing the physical steric
hindrance of the surface layer. Thus, the migration resistance could be further
diminished, whereas the functional groups of the membrane matrix could block the
diffusion of ions caused by the concentration gradient.
Chapter 1
19
Porous materials are defined as solids containing empty voids which can host other
molecules. The fundamental feature of these materials is their porosity, the average
size of the pores and the surface area. Typical values for the surface area of porous
materials applied in technological processes range between 2000 and 8000 m2/g. The
most important applications of such materials are the storage of small molecules and
filtering. The framework defining a porous material can consist of either organic or
inorganic materials. With regard to developing new methods to produce novel and
highly flexible functional materials, the loading of porous materials with functional
molecules or objects is attracting an increasing amount of interest (Shekhah et al.,
2011).
More recently, metal-organic frameworks (MOFs) have been explored as novel
nanoporous crystalline solids combining metal ions with organic linkers resulting in
highly porous frameworks, and are emerging as a new family of molecular sieve
materials (Tranchemontagne et al., 2009). Apart from the versatile applications in
ion-exchange, gas separation and storage, optics, drug delivery and catalysis (Kuppler
et al., 2009), their unique properties including uniform pore size, high surface areas
and specific adsorption affinities make MOFs attractive for assembling into
membranes with excellent performance. Li et al. present a new approach to construct
ion nanochannels by in situ assembly of a poly N-vinylimidazolium (ionic liquid) as
the ion carrier within the highly ordered pores of MOFs (Li et al., 2016). At an
application oriented level, the highly designable nature of MOFs allows for
customizing pore morphologies, and the structural diversity of ionic liquids allows a
wide selection of ions. Because the demand for fabricating such porous coatings is
rather obvious, several studies have either demonstrated or proposed new applications
of MOF thin films. However, most of them were applied in gas separation, seldom
with ion selectivity (Bux et al., 2011; Huang et al., 2013). Tailoring nanochannel
morphologies by constructing efficient ion nanochannels with appropriate geometrical
and chemical structures is a feasible strategy for achieving monovalent selectivity. For
Chapter 1
20
instance, zeolitic imidazolate framework-8 (ZIF-8), a subclass of MOFs with a small
aperture with a size of 3.4 Å, can serve as an effective filter to separate hydrated
cations of Mg2+
(4.28 Å) through a size sieving effect. Consequently, ZIF-8 with the
right size makes it a fine candidate for preparing MIEMs. Thus, membranes with a
pore structure would guarantee a high permeance when nanochannels are explored as
a candidate for membrane fabrication and modification.
1.4 Motivation and contents of the PhD thesis
The aim of this thesis is to prepare ion exchange membrane with improved
electrochemical properties by introducing porous structures. The chapters in this
thesis are organized as follows: Chapter 1 provides a brief review about membrane
technology for water treatment. In view of understanding electrodialysis technology,
the basic concept, the separation mechanism, especially the preparation and
modification of the ion exchange membrane, are highlighted. Chapter 3 provides a
dry-wet phase-inversion strategy for the preparation of porous sulfonated poly(ether
sulfone) (SPES) membranes. The objective of this study was to prepare
laboratory-made porous ion exchange membranes with improved physical and
electrochemical properties. The porosity was tuned by altering the time of membrane
exposure to an elevated temperature environment. The membrane resistance and its
application in electrodialysis were conducted.
By introducing PANI on membrane surface, elevated separation properties can be
obtained based on the size of the molecules or ions and their charge. A previous study
reported that electrodialysis with nanofiltration membrane (EDNF) can be carried out
for fractionation of cations (Ge et al., 2016a). By replacing the cation exchange
membrane with NF membranes, the NF membrane can fractionate ions by rejecting
the multivalent ions with larger Stokes radius. Enlightened by EDNF, an efficient
one-step chemical process to graft a thin polyaniline (PANI) layer on the surface and
Chapter 1
21
pores of an ultrafiltration (UF) membrane is reported in Chapter 4. The influence of
the type of UF membrane (with different molecular cutoff values) on the extraction
and selective permeation of cations was investigated, using Na+ and Mg
2+ as the
monovalent and bivalent cations.
In Chapter 5, a one-pot approach to prepare a monovalent selective cation exchange
membrane by polydopamine (PDA)/PEI co-deposition is reported. PDA, known as
―bio-glue‖ can fix PEI molecular chains on the membrane surface while the side PEI
chains serve as repulsive functional groups that reject multivalent ions. With the
assistance of CuSO4/H2O2, the deposition time can be significantly reduced. The
monovalent selectivity and stability of the PDA/PEI modified membranes were
investigated in this part. For Chapter 6, Mil(53)-Al nanoparticles were incorporated
to the skin surface layer to facilitate the ions transport. Theoretically, the hydrated
radius of Na+ is around 3.0 Å, which is smaller than the Mil(53)-Al pores (8.5 Å).
Therefore, the presence of Mil(53)-Al on the thin film composite surface can provide
extra space to enhance the Na+ diffusion process.
Similarly, an interfacial polymerization (IP) method was applied by anchoring zeolitic
imidazolate framework-8 (ZIF-8) to the skin layer of thin film nanocomposite (TFN)
membranes in order to obtain monovalent selectivity in Chapter 7. Monovalent
selectivity can be achieved by interfacial polymerization while the ion flux across the
ion exchange membrane was enhanced by ZIF-8 incorporation. The transfers of
cations during the electrodialysis process were evaluated by both single salt solution
and mixed ions solutions.
Finally, the general conclusions and future perspectives are presented in Chapter 8.
Chapter 2
22
2. Methods and materials
This section outlines the materials and methods used in this dissertation. The first part
summarizes the chemicals and setups required during the experiments. Then a dry-wet
phase inversion method was introduced to explain the preparation process for porous
cation exchange membranes. After that, a method to prepare monovalent selective ion
exchange membranes based on an ultrafiltration membrane was introduced.
Furthermore, monovalent ion exchange membranes with porous surface structure
prepared by dopamine/polyethylenimine co-deposition and interfacial polymerization
were introduced. This section also covers all the characterization methods for the
obtained membranes.
2.1 Chemicals and methods
2.1.1 Chemicals
Poly(ether sulfone) (PES, Ultrason E6020P, 58000 g/mol) was purchased from BASF
(Antwerp, Belgium). Sulfonated poly(ether sulfone) (SPES) was supplied by Zhejiang
University of Technology (Hangzhou, China). Dimethyl sulfoxide (DMSO) of
analytical grade was purchased from Sigma-Aldrich (Overijse, Belgium). Dopamine
hydrochloride, hydrogen peroxide (H2O2, 30% by weight), polyethylenimine (PEI,
branched, Mn = 600, 10000 Da), sodium chloride (NaCl,99%), magnesium chloride
(MgCl2, 98%), sodium sulfate (Na2SO4, 99%), hydrochloric acid (HCl, 1 M),
sodiumhydroxide (NaOH, 98%), sulfuric acid (H2SO4,95–98%), zinc sulfuric
(ZnSO4, 99%), ammonium persulfate (APS), copper sulfate (CuSO4, 99%),
tris(hydroxymethyl) aminomethane (Tris,99.8%), dimethyl sulfoxide (DMSO, ≥
99%), m-phenylenediamine (MPD, ≥99%), trimesoyl chloride (TMC, ≥98%),
hexane (anhydrous, 95%) and aniline were purchased from Sigma-Aldrich (Diegem,
Chapter 2
23
Belgium). Basolite A100 (Mil(53)-Al) and Basolite® Z1200 (ZIF-8) used in
thisexperiment were produced by BASF (Antwerpen, Belgium) and acquired from
Sigma-Aldrich. All reagents and solvents were commercially available and used as
received. Distilled water was used throughout the thesis.
The ultrafiltration membranes used in Chapter 4 were provided by Ultra Water (USA).
Information about the membranes is given in Table 2.1.
Table 2.1 Properties of polyacrylonitrile (PAN) ultrafiltration membranes
Name Water flux (L m-2
·h-1
·bar-1
) Marker rejection (%)
PAN 200 300 80
PAN 350 1000 80
PAN 400 600 75
Pan 450 1200 75
Cation exchange membranes used in Chapter 3 (FKB, thickness = 0.1–0.12 mm; ion
exchange capacity = 1.01 meq/g) were obtained from Fumatech GmbH (Germany).
Ion exchange membranes used in chapter 5 and chapter 6 were commercial anion
exchange membranes (AEM-80045) and cation exchange membranes (CEM-80050)
purchased from Fujifilm Manufacturing Europe B.V (The Netherlands).
2.1.2 Porous ion exchange membrane preparation
In this part, the strategies of immersion precipitation and dry-casting were combined,
to control the membrane porosity with the purpose of improving the physical and
electrochemical properties of ion-exchange membranes. Briefly, 3 g of SPES was
dissolved in 17 g of DMSO at room temperature, and the mixture was stirred until the
SPES was fully dissolved. Subsequently, the polymer solution was degassed by
sonication. The degassed polymer solution was then cast on a glass with a casting
Chapter 2
24
knife to an initial thickness of 250 μm (K4340 Automatic Film Applicator, Elcometer).
The solution on the plate after casting was dried in an oven at 60 °C for different
heating times before being precipitated in deionized water. The formed membranes
were peeled off from the plate and stored in deionized water for further use.
Membranes with different charge densities were prepared by mixing SPES with
various amounts of PES. The preparation method was the same as described above.
The compositions of the prepared membranes and the fabrication conditions are
summarized in Table 2.2.
Table 2.2 Compositions of PES/SPES cation exchange membranes and preparation conditions
Membrane PES (g) SPES (g) DMSO (g) Heating time (h)
PS0 0 3 17 5
PS1 0 3 17 1
PS2 0 3 17 0.5
PS01 0 3 17 0
PS12 1 2 17 0
PS11 1.5 1.5 17 0
PS21 2 1 17 0
PS10 3 0 17 0
2.1.3 Preparation of monovalent selective ion exchange membrane
based on ultrafiltration membrane
Prior to surface modification, the UF membranes were hydrolyzed via immersion into
a 1 M NaOH aqueous solution at ambient temperature for 24 h. The membranes were
subsequently soaked in deionized water for 24 h to create hydrolyzed PAN
membranes. After that, the PAN films were cut into a specific shape and then
Chapter 2
25
immersed in 200 mL 1.0 M HCl aqueous solution containing 2.5 mL of aniline
monomer. The solution was stored in an ice-water bath for 3 h to fully absorb the
aniline monomer into the PAN film. Then, 50 mL of precooled aqueous solution
containing 2.25 g of APS was poured into the above mixture and then shaken well and
stored in the ice-water bath for 24 h. After polymerization, the modified films were
completely washed with 1 M HCl and deionized water.
2.1.4 Preparation of polydopamine/polyethyleneimine modified
monovalent selective ion exchange membrane
The method used to prepare cation exchange membranes was similar to the method
used in Section 2.1.2: 3 g SPES was dissolved in 17 g DMSO at room temperature.
After degassing by ultrasound, the solution was casted on a glass plate, with a
thickness of 250 μm. The assembled membrane was then placed in an oven at 60 °C
for 12 h and then the membrane was peeled off from the glass plate. The SPES
membranes were stored in DI water for further use.
The monovalent selective cation exchange membrane was prepared by a fast
deposition of dopamine/PEI. In detail, PEI and CuSO4 (39.9 mg) were added into
50 mL Tris buffer solution (pH=8.5). Then, dopamine hydrochloride with designed
mass content was mixed into the above solution. After dopamine hydrochloride was
fully dissolved, 0.1 mL of H2O2 was added. Subsequently, the membrane fixed in a
membrane holder was contacted with the dopamine solution for 4 h at room
temperature. After modification, the membranes were rinsed three times with DI
water and stored in water before further use. For comparison, another group of
membranes was soaked in a dopamine solution without adding PEI by using the same
conditions. During the polymerization process, Fenton-like reactions involving
Cu2+
/H2O2 can generate reactive oxygen species (ROS), including OH·, O2-
and H3O+,
which can significantly accelerate the polymerization process (Poyton et al., 2016).
Chapter 2
26
The oxidized quinone forms catechol groups that can react with amino-terminated PEI
via Michael addition/Schiff base reaction (Xu et al., 2010). Besides, Cu2+
ions in the
solution can strongly chelate with dopamine to form stable dopamine-Cu2+
complexes
(Wang et al., 2017a). The detailed mechanism of the co-deposition process is
presented in Fig. 2.1. The assigned labels with different functionalization parameters
of the modified membranes in this work are listed in Table 2.3.
Fig. 2.1 Proposed mechanisms for the reactions between PDA and PEI
Table 2.3 Surface modification parameters corresponding to the assigned membranes
Membrane PEI (mg) H2O2 (mL) Dopamine (mg) Deposition time (h)
SPES 0 0 0 0
SPES-PDA/PEI-0 0 0.1 60 4
SPES-PDA/PEI-1 60 0.1 60 4
SPES-PDA/PEI-2 120 0.1 60 4
SPES-PDA/PEI-3 180 0.1 60 4
Chapter 2
27
2.1.5 Preparation of monovalent cation exchange membrane
containing hydrophilic MIL53-(Al) framework
A CEM was first soaked in deionized water at room temperature until the membrane
was fully hydrolyzed. Afterwards, the membrane was fixed on the lab-made
membrane holder to contact the water solution. The modification process was similar
as described in section 2.1.4. The cation exchange membrane was modified by the
co-deposition of PDA/PEI and Mil(53)-Al using CuSO4/H2O2 as a trigger, followed
by cross-linking with trimesoyl chloride (TMC). PEI (120 mg) was dissolved in a
mixed solution with CuSO4 (39.9 mg) in a Tris buffer solution (50 mM, pH = 8.5).
Then different mass ratios of MIL (53)-Al (0 mg, 10 mg, 20 mg, 30 mg) were added
to the above solution followed by sonication to achieve a uniform dispersion of
nanoparticles. Finally, 60 mg of dopamine hydrochloride was dispersed to the mixture
followed by the addition of H2O2 (0.1 mL). The fresh solution was transferred to the
holder to contact with membrane for 2 h. Subsequently, the aqueous solution on the
membrane surface was replaced by an equal volume of 0.1 wt % TMC/n-hexane
solution to perform the polymerization reaction for 2 min. After the excess organic
solution was drained, the membrane surface was rinsed with water and n-hexane to
remove the remaining chemicals. Finally, the prepared membranes were air-dried for
further use. The fabrication process of monovalent selective CEMs with Mil(53)-Al is
schematically shown in Fig. 2.2.
Chapter 2
28
Fig. 2.2 Schematic diagram of the codeposition process for preparing monovalent selective CEMs
2.1.6 Preparation of monovalent cation exchange membrane by
interfacial polymerization
The pretreatment process of the Fuji-films substrate was the same as described in
Section 2.1.4. The MPD concentration was fixed at 2.0% (w/v). Organic phase
solutions that were used in these experiments were prepared via adding a specific
amount of ZIF-8 to the TMC/n-hexane solution (0.1 wt %) under sonication for 1 h.
The as-prepared membranes with different ZIF-8 loadings (0.00%, 0.02%, 0.04%,
0.06%, and 0.08% in 50 mL n-hexane solution) were designated as M-1, M-2, M-3,
M-4, and M-5. The commercial cation exchange membrane was first immersed into
the aqueous solutions for 5 min to implement the adsorption of MPD. After removing
the remaining aqueous solutions on the membrane surface, the organic phase solution
with a specific amount of ZIF-8 was poured on the surface of the membrane to carry
out the polymerization. The organic solution was removed after 2 min, and the excess
organic solution was drained at the fume hood. The fresh prepared membrane was
washed three times with n-hexane and dried completely to enhance the surface layer
stability.
Chapter 2
29
2.2 Membrane properties and characterization
2.2.1 Ion exchange capacity
The ion exchange capacity (IEC) is the number of fixed charges inside the ion
exchange membrane per unit weight of dry polymer. It is a crucial parameter that
affects almost all other membrane properties. Since the presence of large quantities of
fixed charges promotes membrane swelling, a high IEC is typically accompanied by a
high swelling degree (SD). While a high IEC tends to increase the membrane
permselectivity, a high swelling degree may reduce the effect of the IEC and
adversely affect the permselectivity. Such competing effects call for a compromise
between the IEC and the SD. The IEC is expressed in milli equivalent of fixed groups
per gram of dry membrane (meq/g) and experimentally tested through determining the
number of counter-ions after turning the CEMs into the H+ saturated form and the
AEMs into the Cl− saturated form. The IEC of the prepared (monovalent) cation
exchange membranes was detected by the titration method. The membrane was
immersed in a 1M HCl solution to saturate it with H+. After the membrane was
saturated by H+, the membrane was taken out and the remaining solution on the
membrane surface was removed by a tissue. After that, the membrane was transferred
to a 1 M NaCl solution for 24 h to liberate the H+ ions. The released H
+ ions
concentration was quantified by a 0.01 M NaOH solution. The IEC was calculated
based on the following equation:
𝐼𝐸𝐶 =𝑛𝐻+
𝑊𝑑𝑟𝑦
where 𝑛𝐻+ is the concentration of the released H+ ions and Wdry is the dry membrane
weight (g). A minimum of five measurements was obtained for each membrane to
calculate its average value.
Chapter 2
30
2.2.2 Water uptake
An ion exchange membrane consists of cross-linked polymers and forms a wet
structure when it absorbs water in a solution. The water content is of crucial
importance for the membrane dimensional stability and ionic transport properties. A
high water content implies a loose mechanical structure and often results in a poor
permselectivity, despite its positive effect on the membrane conductivity. The water
content is influenced by the membrane material, fixed charged groups, cross-linking
degree of membrane matrix and surrounding solution conditions. For example, some
AEMs with relatively low cross-linking degrees tend to have a higher water content
than their more cross-linked CEM counterparts. The water uptakes in these
experiments were measured at room temperature based on the water retention inside
the membranes. The membranes were dried at 60 °C in an oven and then weighed
accurately, after which they were immersed in deionized water for at least 24 h to
ensure complete equilibrium and then reweighed after the surface moisture had been
mopped with filter paper. The water uptake (WU) was calculated as:
𝑊𝑈 =𝑚𝑤𝑒𝑡 − 𝑚𝑑𝑟𝑦
𝑚𝑑𝑟𝑦× 100%
where mwet is the weight of IEMs in wet condition and mdry is the weight of membrane
sample in its corresponding dry phase. The results were obtained based on the mean
average value of three measurements.
2.2.3 Water contact angle
Contact-angle measurements have been frequently used to characterize the polarity or
surface energy of polymeric surfaces (Nabe et al., 1997). The water contact angle on
the membrane surface was measured by the sessile drop method using a DATA
Physics System (OCA20, Dataphysics Instruments, Germany) in which a droplet of
water on the surface was imaged by a precision video camera and displayed on a
monitor. A circle fitting analysis software was utilized to record the contact angle. A
minimum of five contact angle measurements was obtained for each membrane to
Chapter 2
31
calculate the average value.
2.2.4 Zeta potential of membrane surfaces
The zeta potential describes the interaction of the electrical surface charges with their
surroundings, although this potential is somewhat different from an actual surface
potential (Mulyati et al., 2013). As the material comes into contact with solutions,
ions from the solution rapidly migrate to the material surface in order to neutralize the
opposite charge from the materials. The charge behavior at the interfaces was
measured by the zeta potential via a Surpass 3 (Anton Paar, Austria) by using 1mM
KCl as the electrolyte solution (pH=6.0).
2.2.5 Membrane electrical resistance
Ion exchange membranes are widely used in ED for the desalination of brackish water,
the production of table salt, recovery of valuable metals from the effluents of
metal-plating industry, and for many other purposes. (Hwang et al., 1999) Membrane
properties and especially the electrical resistance of the membrane, have a significant
impact on the overall ED stack performance. The electrical resistance is directly
related to the maximum power output in reverse electrodialysis and the energy
consumption in electrodialysis. The membrane resistance is determined by the ion
exchange capacity and the mobility of the ions within the membrane matrix.
Furthermore, the operational temperature is another parameter that can impact the
electrical resistance. With increasing temperature, the electrical resistance decreases.
The specific membrane resistance is in principle reported in Ohm/cm. However, more
useful and most often reported in literature is the membrane resistance in Ohm/cm2.
Membrane resistances are often measured under direct current conditions using a
standard 0.5 M NaCl characterization solution. Using Ohm‘s law, the resistance of the
system (membrane + solution) and the membrane, can be determined. However, direct
Chapter 2
32
current measurements are not able to distinguish between the individual membrane
resistance and the additional resistances of the interfacial ionic charge transfer through
the double layer and the diffusion boundary layer. Furthermore, the membrane
resistance measured under direct current conditions becomes strongly concentration
dependent and strongly increases with decreasing concentration at lower
concentrations (Długołęcki et al., 2010a; Długołęcki et al., 2010b).
Electrochemical impedance spectroscopy (EIS) is a powerful technique to provide
valuable information on the electrochemical properties of the membrane system
(Nikonenko and Kozmai, 2011). The resulting voltage drop over the system is
measured as a function of time U(t) and the phase shift φ (◦) relative to the input
signal is determined. The voltage and current variation with time was defined as:
𝑈(𝑡) = 𝑈0 sin 𝑤𝑡 = 𝑈0𝑒𝑗𝑤𝑡
𝐼(𝑡) = 𝐼0 sin(𝑤𝑡 + 𝜑) = 𝐼0𝑒𝑗(𝑤𝑡+𝜑)
Here, U0 and I0 is the voltage and alternating current in phase (A), j is the imaginary
unity (j = √−1). The symbol w is the circular velocity (1 rad/s) which is also referred
to as circular frequency of the alternating current (Zhao et al., 2017).
By using Euler‘s formula:
𝑒𝑗𝑤 = 𝑐𝑜𝑠φ + 𝑗𝑠𝑖𝑛φ
The impedance Z can then in accordance with Ohm‘s law be calculated as:
𝑍 =𝑈𝑡
𝐼𝑡=
𝐼0𝑒𝑗𝑤𝑡
𝐼0𝑒𝑗(𝑤𝑡+𝜑)= |𝑍|𝑒−φ𝑗 = |𝑍|𝑐𝑜𝑠φ − j|𝑍|sinφ
These fixed charged groups attract ions with opposite charge from the salt solution,
which are distributed over the membrane surface and form the electrical double layer.
The interfacial ionic charge transfer from the solution phase through the electrical
double layer to the membrane is referred to as the electrical double layer resistance.
The thickness of this layer is typically in the order of nanometers (Debye length).
(Strathmann, 2004)
Chapter 2
33
When a current passes through an ion exchange membrane, charge is transported
through the membrane by counter ions as a result of Donnan exclusion. In the bulk
solution, current is carried by both positive and negative ions. The difference in ion
transport number between the bulk solution and the membrane results in the building
up of diffusion boundary layers at the membrane surface. (Krol et al., 1999; Sistat et
al., 2008) The concentration decreases at one side of the membrane and increases at
the other side of the membrane; this phenomenon is called concentration polarization
and occurs within the diffusion boundary layer.
EIS allows distinguishing between these different layers because the different layers
respond differently to the applied signal (current) at different frequencies. At the high
frequency range when there is no phase shift between voltage and current and the
Ohmic relation holds, the response of the single membrane can be extracted from the
EIS measurements. (Coster et al., 1996; Park et al., 2006) In principle this resistance
represents the resistance of the membrane containing the solution resistance (RM+S),
but the pure membrane resistance (RM) can be easily extracted by subtracting the
solution resistance (RS) as determined from a blank measurement without a membrane.
When the frequency is decreased, the contribution of the interfacial ionic charge
transfer from the solution phase through the electric double layer to the membrane can
be extracted. Ions start to migrate through the interfacial double layers and a phase
shift is observed. The resistance (RDL) and capacitance (C) of ionic transport through
these double layers becomes visible. This surface charging is similar to what is
observed for a capacitor and the interfacial ionic charge transfer through the double
layer is represented in the equivalent circuit model as a resistor and capacitor in
parallel. When a very low frequency signal (current) is applied, in addition to the
membrane and the ionic transfer through the electrical double layer, concentration
gradients in the diffusion boundary layers adjacent to the membrane become visible.
At these low frequencies, ions are transported through the membrane, the double layer
and the diffusion boundary layer (Fig. 2.3) and the system responds with a certain
Chapter 2
34
delay to the applied signal (Długołęcki et al., 2010b).
Fig. 2.3 Phenomena occurring in the cation exchange membrane and in the layer adjacent to the
membrane (RM is the membrane resistance, RDL is the resistance of the interfacial ionic transfer
from the solution through the double layer into the membrane, C is the capacitance of the
interfacial ionic charge at the membrane surface (double layer), RDBL is the diffusion boundary
layer resistance and Q is the constant phase element representing a non-ideal capacitor of the
diffusion boundary layer.) (Długołęcki et al., 2010b)
The membrane resistance was measured with a Solartron Electrochemical System by
electrochemical impedance spectroscopy (EIS) over a frequency range from 1 kHz to
1 MHz. The conductivity cell was filled with a NaCl or a MgCl2 solution in each
compartment with an effective area of 1 cm2.
2.2.6 Diffusion dialysis experiments
Conductivity-time curves were used to determine the concentration of ions through
diffusion. These curves reflect the ion concentration directly. Here, diffusion dialysis
experiments with a 1 M NaCl diluated compartment cell and an initial concentrated
Chapter 2
35
compartment cell with deionized water were carried out. The concentration gradient
drives ions across the membrane, which has an effective area of 13.84 cm2.
The dialysis coefficient of the membrane for each component is defined as the amount
of the component transported per unit active membrane area, per unit time, per unit
concentration difference of the component and was calculated from the concentration
of species according to the following equation:
𝑈 =M
A · t · ∆C
where M is the amount of the component transported (moles); A is the effective area
(m2); t is the time (h); and ΔC is the logarithmic average concentration between the
two chambers (moles per cubic meter); given by:
∆𝐶 =(𝐶𝑓
𝑜 − 𝐶𝑓𝑡 + 𝐶𝑑
𝑡)
ln𝐶𝑓
0
𝐶𝑓𝑡 − 𝐶𝑑
𝑡
where 𝐶𝑓0 and 𝐶𝑓
𝑡 are the feed concentrations at times 0 and t, respectively, and 𝐶𝑑𝑡
is the dialysate concentration at time t.
