Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Development of polymer electrolyte membranes for fuel cell applications
(USC Thesis Other)
Development of polymer electrolyte membranes for fuel cell applications
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DEVELOPMENT OF
POLYMER ELECTROLYTE MEMBRANES
FOR FUEL CELL APPLICATIONS
by
Bo Yang
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)
December 2009
Copyright 2009 Bo Yang
ii
Dedication
To my wife and my parents
iii
Acknowledgements
I am heartily thankful to my advisor Professor G. K. Surya Prakash for providing
me with great guidance and support through my studies at the Loker Hydrocarbon
Institute of University of Southern California. His enthusiasm for chemistry and hard-
working attitude deeply inspired me to do research in the area of fuel cells. Besides his
wonderful research guidance, Professor Prakash has great personality and shows genuine
interest in the well being of his graduate students. This thesis would not be possible
without his kind instructions and great patience. In my view he has set a high standard for
a good graduate student advisor. Also, I have been greatly privileged to have the
opportunity to work with Professor George A. Olah in this world-renowned institute. I
consider it as a great honor and thank him sincerely. Professor Olah’s wisdom and his
devotion to science inspire numerous students and make them real scientists. His
kindness to students and his humorous yet instructional talk in the group meetings also
show us how a good human being and a great scientist are blended in a lifetime. I would
also like to thank my committee member Professor Teh Fu Yen for his kind support
despite his health concern, late Professor Robert Bau, (whose sudden demise was a big
loss to all who knew him), Professor Aaron Harper for attending my Qualifying exam.
iv
I am very grateful to Dr. Thomas Mathew for his tremendous assistance in my
research and helpful instructions in my thesis work. I would also like to thank Dr. Patrice
Batamack and Dr. Jinbo Hu for their help in my research work. The support from Dr.
Robert Aniszfeld is also greatly appreciated.
My research in LHI started in collaboration with Dr. Anthony Atti, who helped
me at the early stage of the project and I really learned a lot from his experience. I greatly
appreciate the tremendous help from my colleagues in the fuel cell laboratory and I thank
Dr. Akihisa Saitoh, Dr. Kimberly McGrath, Dr. Frederico, Viva, Dr. Suresh Palale and
Mr. Frederick Krause. I enjoyed working with them in all these pleasant years.
It’s a pleasure to thank all my colleagues in Loker Hydrocarbon Institute. The
friendly people and environment make my present research accomplishments possible.
The list is very long and I would like to thank Dr. Ying Wang, Dr. Alain Goeppert, Dr.
Chiradeep Panja, Dr. Sujit Chakco, Mr. Fang Wang, Mr. Kevin Glinton and all the other
colleagues in LHI. I would also like to thank all the LHI staff, Mrs. Jessy May, Mrs.
Carole Philips, Mr. Ralph Pan and Mr. David Hunter for their help. I also thank the
faculty and staff in the chemistry department and the Loker Hydrocarbon Institute for
making this project possible.
v
Words can’t adequately express my gratitude to my family. My wife Ping has
supported me all these years and it was her love and faith that encouraged me to achieve
these accomplishments. My greatest heroes, my Mom and Dad gave me the best
educations and support with which in achieving my goals. I also owe my sister Li greatly
for her support. I also want to thank my grandma (who just left us) and my grandpa for
their constant encouragement. I am fortunate and grateful to have all the support and
assistance. Thank you!
vi
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables ix
List of Figures xi
List of Schemes xvi
Abstract xvii
Chapter 1: Introduction 1
1.1 Fuel cell history 1
1.2 Background 5
1.3 Membranes development history 22
1.4 Fuel Cell membrane materials 25
1.4.1 Perfluorinated membranes 28
1.4.2 Hydrocarbon membranes 35
1.4.3 Aromatic polymers 36
1.4.4 Acid-base complex 38
1.5 Conclusions 43
1.6 Chapter 1 References 45
Chapter 2 Development of PVDF-PSSA-PMMA composite membranes for
direct methanol fuel cells 52
2.1 PVDF-PSSA-PMMA membrane preparation 52
2.1.1 PVDF precursor study 53
2.1.2 Synthesize PVDF-PSSA using IPN methodology 55
2.1.3 Introduction of PMMA into PVDF-PSSA membranes 58
2.2 Experimental 60
vii
2.3 Membrane characterization 63
2.3.1 Thermal analysis 63
2.3.2 Energy dispersive X-ray analysis (EDAX) 65
2.3.3 Membrane water uptake 66
2.3.4 Proton conductivity measurement 70
2.3.5 Ion Exchange Capacity Test 73
2.3.6 Methanol permeability test using a GC tracking method 75
2.3.7 Methanol permeability test using CO
2
analyzer 81
2.4 Electrical performance of PVDF-PSSA-PMMA membranes 85
2.4.1 Membrane electrode assembly fabrication 85
2.4.2 Ionomer studies with PVDF-PSSA beads 91
2.4.3 Methanol concentration effect on electrical performance 95
2.4.4 Impact of PSSA uptake on electrical performance 97
2.4.5 Impact of membrane thickness on electrical performance 103
2.4.6 Temperature effect on cell performance 105
2.4.7 Gas flow rate effect 107
2.4.8 Catalyst loading effect 109
2.5 PVDF-PSSA-PMMA performance at room temperature 110
2.6 Comparison between PVDF-PSSA-PMMA and Nafion117 112
2.7 Fuel efficiency and fuel cell efficiency 115
2.8 PVDF-PSSA-PMMA performance on a H
2
/O
2
fuel cell 120
2.9 Conclusions 121
2.10 Chapter 2 References 124
Chapter 3 New polymer electrolyte membrane materials 126