2.2.7 Current-voltage and transport number measurements
It has been reported that a cation-exchange membrane modified by formation of a
cationic layer on the membrane surface is preferentially permeable to cations of lower
rather than higher valency and to smaller hydrated cations rather than larger ones. A
method frequently used to characterize the transport properties of cation-exchange
membranes is to study the current-voltage curves corresponding to the membrane
system. The current-voltage curves are associated with the well-known concentration
polarization phenomenon, which is arising at the interface between an ion-exchange
membrane and an electrolyte solution when an electric current passes through the
system. This phenomenon has been widely studied in the last decades with the
purpose of establishing the factors that determine it and resolving the problems that
Chapter 2
36
concentration polarization causes in membrane technology (Mafé et al., 2003;
Rubinstein et al., 1988).
The concentration polarization for cation exchange membranes is due to the fact that
all the charge is carried by cations in the membrane, when in solution the same charge
is carried by cations and anions. In the diluate compartment, the concentration
becomes lower at the membrane surface (Cm) than in the bulk solution (Cb), and in the
concentrate compartment, at the membrane surface, concentration becomes higher
(Cm) (Fig. 2.4). The concentration polarization generates diffusive transport and
creates diffusion boundary layers at membrane surfaces. According to the
concentration polarization theory, the electric current circulating through the
membrane system increases linearly with the voltage increase and eventually reaches
a limiting value, Ilim. In the case of a cation exchange membrane this value is
expressed by:
𝐼𝑙𝑖𝑚 =F · D · 𝐶𝑏
∆𝑡+ · 𝛿
where 𝐶𝑏 is the bulk concentration of the cations, δ is the thickness of the diffusion
boundary layer, D is the diffusion coefficient of the cations, Δt+ is the difference
between the cation transport number in the cation exchange membrane and in the
solution, and F is Faraday‘s constant.
Fig. 2.4 Concentration polarization for cationic exchange membrane (Chamoulaud and Bélanger,
2005)
Chapter 2
37
In practice, the current–voltage curve has a characteristic shape and clearly shows
three regions (Fig. 2.5). First, the Ohmic region that is observed at low current density,
the system resistance could be approximately attributed to ionic transport into the
cation-exchange membrane (Rohm). This region is followed by the current-limiting
region in which current density varies very slowly with the potential to form a
pseudo-plateau. In accordance with the concentration polarization theory, the limiting
current value can give information about the thickness of the diffusion boundary layer,
the diffusion coefficient, or the cation transport number in the membrane. Third, the
current plateau is followed by the electroconvection region, in which the slope of the
current–potential curve increases again (Rec). Those currents larger than the limiting
current are not expected according to the classical theory of concentration polarization
(Barragan and Ruız-Bauzá, 1998).
Fig. 2.5 Typical current-voltage curve for a cation exchange membrane (Chamoulaud and
Bélanger, 2005)
The current-voltage curves of monovalent selective ion exchange membranes were
measured at 25 °C through a two-compartment measuring technique (Fig. 2.6). The
effective monovalent selective ion exchange membrane area was about 13.84 cm2 and
Chapter 2
38
the working electrolytes of electrode were 0.5 mol/L NaCl solutions. The current
across the set-up was supplied by a direct current power system (DF1720SB5A,
Zhejiang Zhongce Electronic Co., Ltd. China). The voltage drop across the membrane
was tested by a voltmeter (ZW1418, Qingdao Qingzhi Instruments Co., China) with
two platinum filaments as electrode probes. The membrane was placed between the
two half-cells, which were equipped with mechanical stirrers in each compartment.
During the experiments, no obvious variation of current voltage curves can be
observed concerning effect of water splitting caused by the electrode reaction.
Fig. 2.6 Schematic diagram of a two-compartment electrolytic cell used for current–voltage
measurement
Membrane transport numbers were determined by measuring membrane potential. A
two-cell apparatus equilibrated a membrane with unequal concentrations (C1 = 0.05
M/C2 =0.01 M) of NaCl solution at both sides were used. The developed potential
across the membrane was measured by multimeter with Ag/AgCl electrodes. During
the experiments, both sections were stirred vigorously to minimize the effects of
boundary layers. The transport number tm was then calculated using the following
equation:
𝐸𝑚 =RT
F(2𝑡𝑚 − 1)𝑙𝑛
𝑎1
𝑎2
where R is the universal gas constant (8.314 J/ (mol·K) ), F is the Faraday constant
Chapter 2
39
(96487 C/mol), T is the absolute temperature (K), and a1 and a2 are the mean activities
of the electrolyte solutions. The mean activity can be expressed as:
𝑎𝑖 = 𝑟𝑖 × 𝑐𝑖
where 𝑐𝑖 is the concentration, and 𝑟𝑖 is the activity coefficient. The mean activities
of the electrolyte can be calculated based on salt concentration. The activity
coefficients were deemed as 1 in the experiments due to the low concentrations.
2.2.8 Electrodialysis experiments
The desalination and selective properties of the membranes were evaluated by ED.
ED is typically operated with a direct current supply, which serves as the driving force
for ions migration. There were three streams in the ED stack: the diluate, the
concentrate, and the electrode rinsing solution. For Chapter 3 and 4, both the
concentrate and diluate compartments were filled with 1 L of a 2 g/L salt solution
while the electrode rinsing solution was 2 L 20 g/L Na2SO4. The experiments were
conducted at constant current conditions and the current applied in Chapter 3 was
0.3 A. The ED stack shown in Fig. 2.7 is applied in the experiments for Chapter 3.
The ED stack applied in Chapter 4 is similar with the setup used in Chapter 3, while
anion exchange membranes in the stack were replaced by cation exchange membranes.
Each compartment had an active membrane area of 64 cm2.
Chapter 2
40
Fig. 2.7 Scheme of a traditional ED setup
For Chapter 5, an ED setup with an active membrane area of 28 cm2 was used. The
concentrate compartment was filled with 150 mL 0.25 M H2SO4 and the diluate
compartment was filled with 150 mL mixed solution (7.5 g ZnSO4 in 1 L 0.25 M
H2SO4 solution). In Chapter 6 and Chapter 7, the diluate and concentrate compartment
were filled with 150 mL 2 g/L salt solutions to analysis the separation performance of
the membranes. For permselectivity experiments, a solution with 0.05 M MgCl2 and
0.5 M NaCl was used in the dilute compartment, and a 0.5 M NaCl was used in the
concentrate compartment. Each compartment has an active membrane area of 19.6
cm2 with an O-ring rubber to avoid leakage during testing. A constant voltage of 20 V
was applied in Chapter 6 for studying desalination properties. The other experiments
were conducted with a constant current of 0.3 A. The electrode rinsing solution was
1 L 20 g/L Na2SO4. The error bar of conductivity-time curves in Chapter 3 and
Chapter 4 indicates an average level of at least three measurements.
The cation flux through the membranes was determined by the concentration change
with time (dCC/dt) in the concentrate compartment according to the following
equation:
𝐽𝑐(𝑚𝑜𝑙/𝑐𝑚2 ∙ 𝑠) =V
𝐴(𝑑𝐶𝐶
𝑑𝑡)
where A is the membrane effective surface area (cm2) and V is the volume (cm
3) of
Chapter 2
41
the concentrate solution.
The desalination efficiency was evaluated in terms of the change in conductivity of
the diluate compartment during the electrodialysis operation. The desalination
efficiency (DE) was calculated using the equation:
𝐷𝐸(%) = 1 −𝜎𝑡
𝜎0
where 𝜎0 is the initial conductivity of the diluate compartment and 𝜎𝑡 is the
conductivity of the diluate compartment at time t.
The permselectivity of a membrane describes the charge selectivity of the ion
exchange membrane. It reflects the ability of the membrane to discriminate between
ions with same charge. The permselectivity of the membranes in Chapter 5 and
Chapter 6 was expressed as follows:
𝑃𝑁+𝑀2+
=𝑡𝑀2+ × 𝐶𝑁+
2𝑡𝑁+ × 𝐶𝑀2+
where 𝑡𝑀2+ and 𝑡𝑁+ are the transport number of multivalent and monovalent ions.
𝐶𝑀2+ and 𝐶𝑁+ are the concentrations of multivalent and monovalent ions in the
diluate compartment. The permselectivity in the other experiments is simply
expressed as the flux ratio of multivalent and monovalent ions.
Current efficiency is an important parameter for assessing the suitability of any
electrochemical process for practical applications. The overall current efficiency (η)
was defined as:
𝜂 =z(𝐶0𝑉0 − 𝐶𝑡𝑉𝑡)F
NIt× 100%
where C0 and Ct are the concentrations of the dilute solution at times 0 and t,
respectively; z is the valence of the ions; V0 and Vt are the volumes of dilute solution
circulated at times 0 and t, respectively; F is the Faraday constant (96500·C·mol−1
); I
is the constant current, N is the number of repeating units; and t is the time allowed
for the electrochemical process.
Chapter 2
42
2.2.9 Structural stability of ion exchange membrane
The stability of the modified layers plays a crucial role in practical applications. The
structural stability was investigated in the light of the performance deterioration. The
membrane was soaked in both strong acid (0.1M HCl) and base environment (0.1M
NaOH) for 7 days. Then, the membranes were taken out and rinsed with DI water to
wipe off residual H+ and OH
-. The permselectivity and ions flux of the membrane
after acid and base treatment were recorded though the aforementioned method.
2.2.10 Morphology and structure of membranes
The cross-sectional and surface morphologies of the resultant membranes were tested
by a field emission scanning electron microscope (Philips XL30-FEG). The samples
for surface imaging were prepared by directly cutting a clean membrane into small
pieces, whereas the samples for cross section were prepared by freezing and breaking
samples in liquid nitrogen. Moreover, the surface morphology and roughness of the
prepared membrane was further analyzed by atomic force microscopy (AFM,
Dimension Icon, Bruker, Germany) with scan areas of 1×1 μm. Approximately 1 cm2
of the prepared membranes was cut and glued on the glass substrate before being
scanned. AFM images were taken in ScanAsyst mode using a FASTSCAN-B probe.
The AFM images were flattened after scanning to remove slope and curvature from
images. After flattening, the root-mean-squared roughness (Rrms) was calculated as:
R𝑟𝑚𝑠 = √∑ (𝑍𝑛 − 𝑍𝑎𝑣𝑒)2𝑁
𝑛=1
𝑁 − 1
where 𝑍𝑎𝑣𝑒 is the arithmetic mean of the height values for all the pixels in the image,
𝑍𝑛 is the height for any given pixel and N is the number of pixels present in the
image. Furthermore, other definitions can also be used to characterize the roughness,
such as the mean roughness (Ra, the mean value of the surface relative to the center
plane) or the peak-to-valley distance (Rm, the distance between the highest data point
Chapter 2
43
and the lowest data point of the surface).
2.2.11 Chemical structure and composition of membranes
The functional groups on the membrane surfaces were verified by Fourier transform
infrared spectroscopy (FTIR, Nicolet Magna-IR 560 Spectrometer). The transmittance
spectra were conducted from 670 to 4000 cm-1
with a resolution of 4 cm-1
at room
temperature. All detections were performed by using air as the background. The
surface chemical compositions of the membranes and the materials were analyzed by
X-ray photoelectron spectroscopy (XPS, AXIS Ultra DLD X-ray photoelectron
spectrometer, Japan) with Mg Kα as the radiation source. The take-off angle of the
photoelectron was set at 90º. Survey spectra of the membranes were collected over a
range of 0 to 1300 eV. The energy dispersive X-ray spectroscopy (EDS)
compositional analysis of membranes surface was performed by a field emission
scanning electron microscope (JEOL Model JSM-6700F, Tokyo, Japan).
2.2.12 Water flux experiments
A dead-end filtration setup was used to study the permeability of the resulting
membranes in Chapter 3. The effective area of the filtration cell was 12.64 cm2. The
water flux was tested at a pressure of 4 bar. The water flux F, used to express the rate
at which the water permeates a membrane, is typically defined as a volume per unit
area per unit time.
F =𝑉
𝐴𝑡
where V is the volume of pure-water permeation, t is the time, and A is the area of the
membrane. A minimum of three measurements was obtained for each membrane to
calculate the average value.
Chapter 3
44
3. Cation exchange membranes with controlled
porosity in electrodialysis application
Adapted from: J. Li, J. Zhu, S. Yuan, J. Lin, J. Shen, B. Van der Bruggen.
Cation-Exchange Membranes with Controlled Porosity in Electrodialysis Application.
Industrial & Engineering Chemistry Research, 28 (2017): 8111-8120.
3.1 Introduction
In general, a homogeneous or heterogeneous ion exchange membrane is a dense
membrane, which defines the transport of ions. Although many strategies have been
developed to accelerate ion transport, these strategies were not economically
competitive or resulted in inferior electrochemical performance (Malik et al., 2016).
Recently, ultrafiltration membranes were used in the electrodialysis process to
facilitate the migration of molecules according to their charge and molecular size.
However, during the application of ultrafiltration membranes in electrodialysis, the
separation of smaller ionic species still remains a challenge due to the relatively large
pore size of these membranes. Pore-filled membranes composed of an ultrafiltration
membrane as the substrate and a polymer with ion-exchange groups can provide both
a high ion conductivity and excellent mechanical properties (Kim et al., 2016) The
application of pore-filled ion exchange membranes has been explored for use in fuel
cells (Lee et al., 2016b), electrochemical energy conversion (Lee et al., 2016c),
vanadium redox flow batteries (Lee et al., 2016c), reverse electrodialysis (Lee et al.,
2017), acid recovery by diffusion dialysis (Lin et al., 2017) and pharmaceutical
preparations (Åkerman et al., 1998). In spite of this, pore filling may have some
restrictions in practice for the following reasons: it is accepted that the introduction of
ion exchange groups onto a membrane is the most effective and practical method for
the preparation of ion exchange membranes; however, the presence of the uncharged
Chapter 3
45
porous substrate would hinder the ion transport through the membranes. Inspired by
methods of electrodialysis with ultrafiltration membranes and preparing pore-filled
membranes, porous membranes prepared from charged polymers with functional
groups represent another option. A porous ion exchange membrane can be defined as
a charged membrane with functional groups and charges. Such membranes with
porous structures might facilitate the migration of ions by reducing the physical steric
hindrance of the surface layer. Thus, the migration resistance could be further
diminished, whereas the functional groups of the membrane matrix could block the
diffusion of ions caused by the concentration gradient. However, the fabrication of
porous ion exchange membranes is still a major challenge limiting the optimization of
the membrane performance referring to pore size, porosity and resistance.
3.2 Results and discussion
3.2.1 SEM results and water flux
The properties of the membrane, such as IEC, water uptake, resistance and
mechanical stability, can be modified by varying preparation conditions. Fig. 3.1
shows the morphologies of the top surface and cross section of membranes prepared
with different heating times. No obvious differences can be seen on the membrane
surface. However, the difference in the cross section is more expressed. With
increasing heating time, membrane surfaces were found to become denser. A
membrane prepared by solvent evaporation generally has a dense surface structure
without any pores, as shown in Fig. 3.2(a1). On the other hand, the membranes
prepared by phase inversion in DI water without curing have a relatively large pore
size, as shown in Fig. 3.2(a4). From the images of samples Fig. 3.2(a2 and a3), it can
be seen that membranes fabricated with the dry-wet phase inversion method have
three layers: two nanoporous layers with a macroporous layer in between. The only
difference is that the membrane with 1 h heating time had a comparatively dense inner
Chapter 3
46
layer that was layered in shape, whereas the membrane with 0.5 h heating time had an
irregular porous inner layer. This is because the polymer concentration in the surface
layer increased as the solvent evaporated, thus leading to a dense surface layer. This
layer on the membrane surface would inhibit the exchange solvent and non-solvent
during the immersion process and facilitates the formation of a thick skin layer with
smaller pores (Buonomenna et al., 2007; Jansen et al., 2005).
Fig. 3.1 SEM images of membranes prepared with different heating time (a1 – a4: cross section of
membranes prepared by heating time of 5 h, 1 h, 0.5 h and 0 h, respectively; b1 – b4: surface
images of membranes prepared by heating time of 5 h, 1 h, 0.5 h and 0 h, respectively)
The water flux is an important parameter in the practical application of polymeric
porous membranes. It is mainly influenced by structural parameters such as pore size
and porosity, with greater pore size and porosity resulting in an enhanced water flux
Fig. 3.2 shows the water flux of porous ion exchange membranes prepared with
different heating times. The results reflect the pore size and porosity to some extent.
As can be seen in Fig. 3.2, the water flux was negligible for dense membranes. This
can be explained by the fact that a dense membrane is not capable of providing
pathways for water transport. For the membrane prepared by the dry-wet phase
inversion method, the water flux was enhanced by the porous interlayer. The structure
a1
a4a3
a2b1
b2
b3
b4
Chapter 3
47
of this interlayer explains the differences in water flux for membranes prepared with
heating times of 0.5 and 1 h. For the membranes prepared by phase inversion, the
highest water flux (of about 43.0 L/(h·m2)) was obtained, which was due to the
increased pore size and porosity throughout the membrane matrix. These water fluxes
are consistent with SEM images of membranes prepared with different heating times.
Fig. 3.2 Effect of heating time on water flux for porous ion exchange membranes
3.2.2 IEC and water uptake
The content of charged functional groups in an ion exchange membrane plays a key
role in providing a hydrophilic and electro-static environment for ion transport. The
ion-exchange capacity, an important factor related to the conductivity and transport
properties, is used to identify the charge density of the membranes. The IEC values of
the as-prepared membranes are shown in Fig. 3.3a. Compared to the membrane
directly immersed in water, membranes prepared by dry-wet phase inversion had a
smaller IEC. This is mainly due to the following reason. The pores inside the porous
ion exchange membrane could stock some acid solutions inside the membrane matrix.
The membrane with higher porosity tends to increase the released H+ concentration,
thus a higher IEC value can be obtained. Incorporating SPES to the porous membrane
Chapter 3
48
is expected to improve the IEC and water uptake. Water uptake is known to have a
profound effect on the membrane conductivity and flux. The water uptake is
significantly improved after the addition of charged SPES (see Fig. 3.3b). The
elevated water uptake is ascribed to the increase of the SPES content in the PES
membrane matrix. The sulfonate groups of SPES are hydrophilic, and an increase of
the IEC from 0.92mmol/g to 1.79 mmol/g with increasing SPES content resulted in
ahigher water content. The water uptake for SPES membrane without PES is around
164%, which is much higher than PES membrane without SPES (132%).
Fig. 3.3 IEC and water uptake change with heating time and radio of PES/SPES
3.2.3 Contact angle measurements
The IEC and water uptake were reduced with increasing heating time because of the
decreased porosity in the polymer matrix. This is consistent with the results of contact
angle measurements, as can be seen in Fig. 3.4, a membrane with high porosity tends
to have a higher surface hydrophilicity. Moreover, the component of membrane
matrix would also have an impact on surface hydrophilicity. By comparing PS01with
PS10, contact angle reduced from 75° to 53°, indicating that a higher surface
hydrophilicity was obtained.
Chapter 3
49
Fig. 3.4 Contact angle change with heating time and ratio of PES/SPES
3.2.4 Membrane resistance and transport number
Compared to porous membranes, dense membranes result in a higher electrical
resistance. On the contrary, membranes prepared by immersion and precipitation have
the greatest pore size and porosities, thereby giving rise to a higher transport number
(0.99). For membranes with different heating time in this work, the order of the
porosity was found to be as follows: 0 h > 0.5 h > 1 h > 5 h. In this case, higher
porosities with higher transport numbers followed the order of 0 h > 0.5 h > 1 h > 5 h.
The resistance of the membranes followed the order of 0 h < 0.5 h < 1 h < 5 h, which
conformed that membranes with higher porosity tended to have a lower resistance.
Table 3.1 Resistance and transport number of membrane with different heating time
PS0 PS1 PS2 PS01
Membrane resistance (Ω cm2) 1.92 1.64 1.27 0.97
Transport number (tm) 0.89 0.92 0.98 0.99
3.2.5 Diffusion dialysis
In diffusion dialysis, solutes pass through an ion exchange membrane from the high to
the low concentration side. Fig. 3.5 shows that the conductivity in the concentrated
cell increases with time during the diffusion process; this was caused by salt diffusion
Chapter 3
50
from the concentrate chamber (1 M NaCl) to the diluate chamber (distilled water).
However, the migration rates for the membranes prepared with different heating times
were quite different. Only a few salts migrated though the dense membrane, whereas
migration through the porous membranes was evident. Dialysis coefficients were
calculated based on changes in concentration in both compartments of the dialysis cell.
The dialysis coefficient of membrane directly immersed in water was found to be
108 mol/(h·m2) at room temperature. As the heating time during membrane
preparation increased, the dialysis coefficient decreased. The dialysis coefficient of a
dense membrane reached as low as 1.4 mol/(h·m2), much smaller than that of the
membrane directly immersed in water. It can be concluded from the slope of the
conductivity–time curves that a porous membrane with smaller pore size and
comparatively dense surface tended to have a lower diffusion of NaCl. This is due to
the steric hindrance effect of the dense surface. The dialysis coefficients of the
different membranes were 1.4, 4.9, 8.6, 108 for dry, 1 h heating time, 0.5 h heating
time, and 0 h heating time membrane, respectively, which are consistent with the
diffusion dialysis results.
Fig. 3.5 Conductivity change of concentrated compartment during diffusion progress for
membrane with different heating times
0 10 20 30 40 50 60
0
1
2
3
4
5
6
T (min)
Co
nd
uctivity (
ms/c
m)
5 h
1 h
0.5 h
0 h
Chapter 3
51
3.2.6 Electrodialysis experiments
As presented in Fig. 3.6, the conductivity of the diluate compartment decreased during
the experiments, but the demineralization rates were different. For the membrane
dried directly, the conductivity change was much faster than for the porous ion
exchange membrane in the first stage. An increasing heating time led to a membrane
with lower porosity and dense structure, so that salt diffusion from the concentrate to
the diluate compartment driven by the concentration gradient was mitigated, and as a
result, the conductivity change was enhanced. For an heating time of 1 h, the
conductivity change during the first 100 min was similar to that for the membrane
dried directly. However, after that, it is found that the demineralization rate of the
dense membrane was surpassed by that of the membrane prepared with 1 h heating
time. In this case, the increased demineralization rate can be explained by the
comparatively high transport number for ions. For the membrane prepared with 0.5 h
and 0 h heating time, the porosities were larger. In this case, the effect of diffusion by
the salt gradient plays a much more important role, thus, the conductivity changes
were smaller. The trend of the conductivity-time curves for the membrane prepared
with 0.5 h and 0 h heating time was similar to that of the membranes prepared with 1
h and 5 h heating time. The conductivity change was similar when the conductivity
was greater than 1.6 mS/cm. After that, the desalination efficiency of the membrane
with 0.5 h heating time became higher. This can be explained by the fact that the
membrane with 0.5 h heating time had a smaller porosity; as a result, steric hindrance
plays an integral part in hindering the salt diffusion.
Chapter 3
52
Fig. 3.6 Conductivity change of diluate compartment during ED process for membrane with
various heating times
The SPES content was also found to have a great influence on the overall properties
of the porous ion exchange membrane. For the membrane prepared with different
SPES content, the changes in conductivity of the SPES/PES composite membranes
are shown in Fig. 3.7. It can be concluded that the desalination efficiency was
enhanced with increasing SPES content. This can be explained by the fact that an
increase in the SPES content enhanced the IEC and water content. As a result, ion
transport was facilitated and the conductivity reduction of diluate compartment
became much more apparent.
0 50 100 150 200 250 300
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Time (min)
Dry
1 h
0.5 h
0 h
Con
du
ctivity (
ms/c
m)
Chapter 3
53
Fig. 3.7 Conductivity change of diluate compartment during ED process for membrane with
different SPES content
To evaluate the performance of the membranes prepared with different heating time
for brine recovery, the current efficiency of the ED stack was calculated for the case
when the conductivity of the diluate compartment reached 1.0 mS/cm, as shown in
Fig. 3.8 (right figure). The current efficiency increased first from 90% to 100%
because of the formation of pores inside the membrane that facilitated the transport of
ions; however, as heating time reduced, the porosity increased further, and the current
efficiency decreased again. This can be explained by the fact that a further increase of
the porosity would enhance the diffusion resulting from the gradient of salt
concentration. Furthermore, the transport of chlorides also contributed to the decrease
of current efficiency. To investigate the influence of the SPES content, the current
efficiency was calculated when the desalination process was conducted for 180 min. It
was found that the SPES content had a profound effect on the current efficiency. With
the increase of SPES, the IEC and was increased. The enhanced IEC could hinder the
transfer of NaCl and Cl- from the concentrate to dilute compartment, thus resulting in
an enhanced current efficiency.
Chapter 3
54
Fig. 3.8 Current efficiency of ED process at different heating times and SPES content
3.3 Conclusions
The combination of immersion precipitation and dry-casting to control the
membrane porosity is an effective method for improving the physical and
electrochemical properties of the resulting membrane. The membrane resistance can
be largely reduced by introducing pores into the membrane matrix. The modified
membranes show an improved IEC and water uptake and a decreased contact angle.
The current-voltage curve and diffusion experiments confirmed that the resistances of
membranes with higher porosities were reduced, and thus, the diffusion of salts
through membranes was also enhanced. A compromise between the membrane
resistance and diffusion should be made to optimize the mechanical stability and
separation capability of the membrane. During desalination by ED, changes in
conductivity were obvious for the membrane with 1 h heating time and for the dense
membrane. However, for the membrane directly immersed in water and for the
membrane prepared with a 0.5 h heating time, the desalination efficiencies were lower;
this can be explained by the enhanced salt diffusion of membrane with high porosities.