3.1 PVDF-Poly(vinylsulfonyl chloride) membrane 126
3.2 PVDF - Poly trifluoro-1-(trifluoromethyl)ethyl vinylsulfonate 128
3.3 Poly(sodium vinylsulfonate) membrane 129
3.4 PVDF-poly(vinyl sulfonic acid tetrabutylammonium salt) membrane 130
3.5 PVDF-PSSA-poly(trifluoroethyl acrylate) 131
3.6 PVDF-PSSA-poly(heptafluorobutyl acrylate) 133
3.7 PVDF-PVP- CF
3
SO
3
H membrane 133
3.8 PVDF-PVP-HF membrane 134
3.9 PVDF-PSSA-PVP membrane 135
3.10 PVDF-PSSA- Poly4-ethenyl- α, α-bis(trifluoromethyl)benzeneethanol 138
3.11 Kynar membranes 139
viii
3.12 Norbornene based Polymer electrolyte membrane 142
3.13 Experimental 146
3.14 Conclusions 147
3.15 Chapter 3 References 149
Chapter 4 Testing of alternative liquid fuels in fuel cells 151
4.1 Crossover test of different fuels 151
4.2 Tracking crossover rates of certain fuels in fuel cells using CO
2
analyzer 155
4.3 Electrical performance of Nafion117 using different fuels 159
4.4 Performance evaluation of fuel cells with ethanol as a fuel 163
4.5 Experimental section of DEFC evaluation 173
4.6 Chapter 4 References 176
Chapter 5 Carbon-based solid acid: A catalyst for intramolecular Friedel–
Crafts acylation 178
5.1 Preparation of cyclic ketones catalyzed by carbon-based solid acid 179
5.2 Experimental section of cyclic ketones synthesis 186
5.3 Potential of carbon-based solid acid in fuel cell applications 187
5.4 Chapter 5 References 189
Bibliography 191
ix
List of Tables
Table 1.1 Thermodynamic data for some common fuels under standard condition 8
Table 1.2 Major fuel cell types 12
Table 1.3 Properties of commercial cation-exchange membranes 24
Table 1.4 Classification of membrane materials 28
Table 2.1 DSC analysis of the various PVDF precursors 64
Table 2.2 Specific proton conductivity values for Nafion117 and PVDF-PSSA
membranes 71
Table 2.3 Equivalent weight of various membranes 74
Table 2.4 Anode and cathode catalyst ink composition I 88
Table 2.5 Anode and cathode catalyst ink composition II 88
Table 2.6 MEA fabrication pressure vs. MEA properties 89
Table 2.7 Physical properties of membranes and related MEA properties 103
Table 3.1 PVDF-Poly(vinylsulfonyl chloride) membranes 127
Table 4.1 Crossover coefficient of DMC 153
Table 4.2 Crossover coefficient of dimethoxymethane 154
Table 4.3 Crossover coefficient of dimethyl oxalate 154
x
Table 4.4 Crossover coefficient of trimethyl orthoformate 155
Table 5.1 Carbon-based solid acid catalyzed intramolecular acylation 185
xi
List of Figures
Figure 1.1 Scheme of energy conversion in a fuel cell and a combustion engine 5
Figure 1.2 Schematic view of a H
2
/O
2
fuel cell 9
Figure 1.3 Schematic drawing of a direct methanol fuel cell 18
Figure 1.4 Chemical structure of Nafion-H 23
Figure 1.5 Structure of poly(butadiene styrene) block copolymer 35
Figure1.6 Structure of grafted membranes: (a) FEP main; (b) Sulfonated
polystyrene side chain 35
Figure 1.7 Structures of (a) SPSU, (b) SPEEK and (c) SPPBP 37
Figure 1.8 Structure of (a) tetraaminobiphenyl, (b) diphenylisopthalate and (c)
poly[2,2
’
-(m-phenylene)-5,5
’
bibenzimidazole] 39
Figure 1.9 Structure of basic polymers (a–d) and acidic polymers (e, f) 42
Figure 2.1 Membrane water content of PVDF-PSSA membrane vs. PSSA
uptakes 68
Figure 2.2 Water collected at the cathode during a scan current experiment 70
Figure 2.3 NaCl sodium exchange effect on the conductivity of Nafion117 72
Figure 2.4 Proton conductivity vs. temperature under humidified conditions 73
Figure 2.5 Apparatus to evaluate methanol permeability 76
Figure 2.6 Methanol concentrations of GC samples vs. time 77
xii
Figure 2.7 Methanol cross-over of Nafion117 and USC membranes using GC
tracking method 78
Figure 2.8 Methanol cross-over of Nafion117 using the CO
2
analyzer tracking
method. 82
Figure 2.9 Methanol cross-over of Nafion117 and USC membranes using the
CO
2
analyzer tracking method at 25
o
C 83
Figure 2.10 Methanol cross-over of Nafion117 and USC membranes using the
CO
2
analyzer tracking method at 55
o
C 83
Figure 2.11 Methanol cross-over of Nafion117 and USC membranes using the
CO
2
analyzer tracking method at 60
o
C 84
Figure 2.12 Methanol cross-over of Nafion117 and USC membranes using the
CO
2
analyzer tracking method at 90
o
C 84
Figure 2.13 MEA fabrication pressure vs. electrical performance 90
Figure 2.14 The effect of PVDF-PSSA beads on cell performance I 93
Figure 2.15 The effect of PVDF-PSSA beads on cell performance II 93
Figure 2.16 Methanol concentration vs. cell performance I 96
Figure 2.17 Methanol concentration vs. cell performance II 97
Figure 2.18 Effect of PSSA uptake on OCV resistance 98
Figure 2.19 Effect of PSSA uptake on methanol crossover 98
Figure 2.20 PSSA uptake on MEA performance at 55
o
C with air 100
Figure 2.21 PSSA uptake on MEA performance at 55
o
C with oxygen 100
xiii
Figure 2.22 PSSA uptake on MEA performance at 90
o
C with air 101
Figure 2.23 PSSA uptake on MEA performance at 90
o
C with oxygen 101
Figure2.24 Effect of membrane thickness on cell performance 104
Figure 2.25 The temperature effect on a 25 cm
2
MEA performance 105
Figure 2.26 Comparison between USC PVDF-PSSA membrane and Nafion117 106
Figure 2.27 The performance of MEA00-72b at different flow rates of air 107
Figure 2.28 The performance of MEA00-72b at different flow rates of oxygen 108
Figure 2.29 Cell performance vs. catalyst loading 109
Figure 2.30 PVDF-PSSA-PMMA MEA performance at room temperature with
air 111
Figure 2.31 PVDF-PSSA-PMMA MEA performance at r. t. with oxygen 111
Figure 2.32 Comparison between USC MEAs with Nafion117 at 25
o
C 113