Compared with the dense membrane, the membrane prepared with 1 h heating time
had a higher desalination efficiency. This trend was similar to the trend in the current
efficiency for the membranes prepared in this work. The desalination efficiency
reached 95% and the current efficiency was around 100% under optimized conditions.
Chapter 3
55
This chapter provides new insights into how to develop porous ion exchange
membranes and lays a foundation for further research on low resistance membranes.
Chapter 4
56
4. Charge-assisted ultrafiltration membranes for
monovalent ions separation in electrodialysis
Adapted from: J. Li, J. Zhu, J. Wang, S. Yuan, J. Lin, J. Shen, B. Van der Bruggen.
Charge-assisted ultrafiltration membranes for monovalent ions separation in
electrodialysis. Journal of Applied Polymer Science. 135(24), 2018:45692.
4.1 Introduction
As mentioned in Chapter 1, monovalent selective ion selective membranes have the
capability to separate monovalent ions from a solution containing both monovalent
and multivalent ions. Different approaches for the fabrication of anion exchange
membranes with permselectivity for monovalent ions have been reported, including
the adjustment of the crosslinking degree. Nevertheless, increasing the degree of
crosslinking would reduce the process efficiency. For this reason, crosslinking is not a
suitable approach for the fabrication of monovalent selective ion exchange
membranes (Kumar et al., 2013). Sieving of ions by the formation of a highly
crosslinked surface layer was found effective in changing the permselectivity between
monovalent and divalent ions (Sata, 1994). The thickness of this layer should be
optimized to balance the tradeoff between the monovalent-ion selectivity and the
electrical resistance (Güler et al., 2014). Recently, electrodialysis with nanofiltration
(EDNF), a process combining electrodialysis with nanofiltration was used for the
separation of multivalent ions from monovalent ions (Ge et al., 2016a). By replacing
the cation exchange membrane with composite ion exchange membrane, the porous
composite membrane can fractionate ions by rejecting the multivalent ions with larger
Stokes radius. In addition, the permselectivity and ion flux were enhanced due to the
porous structure of the nanofiltration membrane.
Chapter 4
57
It should be noted that a similar technology named electrodialysis with ultrafiltration
(EDUF) has been used for selective isolation of proteins (Doyen et al., 2014) and
peptides (Doyen et al., 2011a, b; Firdaous et al., 2009). In this process, a conventional
electrodialysis cell is used, in which some ion exchange membranes are replaced by
ultrafiltration membranes. The common characteristics of these two technologies are
that an electrical field is used as the driving force to separate molecules on the basis of
their electrical charge and size. However, due to the lower molecular cut-off of NF
membranes, UF membranes have a lower resistance compared to NF membranes,
which accelerates the migration of ions (Bazinet and Moalic, 2011a). Ultrafiltration
(UF) membranes are a type of porous and asymmetrical membranes prepared via
phase inversion, having a very thin and nanoporous skin layer and a thick and
macroporous supporting layer (Guillen et al., 2011). The typical molecular weight
cut-off (MWCO) of UF membranes is above 1 kDa, making it extremely challenging
to remove monovalent salts (Mukherjee et al., 2015). By modifying monomers on the
surface or within the pores of a robust microporous host membrane, the modified
porous membrane provides the mechanical strength to mitigate the impact of osmotic
forces. Lin et al. (Lin et al., 2016) modified asymmetrical porous ultrafiltration
membranes by blocking the pores at the top surface. The membranes exhibited a high
acid permeability coefficient, of about 0.041-0.062 m h-1
and a separation factor
(H+/Fe
2+) of 30.4-84.4, which is superior to most reported membranes. With the
difference in ion transport mechanism in the membrane caused by the special
membrane microstructure, the results suggest a significant improvement in salt
diffusion properties; however, such superior properties were never explored in
electrodialysis (Lin et al., 2016).
Polyaniline (PANI), a low cost material with high conductivity in partially oxidized
state, and a high chemical stability in acid medium (Zhu et al., 2016), is facile to be
constructed to different structures such as foams (Xie et al., 2016) and nanorods (Fu et
al., 2016). With PANI, elevated separation properties can be obtained based on the
size of the molecules or ions and their charge. This has been widely used in composite
Chapter 4
58
membranes and surface modification of ion exchange membranes. The influence of
polymerization time, surface charge density and doping agents (Farrokhzad et al.,
2015; Nagarale et al., 2004; Sata et al., 1999) on the electrochemical properties of
membranes has been comprehensively investigated by previous studies. In this
Chapter, a facile and inexpensive method by modifying PANI on the surface of UF
membrane is reported. PANI grown along the network of a UF membrane can
enhance transport, hinder dialysis and ensure an efficacious utilization of energy. With
the inclusion of PANI, separation properties based on the size of the molecules and
ion charge were investigated in electrodialysis for desalination of NaCl and MgCl2
solutions. The influence of the type of UF membrane (with different molecular cut-off
values) on the extraction and selective permeation of cations was investigated, using
Na+ and Mg
2+ as the monovalent and bivalent cations.
4.2 Results and discussion
4.2.1 SEM results
SEM images of the membrane surface were taken to study the morphology changes of
the membrane surface caused by aniline modification. Fig. 4.1 shows the SEM images
of primary and modified PAN 350 ultrafiltration membranes. It can be seen that
before treatment, the pores of PAN 350 are uniformly distributed on the surface. After
the polymerization reaction, the color of the membrane turned from white to dark blue,
which indicates that the surface of PAN 350 was fully covered by polyaniline. In the
SEM images, it can be seen that most PANI is uniformly grown on the surface of PAN
350 membranes. PANI was found to have a unique cluster morphology, and appeared
to be growing on a substrate. Although there were defects on the surface, where the
surface was not uniform after modification, all the pores were occupied. The presence
of polyaniline can greatly shift the transport properties of the membranes, which will
be further discussed.
Chapter 4
59
Fig. 4.1 Surface morphology of PAN 350 (Left) and modified PAN 350 membranes (Right)
4.2.2 FTIR results
Theoretically, PANI has the potential to form hydrogen bonds with polymers
containing carbonyl groups. Because of its H-donating imine groups, PANI has been
found to be compatible or at least partially compatible with some polymer counterpart
such as polycarbonate and nylon 6 (Ouyang et al., 2008; Pan et al., 2005). FTIR
spectroscopy is an extremely useful and convenient technique to detect the formation
of hydrogen bonding. PAN 350 membranes often have a peak at 1666 cm−1
, which is
assigned to the vibration of the C=O bonds derived from the hydroxylation of PAN (Ji
and Zhang, 2008). After modification, the PANI composite shows two peaks of
carbonyl groups at 1693 and 1669 cm-1
, respectively. This can be explained by the
hydrogen bonding between PANI and the hydrolyzed PAN 350 membranes. Among
the two peaks, the carbonyl band at the lower wavenumber of 1669 cm-1
is caused by
the hydrogen bonding between N-H of PANI and C=O of PAN; the band at 1693 cm-1
is the reflection of free carbonyl groups (Jeon et al., 1999). The absorption peaks at
2937, 2243, 1452 and 1354 cm-1
are related to the stretching vibration of methylene (–
CH2–), nitrile groups (–CN–), methylene and hydrocarbon chains, respectively (Li et
al., 2014b; Zhang et al., 2016b). The band at 3323 cm-1
in PAN 350 membrane
corresponds to the stretching vibration of the –OH groups and indicates some water
molecules remaining in the PAN network. These two bands are also observed in the
membrane after modification with polyaniline. After the modification, new peaks at
Chapter 4
60
1609 cm−1
and 1506 cm−1
attributed to the C−C stretching of the quinonoid (Q) and
benzenoid (N) rings of PANI are clearly observed in the curve of the modified PAN
350 membranes. The C-N vibration of PANI was found located at 1248 cm−1
(Bayramoğlu et al., 2010). The FTIR observation (Fig. 4.2) confirms the presence of
PANI on the surface of the membranes.
Fig. 4.2 FTIR of PAN 350 membranes before and after modification
4.2.3 IEC, water uptake and contact angle
For a salt solution, the ion exchange membrane can greatly reject the permeation of
that salt due to its repulsion of one specific ion. Porous UF membranes can serve as
migration media to consolidate the migration of various kinds of ions; however,
during that process, back diffusion from the concentrate to the diluate is a problem.
By introducing PANI to the stereo hierarchical porous composite structure, functional
groups from PANI were added to the modified PAN 350 membranes, which can
greatly elevate its selectivity and slow down penetration during the desalting process.
Considering the porous structure and loose support layer, as well as the low selectivity
toward ions with small hydrated diameter, the IEC of hydrolyzed PAN 350 membrane
is significantly more dependent on the space in membrane matrix rather than on the
total amount of exchangeable anions. It would be pointless here to determine the IEC
Chapter 4
61
of a hydrolyzed PAN 350 membrane. In this chapter, the IEC of modified anion
exchange membranes was optimized to 1.50 mmol/g, while the water uptake and
contact angle were also changed. The water uptake has a profound effect on the
transport behavior of ions across the membrane, as well as the dimensional and
mechanical stability of membranes. In principle, a high water uptake of the ion
exchange membranes would be beneficial to the ion transport. The primary PAN 350
membranes have a water uptake of 123%. After modification, it decreases to 111%.
Such a high water uptake, superior to 100%, can also be explained by the space in
membrane matrix. Fig. 4.3 shows that the contact angle rapidly increases from 22.3°
for the primary membrane to 43.1° for the modified PAN 350 membrane, because of
the highly hydrophilic hydrolyzed PAN and the hydrophobic PANI. Introducing PANI
decreases the water uptake while it increases the contact angle. In addition, the
occupation of PANI in the pore space can also contribute to the decline of the water
uptake.
Fig. 4.3 IEC, water uptake and contact angle of the PAN 350 and the modified PAN 350
membrane
Chapter 4
62
4.2.4 Diffusion dialysis experiments
In Fig 4.4, it can be seen that the conductivity of the concentrated cell increases
during the experiments, which is caused by the diffusion of salt from the concentrate
chamber (1 M NaCl) to the diluate chamber (distilled water). Dialysis coefficients
were calculated based on changes in concentration in both compartments of the
dialysis cell. The dialysis coefficient of the primary PAN 350 membrane is 108
mol/(h·m2) at room temperature. After the addition of PANI, it is noticeable that the
modified membrane impedes the diffusion of NaCl. The dialysis coefficient of the
modified PAN 350 membrane decreases to 9.1 mol/(h·m2). The PAN 350 membrane
with PANI on its surface has a lower diffusion of NaCl compared to the primary
membrane due to the positive charge of PANI, which hinders the transport of NaCl
across the membrane. Furthermore, the effect of sterical hindrance due to the presence
of PANI may play a role.
Fig. 4.4 Conductivity–time curves of diffusion progress
Chapter 4
63
4.2.5 Desalination parameters during ED: conductivity and pH
The evolution of conductivity during ED treatment in different currents is presented in
Fig. 4.5. According to the experimental results, regardless of the current density, the
conductivity of the brine salts solution in the diluate compartment decreases during
the experiments. Fig. 4.5 shows that the current density affects the transport of ions.
When the current is 0.2 A, the conductivity of the diluate compartment decreased
gradually from 3.5 and 4.0 mS/cm to 2.7 and 3.1 mS/cm in the first 3 hours for NaCl
and MgCl2, respectively. At a higher current (0.4 A), the conductivity of the diluate
compartment dropped sharply at the beginning of the experiments and then continues
to decrease until reaching a steady state. At higher current, the transport of ions
accelerated, which led to a dramatic change of the conductivity of the diluate
compartment. Regardless, the conductivity change was not so obvious after 100 min.
It should be pointed out that during the desalting procedure, the driving force is a
compromise between the electrical field and the salinity-gradient power. At the
beginning of the experiments, transport of ions by the electrical field would be
dominant, however, at the end of the experiment at 0.4 A, the conductivity of the
diluate compartment increased again, which implies that ion transport by the electrical
field is surpassed by salinity gradient. The change of pH in the diluate compartment
could be another reason affecting the conductivity change. Although the pH change
was not recorded during the experiment, at 0.3 A the possibility of water splitting at a
higher current for this system could occur (Fig. 4.8). For the experiment at 0.3 A, the
slope of the conductivity-time curve is surrounded by the curve of 0.2 and 0.4 A at the
beginning of the experiment. It can be concluded that for PAN 350 ultrafiltration
membranes used in electrodialysis for desalination, the enhanced current can increase
the migration of ions, however, if the current becomes too high, a reducing
desalination efficiency was obtained due to the back diffusion and water splitting of
electrolyte across the membrane.
Chapter 4
64
Fig. 4.5 Conductivity curves of the diluate compartment at different current density
Conversely, the conductivity of the concentrate increased during the experiments (Fig.
4.6). However, there was a discrepancy with the changes of the diluate compartment
to some extent. Both NaCl and MgCl2 experiments present an increase in
conductivities in the concentrate department; however, the slope of the
conductivity-time curve is different. Table 4.1 lists the conductivity change of the
different compartments at the first 100 min. It was observed that the conductivity
variation of the concentrate compartment is always higher than that of the diluate
compartment, especially at larger current, which suggests that the current efficiency
was decreased. Comparing with NaCl, the variation of the MgCl2 concentration is
much larger, which can be explained by the different transport rates of Na+ and Mg
2+
in cation exchange membranes and PAN 350 membranes. The Na+ from the
electrolyte is able to permeate from the anode to the concentrate compartment and the
diluate compartment, and ultimately to the electrolyte again. Once the Na+ enters the
diluate compartment, it replaces the Mg2+
migration to support the current, so that the
current efficiency decreases. Besides, the protons generated by the electrode reaction
could enter the concentrate compartment, giving rise to a pH change in the
concentrate compartment. Considering the possible water splitting that may occur at
the UF membrane surface, both electrode reactions and water splitting can contribute
Chapter 4
65
to the increase of the conductivity.
Fig. 4.6 Conductivity curves of the concentrate compartment at different current density.
Table 4.1 Conductivity and concentration change of diluate and concentrate compartment
Conductivity change NaCl
0.2 A
NaCl
0.3 A
NaCl
0.4 A
MgCl2
0.2 A
MgCl2
0.3A
MgCl2
0.4A
Diluate compartment (mS/cm) 0.8 1.4 1.5 0.9 0.9 0.7
Concentrate compartment (mS/cm) 0.8 1.6 1.7 1.4 2.2 1.7
Desalination using the modified PAN 350 is summarized in Fig. 4.7. In the presence
of polyaniline, there is no obvious effect on the removal of NaCl. Moreover, while the
current is fixed at 0.2 A, for MgCl2, the conductivity changes in the diluate and
concentrate compartment for the primary PAN 350 membrane is 0.9 and 1.4 mS/cm,
which is almost the same as the modified PAN 350 membranes (1.0 and 1.4 mS/cm,
respectively). However, at 0.4 A, the conductivity changes in the diluate compartment
for primary PAN 350 membrane is 0.7 mS/cm, and it reaches its minimum after 100
min. After this, the conductivity slightly increases again. The main reason expected to
contribute to this could be the water splitting phenomenon. The modified PAN 350
membrane can greatly increase the salt removal from 4.0 mS/cm to 2.1 mS/cm after
Chapter 4
66
180 min desalination. The maximum removal rate was increased from 17.4% to
48.1%, which is almost a factor 3. The presence of polyaniline can hinder the Mg2+
transport from the concentrate to the diluate compartment because of the electrostatic
effect, while there is no obvious effect on Na+ ions. This effect has been applied in
preparation of monovalent selective exchange membrane by Kumar et al. (Kumar et
al., 2013). After the surface modification with PANI, the Na+ transport number across
the membranes was unchanged, whereas the Zn2+
and Al3+
transport numbers
decreased (Kumar et al., 2013). Thus, at lower current there is no obvious change
between the primary and modified PAN 350 membranes; however, at high current, the
desalination ratio significantly increases.
Fig. 4.7 Conductivity change of modified PAN 350 membranes in NaCl and MgCl2 solutions [(a)
NaCl experiments; (b) MgCl2 experiments].
The pH of the diluate and concentrate solution changed differently over the process.
Considering the transport rate and desalination efficiency, a current of 0.3 A is
considered. As shown in Fig. 4.8, in the concentrate compartment, the general trend is
that the pH decreased continuously as the electroseparation process progressed. A
decrease in the pH could be expected as the positive charges of PANI on the surface
of UF membranes could release protons and acidify the solution. The acidification in
the concentrate compartment was caused by the H+ ions migrating from the diluate
Chapter 4
67
compartment. The pH of the diluate solution remained constant for the unmodified
membrane, however, for the modified membranes, the pH slightly changed over the
180 min of electrodialysis. For this reason, the acidification of the concentrate
compartment and comparatively neutral diluate compartment, conductivity
discrepancies of the two compartments can be explained.
Fig. 4.8 pH change of the solutions during NaCl and MgCl2 desalination process using modified
and unmodified UF membranes (U. unmodified membrane M. modified PAN 350 membrane)
4.2.6 Current efficiency
In order to evaluate the performance of the modified PAN 350 membrane in a stack
for brine recovery, the current efficiency of the ED stack at different currents is shown
in Fig. 4.9. For NaCl, the highest current efficiency is at 0.3 A for the primary and for
the modified PAN 350 membrane. An increase in current density to a certain extent
reinforces the migration of ions and shortens the operating time simultaneously, which
can result in an increase of the current efficiency (Shen et al., 2014). Subsequently,
the current efficiency of the ED stack decreases with the current density. Although the
transfer of Na+ and Cl
− ions through the ion exchange membrane was accelerated as
the current density increases, several factors may suppress the current efficiency. With
Chapter 4
68
elevated current density, the splitting of water on ion exchange membranes intensified,
the produced protons and OH- would compromise the current to some content. Co-ion
leakage and permeation through the ion exchange membranes results from the high
concentration gradient, which decreases the current efficiency again (Shen et al.,
2013). However, for MgCl2, it is a totally different condition. For the PAN 350
membrane, the current efficiency decreases with the current density due to the
enhanced transport of Na+ than the Mg
2+ and the diffusion though the membrane.
However, after modification by polyaniline, the modified PAN 350 membrane has an
elevated current efficiency. This can be explained by the observation that the
introduction of polyaniline can greatly reduce the diffusion by an electrostatic
repulsion force between polyaniline and the ions as well as the shrunk pores of the
modified PAN 350 membrane. Fig. 4.9 shows that the current efficiency can be
greatly improved, especially at 0.4 A. The current efficiency in the NaCl concentrate
experiment increased from 31.4% to 36.5%, which is 16.4% higher than before. For
MgCl2, this reached to 52.6%, much higher than that of NaCl. Therefore, it can be
concluded that modified PAN 350 membranes offer a high potential for the
application in electrodialysis aiming at green production.
Fig. 4.9 Current efficiency of the electrodialysis stack at different currents
Chapter 4
69
4.2.7 Monovalent selectivity measurements
The permselectivity and ion flux values were obtained by measuring the Na+ and
Mg2+
ions concentration change in the diluate compartment. The ion flux of Na+ and
Mg2+
was 9.7×10-8
mol·cm–2
·s–1
and 4.5×10-8
mol·cm–2
·s–1
before modification. After
modification, the flux of Na+ was slightly increased to 12.4×10
-8 mol·cm
–2·s
–1 while
the flux of Mg2+
was reduced to 3.1×10-8
mol·cm–2
·s–1
. As a result, the
permselectivity increased from 2.15 to 3.98, which is almost a factor 2. It can be
explained by the fact that the dense layer can reject the divalent ions, with larger
Stokes radius, more effectively.
To further understand the separation process of the experiments, PAN based
ultrafiltration membranes with different molecular cutoff values were measured to
determine whether the mean pore size could influence the permselectivity. The flux
and permselectivity of different membranes are summarized in Table 4.2.
Table 4.2 The flux and permselectivity of different PAN based membranes
(B: before modification A: after modification)
PAN200 PAN350 PAN400 PAN450
B A B A B A B A
Na+ 11.3 13.0 9.7 12.4 12.1 13.3 11.1 15.0
Mg2+
7.1 4.2 4.5 3.1 4.2 3.8 5.4 5.6
Permselectivity 1.60 3.07 2.15 3.98 2.85 2.85 2.05 2.67
Results showed that the permselectivity of ultrafiltration membranes with lower
molecular cut-off is higher. It can be concluded that ultrafiltration membranes with
smaller pore size are more suitable for preparing membranes for monovalent ions
separation in electrodialysis. However, a higher perm-selectivity value of 7 and a Na+
flux of around 2.2×10−7
mol cm−2
s−1
were obtained for NFM in comparison with
those of PANI modified UF membrane. The PANI modified UF membrane still
Chapter 4
70
exhibits an enhanced perm-selectivity for the Na+/Mg
2+ system. To conclude, this
method provides new routes for preparing membranes based on ultrafiltration
membranes for efficient separation of monovalent ions by electro-driven separation
techniques.
4.3 Conclusions
Newly developed PANI modified membranes were successfully applied to produce
concentrated brines. The resulting membranes were evaluated in terms of ion
exchange capacity, water uptake and diffusion dialysis performance. In addition, the
obtained membranes exhibit enhanced properties toward desalination of brine salts by
electrodialysis. This novel technique has advantages not only regarding desalination
but also in terms of the enhanced current efficiency that was observed. This method
provides the possibilities of preparing porous ion exchange membranes to enhance
transport properties of anion exchange membranes, and suggests new routes to
develop monovalent selective membranes with low resistance and high transport
efficiency.
Chapter 5
71
5. Mussel-inspired modification of ion exchange
membrane for monovalent separation
Adapted from: J. Li, S. Yuan, J. Wang, J. Zhu, J. Shen, B. Van der Bruggen.
Mussel-inspired modification of ion exchange membrane for monovalent separation.
Journal of Membrane Science. 553(2018):139-150.
5.1 Introduction
As reviewed in Chapter 1, electrodialysis (ED) is one of the most economic and
advanced separation processes which enables the continuous separation and
concentration of brine water, especially when the concentration of the feed solution is
below 5 g/L. Compared to other technologies such as multistage flash (MSF)
evaporation and reverse osmosis (RO), ED has intrinsic advantages including
selective desalination and enhanced water recovery (Vaselbehagh et al., 2014).
Furthermore, the reduced chemicals usage and low energy consumption can
effectively diminish the cost of production. To date, electrodialysis has been applied
in various applications like biorefinery effluents (Luiz et al., 2017; Luiz et al., 2018),
rare earth elements recycling (Sadyrbaeva, 2015) organic acid recovery (Eliseeva et
al., 2009; Luo et al., 2002), and brackish water/wastewater treatment (Ghaffour et al.,
2013). Nevertheless, in some cases, precipitation caused by scaling compounds (Ca2+
,
Mg2+
, SO42-
, CO32-
) inevitably occurs in the concentrated compartment, which gives
rise to a deleterious effect on the desalination performance (Asraf-Snir et al., 2016).
For a standard ion exchange membrane, the presence of multivalent ions can suppress
the migration of monovalent ions by occupying the ion exchange transfer sites of the
membrane (Firdaous et al., 2007; Liu et al., 2017a). Therefore, the development of
IEMs with the ability to separate multivalent ions from monovalent ions is required.
Furthermore, IEMs with monovalent selectivity could potentially expand the
Chapter 5
72
application scope of ED to disciplines like metallurgy (Reig et al., 2017), sodium
chloride production (Zhang et al., 2017b) and reverse electrodialysis (Güler et al.,
2014).
In view of exploiting the differences in ion valances and hydrated ionic radii, many
approaches have been proposed to design selective cation exchange membranes
(CEMs) capable of separating multivalent ions from a mixed solution. Deposition of a
thin charged surface layer is considered a promising way to prepare monovalent
selective ion exchange membranes because the charged skin surface can increase the
repulsion forces towards multivalent ions. Abdu et al. (Abdu et al., 2014) reported a
layer-by-layer assembly of poly(ethyleneimine) (PEI)/poly(styrenesulfonate) (PSS)
bilayers on commercial CEMs (Astom, Japan); the perm-selectivity that was obtained
is comparable with commercial monovalent CEMs while simultaneously a lower
energy consumption is maintained. Wang et al. (Wang et al., 2013) fabricated a
monovalent CEM by preparation of a photo-induced self-polymerized chitosan layer.
An enhanced monovalent selectivity was obtained for both H+/Zn
2+ and Na
+/Mg
2+
system during the ED process. The leakage of Zn2+
and Mg2+
were reduced by 27.4%
and 62.4%, respectively. Nevertheless, membranes prepared by surface modification
still have disadvantages. Firstly, detachment of the surface functional layer from the
primary membrane may occur due to their weak electrostatic attraction (Wang et al.,
2013). Accordingly, attempts to bind covalently a cation charged layer on CEMs have
been tried to enhance the membrane stability. However, sulfonyl chloride acid
introduced to the reaction is not environmentally friendly. In addition, the
uncontrollable thickness of the modified layer and the reduced functional groups are
still important issues. Thus, developing new methods to fabricate monovalent
selective ion exchange membranes with high selectivity and low resistance in an
economically and environmentally friendly manner is imperative.
Enlightened by mussel-inspired surface chemistry, various attempts have been made
to modify membrane surfaces by this method due to the simplicity, controllability and
Chapter 5
73
extensive applicability (Yang et al., 2014a). Dopamine, a low-molecular-weight
catecholamine, is known as ‗bio-glue‘ derived from mussel adhesive proteins. (Li et
al., 2014a; Liu et al., 2014b) Unlike traditional coating protocols, the inherently
robust adhesion virtues of the catechol structure triggered by self-polymerization has
been broadly utilized in membrane surface functionalization for various purposes such
as structural stability enhancement (Li et al., 2015), antifouling properties
improvement (Li et al., 2014c) and optimization of separation properties (Yang et al.,
2017; Zhao et al., 2016). The self-polymerization of dopamine is a complex process
which involves covalent polymerization and non-covalent interactions (Choi et al.,
2014; Lee et al., 2016a; Wang et al., 2016). Although this strategy is successful in
constructing multifunctional coatings, the air-oxidized formation of PDA takes a long
time (Gao et al., 2013; Zhang et al., 2017a). Besides, aggregation of PDA oligomers
based on non-covalent interactions possibly causes uneven and unstable coating layers
after a long time of deposition (Zhang et al., 2016a). Recently, two main strategies, i.e.
rapid deposition of dopamine triggered by CuSO4/H2O2 and PDA/PEI co-deposition,
were proposed to overcome the above mentioned disadvantages (Wang et al., 2017b).
These two measures can greatly shorten the deposition time and promote a
homogeneous and robust PDA coating, thereby laying the foundation for the potential
development of such bio-inspired deposition process.
In this study, a one-pot approach is reported to prepare a monovalent selective cation
exchange membrane by PDA/PEI co-deposition. PEI is an amorphous polymer with
high thermal and chemical stability, which has been widely applied in water
desalination and membrane modification (Al-Maythalony et al., 2017; Guo et al.,
2016). PDA, known as ―bio-glue‖ can fix PEI molecular chains on the membrane
surface while the side PEI chains serve as repulsive functional groups that reject
multivalent ions. With the assistance of CuSO4/H2O2, the deposition time can be
significantly reduced as CuSO4/H2O2 produces a large amount of reactive oxygen
species (ROS), which is except to accelerate the polymerization of dopamine.
Chapter 5
74
5.2 Results and discussion
5.2.1 Chemical structure of the membrane surface
During the process of dopamine polymerization, interactions between the PDA
coatings and the supporting layer are composed of covalent and non-covalent
interactions, including H-bonding, π-π and electrostatic interactions (Jiang et al.,
2011). Thus, a tightly adherent facial layer is formed on the SPES membrane surface
after being soaked in PDA/PEI aqueous solution for a specific time (Han et al., 2012).
By comparing SEM images for membrane prepared with different PDA/PEI ratios
(Fig. 5.1), a distinct change was observed. For SPES-PDA/PEI-0 membrane, the
surface was covered with numerous nanostructured papillae. Normally, PDA
depositions consist of spherical and linked agglomerates with a size ranging from ca.
100 to 500 nm (Jiang et al., 2011). After introducing PEI, the surface roughness of
PDA-coated membrane was found to be reduced. The PEI/PDA complex can be
formed on the membrane surface because PEI with nucleophilic amine groups can
crosslink with dopamine and other reactive intermediates through Michael addition or
Schiff base reactions (Huang et al., 2015; Lv et al., 2015; Zhang et al., 2014; Zhao et
al., 2015). At this condition, the reduction of non-covalent interactions effectively
diminished the PDA aggregates. When the PEI/dopamine ratio reaches 3, the
membrane surface becomes smooth and no protrusions are observed.
Chapter 5
75
Fig. 5.1 Surface SEM images of SPES and modified SPES membranes (a. Primary SPES
membrane; b. SPES-PDA/PEI-0; c. SPES-PDA/PEI-1; d. SPES-PDA/PEI-2; e. SPES-PDA/PEI-3)
Furthermore, EDAX was applied to identify the element composition of the
membrane surface. The EDAX spectra of different membrane are shown in Fig. 5.2
and the elemental composition of the membrane surface is summarized in Table 5.1. It
was evident that the original SPES substrate contains C and O while no N and Cu can
be detected. Compared with the SPES membrane, a nitrogen peak was found for the
PDA-modified membrane. The atomic content for nitrogen was 33.4%, which is
ascribed to the amine groups of the PDA coatings. After introducing PEI to the
modified procedure, the atomic content of N increased from 33.4% for the
SPES-PDA/PEI-0 membrane to 39.6% for the SPES-PDA/PEI-1 membrane,
demonstrating that PEI was successfully anchored on the membrane surface. The N
ratio further increases for the SPES-PDA/PEI-2 membrane, resulting from the
increased content in PDA/PEI composites. However, when the PEI/dopamine ratio
reached 3, the atom percentage of nitrogen was reduced to 45.6%, which indicated
that the adhesive properties were affected when the concentration of PEI is too high.
Furthermore, with increasing PEI concentration, the Cu2+
concentration reduces
continuously. The catechol groups of PDA can chelate Cu2+
ions from the solution,
which leads to an enhanced stability. Furthermore, catechol/quinone groups present in
PDA are able to covalently couple to nucleophilic amines of PEI (Chien et al., 2012).
Chapter 5
76
With more PEI incorporated to the surface, a reduced copper content would be
obtained. All the changes of element concentration confirm the presence of the PDA
or PDA/PEI coatings on the SPES membrane.
Fig. 5.2 EDAX results of SPES and modified SPES membrane (a. Primary SPES membrane; b.
SPES-PDA/PEI-0; c. SPES-PDA/PEI-2)
Table 5.1 EDAX mapping determined elemental composition (in at %) of membrane surface
C N O Cu
SPES 60 - 40 -
SPES-PDA/PEI-0 32.7 33.4 32.0 1.9
SPES-PDA/PEI-1 28.8 39.6 31.0 0.6
SPES-PDA/PEI-2 25.3 47.8 26.2 0.7
SPES-PDA/PEI-3 28.2 45.6 26.1 0.1
Apart from the electrostatic effect by the positive functional groups of PEI, the
thickness of the modified layer could also impact selectivity due to the steric
Chapter 5
77
hindrance effect. Fig. 5.3(a-d) shows the cross-section morphologies of the composite
membranes. It is obvious that the membrane is covered by a modified layer. The
thickness of the modified layer is typically more than 80 nm, however, it could be
easily adjusted by altering the reaction time or the mass ratio between dopamine and
PEI (Yang et al., 2014b). As a rubber is used to fix the membrane during the
co-deposition process, a boundary is formed on the membrane with and without
dopamine modification. By collecting the height profile along a line across the
boundary, the thickness of the modification layer is obtained. As shown in Fig. 5.3e,
the thickness is obtained from the AFM smooth surface morphology profile along the
line. The thickness firstly increases when PEI was incorporated to the co-deposition
process and then decreases when the mass ratio of PDA/PEI further reduced to 1:3
while the concentration of dopamine is constant. The increased thickness of the
modified layer is due to the accelerating deposition of PDA and PEI composites
generated by Michael addition or Schiff base reaction. Because excessive PEI could
terminate the polymerization of dopamine, at this condition, PEI cannot attach onto
the membrane surface. Thereby, the modified coatings are thinner when the
concentration of PEI is too high (Lv et al., 2015).
Fig. 5.3 The cross-section images of modified SPES membranes (a. SPES-PDA/PEI-0; b.
SPES-PDA/PEI-1; c. SPES-PDA/PEI-2; d. SPES-PDA/PEI-3) and (i. the thickness of the
modified layer observed by AFM smooth surface morphology)
Chapter 5
78
5.2.2 Morphologies of the membrane
The three-dimensional AFM surface topography pictures for the unmodified SPES
and modified SPES membranes are presented in Fig. 5.4. Similar with the results from
SEM, the primary SPES membrane was observed to be relatively smooth; the
morphology was changed after modification. The surface roughness values of the
membranes, including Ra, Rrms and Rm, are listed in Table 5.2. The SPES membrane
has a smoother surface than the SPES-PDA/PEI-0 and SPES-PDA/PEI-1 membranes,
as confirmed by the surface roughness (Ra) results. This is possibly due to the
adhesion and agglomeration of the PDA nanoaggregates. However, Ra and Rrms
remarkably decrease from 0.66 and 1.17 nm to 0.38 and 0.52 nm for
SPES-PDA/PEI-2 membrane, indicating that PEI can be a good candidate to
smoothen the surface structure of the membranes. Nonetheless, a further increase of
the content of PEI has no obvious effect on Ra and Rrms, which indicates that further
increasing the content of PEI may reduce the adhesive properties of PDA (Lv et al.,
2015). The roughness change for the modified membrane was consistent with the
previously discussed SEM results, confirming that the co-deposition of PDA/PEI can
effectually alter the surface structure and morphology of the membranes. The grafted
PEI molecules, of which the chains are rather flexible, endow the membranes with a
potential for monovalent selectivity.
Chapter 5
79
Fig. 5.4 AFM topography of SPES membrane and modified SPES membranes (a. Primary SPES
membrane; b. SPES-PDA/PEI-0; c. SPES-PDA/PEI-1; d. SPES-PDA/PEI-2; e. SPES-PDA/PEI-3)
Table 5.2 AFM surface roughness parameters of the pristine and modified membranes: Ra
(average roughness), Rrms (root mean square roughness), and Rm (maximum vertical difference
between the highest and lowest points)
Membrane Ra (nm) Rrms (nm) Rm (nm)
SPES 0.66 1.17 16.5
SPES-PDA/PEI-0 3.21 4.88 46.8
SPES-PDA/PEI-1 1.25 1.85 16.8
SPES-PDA/PEI-2 0.38 0.52 8.09
SPES-PDA/PEI-3 0.76 1.08 14.4
5.2.3 Zeta potential
Since the permselectivity of monovalent cations can be tuned by adjusting surface
charge characteristics, the zeta potential for modified membranes is indicated in Table
5.3. Both the SPES membrane and SPES-PDA/PEI-0 membranes are negatively
charged, with zeta potentials of -3.9 and -30.1 mV, respectively. Conversely, after
grafting of PEI to the modified layer, the zeta potential of SPES-PDA/PEI-1
membranes shifted to 7.4 mV, greatly increasing the zeta potential. Thus, a positively
charged surface could hinder the transport of multivalent ions through electrostatic
Chapter 5
80
effects. Furthermore, the zeta potential of the modified membranes reaches up to
13.4 mV with increasing PEI content. As previously discussed, the increase of PEI
content on the PDA/PEI coatings can lead to an elevated positive charge density for
the modified membrane. By incorporating PEI to the PDA layer, the increased
positive charge density may result in an improved monovalent selectivity. However,
the resultant SPES-PDA/PEI-3 membranes with a more neutral charge gave rise to a
reduced electro-static effect. The reduced zeta potential is consistent with the reduced
N content results obtained from EDAX.
Table 5.3 Zeta potential of SPES and PDA/PEI modified monovalent cation exchange membranes
Membrane SPES SPES-PDA/PEI-0 SPES-PDA/PEI-1 SPES-PDA/PEI-2 SPES-PDA/PEI-3
Zeta
potential
(mV)
-3.9 -30.1 7.4 13.4 8.3
5.2.4 Water contact angle, ion exchange capacity and water uptake
The SPES membrane exhibits the best hydrophilicity with a water contact angle
decrease from 40° to 31° in 60 s (Fig. 5.5), owing the presence of hydrophilic
functional groups in the membrane matrix. The initial contact angle sharply increases
from 40° to 62° after introducing a PDA layer, which is in agreement with the original
analysis of the hydrophilicity of the PDA layer (Luo et al., 2013). When hydrophilic
groups are covered by PDA/PEI composites, the increase of the PEI mass ratio with
inherent abundance of hydrophilic amine groups can substantially increase the
wettability of the membranes. Furthermore, a high PEI content would impact the
polymerization/deposition rate of dopamine. Therefore, the reduction of thickness for
the coating layer with reduced contact angle can be another factor contributing to an
enhanced hydrophilicity.
Chapter 5
81
Fig. 5.5 Dynamic water contact angles of SPES, SPES-PDA/PEI-0, SPES-PDA/PEI-1,
SPES-PDA/PEI-2 and SPES-PDA/PEI-3 membranes in 1 minute
It can be seen from Table 5.4 that the IEC of SPES was 1.76 mmol/g. It could be
expected that the negative dopamine coating layer caused by the deprotonation of
phenolic hydroxyl groups at neutral pH could increase the IEC. However, the
thickness of the SPES membrane matrix is far more than that of the coating layer, thus
the IEC remains constant (Kim et al., 2014). With a further increase of the PEI ratio, a
reduction trend has been identified concerning the IEC of the modified membrane,
although the variation is not obvious. The reduction of the IEC value is due to the fact
that some –SO3H groups on the membrane surface are neutralized by the electrostatic
effect between -NH2 groups and –SO3H groups. The water uptake has an obvious
effect on the membrane conductivity because water molecules serve as ‗‗vehicles‘‘ for
the transportation of the ions from the anode to the cathode (Muthumeenal et al.,
2014). In the SPES membrane, the sulfonate ions in membrane matrix are hydrated
with absorbed water molecules. The increase in water uptake of SPES-PDA/PEI-0
suggests that the increase in negative charge density attracted more water molecules
inside the membrane matrix. A similar trend for SPES-PDA/PEI-1, SPES-PDA/PEI-2
and SPES-PDA/PEI-3 membranes is caused by the incorporation of hydrophilic PEI
0 10 20 30 40 50 6020
30
40
50
60
70
80
90
Wa
ter
con
tact a
ng
le (
o)
Time (s)
SPES
SPES-PDA/PEI-0
SPES-PDA/PEI-1
SPES-PDA/PEI-2
SPES-PDA/PEI-3
Chapter 5
82
in the membrane matrix.
Table 5.4 Ion exchange capacity and water uptake of the membrane before and after co-deposition
Membrane IEC (mmol/g) Water uptake (%)
SPES 1.76 28.5
SPES-PDA/PEI-0 1.77 30.8
SPES-PDA/PEI-1 1.73 31.4
SPES-PDA/PEI-2 1.71 32.2
SPES-PDA/PEI-3 1.70 32.9
5.2.5 Diffusion experiments
Diffusion dialysis performances, represented by the dialysis coefficient, are
investigated to evaluate mass transport though ion exchange membranes. The
diffusion of salts leads to an increased conductivity in the concentrated compartments
during the experiments. However, the migration rate is quite different. By introducing
a PDA layer to the SPES membrane surface, the IEC increases, but due to the denser
and more compact coating layer, the diffusion dialysis slows down (see Fig. 5.6a) and
thus, the dialysis coefficient was reduced from 0.49 mol/(h·m2) to 0.32 mol/(h·m
2)
(Fig. 5.6b). The transport of NaCl further decreases by introducing PEI in the
co-deposition process, indicating that PEI coated on the modified layer is able to
decelerate the transport of NaCl. Thus, the dialysis coefficient is significantly
decreased. A pronounced increase in dialysis coefficient was observed with a further
increase of the PEI content. The SPES-PDA/PEI-3 membrane has a faster diffusion of
NaCl compared to SPES-PDA/PEI-2 and SPES-PDA/PEI-1 membrane due to the
increased PEI concentration. In this case, the addition of PEI reduces the adhesive
properties, so that PDA/PEI composites fail to attach to membrane substrates. The
enhanced electrostatic interaction of –NH2 with Cl- ions could be another factor to
enhance the dialysis coefficient. Thus, with increased PEI content, the dialysis
Chapter 5
83
coefficients follow the order SPES-PDA/PEI-3 > SPES-PDA/PEI-2 >
SPES-PDA/PEI-1.
Fig. 5.6 Diffusion experiment (a) conductivity variation in concentrate compartment of diffusion
cells employing monovalent selective ion exchange membranes (b) dialysis coefficients
5.2.6 Electrochemical characterization of the monovalent selective
ion exchange membranes
The current voltage curves of the prepared membranes with different PEI content are
shown in Fig. 5.7. The SPES membrane has the lowest limiting current density. With
reduced transport number, the limiting current density increases. The limiting current
density of the I-V curves indicates that the transport number follows the order SPES >
SPES-PDA/PEI-0 > SPES-PDA/PEI-1 > SPES-PDA/PEI-2 > SPES-PDA/PEI-3. This
trend is different from the transport process calculated as dialysis coefficients, which
can be explained by the dominant effect of positive charge density at the existence of
electric field. With respect to the limiting current density, no significant variation can
be observed in terms of the limiting current density. A similar phenomenon was
observed by Belashova et al. and Pismenskaya et al., which can be explained by the
compensation between the effects of the increased hydrophobicity and the conducting
heterogeneity caused by dopamine aggregates (Belashova et al., 2012; Pismenskaya et
Chapter 5
84
al., 2012). The higher surface hydrophobicity, the higher limiting current density,
while the conducting heterogeneities decrease the limiting current density (Volodina
et al., 2005).
The impedance spectra of the monovalent selective ion exchange membranes are
shown by the Nyquist diagram, in which the real impedance (Zreal) is plotted against
the imaginary impedance (-Zimag). The Nyquist plots show that the Ohmic resistance
of SPES membranes was 31.6 Ω. As can be observed, introducing a functional layer
results in a higher area resistance of the membrane (Luo et al., 2008). For the PDA
decorated membrane, the Ohmic resistance increases to 32.8 Ω. The Ohmic resistance
of the modified membranes further increased after incorporating PEI in the coatings.
For the SPES-PDA/PEI-1, SPES-PDA/PEI-2 and SPES-PDA/PEI-3 membranes, the
Ohmic resistances were 33.9 Ω, 35.1 Ω and 35.6 Ω, respectively. This increased
Ohmic resistance is attributed to both the compact coating layer and enhanced
positive charge of PEI. In addition, it would be expected that the monovalent
selectivity would increase by the PDA/PEI co-deposition coating layer, because the
monovalent selectivity is increased with the increase of the total amount of positive
charges on the membrane surface.
Fig. 5.7 (a) Current–voltages for different SPES based monovalent selective ion exchange
membrane (b) Nyquist plot showing the impedance spectra of different membranes
Chapter 5
85
5.2.7 Electrodialysis experiments
In order to investigate the monovalent selectivity of the PDA/PEI modified membrane,
electrodialysis experiments were conducted. The monovalent selectivity of a
membrane was determined by calculating bulk transport numbers of H+ and Zn
2+
based on ionic fluxes under constant direct current density of 10.6 mA/cm2. For the
SPES membrane, the leakage of Zn2+
was 12.7%. It can be observed from Fig. 5.8a
that the Zn2+
leakage decreases in the order SPES-PDA/PEI-0, SPES-PDA/PEI-1, and
SPES-PDA/PEI-2. For SPES-PDA/PEI-0 membrane, the dense surface structure
increased the steric resistance for Zn2+
. In comparison, a slight increase of H+ flux can
be observed, which can be explained by the increased negative charge density and the
smaller Stokes radius of H+. With the increase of the PEI content, Zn
2+ leakage further
decreases, which is ascribed to the increased positive charge density. The flux for H+
increases for SPES-PDA/PEI-1, which is caused by the formation of H+ transfer
channels. The compact acid–base pairs formed by hydroxyl groups and amino groups,
can effectively block Zn2+
, while the transport of H+ is facilitated (Ge et al., 2015).
Based on the Grotthuss mechanism and the pore-size sieving effect, the increased H+
flux can be explained. The reduced H+ flux for SPES-PDA/PEI-2 membrane is due to
the enhanced positive charge density, which hinders the H+ transport through the
electrostatic effect. When the PEI/PDA ratio was 3 during the co-deposition process,
although the positive charge density was increased, both the leakage of Zn2+
and the
flux of H+ increase. The reason can be the reduced attachment, which reduces the
thickness of the functional layer. Thus, dopamine based monovalent selective ion
exchange membranes exhibit a lower Zn2+
leakage compared to an SPES membrane.
Besides, PDA/PEI composites exhibit excellent properties to facilitate H+ transport.
The 𝑃𝐻+𝑍𝑛2+
was greatly reduced after modification.
Chapter 5
86
Fig. 5.8 (a) Zn2+
leakage and perm-selectivity (b) fluxes of H+ of the SPES membrane and the
monovalent selective ion exchange membranes prepared by co-deposition
5.2.8 Stability and effects of molecular weight of PEI
The stability of the modified layer is of significant importance for the practical
applications of ED technology. It has been proven that a PDA layer has a superior
stability in neutral and weak acidic/basic solutions (Yang et al., 2014a). However, the
PDA coatings may become unstable in strong acid and base conditions because the
non-covalent connections disintegrate (Jiang et al., 2013; Wei et al., 2013). To
investigate the stability of membrane modified by PDA/PEI codeposition in harsh
environment, the membrane was immersed in 0.1 M HCl or 0.1 M NaOH for 7 days.
Fig. 5.9 indicates that no significant changes of the perm-selectivity and ion flux take
place when the PDA/PEI modified membranes were rinsed in acid solutions.
Comparatively, the modified membranes show slightly variation after treatment with
alkaline solutions. The slight increase of perm-selectivity and reduction of H+ flux
may be attributed to the loss of some non-covalently bonded components. At this
condition, reduced functional groups and acid-base pairs could cause a variation of the
perm-selectivity and H+ flux. However, the intrinsically robust adhesion of the PDA
catechol structure with the membrane matrix can still maintain the separation
properties in harsh environments. In addition, the residual copper ions on the surface
of modified layer could serve as cross-linking sites by chelation with the amine/imine
Chapter 5
87
groups of PDA. In summary, due to the covalent crosslinking between PDA and PEI,
the affinity properties and structural stability of the co-decorated membranes can be
greatly enhanced.
Fig. 5.9 Permselectivity and H+ flux of the SPES-PDA/PEI-2 membrane. a. before and after acid
treatment; b. before and after alkaline treatment; c. with different PEI molecular weight
The molecular weight of PEI could also have an effect on the membrane
perm-selectivity and H+ flux. In order to ensure the same amount of –CH2CH2NH-,
the mass ratio of dopamine to PEI was fixed at 1:2. Typically, PEI with a higher
molecular weight will resulted in a more hydrophobic surface because large PEI
molecules can impede the deposition process. As a result, a higher selectivity is
obtained (Ran et al., 2017; Yang et al., 2016). In addition, by elongating amine side
chains, the electrostatic interactions between amine groups and Zn2+
are enhanced (Ge
et al., 2017). However, the deposition amount reduces with increasing PEI molecular
weight due to the unfavorable crosslinking between dopamine and PEI, which in turn
leads to a poor separation performance. With increasing mobility of the side chains,
hydrophilic ―channels‖ were constructed by self-assembly of ionic side chains, which
Chapter 5
88
can facilitate the migration of protons (Choi et al., 2005; Ge et al., 2016b; Li and
Guiver, 2014). As a consequence, a slight increase of the proton flux is observed.
In this study, the obtained results demonstrate that the PEI/PDA modified membrane
greatly reduces the Zn2+
leakage. While comparing to monovalent membranes
prepared by other methods (Table 5.5 and Table 5.6), the perm-selectivity of the
prepared membranes was comparable to most membranes. However, this method still
showed a higher Zn2+
leakage compare to monovalent membranes prepared by
annealing treatment. By comparing the modified membrane with commercial
monovalent cation exchange membrane, the modified membrane has a higher H+ flux
while maintaining almost the same Zn2+
flux. To conclude, PDA/PEI codeposition
could be a promising way to prepare monovalent cation exchange membranes.
Table 5.5 Some typical examples about monovalent cation selectivity - Zn2+
leakage
Method Zn
2+ leakage of primary
membrane
Zn2+
leakage of modified
membrane Reference
Quaternized chitosan 8.5% 1.01% (Hu et al.,
2008)
Photo-induced covalent
immobilization of
chitosan
27.4% 4.5% (Wang et
al., 2013)
Annealing treatment 0.3% 0.01% (Ge et al.,
2014)
Chemical modification by
polyquaternium-7 22% 14.2%
-
CSO - 9.1%
This work 12.5% 5.7% -
Chapter 5
89
Table 5.6 Some typical examples about monovalent cation selectivity - Permselectivity
Method Perm-selectivity*
Modified membrane
Reference Zn2+
flux
(mol·cm–2
·s–1
)
H+ flux
(mol·cm–2
·s–1
)
EDNF 354 6.7·10-6
2.3·10-3
(Ge et al.,
2016a) CSO 15 67.9·10-6
1.7·10-3
This work 52 14.1·10-6
1.8·10-7
-
* Perm-selectivity is simply calculated as the ratio of monovalent ion and divalent ion fluxes
5.3 Conclusions
Monovalent selective ion exchange membranes have been prepared by PDA/PEI
co-deposition with the assistance of CuSO4/H2O2. A series of characterizations for the
modified membranes by SEM and EDAX demonstrated the formation of a functional
layer on the membrane surface. With increasing PEI content, the water uptake of the
membranes increased, while the IEC of the membranes slightly decreased. AFM
images indicated that at optimized conditions (SPES-PDA/PEI-2), the surface of the
modified membrane was covered uniformly by the PDA/PEI layer. In addition, the
thickness of the modified layer was first increased and then decreased with the
increase elevation of PEI mass ratio. Membrane transport properties were measured
by diffusion experiments and current–voltage curves. The electrical resistance was
increased after surface modification. Electrodialysis experiments for H+/Zn
2+ system
exhibited that a superior monovalent selectivity could be obtained after introducing
the PDA/PEI layer. After 7 days of acid/alkaline treatment, the modified membrane
shows an excellent stability in terms of permselectivity and H+ flux. With increased
PEI molecular weight, both the permselectivity and H+ flux were increased.
Considering the increased permselectivity, high H+ flux, high hydrophilicity and
long-term stability, PDA/PEI co-deposition can be a promising method to enhance the
monovalent selectivity of a membrane.
Chapter 6
90
6. Mussel-inspired monovalent selective cation
exchange membranes containing hydrophilic
MIL53(Al) framework for enhanced ion flux
Adapted from: J. Li, J. Zhu, S. Yuan, X. Li, Z. Zhao, Y. Liu, Y. Zhao, A. Volodine, J.
Li, J. Shen, B. Van der Bruggen. Mussel-inspired monovalent selective cation
exchange membranes containing hydrophilic mil53 (Al) framework for enhanced ion
flux. Industrial & Engineering Chemistry Research. 57(2018): 6275-6283.