Figure 2.33 Comparison between USC MEAs with Nafion117 at 55
o
C 114
Figure 2.34 Comparison between USC MEAs with Nafion117 at 90
o
C 115
Figure 2.35 Fuel efficiency at 25
o
C 116
Figure 2.36 Fuel efficiency at 90
o
C 117
Figure 2.37 Fuel cell efficiency at 25
o
C 118
Figure 2.38 Fuel cell efficiency at 90
o
C 119
xiv
Figure2.39 Comparison between PVDF-PSSA-PMMA and Nafion117 on H
2
/O
2
fuel cell 120
Figure 3.1 Cell performance of PVDF-PSSA-poly(trifluoroethyl acrylate) 132
Figure 3.2 Polyvinylpyridine doped with triflic acid 134
Figure3.3 Polyvinylpyridine doped with HF 134
Figure 3.4 Structure of PSSA-PVP 135
Figure 3.5 Cell performance of PVDF-PSSA-PVP at 25
o
C 136
Figure 3.6 Cell performance of PVDF-PSSA-PVP at 60
o
C 136
Figure 3.7 Cell performance of PVDF-PSSA-PVP at 90
o
C 137
Figure 3.8 PSSA- poly4-ethenyl- α,α-bis(trifluoromethyl)benzeneethanol 138
Figure3.9 Electrical performance of Kynar460 4 mil MEA 140
Figure3.10 Kynar460 4 mil MEA performance at high temperature 141
Figure3.11 Kynar460 4 mil MEA performance compared with Nafion 117 141
Figure 3.12 Structures of Grubb’s catalyst, dicyclopentadiene and Norbornene 144
Figure 3.13 Structures of (a) FPHS and (b) PHS 144
Figure 4.1 Alternative organic fuels in fuel cell 152
Figure 4.2 Methanol crossover density 156
Figure 4.3 Glyoxal crossover density 157
Figure 4.4 Glyoxal trimer dihydrate crossover density 157
xv
Figure 4.5 Dioxolane crossover density 158
Figure 4.6 Cell performance of Nafion117 using glyoxal trimer dihydrate 160
Figure 4.7 Cell performance of Nafion117 using glyoxal as fuel 161
Figure 4.8 Cell performance of Nafion117 using 1,3-dioxolane as fuel 161
Figure 4.9 Cell performance of Nafion117 using trimethyl orthoformate 162
Figure 4.10 Cell performance of Nafion117 using acetaldehyde 163
Figure 4.11 Polarization curves at different temperatures of DMFC 166
Figure 4.12 Polarization curves at different temperatures of DEFC 167
Figure 4.13 Anode efficiency of DEFC compared to DMFC 168
Figure 4.14 Voltage vs. constant current of DMFC at 90
o
C 171
Figure 4.15 Voltage vs. constant current of DEFC at 90
o
C 171
Figure 4.16 Specific energy vs. current density of DMFC 172
Figure 4.17 Specific energy vs. current density of DEFC 173
Figure 5.1 Proposed schematic structure of the carbon-based solid acid 182
Figure 5.2 Thermogravimetric analysis (TGA) of the carbon-based solid acid 183
xvi
List of Schemes
Scheme 1.1 Electrochemical reactions in a H
2
/O
2
fuel cell 8
Scheme 1.2 Electrochemical reactions in a direct methanol fuel cell 19
Scheme 2.1 Synthetic scheme of PVDF-PSSA-PMMA membrane 62
Scheme 3.1 Synthesis of vinylsulfonyl chloride 127
Scheme 3.2 Synthesis of trifluoro-1-(trifluoromethyl)ethyl vinylsulfonate 128
Scheme 3.3 Synthesis of tetrabutylammonium vinylsulfonate 130
Scheme 3.4 Synthesis of PVDF-PSSA-poly(trifluoroethyl acrylate) 131
Scheme 3.5 Synthesis of PVDF-PSSA-poly(heptafluorobutyl acrylate) 133
Scheme 3.6 Synthesis of PSSA-PEFBE 139
Scheme 3.7 Scheme of synthesis of norbornene-based PEM 143
Scheme 5.1 Nafion-H catalyzed intramolecular cyclization 181
Scheme 5.2 Carbon-based solid acid catalyzed intramolecular cyclization 184
xvii
Abstract
Fuel cell has long been viewed as a promising scientific technology for power
generation in many applications. Polymer electrolyte membrane (PEM) is one of the key
components which determine the electronic performance of a PEM fuel cell. Significant
effort has been made in our research group to develop a suitable polymer electrolyte
membrane in direct methanol fuel cells (DMFC). The concept of interpenetrating
polymer network (IPN) was applied in the synthesis of poly(vinylidenefluoride)-
poly(styrenesuflonic acid)-poly(methyl methacrylate) (PVDF-PSSA-PMMA) composite
membranes. IPN technology retained the excellent physical and chemical properties of
PVDF and insured a more uniformed distribution of PSSA moiety in the membranes.
PMMA was introduced into IPN network as compatibilizer to reduce the interfacial
tension between PVDF and polystyrene, thus alleviating phase separation problem often
encountered in styrene-grafted membranes. Optimization of experimental condition led to
great improvement of the mechanical properties of PEM during membrane electrode
assembly (MEA) fabrication. PVDF-PSSA-PMMA membranes have been demonstrated
to exhibit substantially low methanol crossover rates compared to state of art materials
and high electrical performance in direct methanol applications.
xviii
In order to increase the performance of the MEA and reduce the cost, various
attempts have been made in our laboratories to develop several different polymer
electrolyte materials. It was observed that introducing a copolymer with polystyrene
increase the mechanical stability and enhances the PSSA distribution in an IPN system.
Present identification work of new monomers with functional groups which can be
polymerized and then hydrolyzed to sulfonic acid in an attempt to avoid the sulfonation
step is described. Several new polymer electrolyte membranes were synthesized and
tested in direct methanol fuel cell, some of them exhibited promising results.
Some organic compounds with non-toxic and environmentally friendly properties
have been considered as alternative fuels in fuel cell. Among them, acetaldehyde, ethanol,
glyoxal, dioxolane and trimethyl orthoformate all show promising electrical performance
results in fuel cell operations. We also prepared a newly reported carbon-based acid and
explored the new catalytic application and its potential as a proton conducting material in
fuel cells.