6.1 Introduction
As indicated in Chapter 5, it is of interest to discover novel materials and
methodologies to prepare monovalent CEMs. The method of rapid co-deposition of
PDA and PEI by using CuSO4/H2O2 as trigger can form a thin selective layer. As a
consequence, the surface properties and the monovalent selectivity of the resultant
membranes can be easily tailored. The optimum membranes, with 4 h co-deposition
of 60 mg PDA and 120 mg PEI, exhibited a 2.5 times higher permselectivity than the
primary membrane. Especially, the flux of H+ was enhanced in the binary system of
H+/Zn
2+. Meantime, the PDA/PEI modified ion exchange membrane shows excellent
operation stability for monovalent separation performance in both acid and alkaline
situation. However, the selectivity of the as-prepared membrane in other systems,
especially Na+/Mg
2+ system, is unknown.
To obtain a promising monovalent selective ion exchange membrane, the monovalent
selectively is not the only factor to be taken into consideration. The membrane
resistance, current efficiency, and energy consumption are the key parameters to
evaluate an electrodialysis process. MIL-53(Al), a sub-branch of metal-organic
frameworks (MOFs), contains 1D diamond-shaped channels with pores of nanometer
Chapter 6
91
dimensions (Ramsahye et al., 2007). The mesoporous structure enables a high mass
transfer efficiency, and the hydrophilic characteristics facilitate a uniform dispersion
of MIL-53(Al) in aqueous solution (Pashley et al., 2005). These superiorities open up
the possibility of preparing monovalent selective CEMs with enhanced ion flux by
incorporating MIL-53(Al). In this study, novel organic-inorganic thin film composite
(TFC) monovalent CEMs were fabricated by a fast co-deposition of PDA/PEI
composites with MIL-53(Al). MIL-53(Al) was adopted as mesoporous component to
maintain a high ion flux while the positively charged PDA/PEI coatings ensure the
rejection of multivalent ions.
6.2 Results and discussion
6.2.1 Surface morphology and chemical structure of the membrane
The surface morphology of PDA/PEI based thin film nanocomposite (TFN)
monovalent selective CEMs with different contents of MIL (53)-Al (0%, 0.2%, 0.4%
and 0.6%) are shown in Fig. 6.1. It can be observed that co-deposition with a
PDA/PEI skin layer has no significant change on the surface morphology. After the
addition of MIL (53)-Al, small rougher dots can be observed, indicating an
inhomogeneous decoration of MIL (53)-Al. With increasing Mil(53)-Al loading to 0.4%
w/v, many nodules were found. The nodules on the membrane surface are attributed
to the aggregation of Mil(53)-Al. Despite the aggregation, the coverage becomes
higher for the membrane with higher Mil(53)-Al content. However, too much
Mil(53)-Al will result in a serious aggregation, which is difficult to deposit on the
substrate surface. As a consequence, defects on the membrane surface can be found as
indicated by the red circle in Fig. 2 (PDA-mil-30), while no obvious
Mil(53)-Alaggregration can be found in the blue circle. Therefore, the fabrication of
PDA/PEI modified membrane with suitable Mil(53)-Al nanoparticles incorporation is
particularly important.
Chapter 6
92
Fig. 6.1 SEM images of (M-0) unmodified ion exchange membrane and TFC monovalent
selective membrane surfaces. Nanoparticle loadings for PDA-0, PDA-mil-10, PDA-mil-20 and
PDA-mil-30 are 0.0%, 0.2%, 0.4% and 0.6% (w/v), respectively (The scale bar represents 2 µm)
During the modification process, a tight skin facial layer is formed on the membrane
surface through interactions such as π-π interaction, hydrogen bonding interaction,
and electrostatic interaction. According to the three-dimensional AFM surface
topography pictures presented in Fig. 6.2, the PDA/PEI decorated membrane has no
obvious changes on the surface morphology, which is in agreement with the SEM
results. The surface root mean square height (Rsq) was similar after Mil(53)-Al
decoration, whereas the maxium height (Rsz) increased from 15.4 nm to 23.0 nm
(Table 6.1). The nanoparticles on the membrane surface improve the Rsz; meantime,
the Rsq is stable due to the low loading degree. A higher Mil(53)-Al coverage with
more nanoparticles incorporation causes a further increase of the surface roughness.
When the amount of decorated Mil(53)-Al reached 0.6% (w/v), a distinct increase of
roughness was detected. This is in accordance with the aggregation of nanoparticles
found in the SEM images.
Chapter 6
93
Fig. 6.2 AFM topography of primary membrane and modified membranes (a. M-0; b. PDA-0; c.
PDA-mil-10; d. PDA-mil-20; e. PDA-mil-30)
Table 6.1 AFM surface roughness parameters of the pristine and modified membranes
Membrane type M-0 PDA-0 PDA-mil-10 DA-mil-20 DA-mil-30
Root mean square height (Rsq) 1.81 nm 1.86 nm 1.87 nm 1.99 nm 3.77 nm
Maximum height
(Rsz) 15.7 nm 15.4 nm 23.0 nm 24.6 nm 31.6 nm
Typical XPS spectra of M-0, PDA-0, PDA-mil-20 are shown in Fig. 6.3. For all the
membranes, C 1s, N 1s and O 1s characteristic peaks were obtained. In comparison
with M-0 membranes, the emission peak for N 1s is more intensive for the modified
membrane, which is assigned to the increased nitrogen content of PEI. For the
PDA-mil-20 membrane, although the Al and Cu concentration is too low to obtain Al
2s, Al 2p, Cu 2p1/2 and Cu 2p3/2 peaks, the elemental concentration given by Table 6.2
confirms the presence of Mil (53)-Al and Cu. For the method of PDA/PEI
co-deposition, the thickness of the modified function layer is generally around 100 nm
(Lv et al., 2015, 2016), which means a limited anchor effect while the size of the
nanoparticles is large enough. As indicated by SEM and AFM results, no obvious
aggregates with large size were observed. As a consequence, Mil (53)-Al
nanoparticles with large size failed to be fixed on the membrane surface, ultimately
leading to a low Al concentration.
Chapter 6
94
Fig. 6.3 XPS spectra of M-0, PDA-0 and PDA-mil-20 membranes
Table 6.2 Atomic concentrations of C, N, O and Al for M-0, PDA-0 and PDA-mil-20 membranes
Membrane
Element atomic concentrations (%)
C N O Al
M-0 73.5 4.0 19.7 0.0
PDA-0 69.0 6.2 23.2 0.0
PDA-mil-20 74.3 4.8 19.6 0.3
6.2.2 Contact angle, ion exchange capacity and water uptake
The water contact angle, ion exchange capacity and water uptake of the newly
developed membranes were studied to explore the effects of Mil (53)-Al nanoparticles.
The commercial cation exchange membranes showed an initial water contact angle of
31°, which indicates the high hydrophilic properties of the primary CEMs (Fig. 6.4).
After decoration by PDA/PEI composites, the membrane surface became more
hydrophobic as water contact angles increased to 77°. The decreased hydrophilicity
can be explained by the more hydrophilic nature of the primary CEMs and the
depletion of amine groups in PEI to react with acryl chloride groups in TMC (Yang et
Chapter 6
95
al.). The addition of hydrophilic Mil (53)-Al nanoparticles to TFC membranes greatly
enhances the surface hydrophilicity. However, no obvious variation was observed
with further increase the decorated Mil (53)-Al concentration.
Fig. 6.4 Contact angle, IEC and water uptake of PDA/PEI modified membranes with Mil(53)-Al
at different loadings (a. contact angle, b. IEC and water uptake)
The ion exchange capacity is responsible for the ionic conductivity of the membranes,
while the water uptake can affect the transport behavior of ions across the membrane.
After surface modification, both the IEC and water uptake exhibited a slight increase
from 1.34 mmol/g to 1.38 mmol/g and 33.4% to 37.6%, respectively. Since PDA/PEI
composites are positive charged via a synergetic effect of -NH2, -OH and -COOH
groups, the change of IEC is not as obvious as the water uptake. For the M-0 and
PDA-0 membranes, the surface composition was uniform while the surface of
PDA-mil-10, PDA-mil-20, and PDA-mil-30 was heterogeneous, the strong affinity of
Mil(53)-Al for water would give rise to hydrophilic regions on the membrane surface.
These hydrophilic regions formed around the cluster of chains lead to absorption of
water and attraction of protons. Furthermore, the established pores of Mil(53)-Al can
accommodate water molecules due to their relatively large sizes. The prepared
membrane with Mil(53)-Al shows an enhanced hydrophilicity, and a more positive
charge density, indicating a strong potential use for separation of monovalent ions.
Chapter 6
96
6.2.3 Diffusion dialysis experiments
Diffusion dialysis experiments were undertaken to understand the diffusional
ion-transport process, and particularly the effect of structural parameters with
different Mil(53)-Al content. The diffusion of salts from the concentrate chamber to
the diluate chamber caused an increase of the conductivity in the concentrated cell
(Fig. 6.5). For the diffusion experiments of the primary membrane M-0, after 1 h
self-diffusion, the conductivity of the concentrated compartment changed from
15.8 µS/cm to 258.8 µS/cm. The pristine interfacial polymerization between PDA/PEI
and TMC limits the diffusion process. As a result, the diffusion of NaCl for the
PDA-0 membrane becomes slower, and thus the conductivity change was reduced.
Theoretically, the hydrated radius of Na+ is around 3.0 Å and radius of Cl
- is 1.8 Å
(Kang et al., 2014; Tansel et al., 2006), which is smaller than the Mil(53)-Al pores
(8.5 Å) (Yang et al., 2013). Therefore, the presence of Mil(53)-Al on the TFC surface
can provide extra space to enhance the salts diffusion process. Furthermore, the
improvement of IEC and water uptake could form ionic transfer pathways on the
functional layer of the membrane surface, and facilitate the transport of salts.
However, the diffusion was mitigated when the concentration increased to 0.4% (w/v).
At this condition, ionic pathways on the membrane surface are occupied by the
increase of additive particles and narrowed as space limiting factors (Nemati et al.,
2015). When the Mil(53)-Al incorporation reaches 0.6% (w/v), the high concentration
of Mil(53)-Al particles tends to agglomerate to form larger particles, thus large filler
clusters are formed and salts can be easily transported. Although the conductivity
variation in the concentrate compartment was not obvious during the diffusion
experiments, the performance of the membranes after modification changed
significantly, which were confirmed by the following characterizations.
Chapter 6
97
Fig. 6.5 NaCl diffusion process on the surface of TFC membranes with different Mil(53)-Al
loadings
6.2.4 Electrochemical properties of membranes
Electrochemical impedance spectroscopy (EIS) is an important tool to enlighten
electrochemical phenomena related to membranes, allowing to quantify the resistance
of the membrane matrix. EIS results obtained in both NaCl and MgCl2 solutions are
shown in Fig. 6.6. However, the phenomena are very different. For the experiments
conducted in NaCl solution, the increased resistance indicates that the presence of the
PDA/PEI layer atop the CEMs hinders the ionic transport. A reduction of the
membrane resistance was observed after introduction of Mil(53)-Al, because porous
structures greatly facilitate the Na+ mitigation. As a consequence, Na
+ permeating
through the CEMs becomes easier and the membrane resistance is reduced. An
increase in electrical resistance arising from the increasing Mil(53)-Al loadings
implies that a hybrid membrane containing proper inorganic materials can
significantly improve the membrane performance, but an excessive proportion of the
inorganic materials leads to a high resistance (Mistry et al., 2008). Such a variability
in conductivity is in good agreement with the previous results of diffusion dialysis,
confirming that an excess of Mil(53)-Al disrupts the ion transfer pathways in this
particular system (Mistry et al., 2008). For the EIS results conducted in MgCl2
0 10 20 30 40 50 60
0
50
100
150
200
250
300
Co
nd
uctivity (s/c
m2)
T (min)
M-0
PDA-0
PDA-10
PDA-20
PDA-30
Chapter 6
98
solutions, two arcs appeared on the impedance spectrum. Typically, the geometric
arch at low frequencies is determined by the ionic migration while the high
frequencies of the Nyquist plot are the sum of the membrane resistance and the
solution resistance (Zhao et al., 2018). The total Ohmic resistance of the membrane
with different Mil(53)-Al loadings was in this order: PDA-mil-30 < PDA-0 <
PDA-mil-10 < PDA-mil-20 < M-0. It is remarkable that the resistance of M-0 is
higher than PDA-0. The lower resistance in these circumstances may be related to the
enhanced shielding effect of double layer compression between the positive PDA/PEI
modified layer and the negative CEMs, which reduced the electrostatic repulsion
between the modified layer and Mg2+
(Mo et al., 2008). In addition, the great affinity
of Mg2+
for the functional groups of the CEMs gives the membrane a high
electroconductance (Rodzik, 2005). The only difference of the geometric arch at high
frequencies is the diameter of the additional arch, which indicates the formed
functional layer on membrane surface can restrict the ions transport (Zhao et al.,
2017). With insight into the second diffusional arc at low frequencies, despite no
significant difference can be observed concerning the diameter of arch, the slight
reduction trend for membrane after modification confirms the restriction of the ion
migration through the diffusion layer, electrical double layer and the CEMs.
Fig. 6.6 Mil(53)-Al effect on EIS of membrane for NaCl and MgCl2 solutions
The I–V curves of membranes with different Mil(53)-Al loadings are shown in Fig.
6.7. For the M-0 membrane, the limiting current density did not clearly appear within
Chapter 6
99
measuring conditions. A membrane with low resistance could facilitate the transfer of
ions, therefore, concentration polarization can be effectively avoided. By this means,
the increased limiting current density ensured the application of ED under higher
current density and thus an increased efficiency (Ge et al., 2016a). After modification
by PDA/PEI, a plateau appears. This behavior corresponds to the formation of the
modified layer, which enhances the concentration polarization. Moreover, the
enhanced surface hydrophobicity can be another factor contributing to the increase of
the plateau length (Balster et al., 2007; Güler et al., 2014). For the Mil(53)-Al
incorporated membrane, no apparent plateau can be found, which confirms the
promotion effect of Mil(53)-Al for ion migration.
Fig. 6.7 Current voltage of M-0, PDA-0 and PDA-mil-20 membranes
6.2.5 Electrodialysis experiments
The desalination performance of the membranes was first investigated by
electrodialysis using single salt solutions (NaCl system and MgCl2 system,
respectively). In these experiments, a constant-voltage strategy was applied. When the
desalination process continues, the conductivity of the diluate compartment was
Chapter 6
100
recorded as a function of the desalination time. During the experiments, the volume
was stable for the diluate and concentrate compartments, thus the water flow across
the membrane can be neglected. The flux of Na+ and Mg
2+ reduced significantly after
the interfacial polymerization of PDA/PEI, nevertheless, no distinct effect on total
desalination ratio can be observed. Since the current density tends to decrease with the
increasing resistance of the diluate compartment, the concentration variation in the
diluate compartment would be expected to be minimized. To better understand this
specific process, the concentrations of Na+ and Mg
2+ at 30 min were used to calculate
the ion flux due to the large variations of concentrations in the first 30 min. As shown
in Fig. 6.8, higher Na+ and Mg
2+ fluxes were obtained after introducing the porous
Mil(53)-Al nanoparticles. The steric hindrance effect becomes much more obvious
with increasing the Mil(53)-Al content, thus the flux of Na+ and Mg
2+ was reduced.
However, the reduction of the flux of Mg2+
was not obvious while the flux of Na+ was
significantly reduced, which suggests a more obvious effect of the Mil(53)-Al content
on Na+ with smaller hydrated radius.
Fig. 6.8 Conductivity change of diluate compartment for different membranes at a. NaCl, b.MgCl2
systems and c. ion flux
Chapter 6
101
For the separation of monovalent from divalent cations, a comparison of monovalent
and multivalent fluxes is presented in Fig. 6.9. Likewise, PDA/PEI modified
membranes exhibited a much lower Mg2+
flux than the untreated membrane,
demonstrating the improvement of the steric hindrance effect. In contrast, the flux of
Na+ is comparatively higher to compensate for the reduction in the migration current
and the permselectivity of the modified membrane decreased from 1.26 to 0.37.
Incorporating Mil(53)-Al nanoparticles slightly enhances the flux of Na+, while the
Mg2+
flux is sustained. Improving the selectivity is typically at the expense of the flux
of monovalent ions; Mil(53)-Al nanoparticles incorporation can be an alternative way
to solve this problem. A further increase of the Mil(53)-Al content has no obvious
effect on the permselectivity; however, in this case, both the Na+ and Mg
2+ permeance
increased. This proves the desalination contribution from Mil(53)-Al, indicating a
higher current efficiency. For PDA-mil-30, the transfer resistance was further
decreased and the voids facilitated the Mg2+
transfer from the interior of the
membrane to the solution. In this case, Mg2+
with higher electrostatic interaction
occupied the ion exchange transfer sites, resulting in a lower Na+ flux and a higher
permselectivity. Furthermore, aggregates resulted in an uneven surface, so that the
difference in PDA/PEI polymer thickness also contributed to the lower selectivity.
Fig. 6.9 The ion flux and permselectivity of Na+/Mg
2+ system during ED
Chapter 6
102
Table 6.3 lists the ion selectivity values of commercial monovalent ion selective
CEMs (Luo et al., 2018). The ionic radii for Mg2+
and Ca2+
are 1.0 Å and 0.72 Å,
respectively. Thus, separation of Na+ with Ca
2+ tend to be easier than the separation of
Na+ with Mg
2+. By comparing the selectivity of the membrane modified by dopamine
and a commercial CSO membrane, it is obvious that the membrane prepared by
dopamine has a high selectivity.
Table 6.3 Reported selectivity of commercial monovalent ion selective IEMs
CMX CMS CSO
𝑃𝑁𝑎+𝐶𝑎2+
1.56 0.81 0.58
6.3 Conclusions
In conclusion, novel monovalent selective ion exchange membranes were fabricated,
with potential for large scale application. Moreover, the usage of PDA/PEI solution
can be performed in an economic and environmentally friendly way (Yang et al.,
2014a). Because of the high cationic charge density of PEI, an ultrathin PDA/PEI
selective layer was constructed to reject multivalent ions while Mil(53)-Al
nanoparticles would serve as a porous additive to enhance the ion flux. In particular,
the selectivity is maintained at a high level, demonstrating an excellent monovalent
selectivity. This study can provide new insights into utilizing mussel-inspired
materials for creating ion channels for various promising applications.
Chapter 7
103
7. Thin-Film-Nanocomposite Cation
ExchangeMembranes Containing Hydrophobic
Zeolitic Imidazolate Framework for Monovalent
Selectivity
Adapted from: J. Li, Z. Zhao, S. Yuan, J. Zhu, B. Van der Bruggen. High-Performance
Thin-Film-Nanocomposite Cation Exchange Membranes Containing Hydrophobic
Zeolitic Imidazolate Framework for Monovalent Selectivity. Applied Sciences.
135(24), 2018:45692.
7.1 Introduction
As mentioned in Chapter 5 and Chapter 6, monovalent selectivity is governed by the
affinity towards the fixed charge groups and the migration speed within the membrane
matrix (Ge et al., 2017). Based on these effects, IEMs with selectivity for specific ions
have been explored by including a thin charged skin layer or by generating a compact
functional layer on the surface of IEMs. Thin film composite (TFC) membranes,
comprising an ultrathin separating barrier layer prepared by interfacial polymerization
on top of a membrane, could be used in electrodialysis for the purpose of separating
multivalent ions from a mixed solution containing monovalent and multivalent ions
(Ge et al., 2016a). However, the inevitable higher area resistance caused by a surficial
functional layer increases the energy consumption at the same time. In recent years,
thin film nanocomposite (TFN) membrane emerged as a new type of composite
membranes, and have been widely studied and industrially applied (Jeong et al., 2007).
Nanoparticles are incorporated within the interfacial layer of the TFC membrane, with
the aim of enhancing the properties of the surface layer such as hydrophilicity,
permeability, selectivity, stability and surface charge density (Lau et al., 2015).
Currently, TFNs are widely used in forward osmosis (FO), reverse osmosis (RO) and
Chapter 7
104
nanofiltration (NF) (Amini et al., 2013; Peyravi et al., 2014; Safarpour et al., 2015).
However, they are seldom applied in electromembrane processes.
Metal-organic frameworks (MOFs), as a class of hybrid inorganic-organic solid
compounds, have gained interest due to their structural and functional tunability
(Kitagawa, 2014). They can serve as porous materials similar to zeolites while having a
better affinity for the polymeric chains (Sorribas et al., 2013). In addition, the flexibility
in pore size of MOFs can be controlled by choosing appropriate organic ligands and
inorganic secondary building units, which significantly broadens their application in
molecular sieving. Zeolitic imidazolate frameworks (ZIFs) are a subclass of MOFs
with large surface areas and pore volumes. Notably, ZIF-8, obtained by the reaction of
Zn2+
with 2-methylimidazole as linker, has gained great interest in membrane
utilization due to its high chemical and thermal stability (Sorribas et al., 2013; Van
Goethem et al., 2016; Wee et al., 2013). Duan et al. added 0.4 w/v% ZIF-8
nanoparticles to a TFN membrane with an 162% permeance increase while maintaining
a high salt rejection (Duan et al., 2015). However, the interfacial nanogaps present in
the functional layer of the decorated membrane cannot be fully avoided (Wang et al.,
2015b). ZIF-8 has a small aperture with a size of 3.4 Å and a comparatively large cavity
with a size of 11.6 Å (Shi et al., 2012). The small aperture of ZIF-8 can serve as an
effective filter to separate hydrated cations of Mg2+
(4.28 Å) through a size sieving
effect (Bazinet and Moalic, 2011a; Kang et al., 2014). Although there is a significant
variation on the reported size of hydrated Na+ (between 2.99 and 3.58 Å), in general,
the hydrated cations with larger crystal radii have weaker hydration shells, so that the
detachment of the hydration shell would occur when ions pass the solution-membrane
interface (Kotov et al., 2002; Tansel et al., 2006). Consequently, ZIF-8 has the right
size to separate dehydrated Na+ (0.95 Å) and hydrated Mg
2+ (Firdaous et al., 2007; Han
et al., 2015).
Different from the previously reported strategy to form a dense cationic charged layer
by chemical modification, this chapter presents an interfacial polymerization strategy
with nanoparticles to separate monovalent and multivalent ions. Porous ZIF-8 was used
as nanofiller underneath the surficial functional layer to separate monovalent ions. The
influence of the ZIF-8 content on the desalination performance was explored to
determine the optimal preparation parameters.
Chapter 7
105
7.2 Results and Discussion
7.2.1 Surface morphology and zeta potential
The surface morphology of the prepared membranes with different ZIF-8 contents is
displayed in Fig. 7.1. The surface of the primary Fujifilm membrane was flat, whereas
that of a MPD/TMC membrane (M-1) exhibited the typical ridge-and-valley feature.
This phenomenon resulted from the violent reaction between the small molecular
amine and acyl chloride. With increasing ZIF-8 loading to the surface layer, the
membrane showed a more net-like structure with denser and smoother zones. At higher
ZIF-8 loadings, especially at 0.08%, more ―cubic‖ structures are visible, which
represents that ZIF-8 nanoparticles are covered by polyamide. No direct relation
between the morphology and ZIF-8 concentration was observed, indicating that
nanoparticles were covered by a continuous PA film. It is believed that interfacial
polymerization is a diffusion controlled process. MPD adheres on the membrane
surface, diffuses and reacts with TMC in the organic phase. When no MPD permeates
across the barrier layer, the reaction stops. For the monovalent selective ion exchange
membrane that was prepared with ZIF-8 nanoparticles, ZIF-8 deposits on the
membrane first and then embedded under the PA layer, leading to similar SEM images
(Van Goethem et al., 2016). The elemental weight distribution on the membrane
surface was determined by EDAX to confirm the presence of ZIF-8. In Table 7.1, a low
Zn content can be observed. With increasing concentration of ZIF-8 nanoparticles, a
higher Zn content can be achieved. The EDAX mapping (Fig. 7.2) demonstrated a
uniform distribution of ZIF-8 on the membrane surface. However, the variation of the
Zn concentration is not obvious. Thus, the results of SEM and EDAX suggest a
successful encapsulation of ZIF-8 nanoparticles. The PA surface structures with
various crosslinking degree can be explored by further analyzing the element ratios
between O and N. For the modified membrane without ZIF-8 nanoparticles, the O/N
ratio is about 1.75. It was obvious that no distinct difference can be seen in the O/N
ratio with the increase of ZIF-8 concentration. This indicates that the crosslinking
degree of MPD/TMC copolymer was maintained stable when the ZIF-8 concentration
is blow 0.08%. The zeta potential of the as-prepared membranes is listed in Table 7.2.
Both the primary ion exchange membrane and the MPD/TMC modified membrane
Chapter 7
106
are negatively charged, with zeta potentials of −16.4 and −20.3 mV, respectively.
After introducing ZIF-8 into the selective layer, the zeta potential of the modified
membrane shifted to −22.3 mV. With a further increasing ZIF-8 content, the zeta
potential continued to decrease. It should be noticed that the slight reduction of zeta
potential is not consistent with the O/N ratio results that were obtained from EDAX,
which can be explained by the detection depth of EDAX. A detection depth around 1
μm for EDAX with complex background can reduce the impact of carboxylic acid
groups on O/N ratio. Thus, the ZIF-8 loading inside the MPD/TMC surface layer
could generate more carboxylic acid groups.
Fig. 7.1 SEM images of the unmodified and modified ion exchange membrane.
Nanoparticle loadings are (M-1) 0.00%, (M-2) 0.02%, (M-3) 0.04%, (M-4) 0.06%, and
(M-5) 0.08% (w/v).
Chapter 7
107
Fig. 7.2 EDAX mapping for the membrane after ZIF-8 incorporation (Light dots
are of Zn)
Table 7.1 Atomic concentrations of C, N, O and Zn obtained by EDAX results
C (%) N (%) O (%) Zn (%)
M-0 56.80 14.31 28.89 -
M-1 59.35 14.78 25.87 -
M-2 57.59 16.60 25.78 0.03
M-3 57.77 16.14 26.01 0.08
M-4 58.30 16.92 24.70 0.08
M-5 59.07 15.49 25.32 0.12
Table 7.2 Zeta potential results of the primary and modified membranes
Membrane M-0 M-1 M-2 M-3 M-4 M-5
Zeta potential (mV) -16.4 -20.7 -22.3 -23.7 -23.8 -24.5
The water contact angle measurements also confirmed the presence of ZIF-8 under the
PA film (Fig. 7.3). After modification, contact angles increased from 27.6° to 43.7°.
This indicates that the primary commercial cation exchange membranes are highly
hydrophilic; by introducing the MPD/TMC layer, the surface becomes more
hydrophobic. The contact angle increased from 43.7° to 78.9° for M-2 after
incorporation ZIF-8 nanoparticles. The addition of a certain amount of ZIF-8
nanoparticles to TFN membranes greatly reduces the surface hydrophilicity, which is in
accordance with previous research (Lind et al., 2009). Assuming that some ZIF-8
nanoparticles are bared on the modified layer, water contact angles would increase with
the increase of the ZIF-8 nanoparticles content. However, a comparatively more
Chapter 7
108
hydrophilic surface was obtained with increasing ZIF-8 nanoparticles content, which
means that ZIF-8 nanoparticles tend to be covered by the polyamide layer (Duan et al.,
2015; Lind et al., 2009). As was proved by previous zeta potential results, increased
ZIF-8 loadings suggest an increased surface charge density of the MPD/TMC surface
layer, and thus increased carboxylic acid groups lead to an increased hydrophilicity.
Furthermore, For hydrophilic materials, the increased roughness could also contribute
to the enhancement of the hydrophilicity (Wang et al., 2015a). However, the slight
change of the zeta potential as well as the comparative poor hydrophilicity of the
membrane surface finally caused a small variation of the contact angle. As shown in
Fig. 7.4 and Table 7.3, the pristine membrane has a smooth surface, with a roughness of
1.0 nm. After modification, the Rsq significantly increases to 36.3 nm. By raising the
initial concentration of ZIF-8 to 0.08%, the surface roughness increased to 70.8 nm,
with the contact angles decreasing from 78° to 71° (Erbil et al., 2003; Nabe et al., 1997).
As a consequence, a further increase of ZIF-8 nanoparticles loadings tends to reduce
the contact angles.
Fig. 7.3 Contact angle of the modified ion exchange membranes
0
10
20
30
40
50
60
70
80
90
100
M-5
M-4M-3M-2
M-1
M-0
Co
nta
ct
an
gle
()
Chapter 7
109
Fig. 7.4 AFM topography of the membrane with and without modification
Table 7.3 AFM surface roughness parameters of the pristine and modified membranes
Membrane type M-0 M-1 M-2 M-3 M-4 M-5
Root mean square height (Rsq) 1.0 nm 36.3 nm 43.4 nm 51.5 nm 65.2 nm 70.8 nm
Maximum height (Rsz) 4.4 nm 210 nm 230 nm 274 nm 347 nm 335 nm
7.2.2 IEC and water uptake
The ion-exchange capacity yields the ionic conductivity of the membranes, while the
water uptake can affect the transport behavior of ions across the membrane. The IEC
and water uptake as a function of ZIF-8 content for all the prepared membranes are
presented in Fig. 7.5. Changes of the IEC and water uptake can be observed after
surface modification of the MPD/TMC layer from 1.34 mmol/g to 1.42 mmol/g and
33.4% to 30.6%, respectively. Since MPD/TMC composites are negatively charged
(Kang and Cao, 2012), the IEC increases after introducing more negative functional
groups. The dense structure of the MPD/TMC layer is more hydrophobic, which
indicates the reduced water molecular accessibility to the surface matrix. Therefore, an
Chapter 7
110
increased IEC and a reduced water uptake for the MPD/TMC modified membrane were
obtained. Furthermore, all the membranes exhibited an increase in IEC and water
uptake with increasing ZIF-8 content. Besides, the large cavity of ZIF-8 can
accommodate water molecules inside the nanoparticles, which can be another factor
contributed to the higher water uptake. Consequently, the prepared PA/ZIF-8
membrane surface has an enhanced water uptake, and a more negative charge density
with increased ZIF-8 content.
Fig. 7.5 IEC and water uptake of the membranes with different ZIF-8 loadings
7.2.3 Diffusion dialysis experiments
Diffusion dialysis using the synthesized membranes with different ZIF-8 content was
carried out to study the diffusional transport process. In Fig. 7.6, it can be seen that the
conductivity of the permeate side increases during the experiments, which is caused by
the diffusion of salts from the high concentration chamber to the low concentration
chamber. However, the conductivity change rate is different. After 1 h self-diffusion,
the conductivity of the M-0 membrane changed from 10 µS/cm to 275 µS/cm. The
interfacial polymerization between MPD and TMC limited the diffusion process; as a
consequence, the diffusion of NaCl for the M-1 membrane becomes slower, and the
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
IEC
(m
mo
l/g
)
28
30
32
34
36
38
40
M-0 M-5M-4M-3M-1 M-2
Wa
ter u
pta
ke
(%)
Chapter 7
111
conductivity change reduced. Theoretically, the radius of Cl- is around 1.8 Å (Kang et
al., 2014; Tansel et al., 2006), which is smaller than the ZIF-8 pores (3.4 Å). Therefore,
adding a small amount of ZIF-8 to the TFN increased the NaCl flux. With a higher
ZIF-8 loading in the surface selective layer, the salt flux further increases, which can be
explained by the following two reasons. First, increased ZIF-8 loadings introduce more
free space that can facilitate the salt migration. More importantly, the structural
changes with ZIF-8 incorporation and the possible voids between the organic/inorganic
interphase could facilitate the NaCl permeance (Chung et al., 2007; Duan et al., 2015).
Fig. 7.6 Conductivity change of diluate compartment during diffusion process
7.2.4 Membrane resistance
As can be observed in Fig. 7.7, for the M-0 membrane, the membrane resistance
increased by a factor 1.5 for MgCl2 solution compared to the NaCl solution. The
increased resistance for NaCl solutions indicates that the presence of the MPD/TMC
layer near the cation-exchange membrane hinders the ionic transport. Since the similar
IEC and water uptake yield a relative constant resistance, the transport of Na+ is not
much affected by the ZIF-8 incorporation. In the case of Mg2+
, a reduction of the
membrane resistance was observed, because carboxylic acids are week acids so that the
interaction with Mg2+
was greatly mitigated. As a consequence, the Mg2+
permeation
through the cation exchange membrane becomes easier and the membrane resistance is
0 10 20 30 40 50 600
50
100
150
200
250
300
350
400
Co
nd
uctivity (s/c
m)
T (min)
M-0
M-1
M-2
M-3
M-4
M-5
Chapter 7
112
reduced. The variation of resistance with ZIF-8 content on the membrane surface was
large. Since incorporation of ZIF-8 would increase the water uptake, in this regard, a
further enhancement of membrane conductivity is obtained. However, with the increase
of ZIF-8 loading, the membrane resistances continue to increase. In this condition, the
steric hindrance effect of ZIF-8 and the enhanced crosslinking play a much more
important role, rather than the electrostatic effect. In contrast, while the ZIF-8 loading
reached 0.8%, a higher filler concentration caused unselective voids with a significant
drop in rejection. As a result, the membrane resistance in MgCl2 solution drops again.
Fig. 7.7 ZIF-8 effect on EIS of membrane for NaCl and MgCl2 solutions
7.2.5 Electrodialysis experiments
The electrochemical behavior of the modified membranes was investigated by
electrodialysis using single NaCl and MgCl2 systems. During the NaCl desalination
experiment, 2 g/L NaCl solutions were used as diluate and concentrate compartment,
respectively; while they were replaced by 2 g/L MgCl2 solutions during the MgCl2
purification experiment. A Na2SO4 solution with a concentration of 20 g/L was used as
the electrode rinsing solution while the current density was maintained at 15.3 mA
cm−2
. The conductivity of the diluate compartment decreased with the purification time
for both systems (Fig. 7.8). No obvious variation on desalination performance can be
observed after surface modification and the incorporation of nanoparticles. For ED,
constant-current or constant-voltage can be applied as operating mode. In previous
studies using a constant-voltage strategy, the conductivity of the diluate compartment
Chapter 7
113
continually decreased. The elevated system resistance, and thus lower current density,
retards the transfer of ions through the ion exchange membrane. Since a
constant-current system could maintain a stable current, the effect of a variation of the
current density on the concentration change of the diluate compartment would be
expected to be minimized. To obtain a better understanding of this specific process, the
concentrations of Na+ and Mg
2+ at 15 min were considered to calculate the cation flux,
due to the fact that the voltage of the system would exceed the maximum voltage of the
power supply after 15 min. It can be seen from Fig. 7.8c that a reduction of the Na+ and
Mg2+
ion flux is obtained after surface modification. Conversely, a higher Na+ transport
and a lower Mg2+
transport were found after introducing the ZIF-8 nanoparticles.
Generally, the increased IEC could facilitate the transport of Na+ and Mg
2+, however,
ZIF-8 under the PA surface layer could hinder the Mg2+
transport. The results were
different from previous results, which were carried out by incorporating Mil53-(Al)
nanoparticles in Chapter 6. The free diameter of Mil53-(Al) is close to 0.85 nm, which
makes it easier to transfer Na+ and Mg
2+ than ZIF-8. With more ZIF-8 incorporated into
the membrane matrix, both the Na+ and Mg
2+ transport increased, which can be
explained by the higher IEC and the formation of unselective voids. When compared
with the membrane with Mil53-(Al), the interfacial polymerization with ZIF-8 method
provides an enhanced IEC and diffusion ability with a lower water uptake. Furthermore,
a higher electro-resistance to Mg2+
ions could impose the membrane with the
possibility to separate monovalent ions. The energy consumption that was required
during the ED process was also considered (Fig. 7.8d). The energy consumption was
found to decrease with increasing ZIF-8 loadings, which confirmed the facilitated
migration of cations.
Chapter 7
114
Fig. 7.8 a. Conductivity change of diluate compartment for NaCl system; b.
Conductivity change of diluate compartment for MgCl2 system; c. Flux of ions in 15 min
and d. Energy consumption for different systems.
7.2.6 Monovalent selectivity
For the separation of monovalent and divalent cations, a comparison of monovalent and
multivalent cations fluxes is presented in Fig. 7.9. In the binary mixtures, MPD/TMC
modified membranes show a much lower Na+ and Mg
2+ flux than the untreated
membrane, demonstrating the improvement of the steric hindrance effect.
Simultaneously, the monovalent selectivity of the modified membrane increased from
1.77 to 3.66. Particularly, with incorporating ZIF-8 nanoparticles, the monovalent
selectivity notably increased to 4.03. However, the flux of Na+ reduces, which is
different from the observations in single salt desalination. The larger affinity of Mg2+
for the ion exchange groups inside the membrane matrix would allow them to occupy
more ions exchange transfer sites; as a consequence, a strong suppression was imposed
on the transfer of Na+ ions. Furthermore, the more hydrophobic membrane surface
could reduce the permeation of strongly hydrated cations, while facilitating the less
hydrated ones (Sata, 2000). Combining the contribution of the dense MPD/TMC layer
as well as the size sieving effect of ZIF-8, the monovalent selectivity was greatly
enhanced. Further increasing the ZIF-8 content has no obvious effect on the
Chapter 7
115
monovalent selectivity; however, in this case, both the Na+ and Mg
2+ permeance
increased. At this condition, the increased amount of ZIF-8 reduced the thickness of the
dense surface, which would also contribute to the increase of the ion flux. For M-5,
although the transfer resistance was further decreased, the voids facilitated both the Na+
and Mg2+
migration from the membrane matrix to the solution. In this case, a lower
monovalent selectivity was obtained.
Fig. 7.9 The ion flux and monovalent selectivity of Na+/Mg
2+ system during ED
The results clearly demonstrated that porous structure of the surface skin layer
decreased the transfer resistance of ions and improved the flux of monovalent ions.
The dense layer could reject the divalent ions with larger Stokes radius effectively.
The selectivity of the modified membrane is similar to commercial CSO
monovalent ion exchange membranes, however, the flux is restricted due to the
dense polyamide layer (Fig. 7.10).
Mg2+
Flux
Na+ flux
0
1
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
M-5M-4M-3M-2M-1M-0
Monovalent
selectivity
Flu
x 1
08 (
mo
lcm
-2s
-1)
Pe
rm-s
ele
ctiv
ity
Chapter 7
116
Fig. 7.10 The ion flux and perm-selectivity of Na+/Mg
2+ system (Ge et al., 2016a)
7.3 Conclusions
ZIF-8 was successfully anchored under the skin layer of commercial ion exchange
membranes by interfacial polymerization. The monovalent selectivity of the modified
membrane increased from 1.77 to 4.03, which is more than a twofold increase.
Furthermore, during the separation process, the Na+ flux maintained a similar level
compared to the primary membrane, while the Mg2+
flux was significantly reduced.
When single salt solution purification experiments were conducted, the ZIF-8
incorporation could decrease the energy consumption by facilitating the ion transport.
Considering the increased monovalent selectivity and the enhanced Na+ flux,
introducing nanoparticles to the surface functional layer could be a promising way to
enhance the ion flux in monovalent selective ion exchange membrane applications.
Chapter 8
117
8. Conclusions and recommendations for further
research
8.1 General conclusions
Separation processes play an extraordinary role in modern industries (where they
comprise 40-70% of both capital and operational costs) due to their three primary
functions: concentration, fractionation, and purification. The ultimate goal of
separation technologies is to achieve the precise and rapid separation of different
molecules from aqueous solutions, organic solutions and gas mixtures. These features
are necessary and significant for modern industries, as well as for the environment in
view of resource saving and recovery, and sustainable development. Simultaneously,
the pursuit of advanced separation technologies is never-ending.
Membrane-based separation technology, emerging as a promising tool to fulfill
separations, has aroused huge interests in recent years. ED, a type of technology
which arranges ion-exchange membranes alternately in a direct current field, has been
widely used to demineralize, concentrate and/or convert salt-containing solutions.
Over the past decade, the development of IEMs has attracted a large amount of
research attention in fields of materials, preparation and applications, due to their
academic and industrial values. The properties of ion-exchange membranes are
determined by different parameters, such as the density of the polymer network, the
hydrophobic or hydrophilic character of the matrix polymer, the type and
concentration of the fixed charges in the polymer, and the morphology of the
membrane itself. The most desired properties of IEMs are as follows: high
permselectivity, low electrical resistance, good mechanical and form stability, high
chemical and thermal stability and low production costs. Many of today's available
membranes meet most of the required properties mentioned above, however, more
attention should be paid to the development of ion-exchange membranes with higher
Chapter 8
118
permselectivity, lower electrical resistance and better chemical and thermal stability at
lower cost. To date, most of the IEMs consist of polymeric backbones prepared by
either post-functionalization of pre-existing polymers or direct polymerization of
functionalized monomers. Particularly, with the rapid progress in nano-science, the
regulation and control of polymer structures allow for the formation of ionic channels,
which is a new development in this field.
In this thesis, new preparation methods and rationally designed nanomaterials are
explored to effectively fabricate porous ion exchange membranes and monovalent
selective ion exchange membrane, aiming at reducing the limitations of low
permselectivity, in-sufficient ion flux, and high electrical resistance. First, a dry-wet
phase-inversion strategy by combining immersion precipitation and dry-casting was
used to control the membrane porosity with the purpose of improving the physical and
electrochemical properties of ion-exchange membranes. In addition to control the
porosity to balance the membrane electrical resistance with the diffusion caused by
the concentration gradient, it was experimentally shown that the porosity can
influence the IEC and water uptake of the membrane and, thus, further affect the
resistance. As demonstrated by experimental data for desalination by electrodialysis, it
was found that a membrane dried at 60 °C for 1 h had the highest desalination
efficiency. This is mainly because porous membranes facilitate the transport of ions.
The membrane prepared with 1-h heating time had more steric hindrance, which can
decrease the diffusion of ions, so that a superior desalination efficiency can be
obtained. Furthermore, a membrane with higher density of functional groups was
found to have a higher desalination efficiency, because of the electrostatic effect of the
membrane. Remarkably, under optimal membrane preparation conditions, the
desalination efficiency reached 95%, and the current efficiency reached 100%. It was
concluded that the performance of a porous membrane with controllable porosity can
enhance the ED process with respect to energy efficiency and desalination efficiency.
New methods of fabricating membranes with pores such as immersion precipitation
and dry-casting are thought to be potential routes to decreasing the electrical
Chapter 8
119
resistance.
On the other hand, an efficient one-step chemical process to graft a thin PANI layer on
the surface and pores of an UF membrane is reported. During the desalting procedure,
the driving force has two contributions, namely the electrical field and the
salinity-gradient. Initially, transport of ions by the electrical field is dominant, while at
the end of the experiment, diffusion dialysis by the salinity gradient plays a larger role
in ion transport. In single salt solutions, the presence of PANI can hinder Mg2+
transport from the concentrate to the diluate compartment because of the electrostatic
effect, while there is no obvious effect on Na+ ions. In the binary system, the flux of
Na+ slightly increased to 12.4×10
-8 mol·cm
–2·s
–1 while the flux of Mg
2+ reduced to
3.1×10-8
mol·cm–2
·s–1
, so that the permselectivity is almost doubled. It can be
concluded that this method is suitable for preparing membranes based on UF
membranes for efficient separation of monovalent ions by electro-driven separation
techniques.
Furthermore, a simple method is proposed for fabricating monovalent selective ion
exchange membranes based on the rapid co-deposition of biomimetic adhesive
polydopamine and poly(ethylene imine) by using CuSO4/H2O2 as a trigger. Through
this strategy, the surface properties and the permselectivity of the membranes can be
easily tailored by the addition of PEI and by tuning the PEI molecular weight. Surface
characterization revealed that overall enhanced surface properties including low
roughness, favorable hydrophilicity, and enhanced positive charge can be achieved
after the addition of PEI. The optimum membranes, with 4 h co-deposition of 60 mg
PDA and 120 mg PEI in 50 mL Tris buffer solution, permselectivity of
SPES-PDA/PEI-2 was 2.5 times higher than that of the SPES membrane, especially,
the flux of H+ was enhanced. In addition, the PDA/PEI modified ion exchange
membrane shows an excellent operational stability for monovalent separation
performance after immersion in acid and alkaline solution for 7 days. Comparing
membranes prepared with different molecular weights of PEI, results revealed that
Chapter 8
120
modification with a lower molecular weight PEI yields a higher selectivity. This facile
strategy may provide new opportunities not only to develop monovalent selective ion
exchange membranes but also to engineer the surface of numerous materials in energy
and environmental applications.
Similarly, taking advantage of the nanochannels of MIL(53)-Al, monovalent selective
membranes were prepared through rapid codeposition of PDA/PEI and Mil(53)-Al,
followed by cross-linking with TMC. The positive −NH2 allows to reject multivalent
cat ions, while porous Mil(53)-Al can accelerate the migration of Na+. The effects of
the embedded nanoparticles on the physicochemical properties of the prepared
membranes, and on the monovalent selective performance were investigated. A mass
ratio of 0.2–0.4% (w/v) for Mil(53)-Al is the optimum protocol, yielding a membrane
with a permselectivity of about 0.3 and an ion flux of about 22.0 and 0.6 mol·cm–2
·s–1
for Na+ and Mg
2+, respectively. At this condition, the PDA-coated membrane
maintains a high monovalent selectivity with enhanced Na+ and Mg
2+ flux in single
salt solutions.
In addition, a similar material ZIF-8 was used to replace MIL(53)-Al for fabricating
monovalent selective ion exchange membrane via interfacial polymerization. No
significant changes of the surface structure of the PA/ZIF-8 based membranes were
observed. Nevertheless, the presence of ZIF-8 under the PA layer plays a key role in
the separation process. For single salt solutions that were applied in electrodialysis
(ED), faster transport of Na+ and Mg
2+ was obtained after introducing the ZIF-8
nanoparticles, however, the desalination efficiency remained constant. When the
hybrid membranes were applied to electrodialysis for binary mixtures containing Na+
as well as Mg2+
, it was demonstrated that the monovalent selectivity and Na+ flux
were enhanced by a higher ZIF-8 loading. Considering the superior performance
derived from ZIF-8 hybrid surface layer, a promising future of ZIF-8 based
nanocomposites as a surface functional agent for versatile applications is anticipated.
Chapter 8
121
In conclusion, ED is an economic process using ion exchange membranes for
producing drinking water when the salinity of target water is below 5 g/L. In order to
generate high quality water to meet the requirements of specific industrial processes,
membranes with low resistance and/or selectivity for given ions are required.
Introducing the porosity to the membrane matrix and skin layer by suitable membrane
formation techniques or nanoparticles incorporation can be an efficient way to reduce
the resistance and enhance the flux of ions.
8.2 Recommendations for further research
In view of the experiments and conclusions obtained in this thesis, some
recommendations can be made for further research.
1. In Chapter 3, a dry-wet phase-inversion method was adapted to fabricate porous
ion exchange membranes with the purpose of improving the physical and
electrochemical properties of ion-exchange membranes. However, the pore size
and the size distribution of porous ion exchange membrane should also be
intensively evaluated. In addition, nanoparticles incorporation could be another
option to preciously control the porosity.
2. In Chapter 4, a monovalent selective ion exchange membrane was prepared based
on a UF membrane. If the UF membrane was replaced by a porous ion exchange
membrane with good properties, the results may be more promising. Such as the
membrane used in Chapter 3 or the membrane prepared with other methods to
control the ionic channels.
3. In Chapter 6 and Chapter 7, beyond the effect of the nanocomposite amount, the
size and the size distribution of the nanocomposite should also be intensively
evaluated as they could have a large influence on the resultant membrane.
Chapter 8
122
4. Except ZIF-8 and MIL(53)-Al, the potential of other nanoparticles such as MOFs
and carbon nanotubes should also be investigated. Two-dimensional (2D)
materials have emerged as nano-building blocks to develop high-performance
separation membranes that feature unique nanopores and/or nanochannels. Other
two-dimensional (2D) materials such as porous graphene, 2D MOFs,
molybdenum disulfide (MoS2), MXene and C3N4 should also be studied as
options for monovalent selective ion exchange membrane fabrication.
5. Nanoparticles with different pore size can be chosen to separate specific ions.
6. Since the membranes were prepared on lab scale, the membranes should be
investigated in view of industrial applications.
7. In this thesis, the binary mixture solution were H+/Zn
2+ or Na
+/Mg
2+ solutions.
However, real waste water is much more complicated. Therefore, the applicability
of monovalent selective ion exchange membranes in real industrial waste water
should be evaluated.
Through the invention of new methods, configurations and processes as well as by the
improvement of membrane performances, as presented here, it can be expected that
electrodialysis with novel ion exchange membrane will play a more important role in
waste water purification/reclamation.
References
123
References
Abdu, S., Martí-Calatayud, M.-C.s., Wong, J.E., García-Gabaldon, M., Wessling, M. (2014)
Layer-by-layer modification of cation exchange membranes controls ion selectivity and water splitting.
ACS Appl Mater Interfaces 6, 1843-1854.
Aider, M., Brunet, S., Bazinet, L. (2008) Electroseparation of chitosan oligomers by electrodialysis
with ultrafiltration membrane (EDUF) and impact on electrodialytic parameters. Journal of Membrane
Science 309, 222-232.
Aider, M., Brunet, S., Bazinet, L. (2009) Effect of solution flow velocity and electric field strength on
chitosan oligomer electromigration kinetics and their separation in an electrodialysis with ultrafiltration
membrane (EDUF) system. Separation and Purification Technology 69, 63-70.
Åkerman, S., Viinikka, P., Svarfvar, B., Järvinen, K., Kontturi, K., Näsman, J., Urtti, A., Paronen, P.
(1998) Transport of drugs across porous ion exchange membranes. Journal of controlled release 50,
153-166.
Al-Maythalony, B.A., Alloush, A.M., Faizan, M., Dafallah, H., Elgzoly, M.A., Seliman, A.A.,
Al-Ahmed, A., Yamani, Z.H., Habib, M.A., Cordova, K.E. (2017) Tuning the Interplay between
Selectivity and Permeability of ZIF-7 Mixed Matrix Membranes. ACS Appl Mater Interfaces.
Alvarez, F., Alvarez, R., Coca, J., Sandeaux, J., Sandeaux, R., Gavach, C. (1997) Salicylic acid
production by electrodialysis with bipolar membranes. Journal of Membrane Science 123, 61-69.
Amara, M., Kerdjoudj, H. (2003) Modification of cation-exchange membrane properties by
electro-adsorption of polyethyleneimine. Desalination 155, 79-87.
Amini, M., Jahanshahi, M., Rahimpour, A. (2013) Synthesis of novel thin film nanocomposite (TFN)
forward osmosis membranes using functionalized multi-walled carbon nanotubes. Journal of
Membrane Science 435, 233-241.
Asraf-Snir, M., Gilron, J., Oren, Y. (2016) Gypsum scaling of anion exchange membranes in
electrodialysis. Journal of Membrane Science 520, 176-186.
Balster, J., Yildirim, M., Stamatialis, D., Ibanez, R., Lammertink, R.G., Jordan, V., Wessling, M. (2007)
Morphology and microtopology of cation-exchange polymers and the origin of the overlimiting current.
The Journal of Physical Chemistry B 111, 2152-2165.
References
124
Barragan, V., Ruız-Bauzá, C. (1998) Current–voltage curves for ion-exchange membranes: a method
for determining the limiting current density. J Colloid Interface Sci 205, 365-373.
Bayramoğlu, G., Metin, A.Ü., Altıntas, B., Arıca, M.Y. (2010) Reversible immobilization of glucose
oxidase on polyaniline grafted polyacrylonitrile conductive composite membrane. Bioresour Technol
101, 6881-6887.