1
Chapter 1: Introduction
1.1 Fuel cell history
Fuel cell technology represents one of the most promising scientific tools for
mankind to improve life. Imagine driving home a fuel cell car with nothing but pure
water dripping from the tailpipe, a laptop that runs for several days on a single charge and
a house powered by fuel cell station with no need of long distance power lines. These
dreams motivate today’s fuel cell research. With the increasing threat by the fast
depleting of petroleum, coal and natural gas and the green house effect caused by burning
fossil fuels, people are forced to seek regenerative energy sources and develop more
efficient energy conversion devices fuel cells to save the valuable natural resources. Fuel
cell research and development are gaining more attention for its higher energy conversion
efficiency and low green house gas emission compared to other devices using thermal
engines. The power and energy efficiency of a fuel cell depend on the thermodynamics,
electrode kinetics, reactant mass transfer and materials for assembling the fuel cell. These
factors have been addressed throughout the fuel cell history and many challenges still
remain for fuel cell researchers.
2
Every new technology development story has a beginning and the fuel cell
concept dates back to 1839, when William Grove, a Scotsman, developed a prototype
H
2
/O
2
fuel cell using porous platinum electrode immersed in sulfuric acid. Then 120
years later, Francis Bacon developed a 5 kW fuel cell for powering a welding machine in
1959. In the 1960s, the U.S. Aerospace program demonstrated alkaline fuel cells using
high purity H
2
and O
2
[1]. The requirement of fuel purity to avoid poisoning the anode
and carbonization of the electrolyte limited the practical use of fuel cell for commercial
or residential purposes. There were increased research efforts in 1960s, which led to the
use of acid electrolytes capable of using fuels alternative to pure hydrogen. However,
typical acid electrolyte like H
2
SO
4
and H
3
PO
4
began to lose proton conductivity below
100 °C and the corrosive nature of the electrolyte further limited the cell hardware choice.
The challenge for better electrolyte led to one of the most important milestone in
fuel cell history, the discovery of polymer electrolyte membranes (PEM). PEM is more
compact than aqueous system and has the ability to perform well at low temperatures
with good proton conductivity. These advantages make PEM a promising candidate for
portable devices and vehicular applications. In 1955, Willard Thomas Grubb in GE
modified the original fuel cell design with a sulfonated polystyrene membrane as the
electrolyte [2]. However, this kind of membrane had undesirable physical properties and
3
poor lifetime stabilities. When hydrocarbon based ion exchange films were used, polymer
degradation at the anode side became noticeable. The reason is that attack on the weak
alpha C-H bond of polystyrene was caused by the highly oxidative anode environment.
Sulfonated tetrafluorethylene based copolymer (Nafion
®
) was a better PEM material. It
was developed by DuPont in the late 1960s. With the good mechanical properties and
thermal stabilities, Nafion
®
became the most widely used electrolyte material for PEM.
Hydrogen has been used as the fuel since the early stage of fuel cell development.
One source of H
2
came from the reformation of hydrocarbons. During the reformation, a
mixture of hydrogen and carbon monoxide (syngas) is produced. The carbon monoxide is
generally separated to obtain a pure stream of hydrogen. Furthermore, the residual CO
would strongly bind to the Noble catalyst and poison the catalyst layer. Hydrogen storage
requires high compression and therefore stringent safety measures have to be taken.
Hydrogen fuel cells also require gas humidification to maintain the conductivity of the
PEM material, which increases the cost and complexity of the system.
Direct methanol fuel cell (DMFC) development has been motivated by US
department of Defense in the early 1960s. The main advantage of the DMFC is the
liquid-feed delivery of methanol aqueous solution as a fuel. The discovery of Pt-Ru
binary metal alloy has greatly reduced the anode poisoning caused by the formation of
4
CO during methanol oxidation. With rapidly increasing demands for portable power
sources, research on DMFC has received much attention in the last decade. University of
Southern California (USC) and the Jet Propulsion Laboratory (JPL) demonstrated the
DMFC concept in the early 1990s and several advantages over H
2
/O
2
fuel cells were
demonstrated [3-4].
A fuel cell stack is formed by several single fuel cells connected together for
practical uses. Therefore, many complex components have to added and integrated into a
fuel cell system. With the rapid development in fuel cell manufacture and design, several
DMFC system prototypes have emerged recently [5-6].
Fuel cells represent a cleaner energy alternative to current devices, which utilize
hydrocarbon fuel resources. With higher efficiencies, inexpensive and renewable fuels,
and a relatively lower cost, the direct methanol fuel cell (DMFC) has the potential to be
used in automobiles and some portable electronic devices with the required efficiency to
offer more than 10 times higher power densities compared to current lithium-ion
rechargeable batteries [7]. Recently, several companies including Toshiba, Hitachi,
Fujitsu, Sony and Sanyo have developed some prototype laptops, cellular phones, and
personal digital electronics that are powered by DMFCs [8].
5
1.2 Background
In general, a fuel cell works by converting chemical energy of a fuel directly into
electrical energy. A fuel cell is like a chemical battery. It will continue to produce
electricity as long as the fuel is supplied. This is the key difference between a fuel cell
and a battery. Combustion engines also take the chemical energy stored in fuel and
transform it into electrical energy. But to do this, it needs to first burn the fuels to
generate heat, which is then converted to mechanical energy before finally transformed
into electrical energy through an alternator. The overall process involving all these steps
is potentially complex and thermodynamically inefficient. Fuel cell has the advantage of
being non-Carnot limited with theoretical efficiency much higher than that of the typical
internal combustion engine (Figure 1.1).
Combustion engine: chemical energy → heat → mechanical energy →electrical energy
Fuel cell: chemical energy --------> electrical energy
Figure 1.1 Scheme of energy conversion in a fuel cell and a combustion engine
The maximum thermal efficiency of a heat engine is determined by the theoretical
Carnot cycle, which is thermodynamically reversible. Equation 1.1 shows that η
th, Carnot
,
6
the thermal efficiency of reversible heat engines, depends on the ratio of the low (T
L
) and
high (T
H
) temperatures in the thermodynamic cycle. Since the low temperature is usually
fixed at the ambient condition, the efficiency is therefore determined by the highest
temperature in the cycle: the higher the temperature, the higher the efficiency. For a heat
engine that operates at 400
o
C and releases heat at 50
o
C, the reversible efficiency is 52%.
η
th, Carnot
= 1- (T
L
/ T
H
) (1.1)
For a fuel cell, the maximum amount of energy available to do work is given by
the Gibbs free energy. ΔH here represents the total energy, which is often called higher
heating value (HHV) when the fuel is burned. Thus the reversible efficiency of a fuel cell
can be written as Equation 1.2. For a hydrogen oxygen fuel cell, the efficiency is 83%.
For a Carnot cycle heat engine to match this thermal efficiency, the high temperature of
the cycle has to be 1480
o
C, with the low temperature being 25
o
C. Fuel cells hold a
significant thermodynamic efficiency at the low temperature (up to 950 K for a H
2
/O
2
fuel cell) compared to combustion engines.
7
η
th, fc
= ΔG / ΔH = n
e
FE / HHV (1.2)
η
fc
= η
th, fc ×
η
voltage ×
η
fuel
(1.3)
η
fc
= [ΔG / ΔH]
×
[V/ E]
×
[i / (nFνfuel)] (1.4)
The real efficiency of a fuel cell is less than the thermodynamic efficiency
considering voltage efficiency (η
voltage
) and the fuel utilization efficiency (η
fuel
) (Equation
1.3 and 1.4). The voltage efficiency is the ratio of the real operating voltage (V) to the
thermodynamically reversible voltage of the fuel cell (E). The fuel utilization efficiency
is the ratio of the fuel used by the cell to generate electric current (i) versus the total fuel
provided to the fuel cell (νfuel). F is the Faraday’s constant and n is the number of
electrons for the particular reaction. The fuel utilization efficiency accounts for the fact
that not all of the fuel provided will participate in the electrochemical reaction in a fuel
cell. Some of the fuel may undergo side reactions while some will flow through the fuel
cell without reacting. Table 1.1 shows thermodynamic data for some common fuels used
in fuel cells. It is interesting to note that the thermodynamic efficiencies of most fuels are
close to 100% at 298 K, which is significantly higher than the values of combustion
engines at similar conditions. However, the kinetics of electrochemical reactions will
8
greatly affect the performance of a fuel cell. To maximize the overall efficiency, kinetics
has to be studied extensively as well as fuel utilization.
Table 1.1 Thermodynamic data for some common fuels under standard condition
Fuel ΔH
o
(KJ/mol) ΔG
o
(KJ/mol) E
o
cell (V) n η
th, fc
(%)
Hydrogen -286 -237 1.229 2 83
Methane -891 -818 1.060 8 92
Methanol -727 -703 1.214 6 97
Ammonia -383 -338 1.170 3 88
Hydrazine -311 -312 1.610 4 100
Anode half reaction: H
2
———> 2 H
+
+ 2 e
-
E
o
= 0 V
Cathode half reaction: ½ O
2
+ 2 H
+
+ 2 e
-
———> H
2
O E
o
= 1.23 V
Overall reaction: H
2
+ ½ O
2
——— > H
2
O E
cell
= 1.23 V
Scheme 1.1 Electrochemical reactions in a H
2
/O
2
fuel cell
Figure 1.2 gives you an idea what a simple H
2
/O
2
fuel cell looks like. In a H
2
/O
2
fuel cell, the hydrogen combustion reaction is split into two electrochemical half cell
reactions (Scheme 1.1).
9
Figure 1.2 Schematic view of a H
2
/O
2
fuel cell
By separating these reactions, the electrons produced at the anode side are forced
to flow through an external circuit and do useful work before they complete the reaction
with oxygen at the cathode side. Polymer electrolyte membrane will separate the anode
from cathode, allowing only ions flow through but not electrons.
As mentioned before, a fuel cell using pure hydrogen as fuel operates with zero
emissions except water. Compared to combustion engines, fuel cells can produce
PEM
L
-
+
H
+
H
+
H
2
+ H
2
O
H
2
O
O
2
Anode
Catalyst, Pt
Cathode
Catalyst, Pt
Porous Gas
Diffusion Electrode
H
+
H
+
e-
Overall: H
2
+ ½ O
2
→ H
2
O E
0
= 1.23 V
10
electricity with a much higher efficiency. Fuel cells can be very simple with no moving
parts, which lead to a reliable and long-lasting system with low noise level. It allows easy
scaling between power (decided by the fuel cell stack size) and capacity (decided by the
fuel reservoir), which is difficult to do with batteries. Fuel cells offer higher energy
densities than batteries and can be recharged much faster by simple refilling fuel.
The actual fuel cell potential can not reach its ideal potential because of some
irreversible losses. Several factors contribute to irreversible losses in an actual fuel cell.
Being called polarization, overvoltage or overpotential, the losses arise primarily from
three sources: activation polarization, ohmic polarization and concentration polarization.
These losses combined result in a decreased cell voltage.
At low current density in the fuel cell operation, the activation polarization loss is
dominant. Electronic barriers must be overcome to obtain current and ion flow.
Activation loss increases with current. Ohmic polarization is dependent on current. It
increases over the entire range of current because of the constant cell resistance.
Concentration polarization also occurs over the entire range of current density and these
losses become prominent at high current densities where it becomes difficult to provide
enough reactant to the reaction sites.
11
Activation Polarization: Activation polarization happens when sluggish electrode
kinetics control the rate of electrochemical reactions at electrode surfaces. Therefore,
activation polarization is correlated with the rates of electrochemical reactions. It is
interesting to note the similarity between electrochemical and chemical reactions: Both
need to overcome activation energy by the reacting species.
Ohmic Polarization: Ohmic polarization is caused by the resistance to conduction of
ions in the electrolyte and the resistance to conduction of electrons through the electrode.
By enhancing the ionic conductivity of the electrolyte and decreasing the electrode
separation, the ohmic losses through the electrolyte can be reduced.
Concentration Polarization: Electrochemical reactions will consume the reactants at the
electrodes. If the surrounding material can not maintain the initial concentration of the
bulk fluid, there will be a decrease of potential. In another word, a concentration
polarization happened. Several processes contribute to the occurrence of concentration
polarization: solution/dissolution of reactants and products through the electrolyte, or
diffusion of reactants and products to and from the electrochemical reaction site in the
electrolyte, slow diffusion in the electrode pores. At the electrochemical reaction sites,
slow diffusion of reactants and products is a dominant factor in determining
concentration polarization.