Bazinet, L., Moalic, M. (2011a) Coupling of porous filtration and ion-exchange membranes in an
electrodialysis stack and impact on cation selectivity: A novel approach for sea water demineralization
and the production of physiological water. Desalination 277, 356-363.
Bazinet, L., Moalic, M. (2011b) Coupling of porous filtration and ion-exchange membranes in an
electrodialysis stack and impact on cation selectivity: A novel approach for sea water demineralization
and the production of physiological water. Desalination 277, 356-363.
Belashova, E., Melnik, N., Pismenskaya, N., Shevtsova, K., Nebavsky, A., Lebedev, K., Nikonenko, V.
(2012) Overlimiting mass transfer through cation-exchange membranes modified by Nafion film and
carbon nanotubes. Electrochimica Acta 59, 412-423.
Boyaval, P., Seta, J., Gavach, C. (1993) Concentrated propionic acid production by electrodialysis.
Enzyme and microbial technology 15, 683-686.
Buonomenna, M., Macchi, P., Davoli, M., Drioli, E. (2007) Poly (vinylidene fluoride) membranes by
phase inversion: the role the casting and coagulation conditions play in their morphology, crystalline
structure and properties. European Polymer Journal 43, 1557-1572.
Bux, H., Chmelik, C., Krishna, R., Caro, J. (2011) Ethene/ethane separation by the MOF membrane
ZIF-8: molecular correlation of permeation, adsorption, diffusion. Journal of Membrane Science 369,
284-289.
Chamoulaud, G., Bélanger, D. (2005) Modification of ion-exchange membrane used for separation of
protons and metallic cations and characterization of the membrane by current–voltage curves. J Colloid
Interface Sci 281, 179-187.
Chien, H.-W., Kuo, W.-H., Wang, M.-J., Tsai, S.-W., Tsai, W.-B. (2012) Tunable micropatterned
substrates based on poly (dopamine) deposition via microcontact printing. Langmuir 28, 5775-5782.
Chlanda, F.P., Lee, L.T., Liu, K.-J., (1978) Bipolar membranes and method of making same. Google
Patents.
Choi, P., Jalani, N.H., Datta, R. (2005) Thermodynamics and proton transport in Nafion I. Membrane
References
125
swelling, sorption, and ion-exchange equilibrium. Journal of the Electrochemical Society 152,
E84-E89.
Choi, Y.-S., Kang, H., Kim, D.-G., Cha, S.-H., Lee, J.-C. (2014) Mussel-inspired dopamine-and
plant-based cardanol-containing polymer coatings for multifunctional filtration membranes. ACS Appl
Mater Interfaces 6, 21297-21307.
Chung, T.-S., Jiang, L.Y., Li, Y., Kulprathipanja, S. (2007) Mixed matrix membranes (MMMs)
comprising organic polymers with dispersed inorganic fillers for gas separation. Progress in Polymer
Science 32, 483-507.
Coster, H.G., Chilcott, T.C., Coster, A.C. (1996) Impedance spectroscopy of interfaces, membranes and
ultrastructures. Bioelectrochemistry and Bioenergetics 40, 79-98.
Długołęcki, P., Anet, B., Metz, S.J., Nijmeijer, K., Wessling, M. (2010a) Transport limitations in ion
exchange membranes at low salt concentrations. Journal of Membrane Science 346, 163-171.
Długołęcki, P., Ogonowski, P., Metz, S.J., Saakes, M., Nijmeijer, K., Wessling, M. (2010b) On the
resistances of membrane, diffusion boundary layer and double layer in ion exchange membrane
transport. Journal of Membrane Science 349, 369-379.
Donnan, F.G. (1911) Theorie der Membrangleichgewichte und Membranpotentiale bei Vorhandensein
von nicht dialysierenden Elektrolyten. Ein Beitrag zur physikalisch‐chemischen Physiologie. Berichte
der Bunsengesellschaft für physikalische Chemie 17, 572-581.
Doyen, A., Beaulieu, L., Saucier, L., Pouliot, Y., Bazinet, L. (2011a) Demonstration of in vitro
anticancer properties of peptide fractions from a snow crab by-products hydrolysate after separation by
electrodialysis with ultrafiltration membranes. Separation and purification technology 78, 321-329.
Doyen, A., Beaulieu, L., Saucier, L., Pouliot, Y., Bazinet, L. (2011b) Impact of ultrafiltration membrane
material on peptide separation from a snow crab byproduct hydrolysate by electrodialysis with
ultrafiltration membranes. Journal of agricultural and food chemistry 59, 1784-1792.
Doyen, A., Udenigwe, C.C., Mitchell, P.L., Marette, A., Aluko, R.E., Bazinet, L. (2014) Anti-diabetic
and antihypertensive activities of two flaxseed protein hydrolysate fractions revealed following their
simultaneous separation by electrodialysis with ultrafiltration membranes. Food chemistry 145, 66-76.
Duan, J., Pan, Y., Pacheco, F., Litwiller, E., Lai, Z., Pinnau, I. (2015) High-performance polyamide
thin-film-nanocomposite reverse osmosis membranes containing hydrophobic zeolitic imidazolate
framework-8. Journal of Membrane Science 476, 303-310.
References
126
Elimelech, M., Phillip, W.A. (2011) The future of seawater desalination: energy, technology, and the
environment. Science 333, 712-717.
Eliseeva, T., Shaposhnik, V., Krisilova, E., Bukhovets, A. (2009) Transport of basic amino acids
through the ion-exchange membranes and their recovery by electrodialysis. Desalination 241, 86-90.
Eliseeva, T., Tekuchev, A.Y., Shaposhnik, V., Lushchik, I. (2001) Electrodialysis of amino acid
solutions with bipolar ion-exchange membranes. Russian Journal of Electrochemistry 37, 423-426.
Erbil, H.Y., Demirel, A.L., Avcı, Y., Mert, O. (2003) Transformation of a simple plastic into a
superhydrophobic surface. Science 299, 1377-1380.
Farrokhzad, H., Moghbeli, M., Van Gerven, T., Van der Bruggen, B. (2015) Surface modification of
composite ion exchange membranes by polyaniline. Reactive and Functional Polymers 86, 161-167.
Ferrer, J.J., Laborie, S., Durand, G., Rakib, M. (2006) Formic acid regeneration by electromembrane
processes. Journal of Membrane Science 280, 509-516.
Fiksdal, L., Leiknes, T. (2006) The effect of coagulation with MF/UF membrane filtration for the
removal of virus in drinking water. Journal of Membrane Science 279, 364-371.
Firdaous, L., Dhulster, P., Amiot, J., Gaudreau, A., Lecouturier, D., Kapel, R., Lutin, F., Vézina, L.-P.,
Bazinet, L. (2009) Concentration and selective separation of bioactive peptides from an alfalfa white
protein hydrolysate by electrodialysis with ultrafiltration membranes. Journal of membrane science 329,
60-67.
Firdaous, L., Malériat, J.P., Schlumpf, J.P., Quéméneur, F. (2007) Transfer of monovalent and divalent
cations in salt solutions by electrodialysis. Separation Science and Technology 42, 931-948.
Fu, X., Wang, S., Xia, Z., Li, Y., Jiang, L., Sun, G. (2016) Aligned polyaniline nanorods in situ grown
on gas diffusion layer and their application in polymer electrolyte membrane fuel cells. International
Journal of Hydrogen Energy 41, 3655-3663.
Güler, E., van Baak, W., Saakes, M., Nijmeijer, K. (2014) Monovalent-ion-selective membranes for
reverse electrodialysis. Journal of Membrane Science 455, 254-270.
Gao, H., Sun, Y., Zhou, J., Xu, R., Duan, H. (2013) Mussel-inspired synthesis of
polydopamine-functionalized graphene hydrogel as reusable adsorbents for water purification. ACS
Appl Mater Interfaces 5, 425-432.
Ge, L., Liu, X., Wang, G., Wu, B., Wu, L., Bakangura, E., Xu, T. (2015) Preparation of proton selective
membranes through constructing H+ transfer channels by acid–base pairs. Journal of Membrane
References
127
Science 475, 273-280.
Ge, L., Wu, B., Li, Q., Wang, Y., Yu, D., Wu, L., Pan, J., Miao, J., Xu, T. (2016a) Electrodialysis with
nanofiltration membrane (EDNF) for high-efficiency cations fractionation. Journal of Membrane
Science 498, 192-200.
Ge, L., Wu, B., Yu, D., Mondal, A.N., Hou, L., Afsar, N.U., Li, Q., Xu, T., Miao, J., Xu, T. (2017)
Monovalent cation perm-selective membranes (MCPMs): New developments and perspectives.
Chinese Journal of Chemical Engineering.
Ge, L., Wu, L., Wu, B., Wang, G., Xu, T. (2014) Preparation of monovalent cation selective membranes
through annealing treatment. Journal of Membrane Science 459, 217-222.
Ge, X., He, Y., Guiver, M.D., Wu, L., Ran, J., Yang, Z., Xu, T. (2016b) Alkaline Anion‐Exchange
Membranes Containing Mobile Ion Shuttles. Adv Mater 28, 3467-3472.
Ghaffour, N., Missimer, T.M., Amy, G.L. (2013) Technical review and evaluation of the economics of
water desalination: current and future challenges for better water supply sustainability. Desalination
309, 197-207.
Gohil, G., Binsu, V., Shahi, V.K. (2006a) Preparation and characterization of mono-valent ion selective
polypyrrole composite ion-exchange membranes. Journal of Membrane Science 280, 210-218.
Gohil, G., Nagarale, R., Binsu, V., Shahi, V.K. (2006b) Preparation and characterization of monovalent
cation selective sulfonated poly (ether ether ketone) and poly (ether sulfone) composite membranes. J
Colloid Interface Sci 298, 845-853.
Gohil, G., Shahi, V.K., Rangarajan, R. (2004) Comparative studies on electrochemical characterization
of homogeneous and heterogeneous type of ion-exchange membranes. Journal of Membrane Science
240, 211-219.
Grot, W., (1973) Laminates of support material and fluorinated polymer containing pendant side chains
containing sulfonyl groups. Google Patents.
Guesmi, F., Hannachi, C., Hamrouni, B. (2012) Selectivity of anion exchange membrane modified with
polyethyleneimine. Ionics 18, 711-717.
Guillen, G.R., Pan, Y., Li, M., Hoek, E.M. (2011) Preparation and characterization of membranes
formed by nonsolvent induced phase separation: a review. Industrial & Engineering Chemistry
Research 50, 3798-3817.
Guo, H., Ma, Y., Qin, Z., Gu, Z., Cui, S., Zhang, G. (2016) One-Step Transformation from
References
128
Hierarchical-Structured Superhydrophilic NF Membrane into Superhydrophobic OSN Membrane with
Improved Antifouling Effect. ACS Appl Mater Interfaces 8, 23379-23388.
Guo, W., Ngo, H.-H., Li, J. (2012) A mini-review on membrane fouling. Bioresour Technol 122, 27-34.
Gutiérrez, L.-F., Bazinet, L., Hamoudi, S., Belkacemi, K. (2013) Production of lactobionic acid by
means of a process comprising the catalytic oxidation of lactose and bipolar membrane electrodialysis.
Separation and Purification Technology 109, 23-32.
Han, G., Zhang, S., Li, X., Widjojo, N., Chung, T.-S. (2012) Thin film composite forward osmosis
membranes based on polydopamine modified polysulfone substrates with enhancements in both water
flux and salt rejection. Chemical Engineering Science 80, 219-231.
Han, L., Galier, S., Roux-de Balmann, H. (2015) Ion hydration number and electro-osmosis during
electrodialysis of mixed salt solution. Desalination 373, 38-46.
Horowitz, S.T., Roseman, S., Blumenthal, H.J. (1957) The Preparation of glucosamine
oligosaccharides. I. Separation1, 2. J Am Chem Soc 79, 5046-5049.
Hou, L., Wu, B., Yu, D., Wang, S., Shehzad, M.A., Fu, R., Liu, Z., Li, Q., He, Y., Afsar, N.U. (2018)
Asymmetric porous monovalent cation perm-selective membranes with an ultrathin polyamide
selective layer for cations separation. Journal of Membrane Science 557, 49-57.
Hou, X. (2016) Smart gating multi ‐ scale pore/channel‐ based membranes. Adv Mater 28,
7049-7064.
Hu, M., Mi, B. (2013) Enabling graphene oxide nanosheets as water separation membranes. Environ
Sci Technol 47, 3715-3723.
Hu, Y., Wang, M., Wang, D., Gao, X., Gao, C. (2008) Feasibility study on surface modification of
cation exchange membranes by quaternized chitosan for improving its selectivity. Journal of Membrane
Science 319, 5-9.
Huang, K., Liu, S., Li, Q., Jin, W. (2013) Preparation of novel metal-carboxylate system MOF
membrane for gas separation. Separation and Purification Technology 119, 94-101.
Huang, N., Zhang, S., Yang, L., Liu, M., Li, H., Zhang, Y., Yao, S. (2015) Multifunctional
electrochemical platforms based on the michael addition/schiff base reaction of polydopamine
modified reduced graphene oxide: Construction and application. ACS Appl Mater Interfaces 7,
17935-17946.
Hwang, G.-J., Ohya, H., Nagai, T. (1999) Ion exchange membrane based on block copolymers. Part III:
References
129
preparation of cation exchange membrane. Journal of Membrane Science 156, 61-65.
Jansen, J.C., Macchione, M., Oliviero, C., Mendichi, R., Ranieri, G.A., Drioli, E. (2005) Rheological
evaluation of the influence of polymer concentration and molar mass distribution on the formation and
performance of asymmetric gas separation membranes prepared by dry phase inversion. Polymer 46,
11366-11379.
Jeon, B.H., Kim, S., Choi, M.H., Chung, I.J. (1999) Synthesis and characterization of polyaniline–
polycarbonate composites prepared by an emulsion polymerization. Synthetic Metals 104, 95-100.
Jeong, B.-H., Hoek, E.M., Yan, Y., Subramani, A., Huang, X., Hurwitz, G., Ghosh, A.K., Jawor, A.
(2007) Interfacial polymerization of thin film nanocomposites: a new concept for reverse osmosis
membranes. Journal of Membrane Science 294, 1-7.
Ji, L., Zhang, X. (2008) Ultrafine polyacrylonitrile/silica composite fibers via electrospinning.
Materials Letters 62, 2161-2164.
Jiang, J., Zhu, L., Zhu, L., Zhang, H., Zhu, B., Xu, Y. (2013) Antifouling and antimicrobial polymer
membranes based on bioinspired polydopamine and strong hydrogen-bonded poly (N-vinyl
pyrrolidone). ACS Appl Mater Interfaces 5, 12895-12904.
Jiang, J., Zhu, L., Zhu, L., Zhu, B., Xu, Y. (2011) Surface characteristics of a self-polymerized
dopamine coating deposited on hydrophobic polymer films. Langmuir 27, 14180-14187.
Johnson, R.J., Stenvinkel, P., Jensen, T., Lanaspa, M.A., Roncal, C., Song, Z., Bankir, L.,
Sánchez-Lozada, L.G. (2016) Metabolic and kidney diseases in the setting of climate change, water
shortage, and survival factors. Journal of the American Society of Nephrology 27, 2247-2256.
Kang, G.-d., Cao, Y.-m. (2012) Development of antifouling reverse osmosis membranes for water
treatment: a review. Water Res 46, 584-600.
Kang, G.-d., Cao, Y.-m. (2014) Application and modification of poly (vinylidene fluoride)(PVDF)
membranes–A review. Journal of Membrane Science 463, 145-165.
Kang, K.C., Linga, P., Park, K.-n., Choi, S.-J., Lee, J.D. (2014) Seawater desalination by gas hydrate
process and removal characteristics of dissolved ions (Na+, K+, Mg 2+, Ca 2+, B 3+, Cl−, SO 4 2−).
Desalination 353, 84-90.
Khan, M.I., Luque, R., Akhtar, S., Shaheen, A., Mehmood, A., Idress, S., Buzdar, S.A. (2016) Design
of anion exchange membranes and electrodialysis studies for water desalination. Materials 9, 365.
Khawaji, A.D., Kutubkhanah, I.K., Wie, J.-M. (2008) Advances in seawater desalination technologies.
References
130
Desalination 221, 47-69.
Kim, D.-H., Park, J.-S., Choun, M., Lee, J., Kang, M.-S. (2016) Pore-filled anion-exchange membranes
for electrochemical energy conversion applications. Electrochimica Acta 222, 212-220.
Kim, K.-Y., Yang, E., Lee, M.-Y., Chae, K.-J., Kim, C.-M., Kim, I.S. (2014) Polydopamine coating
effects on ultrafiltration membrane to enhance power density and mitigate biofouling of ultrafiltration
microbial fuel cells (UF-MFCs). Water Res 54, 62-68.
Kitagawa, S. (2014) Metal–organic frameworks (MOFs). Chem Soc Rev 43, 5415-5418.
Kotov, V., Peregonchaya, O., Selemenev, V. (2002) Electrodialysis of binary electrolyte mixtures with
membranes modified by organic species. Russian Journal of Electrochemistry 38, 927-929.
Krol, J., Wessling, M., Strathmann, H. (1999) Concentration polarization with monopolar ion exchange
membranes: current–voltage curves and water dissociation. Journal of Membrane Science 162,
145-154.
Kumar, M., Khan, M.A., AlOthman, Z.A., Siddiqui, M.R. (2013) Polyaniline modified organic–
inorganic hybrid cation-exchange membranes for the separation of monovalent and multivalent ions.
Desalination 325, 95-103.
Kuppler, R.J., Timmons, D.J., Fang, Q.-R., Li, J.-R., Makal, T.A., Young, M.D., Yuan, D., Zhao, D.,
Zhuang, W., Zhou, H.-C. (2009) Potential applications of metal-organic frameworks. Coordination
Chemistry Reviews 253, 3042-3066.
Lattemann, S., Höpner, T. (2008) Environmental impact and impact assessment of seawater
desalination. Desalination 220, 1-15.
Lau, W., Gray, S., Matsuura, T., Emadzadeh, D., Chen, J.P., Ismail, A. (2015) A review on polyamide
thin film nanocomposite (TFN) membranes: history, applications, challenges and approaches. Water
Res 80, 306-324.
Lee, E.G., Moon, S.-H., Chang, Y.K., Yoo, I.-K., Chang, H.N. (1998) Lactic acid recovery using
two-stage electrodialysis and its modelling. Journal of Membrane Science 145, 53-66.
Lee, H., Han, T., Cho, K.Y., Ryou, M.-H., Lee, Y.M. (2016a) Dopamine as a novel electrolyte additive
for high-voltage lithium-ion batteries. ACS Appl Mater Interfaces 8, 21366-21372.
Lee, J.-H., Lee, J.-Y., Kim, J.-H., Joo, J., Maurya, S., Choun, M., Lee, J., Moon, S.-H. (2016b) SPPO
pore-filled composite membranes with electrically aligned ion channels via a lab-scale continuous
caster for fuel cells: An optimal DC electric field strength-IEC relationship. Journal of Membrane
References
131
Science 501, 15-23.
Lee, M., Kang, H., Jeon, J., Choi, Y., Yoon, Y. (2016c) A novel amphoteric ion-exchange membrane
prepared by the pore-filling technique for vanadium redox flow batteries. Rsc Advances 6,
63023-63029.
Lee, M.S., Kim, H.K., Kim, C.S., Suh, H.Y., Nahm, K.S., Choi, Y.W. (2017) Thin Pore‐Filled Ion
Exchange Membranes for High Power Density in Reverse Electrodialysis: Effects of Structure on
Resistance, Stability, and Ion Selectivity. ChemistrySelect 2, 1974-1978.
Li, M., Xu, J., Chang, C.-Y., Feng, C., Zhang, L., Tang, Y., Gao, C. (2014a) Bioinspired fabrication of
composite nanofiltration membrane based on the formation of DA/PEI layer followed by cross-linking.
Journal of Membrane Science 459, 62-71.
Li, N., Guiver, M.D. (2014) Ion transport by nanochannels in ion-containing aromatic copolymers.
Macromolecules 47, 2175-2198.
Li, W., Meng, Q., Li, X., Zhang, C., Fan, Z., Zhang, G. (2014b) Non-activation ZnO array as a
buffering layer to fabricate strongly adhesive metal–organic framework/PVDF hollow fiber membranes.
Chem Commun (Camb) 50, 9711-9713.
Li, Y., Su, Y., Li, J., Zhao, X., Zhang, R., Fan, X., Zhu, J., Ma, Y., Liu, Y., Jiang, Z. (2015) Preparation
of thin film composite nanofiltration membrane with improved structural stability through the
mediation of polydopamine. Journal of Membrane Science 476, 10-19.
Li, Y., Su, Y., Zhao, X., He, X., Zhang, R., Zhao, J., Fan, X., Jiang, Z. (2014c) Antifouling, high-flux
nanofiltration membranes enabled by dual functional polydopamine. ACS Appl Mater Interfaces 6,
5548-5557.
Li, Z., Wang, W., Chen, Y., Xiong, C., He, G., Cao, Y., Wu, H., Guiver, M.D., Jiang, Z. (2016)
Constructing efficient ion nanochannels in alkaline anion exchange membranes by the in situ assembly
of a poly (ionic liquid) in metal–organic frameworks. Journal of Materials Chemistry A 4, 2340-2348.
Lin, X., Kim, S., Zhu, D.M., Shamsaei, E., Xu, T., Fang, X., Wang, H. (2017) Preparation of porous
diffusion dialysis membranes by functionalization of polysulfone for acid recovery. Journal of
Membrane Science 524, 557-564.
Lin, X., Shamsaei, E., Kong, B., Liu, J.Z., Zhao, D., Xu, T., Xie, Z., Easton, C.D., Wang, H. (2016)
Asymmetrically porous anion exchange membranes with an ultrathin selective layer for rapid acid
recovery. Journal of Membrane Science 510, 437-446.
References
132
Lind, M.L., Ghosh, A.K., Jawor, A., Huang, X., Hou, W., Yang, Y., Hoek, E.M. (2009) Influence of
zeolite crystal size on zeolite-polyamide thin film nanocomposite membranes. Langmuir 25,
10139-10145.
Liu, G., Luo, H., Wang, H., Wang, B., Zhang, R., Chen, S. (2014a) Malic acid production using a
biological electrodialysis with bipolar membrane. Journal of Membrane Science 471, 179-184.
Liu, H., Ruan, H., Zhao, Y., Pan, J., Sotto, A., Gao, C., Van der Bruggen, B., Shen, J. (2017a) A facile
avenue to modify polyelectrolyte multilayers on anion exchange membranes to enhance monovalent
selectivity and durability simultaneously. Journal of Membrane Science 543, 310-318.
Liu, J., Yang, H., Gosling, S.N., Kummu, M., Flörke, M., Pfister, S., Hanasaki, N., Wada, Y., Zhang, X.,
Zheng, C. (2017b) Water scarcity assessments in the past, present, and future. Earth's Future 5,
545-559.
Liu, X., Li, Q., Jiang, C., Lin, X., Xu, T. (2015) Bipolar membrane electrodialysis in aqua–ethanol
medium: Production of salicylic acid. Journal of Membrane Science 482, 76-82.
Liu, Y., Meng, H., Konst, S., Sarmiento, R., Rajachar, R., Lee, B.P. (2014b) Injectable
dopamine-modified poly (ethylene glycol) nanocomposite hydrogel with enhanced adhesive property
and bioactivity. ACS Appl Mater Interfaces 6, 16982-16992.
Liu, Z., Wang, W., Xie, R., Ju, X.-J., Chu, L.-Y. (2016) Stimuli-responsive smart gating membranes.
Chem Soc Rev 45, 460-475.
Luiz, A., McClure, D.D., Lim, K., Leslie, G., Coster, H.G., Barton, G.W., Kavanagh, J.M. (2017)
Potential upgrading of bio-refinery streams by electrodialysis. Desalination 415, 20-28.
Luiz, A., Spencer, E., McClure, D.D., Coster, H.G., Barton, G.W., Kavanagh, J.M. (2018) Membrane
selection for the desalination of bio-refinery effluents using electrodialysis. Desalination 428, 1-11.
Luo, G., Pan, S., Liu, J. (2002) Use of the electrodialysis process to concentrate a formic acid solution.
Desalination 150, 227-234.
Luo, Q., Zhang, H., Chen, J., Qian, P., Zhai, Y. (2008) Modification of Nafion membrane using
interfacial polymerization for vanadium redox flow battery applications. Journal of Membrane Science
311, 98-103.
Luo, R., Tang, L., Zhong, S., Yang, Z., Wang, J., Weng, Y., Tu, Q., Jiang, C., Huang, N. (2013) In vitro
investigation of enhanced hemocompatibility and endothelial cell proliferation associated with
quinone-rich polydopamine coating. ACS Appl Mater Interfaces 5, 1704-1714.
References
133
Luo, T., Abdu, S., Wessling, M. (2018) Selectivity of ion exchange membranes: A review. Journal of
Membrane Science.
Lv, Y., Yang, H.-C., Liang, H.-Q., Wan, L.-S., Xu, Z.-K. (2015) Nanofiltration membranes via
co-deposition of polydopamine/polyethylenimine followed by cross-linking. Journal of Membrane
Science 476, 50-58.
Lv, Y., Yang, H.-C., Liang, H.-Q., Wan, L.-S., Xu, Z.-K. (2016) Novel nanofiltration membrane with
ultrathin zirconia film as selective layer. Journal of Membrane Science 500, 265-271.
Mafé, S., Manzanares, J.A., Ramirez, P. (2003) Modeling of surface vs. bulk ionic conductivity in fixed
charge membranes. Physical Chemistry Chemical Physics 5, 376-383.
Malik, M.S., Qaiser, A.A., Arif, M.A. (2016) Structural and electrochemical studies of heterogeneous
ion exchange membranes based on polyaniline-coated cation exchange resin particles. Rsc Advances 6,
115046-115054.
Mei, Y., Tang, C.Y. (2018) Recent developments and future perspectives of reverse electrodialysis
technology: A review. Desalination 425, 156-174.
Mikhaylin, S., Bazinet, L. (2016) Fouling on ion-exchange membranes: classification, characterization
and strategies of prevention and control. Adv Colloid Interface Sci 229, 34-56.
Mistry, M.K., Choudhury, N.R., Dutta, N.K., Knott, R., Shi, Z., Holdcroft, S. (2008) Novel organic−
inorganic hybrids with increased water retention for elevated temperature proton exchange membrane
application. Chemistry of Materials 20, 6857-6870.
Mo, H., Tay, K.G., Ng, H.Y. (2008) Fouling of reverse osmosis membrane by protein (BSA): effects of
pH, calcium, magnesium, ionic strength and temperature. Journal of Membrane Science 315, 28-35.
Mukherjee, R., Sharma, R., Saini, P., De, S. (2015) Nanostructured polyaniline incorporated
ultrafiltration membrane for desalination of brackish water. Environmental Science: Water Research &
Technology 1, 893-904.
Mulyati, S., Takagi, R., Fujii, A., Ohmukai, Y., Matsuyama, H. (2013) Simultaneous improvement of
the monovalent anion selectivity and antifouling properties of an anion exchange membrane in an
electrodialysis process, using polyelectrolyte multilayer deposition. Journal of Membrane Science 431,
113-120.
Muthumeenal, A., Neelakandan, S., Rana, D., Matsuura, T., Kanagaraj, P., Nagendran, A. (2014)
Sulfonated polyethersulfone (spes)–charged surface modifying macromolecules (csmms) blends as a
References
134
cation selective membrane for fuel cells. Fuel cells 14, 853-861.
Nabe, A., Staude, E., Belfort, G. (1997) Surface modification of polysulfone ultrafiltration membranes
and fouling by BSA solutions. Journal of Membrane Science 133, 57-72.
Nagarale, R., Gohil, G., Shahi, V.K. (2006) Recent developments on ion-exchange membranes and
electro-membrane processes. Adv Colloid Interface Sci 119, 97-130.
Nagarale, R., Gohil, G., Shahi, V.K., Trivedi, G., Rangarajan, R. (2004) Preparation and
electrochemical characterization of cation-and anion-exchange/polyaniline composite membranes.
Journal of colloid and interface science 277, 162-171.
Nemati, M., Hosseini, S., Bagheripour, E., Madaeni, S. (2015) Electrodialysis Heterogeneous Anion
Exchange Membranes Filled with TiO2 Nanoparticles: Membranes' Fabrication and Characterization.
Journal of Membrane Science and Research 1, 135-140.
Nikonenko, V.V., Kozmai, A.E. (2011) Electrical equivalent circuit of an ion-exchange membrane
system. Electrochimica Acta 56, 1262-1269.
Ouyang, Q., Cheng, L., Wang, H., Li, K. (2008) Mechanism and kinetics of the stabilization reactions
of itaconic acid-modified polyacrylonitrile. Polymer Degradation and Stability 93, 1415-1421.
Paidar, M., Fateev, V., Bouzek, K. (2016) Membrane electrolysis—History, current status and
perspective. Electrochimica Acta 209, 737-756.
Pan, W., Yang, S.L., Li, G., Jiang, J.M. (2005) Electrical and structural analysis of conductive
polyaniline/polyacrylonitrile composites. European Polymer Journal 41, 2127-2133.
Park, J.-S., Choi, J.-H., Woo, J.-J., Moon, S.-H. (2006) An electrical impedance spectroscopic (EIS)
study on transport characteristics of ion-exchange membrane systems. J Colloid Interface Sci 300,
655-662.
Pashley, R., Rzechowicz, M., Pashley, L., Francis, M. (2005) De-gassed water is a better cleaning agent.
The Journal of Physical Chemistry B 109, 1231-1238.
Pendergast, M.M., Hoek, E.M. (2011) A review of water treatment membrane nanotechnologies.
Energy & Environmental Science 4, 1946-1971.
Peyravi, M., Jahanshahi, M., Rahimpour, A., Javadi, A., Hajavi, S. (2014) Novel thin film
nanocomposite membranes incorporated with functionalized TiO 2 nanoparticles for organic solvent
nanofiltration. Chemical Engineering Journal 241, 155-166.
Pismenskaya, N.D., Nikonenko, V.V., Melnik, N.A., Shevtsova, K.A., Belova, E.I., Pourcelly, G., Cot,
References
135
D., Dammak, L., Larchet, C. (2012) Evolution with time of hydrophobicity and microrelief of a
cation-exchange membrane surface and its impact on overlimiting mass transfer. The Journal of
Physical Chemistry B 116, 2145-2161.
Poyton, M.F., Sendecki, A.M., Cong, X., Cremer, P.S. (2016) Cu2+ binds to phosphatidylethanolamine
and increases oxidation in lipid membranes. J Am Chem Soc 138, 1584-1590.
Ramsahye, N.A., Maurin, G., Bourrelly, S., Llewellyn, P.L., Loiseau, T., Serre, C., Férey, G. (2007) On
the breathing effect of a metal–organic framework upon CO 2 adsorption: monte carlo compared to
microcalorimetry experiments. Chem Commun (Camb), 3261-3263.
Ran, J., Wu, L., He, Y., Yang, Z., Wang, Y., Jiang, C., Ge, L., Bakangura, E., Xu, T. (2017) Ion
exchange membranes: New developments and applications. Journal of Membrane Science 522,
267-291.
Reig, M., Vecino, X., Valderrama, C., Gibert, O., Cortina, J. (2017) Application of selectrodialysis for
the removal of As from metallurgical process waters: recovery of Cu and Zn. Separation and
Purification Technology.
Rodzik, I. (2005) Selectivity of Transport of Sodium, Magnesium, and Calcium Ions through a
Sulfo-Cationite Membrane in Mixtures of Solutions of Their Chlorides. Russian Journal of
Electrochemistry 41, 888-891.
Rubinstein, I., Staude, E., Kedem, O. (1988) Role of the membrane surface in concentration
polarization at ion-exchange membrane. Desalination 69, 101-114.
Sadrzadeh, M., Mohammadi, T. (2008) Sea water desalination using electrodialysis. Desalination 221,
440-447.
Sadyrbaeva, T.Z. (2015) Separation of cobalt (II) from nickel (II) by a hybrid liquid membrane–
electrodialysis process using anion exchange carriers. Desalination 365, 167-175.
Safarpour, M., Khataee, A., Vatanpour, V. (2015) Thin film nanocomposite reverse osmosis membrane
modified by reduced graphene oxide/TiO 2 with improved desalination performance. Journal of
Membrane Science 489, 43-54.
Sata, T. (1994) Studies on ion exchange membranes with permselectivity for specific ions in
electrodialysis. Journal of Membrane Science 93, 117-135.
Sata, T. (2000) Studies on anion exchange membranes having permselectivity for specific anions in
electrodialysis—effect of hydrophilicity of anion exchange membranes on permselectivity of anions.
References
136
Journal of Membrane Science 167, 1-31.
Sata, T., Ishii, Y., Kawamura, K., Matsusaki, K. (1999) Composite membranes prepared from cation
exchange membranes and polyaniline and their transport properties in electrodialysis. Journal of the
Electrochemical Society 146, 585-591.
Shannon, M.A., Bohn, P.W., Elimelech, M., Georgiadis, J.G., Marinas, B.J., Mayes, A.M. (2008)
Science and technology for water purification in the coming decades. Nature 452, 301.
Shekhah, O., Liu, J., Fischer, R., Wöll, C. (2011) MOF thin films: existing and future applications.
Chem Soc Rev 40, 1081-1106.
Shen, J., Huang, J., Liu, L., Ye, W., Lin, J., Van der Bruggen, B. (2013) The use of BMED for
glyphosate recovery from glyphosate neutralization liquor in view of zero discharge. Journal of
hazardous materials 260, 660-667.
Shen, J., Yu, J., Liu, L., Lin, J., Van der Bruggen, B. (2014) Synthesis of quaternary ammonium
hydroxide from its halide salt by bipolar membrane electrodialysis (BMED): effect of molecular
structure of ammonium compounds on the process performance. Journal of Chemical Technology and
Biotechnology 89, 841-850.
Shi, G.M., Yang, T., Chung, T.S. (2012) Polybenzimidazole (PBI)/zeolitic imidazolate frameworks
(ZIF-8) mixed matrix membranes for pervaporation dehydration of alcohols. Journal of Membrane
Science 415, 577-586.
Sistat, P., Kozmai, A., Pismenskaya, N., Larchet, C., Pourcelly, G., Nikonenko, V. (2008)
Low-frequency impedance of an ion-exchange membrane system. Electrochimica Acta 53, 6380-6390.
Sivaraman, P., Chavan, J., Thakur, A., Hande, V., Samui, A. (2007) Electrochemical modification of
cation exchange membrane with polyaniline for improvement in permselectivity. Electrochimica Acta
52, 5046-5052.
Sorribas, S., Gorgojo, P., Tellez, C., Coronas, J., Livingston, A.G. (2013) High flux thin film
nanocomposite membranes based on metal–organic frameworks for organic solvent nanofiltration. J
Am Chem Soc 135, 15201-15208.
Souzy, R., Ameduri, B., Boutevin, B. (2004) Synthesis and (co) polymerization of monofluoro, difluoro,
trifluorostyrene and ((trifluorovinyl) oxy) benzene. Progress in Polymer Science 29, 75-106.
Strathmann, H. (2004) Ion-exchange membrane separation processes. Elsevier.
Strathmann, H. (2010) Electrodialysis, a mature technology with a multitude of new applications.
References
137
Desalination 264, 268-288.
Takata, K., Yamamoto, Y., Sata, T. (2000) Modification of transport properties of ion exchange
membranes: XIV. Effect of molecular weight of polyethyleneimine bonded to the surface of cation
exchange membranes by acid–amide bonding on electrochemical properties of the membranes. Journal
of Membrane Science 179, 101-107.
Tansel, B., Sager, J., Rector, T., Garland, J., Strayer, R.F., Levine, L., Roberts, M., Hummerick, M.,
Bauer, J. (2006) Significance of hydrated radius and hydration shells on ionic permeability during
nanofiltration in dead end and cross flow modes. Separation and Purification Technology 51, 40-47.
Tongwen, X., Weihua, Y. (2002) Citric acid production by electrodialysis with bipolar membranes.
Chemical Engineering and Processing: Process Intensification 41, 519-524.
Tranchemontagne, D.J., Mendoza-Cortes, J.L., O‘Keeffe, M., Yaghi, O.M. (2009) Secondary building
units, nets and bonding in the chemistry of metal–organic frameworks. Chem Soc Rev 38, 1257-1283.
Tzabiras, J., Vasiliades, L., Sidiropoulos, P., Loukas, A., Mylopoulos, N. (2016) Evaluation of water
resources management strategies to overturn climate change impacts on Lake Karla watershed. Water
Resources Management 30, 5819-5844.
Van Goethem, C., Verbeke, R., Hermans, S., Bernstein, R., Vankelecom, I. (2016) Controlled
positioning of MOFs in interfacially polymerized thin-film nanocomposites. Journal of Materials
Chemistry A 4, 16368-16376.
Vaselbehagh, M., Karkhanechi, H., Mulyati, S., Takagi, R., Matsuyama, H. (2014) Improved
antifouling of anion-exchange membrane by polydopamine coating in electrodialysis process.
Desalination 332, 126-133.
Vermaas, D.A., Saakes, M., Nijmeijer, K. (2014) Early detection of preferential channeling in reverse
electrodialysis. Electrochimica Acta 117, 9-17.
Volodina, E., Pismenskaya, N., Nikonenko, V., Larchet, C., Pourcelly, G. (2005) Ion transfer across
ion-exchange membranes with homogeneous and heterogeneous surfaces. J Colloid Interface Sci 285,
247-258.
Wang, H.-B., Li, Y., Dong, G.-L., Gan, T., Liu, Y.-M. (2017a) A convenient and label-free colorimetric
assay for dopamine detection based on the inhibition of the Cu (ii)-catalyzed oxidation of a 3, 3′, 5,
5′-tetramethylbenzidine–H 2 O 2 system. New Journal of Chemistry 41, 14364-14369.
Wang, J., Zhu, J., Tsehaye, M.T., Li, J., Dong, G., Yuan, S., Li, X., Zhang, Y., Liu, J., Van der Bruggen,
References
138
B. (2017b) High flux electroneutral loose nanofiltration membranes based on rapid deposition of
polydopamine/polyethyleneimine. Journal of Materials Chemistry A 5, 14847-14857.
Wang, L., Fang, M., Liu, J., He, J., Deng, L., Li, J., Lei, J. (2015a) The influence of dispersed phases
on polyamide/ZIF-8 nanofiltration membranes for dye removal from water. Rsc Advances 5,
50942-50954.
Wang, L., Fang, M., Liu, J., He, J., Li, J., Lei, J. (2015b) Layer-by-layer fabrication of
high-performance polyamide/ZIF-8 nanocomposite membrane for nanofiltration applications. ACS
Appl Mater Interfaces 7, 24082-24093.
Wang, M., Jia, Y.-x., Yao, T.-t., Wang, K.-k. (2013) The endowment of monovalent selectivity to cation
exchange membrane by photo-induced covalent immobilization and self-crosslinking of chitosan.
Journal of Membrane Science 442, 39-47.
Wang, T., Qiblawey, H., Sivaniah, E., Mohammadian, A. (2016) Novel methodology for facile
fabrication of nanofiltration membranes based on nucleophilic nature of polydopamine. Journal of
Membrane Science 511, 65-75.
Wee, L.H., Lescouet, T., Ethiraj, J., Bonino, F., Vidruk, R., Garrier, E., Packet, D., Bordiga, S.,
Farrusseng, D., Herskowitz, M. (2013) Hierarchical Zeolitic Imidazolate Framework‐8 Catalyst for
Monoglyceride Synthesis. ChemCatChem 5, 3562-3566.
Wei, H., Ren, J., Han, B., Xu, L., Han, L., Jia, L. (2013) Stability of polydopamine and poly (DOPA)
melanin-like films on the surface of polymer membranes under strongly acidic and alkaline conditions.
Colloids and Surfaces B: Biointerfaces 110, 22-28.
Xie, P., Rong, M.Z., Zhang, M.Q. (2016) Moisture Battery Formed by Direct Contact of Magnesium
with Foamed Polyaniline. Angewandte Chemie International Edition 55, 1805-1809.
Xu, L.Q., Yang, W.J., Neoh, K.-G., Kang, E.-T., Fu, G.D. (2010) Dopamine-induced reduction and
functionalization of graphene oxide nanosheets. Macromolecules 43, 8336-8339.
Xu, Q., Zhang, W., (2016) Next-Generation Graphene-Based Membranes for Gas Separation and Water
Purifications, Advances in Carbon Nanostructures. InTech.
Xu, T. (2005) Ion exchange membranes: state of their development and perspective. Journal of
Membrane Science 263, 1-29.
Yamada, K., Odomi, M., Okada, N., Fujita, T., Yamamoto, A. (2005) Chitosan oligomers as potential
and safe absorption enhancers for improving the pulmonary absorption of interferon‐α in rats. Journal
References
139
of pharmaceutical sciences 94, 2432-2440.
Yang, C.-X., Ren, H.-B., Yan, X.-P. (2013) Fluorescent metal–organic framework MIL-53 (Al) for
highly selective and sensitive detection of Fe3+ in aqueous solution. Analytical chemistry 85,
7441-7446.
Yang, H.-C., Liao, K.-J., Huang, H., Wu, Q.-Y., Wan, L.-S., Xu, Z.-K. (2014a) Mussel-inspired
modification of a polymer membrane for ultra-high water permeability and oil-in-water emulsion
separation. Journal of Materials Chemistry A 2, 10225-10230.
Yang, H.-C., Xu, W., Du, Y., Wu, J., Xu, Z.-K. (2014b) Composite free-standing films of
polydopamine/polyethyleneimine grown at the air/water interface. Rsc Advances 4, 45415-45418.
Yang, H.C., Wu, M.B., Li, Y.J., Chen, Y.F., Wan, L.S., Xu, Z.K. (2016) Effects of polyethyleneimine
molecular weight and proportion on the membrane hydrophilization by codepositing with dopamine.
Journal of Applied Polymer Science 133.
Yang, X., Du, Y., Zhang, X., He, A., Xu, Z.-K. (2017) Nanofiltration Membrane with a Mussel-Inspired
Interlayer for Improved Permeation Performance. Langmuir 33, 2318-2324.
Yu, L., Lin, A., Zhang, L., Jiang, W. (2002) Large scale experiment on the preparation of vitamin C
from sodium ascorbate using bipolar membrane electrodialysis. Chemical Engineering
Communications 189, 237-246.
Zhang, C., Ma, M.-Q., Chen, T.-T., Zhang, H., Hu, D.-F., Wu, B.-H., Ji, J., Xu, Z.-K. (2017a)
Dopamine-Triggered One-Step Polymerization and Codeposition of Acrylate Monomers for Functional
Coatings. ACS Appl Mater Interfaces 9, 34356-34366.
Zhang, C., Ou, Y., Lei, W.X., Wan, L.S., Ji, J., Xu, Z.K. (2016a) CuSO4/H2O2‐Induced Rapid
Deposition of Polydopamine Coatings with High Uniformity and Enhanced Stability. Angewandte
Chemie 128, 3106-3109.
Zhang, F., Gao, S., Zhu, Y., Jin, J. (2016b) Alkaline-induced superhydrophilic/underwater
superoleophobic polyacrylonitrile membranes with ultralow oil-adhesion for high-efficient oil/water
separation. Journal of Membrane Science 513, 67-73.
Zhang, R., Liu, Y., He, M., Su, Y., Zhao, X., Elimelech, M., Jiang, Z. (2016c) Antifouling membranes
for sustainable water purification: strategies and mechanisms. Chem Soc Rev 45, 5888-5924.
Zhang, R., Su, Y., Zhao, X., Li, Y., Zhao, J., Jiang, Z. (2014) A novel positively charged composite
nanofiltration membrane prepared by bio-inspired adhesion of polydopamine and surface grafting of
References
140
poly (ethylene imine). Journal of Membrane Science 470, 9-17.
Zhang, W., Miao, M., Pan, J., Sotto, A., Shen, J., Gao, C., Van der Bruggen, B. (2017b) Separation of
divalent ions from seawater concentrate to enhance the purity of coarse salt by electrodialysis with
monovalent-selective membranes. Desalination 411, 28-37.
Zhang, X., Li, C., Wang, Y., Luo, J., Xu, T. (2011) Recovery of acetic acid from simulated acetaldehyde
wastewaters: Bipolar membrane electrodialysis processes and membrane selection. Journal of
Membrane Science 379, 184-190.
Zhao, F.-Y., Ji, Y.-L., Weng, X.-D., Mi, Y.-F., Ye, C.-C., An, Q.-F., Gao, C.-J. (2016) High-flux
positively charged nanocomposite nanofiltration membranes filled with poly (dopamine) modified
multiwall carbon nanotubes. ACS Appl Mater Interfaces 8, 6693-6700.
Zhao, J., Fang, C., Zhu, Y., He, G., Pan, F., Jiang, Z., Zhang, P., Cao, X., Wang, B. (2015) Manipulating
the interfacial interactions of composite membranes via a mussel-inspired approach for enhanced
separation selectivity. Journal of Materials Chemistry A 3, 19980-19988.
Zhao, Z., Shi, S., Cao, H., Li, Y. (2017) Electrochemical impedance spectroscopy and surface
properties characterization of anion exchange membrane fouled by sodium dodecyl sulfate. Journal of
Membrane Science 530, 220-231.
Zhao, Z., Shi, S., Cao, H., Shan, B., Sheng, Y. (2018) Property characterization and mechanism
analysis on organic fouling of structurally different anion exchange membranes in electrodialysis.
Desalination 428, 199-206.
Zhu, K., Gao, Y., Tan, X., Chen, C. (2016) Polyaniline-Modified Mg/Al Layered Double Hydroxide
Composites and Their Application in Efficient Removal of Cr (VI). ACS Sustainable Chemistry &
Engineering 4, 4361-4369.
Curriculum Vitae
141
Curriculum Vitae
Jian Li
PERSONAL INFORMATION
Date of Birth: July 25th
, 1990
Nationality: Chinese
Gender: Male
CONTACT
Process Engineering for Sustainable Systems (ProcESS)
Department of Chemical Enigineering
KU Leuven Chem & Tech
Celestijnenlaan 200F
3001 Heverlee, Leuven, Belgium
Tel: (+32) 485632238
Email: [email protected]; [email protected]
EDUCATION
Sept. 2015-Present: Ph.D in Chemical Engineering
KU Leuven, Belgium
Sept. 2012-Jun. 2015 Master degree in Chemical Engineering
Zhejiang University of Technology, China
Curriculum Vitae
142
Sept. 2008-Jun. 2012 Bachelor in Chemical Engineering
Yancheng Institude of Technology, China
Publications
143
PUBLICATIONS
1. J Li, S Yuan, J Wang, J Zhu, J Shen, B Van der Bruggen. Mussel-inspired
modification of ion exchange membrane for monovalent separations. Journal of
Membrane Science. 553 (2018): 139-150.
2. J Li, J Zhu, S Yuan, J Wang, J Shen, B Van der Bruggen. Mussel-inspired
monovalent cation exchange membranes containing hydrophilic MIL53(Al)
framework for enhanced ion flux. Industrial & Engineering Chemistry
Research 57. 18(2018):6275-6283.
3. J Li, Z Zhao, S Yuan, J Zhu, B Van der Bruggen. High-performance
thin-film-nanocomposite cation exchange membranes containing hydrophobic
zeolitic imidazolate framework for monovalent selectivity. Applied sciences.
8.5(2018): 759
4. J Li, J Zhu, J Wang, S Yuan, J Lin, J Shen, B Van der Bruggen. Charge‐assisted
ultrafiltration membranes for monovalent ions separation in electrodialysis.
Journal of Applied Polymer Science 135.24 (2018): 45692.
5. J Li, J Zhu, S Yuan, J Lin, J Shen, B Van der Bruggen. Cation-Exchange
Membranes with Controlled Porosity in Electrodialysis Application. Industrial &
Engineering Chemistry Research 56.28 (2017): 8111-8120.
6. J Li, ST Morthensen, J Zhu, S Yuan, J Wang, A Volodine, J Lin, J Shen, B van
der Bruggen. Exfoliated MoS 2 nanosheets loaded on bipolar exchange
membranes interfaces as advanced catalysts for water dissociation. Separation
and Purification Technolog 194 (2018): 416-424.
7. J Li, Y Xu, M Hu, J Shen, C Gao, B van der Bruggen. Enhanced conductivity of
monovalent cation exchange membranes with chitosan/PANI composite
modification. RSC Advances 5.110 (2015): 90969-90975.
8. J Li, M Zhou, J Lin, W Ye, Y Xu, J Shen, C Gao, B Van der Bruggen.
Mono-valent cation selective membranes for electrodialysis by introducing
polyquaternium-7 in a commercial cation exchange membrane. Journal of
Membrane Science 486 (2015): 89-96.
Publications
144
9. S Yuan, J Li, J Zhu, A Volodine, J Li, G Zhang, P Van Puyvelde, B Van der
Bruggen. Hydrophilic nanofiltration membranes with reduced humic acid fouling
fabricated from copolymers designed by introducing carboxyl groups in the
pendant benzene ring. Journal of Membrane Science. 563(2018): 655-663
10. J Lin, C Y. Tang, C Huang, Y Tang, W Ye, J Li, J Shen, R Van den Broeck, J
Van Impe, A Volodin, C Van Haesendonck, A Sotto, P Luis, B Van der Bruggen.
A comprehensive physico-chemical characterization of superhydrophilic loose
nanofiltration membranes. Journal of Membrane Science. 501 (2016): 1-14.
11. L Hao, J Liao, Y Jiang, J Zhu, J Li, Y Zhao, B Van der Bruggen, A Sotto, J Shen.
Sandwich‖-like structure modified anion exchange membrane with enhanced
monovalent selectivity and fouling resistant. Journal of Membrane Science,
556(2018): 98-106.
12. B Han, J Pan, S Yang, M Zhou, J Li, B Van der Bruggen, A. Sotto, C Gao, J Shen.
Novel composite anion exchange membranes based on quaternized
polyepichlorohydrin for electromembrane application. Industrial & Engineering
Chemistry Research. 2016, 55(26): 7171-7178.
13. J Wang, J Zhu, M T Tsehaye, J Li, S Yuan, G Dong, X Li, Y. Zhang, J Liu, B
Van der Bruggen. High flux electroneutral loose nanofiltration membranes based
on rapid deposition of polydopamine/polyethyleneimine. Journal of Materials
Chemistry A, 2017, 5(28): 14847-14857.
14. J Zhu, J Hou, R Zhang, S Yuan, J Li, M Tian, P Wang, Y Zhang, A Volodin, B
Van der Bruggen. Rapid water transport through controllable, ultrathin polyamide
nanofilms for high-performance nanofiltration[J]. Journal of Materials Chemistry
A, 2018,6, 15701-15709
15. Y Jiang, J Liao, S Yang, J Li, Y. Xu, H. Ruan, A. Sotto, B Van der Bruggen, J
Shen. Stable cycloaliphatic quaternary ammonium-tethered anion exchange
membranes for electrodialysis. Reactive and Functional Polymers, 130(2018):
61-69
Publications
145
INTERNATIONAL CONFERENCES
J Li, B Van der Bruggen. Mussel-inspired modification of ion exchange membrane
for monovalent separations. MELPRO 2018, May 13-16, 2018. Prague, Czech
Republic. (Poster)