12
Table 1.2 Major fuel cell types
Fuel cell
types
AFC
PEMFC
DMFC
PAFC
MCFC
SOFC
Operating
temp. (°C)
<100 60-120 20-120 160-220 600-800 800-1000
Application
Energy storage, Space mission,
Military, Transport,
Portable devices
Decentralized stationary power system,
Transport
Fuel
H
2
H
2
Methanol H
2
H
2
,
Hydrocarbons
H
2
,
Hydrocarbons
Charge
carrier
OH
-
H
+
H
+
H
+
CO
3
2-
O
2-
Catalyst
Pt Pt Pt-Ru Pt Ni Ceramic
Cell
component
Carbon
based
Carbon
based
Carbon
based
Carbon
based
Stainless steel
based
Ceramic based
There are six major types of fuel cells based on different electrolytes. They all
operate at different range, use different materials, differ in fuel tolerance and still share
the same underlying electrochemical principle (Table 1.2).
1. Polymer electrolyte membrane fuel cell (PEMFC).
2. Direct methanol fuel cell (DMFC).
3. Alkaline fuel cell (AFC).
13
4. Phosphoric acid fuel cell (PAFC).
5. Molten carbonate fuel cell (MCFC).
6. Solid oxide fuel cell (SOFC).
Polymer electrolyte membrane (PEM) can also be called proton exchange
membrane. PEM fuel cells can produce high power density with the advantages of having
low weight and volume. Platinum containing porous carbon electrodes and solid polymer
electrolytes are often used. Some fuel cells need corrosive fluids, while PEM fuel cells
need only hydrogen, water and oxygen from the air to operate. Hydrogen fuel can be
provided by on-board hydrocarbon reformers or directly by storage tanks. Low-
temperature operation of polymer electrolyte membrane fuel cells allows them to start
quickly, thus reducing the system components wear and enhance the system durability.
However, PEMFC uses a very expensive noble metal catalyst (typically platinum), and Pt
is also extremely sensitive to CO poisoning. If the hydrogen is reformed from methanol
or hydrocarbons sources, Additional separators have to be used to reduce CO moiety in
the fuel gas. PEM fuel cells have great potential in some stationary power plant
applications and some transportation applications. Hydrogen-based PEM fuel cells are
very promising for application in passenger vehicles, such as cars and buses, because
14
they have many advantages like their favorable power-to-weight ratio, fast start-up time
at relatively low temperature and low sensitivity to orientation.
Hydrogen production and storages can be a significant problem in the practical
application of these fuel cells in transportation. Hydrogen must be stored in highly
pressurized tanks for most fuel cell vehicles (FCVs) powered by pure hydrogen. To allow
cars to travel the same distance as gasoline-powered cars before refueling, enough
hydrogen should be stored on board. However, it’s difficult to do due to the low
volumetric energy density of hydrogen. Hydrogen can be reformed from higher density
liquid fuels, such as methanol, ethanol, gasoline, natural gas and liquefied petroleum gas.
However, the vehicles must have an on-board processor to reform the fuels to hydrogen.
This requirement increases costs and additional maintenance.
One of the first fuel cell technologies developed were alkaline fuel cells (AFCs).
AFCs were the first type of fuel cells used in the U.S. space program (including space
shuttle) to produce water and electrical energy on spacecrafts. The byproduct water was
consumed by astronauts. A solution of potassium hydroxide in water is used as the
electrolyte and a variety of non-precious metals are used as catalysts for AFCs. High
temperature AFCs work at a temperature as high as 250 °C, while newer AFC designs
operate at relatively lower temperatures between 23 °C and 70 °C.
15
High performance of AFC is obtained due to the high chemical reactions that take
place in the cell. One of the disadvantages of AFC is that it can be poisoned by carbon
dioxide (CO
2
). Even the small amount of CO
2
in the air can affect cell operation by
forming carbonate. Both the hydrogen and oxygen used in the cell have to be purified.
Susceptibility to CO
2
poisoning also reduces the cell's lifetime, adding to its cost.
Liquid phosphoric acid is used as an electrolyte in phosphoric acid fuel cells. The
platinum catalyst is supported by a Teflon-bonded silicon carbide matrix. This silicon
carbide contains the acid and porous carbon electrodes. The phosphoric acid fuel cell
(PAFC) is regarded as one of the earliest version of modern fuel cells. It has been well
developed has been used commercially as the first type of fuel cells for stationary power
generation. Special interest has been paid to PAFC because of its application in powering
large vehicles like city buses.
PAFCs are more tolerant to impurities in fossil fuels that have been reformed into
hydrogen than PEMFCs, because of their higher operating temperatures. They have 85%
efficiency when used for the co-generation of electricity and heat but less efficient in
generating electricity alone (37%–42%). This is only slightly more efficient than
combustion-based power plants, which typically operate at 33%–35% efficiency.
16
However, the large size and heavy weight render PAFCs lower efficiency compared to
other fuel cells.
Being developed for natural gas and coal based power plant, molten carbonate
fuel cells (MCFCs) have great potential in electrical utility, industrial and military
applications. MCFCs have a very high operating temperature of 600-800 °C. The
electrolyte of a MCFC is a molten carbonate salt mixture, which is suspended in a porous
and chemically inert ceramic lithium aluminum oxide (LiAlO
2
) matrix. One advantage of
MCFC is that it can use non-precious metals as catalysts, thus cutting costs. High
efficiency is another advantage of MCFCs. An efficiency as high as 60% can be obtained
using a molten carbonate fuel cell, which is much higher than the 43% efficiency of a
plant using a phosphoric acid fuel cell. Overall fuel efficiencies can be as high as 85%
when the waste heat is properly used.
Unlike other fuel cells, MCFCs do not need an external reformer to convert other
hydrocarbon fuels to hydrogen. With the high operating temperatures of MCFCs, these
hydrocarbon fuels are converted directly to hydrogen within the fuel tank through an
internal reforming process. Thus, molten carbonate fuel cells are not subject to carbon
monoxide or carbon dioxide poisoning. In fact, carbon monoxide can be used as a fuel in
MCFCs, making MCFCs more attractive when coal-based gases are used.
17
However, MCFCs has one primary disadvantage, which is the cell durability.
Because MCFCs use very high operating temperatures and corrosive electrolytes, cell
components breakdown is fast and the cell life is shortened. Corrosion-resistant materials
for components are needed to improve the cell life, and new fuel cell designs needs to be
explored without decreasing fuel cell efficiency.
Solid oxide fuel cells (SOFCs) use some ceramic compounds as the electrolytes.
The SOFCs can adopt different configurations without sticking to the plate-like
configuration typical of other fuel cell types because of the solid electrolyte. The
efficiency of SOFCs is expected to be around 50%. Utilizing the system's waste heat, the
overall fuel efficiencies could be 80%. Solid oxide fuel cells work at temperatures as high
as 1,000 °C. Like MCFCs, precious-metal catalyst is not needed for SOFCs. A variety of
fuels can be used because the high operating temperature allows SOFCs to reform fuels
internally. SOFCs also have the advantage of being sulfur-resistant fuel cell. They can
tolerate carbon monoxide (CO) moiety, which can be utilized as a fuel. These
properties allow SOFCs to use coal-made gases. The high operating temperature results
in a slow start-up and additional thermal shielding are required to retain heat, which will
be a disadvantage for small portable applications and transportation. Stringent durability
requirements on materials are needed because of the high operating temperatures.
18
Most fuel cells use hydrogen as fuel, which can be directly fed to the fuel cell
system or can be generated within the fuel cell system by reforming hydrocarbon fuels.
Hydrogen has a low volumetric energy density and is not widely available. The storage
problem combined with the complex humidification system make people turn to other
alternative fuels such as methanol.
Figure 1.3 Schematic drawing of a direct methanol fuel cell
PEM
L
-
+
H
+
H
+
CH
3
OH +
H
2
O
H
2
O
O
2
Anode
Catalyst, Pt-Ru
Cathode
Catalyst, Pt
Porous Gas
Diffusion Electrode
CO
2
H
+
H
+
e-
Overall: CH
3
OH + 3/2 O
2
→ CO
2
+ 2 H
2
O E
0
= 1.21 V
19
Due to the higher energy density of methanol compared to hydrogen, direct
methanol fuel cells do not have many of the fuel storage problems typical of some fuel
cells. Methanol is also easier to distribute and transport using our current
infrastructure because it is a liquid fuel just like gasoline.
With methanol as a fuel, the volumetric energy density is about 16 MJ/L
compared with 4 MJ/L of a 7500 psi hydrogen fuel. The liquid feeding character of a
direct methanol fuel cell (DMFC) avoids a complex humidification system used in a
H
2
/O
2
fuel cell. In the DMFC, energy is converted when liquid methanol is directly
oxidized at the anode and produces protons and electrons. Protons diffuse through a
polymer electrolyte membrane (PEM) to the cathode, where they combine with electrons
to produce electrical energy with water and carbon dioxide as byproducts (Figure1.3).
The anode, cathode, and overall cell reactions, respectively, are shown in Scheme 1.2.
Anode half reaction: CH
3
OH + H
2
O ——> CO
2
+ 6 H
+
+ 6 e
-
E
o
= 0.02 V
Cathode half reaction: 3/2 O
2
+ 6 H
+
+ 6 e
-
——> 3 H
2
O E
o
= 1.23 V
Overall reaction: CH
3
OH
+ 3/2 O
2
——> CO
2
+ 2 H
2
O E
cell
= 1.21 V
Scheme 1.2 Electrochemical reactions in a direct methanol fuel cell
20
The anode catalyst/PEM/cathode catalyst composite is referred to as a membrane
electrode assembly (MEA). Catalyst usually adhered to the membrane and carbon fiber
cloth or gas diffusion layer (GDL), and is placed on each side of the MEA.
Many parameters will affect the performance of PEM fuel cells. It is necessary to
understand the effect of operating parameters on fuel cell performance to improve the
fuel cell efficiency. It was reported that current DMFCs power densities double that of
current lithium-ion rechargeable batteries with an overall efficiency around 25%. There
are two factors that hinder the ability of the DMFC to reach its maximum efficiency:
slow reaction kinetics at the anode and methanol crossover [10]. Significant power
density and efficiency losses will occur with high methanol crossover through polymer
electrolyte membranes (PEMs). Many efforts have been made in solving methanol
crossover. Although many PEMs do not exhibit much higher efficiency than Nafion, the
benchmark PEM in DMFCs, some new PEMs show lower methanol crossover at similar
proton conductivities and higher DMFC power densities [11-13]. The overall six-electron
oxidation of methanol is composed of a series of successive reactions forming
formaldehyde and formic acid as intermediates which contribute to slow reaction rates
and decreased cell voltage [14]. The switch from platinum to platinum/ruthenium for the
anode catalyst provided some initial improvement in methanol oxidation rates [10].
21
Currently, the investigation of new anode catalysts to improve oxidation reaction rates is
an active area of research in DMFC [15].
Methanol crossover also contributes to decreased overall cell efficiency and cell
lifetime [16-17]. The reaction of methanol at the cathode results in a loss of fuel and
cathode voltage and is referred to as a mixed potential. The crossover can also result in
cathode flooding. To solve this problem, low methanol concentrations are often used in
the DMFC, which will limit the overall cell power density. Practical DMFCs have an
open circuit voltage (0.7 V) approximately half of the reversible cell voltage (1.21 V),
which is determined from the molar Gibbs free energy change in the DMFC overall
reaction. Besides cell operating temperature, anode and cathode flow rates can also
impact DMFC performance. One thing need to be mentioned is that methanol
concentration has a significant effect on the fuel cell performance. When the methanol
concentration in the anode side is increased from 2 M to 6 M, a 60 percent reduction in
cell voltage at a certain operating current density is observed. At the same time, open
circuit voltage decreases with the increase of methanol concentration [18]. Therefore, the
development of PEM with low methanol crossover is crucial.
22
1.3 Membranes development history
The development of membranes for fuel cell applications started in 1950s by GE
with the introduction of phenolic films, prepared by condensation polymerization of
phenolsulfonic acid with formaldehyde. These membranes had poor mechanical property
and a short lifetime less than 1000 hours [19]. During 1960s, GE developed partially
sulfonated polystyrene sulfonic acid membranes in an effort to improve the power density.
This type of membrane was synthesized by dissolving polystyrene sulfonic acid in
ethanol-stabilized chloroform followed by sulfonation at about 25
o
C. With a better water
uptake and an improved power density of 0.5 kWm
−2
, this membrane was used in seven
NASA's Gemini space missions [20]. After some initial unsatisfying attempts, GE
immediately redesigned their PEM cell and established a new model. This new model
served adequately for the subsequent Gemini flights despite poor performance and some
malfunctions on Gemini 5. However, polymer degradation of this membrane caused
brittleness in the dry state and membrane lifetime was not satisfactory. GE then
developed “D” series of membranes made by preparing crosslinked polystyrene sulfonic
acid polymer in an inert matrix to improve the mechanical strength and the life of the
membrane in the late sixties. The lifetime of the membrane can reach 10,000 hrs and the
power density can be as high as 0.8 kWm
−2
[21]. The main issue encountered with all the
23
above mentioned membranes was the difficulty to obtain a power density even as low as
100 mWcm
−2
because of the insufficient proton conductivities [22].
To address the stability issue, GE developed “S” series membranes from
homopolymers of trifluorostyrenesulfonic acid. The chemical stability of this membrane
is much higher than that of hydrocarbon based polystyrenesulfonic acid. This was
because of the strength of the C-F bond and its resistance to degradative cleavage.
However, the high degree swelling of the membrane still existed. To solve the problem,
the homopolymer was blended with polyvinylidene fluoride (PVDF) and plasticizer
triethyl phosphate. The result showed a life time extension to 2000 hours at 80 °C.
Improved lifetime was obtained by grafting trifluorostyrenesulfonic acid into an inert
fluorocarbon matrix. This established the foundation for future work into grafted
membrane systems.
CF
2
CF
2
CF
2
CF
2
OCF
2
CF
CF
3
OCF
2
CF
2
SO
3
H
x
y
n
Figure 1.4 Chemical structure of Nafion-H
24
Table 1.3 Properties of commercial cation-exchange membranes (reprinted from
reference 9)
Membrane
Membrane type
IEC
(mequiv./g)
Thickness
(mm)
Gel
water
(%)
Conductivity (S/cm) at
30 °C and 100% RH
Asahi Chemical Industry Company Ltd., Chiyoda-ku, Tokyo, Japan
K 101 Sulfonated polyarylene 1.4 0.24 24 0.0114
Asahi Glass Company Ltd., Chiyoda-ku, Tokyo, Japan
CMV Sulfonated polyarylene 2.4 0.15 25 0.0051
DMV Sulfonated polyarylene – 0.15 – 0.0071
Flemion Perfluorinated – 0.15 – –
DuPont Company, Wilmington, DE 19898, USA
Nafion 117 Perfluorinated 0.9 0.2 16 0.0133
Nafion901 Perfluorinated 1.1 0.4 5 0.01053
Pall RAI Inc., Hauppauge, NY 11788, USA
R-1010 Perfluorinated 1.2 0.1 20 0.0333
In 1970s, DuPont developed Nafion, a perfluorosulfonic acid membrane, which
not only showed a significant increase in the proton conductivity of the membrane but
also greatly extended the membrane lifetime by almost four orders of magnitude. The
membrane was used extensively in chloralkali cells for chlorine production. Nafion soon
25
became a standard material for PEMFC and is still in today’s fuel cell applications.
Nafion membranes were incorporated into some fuel cell applications such as the
“biosatellite missions” in the lat 1960s by GE. The Dow Chemical Company and Asahi
Chemical Company synthesized advanced perfluorosulfonic acid membranes with shorter
side chains and a higher ratio of SO
3
H to CF
2
groups [23]. Table1.2 provides a
comparison of some commercial cation-exchange membranes.
1.4 Fuel Cell membrane materials
As the key component of the PEM fuel cell, the membrane must possess the
following desirable properties to achieve high efficiency:
• High proton conductivity to support high currents with minimal resistive losses and zero
electronic conductivity;
• Strong mechanical strength and stability;
• Low fuel cross over;
• Chemical and electrochemical stability under operating conditions;
• Low costs for commercial applications.
Many PEM have been developed in the past [24], the main factors affecting the
performance of these membranes are the level of hydration and thickness of the
26
membrane. The performance of a membrane depends on proton conductivity, which in
turn attributes to the hydration levels. Higher levels of hydration of membranes are
needed to support higher membrane conductivity. When the water contents of the
membranes are too high, the cathode side might getting flooded which will block the
active site of catalyst hence slow down the oxygen reduction kinetics. Zadowzinski et al
have studied this particular problem related to Nafion, caused by a phenomenon known
as electro-osmotic drag [25]. The electro-osmotic drag coefficient [EODC] is defined as
the number of water molecules transported per proton. It is a quantitative measure of
hydration. Their studies find that if Nafion117 (175 μm) is in equilibrium with water
vapor, the EODC is about 1. When Nafion117 is immersed in water, EDOC is about 2.5.
Zadowzinski et al. also indicated that the drag is mainly a function of membrane water
content no matter which type of Nafion membrane is used. The amount of water within
Nafion has been used to explain the corresponding increase in proton conductivity.
Researchers are trying many methods to improve the fuel cell performance. One
way to avoid water drag or water crossover is to reduce the membrane thickness.
Reduced membrane thickness will decrease membrane resistance to enhance membrane
conductivity, cut cost and accelerate hydration speed. However, because of difficulties
with membrane durability and mechanical stability, there is a limit to the extent to which
27
membrane thickness can be reduced. The appropriate way to balance this would be to
control the acidic group’s distribution in the membrane and increase the charge density in
the microstructure of the proton exchange membrane to obtain highly conductive
materials. By preparing the membranes in asymmetric or thin film composite form, high
charge density can be obtained. For example, solvent casting using chloroform as solvent
and methanol as non-solvent can produce asymmetric films of partially sulfonated
polystyrene or poly (phenylene oxide) [26]. Surface modification of the membranes can
obtain the desired spatial control of the acidic regions. For example, the desired number
of acidic regions can be obtained by surface modification of chitosan using aqueous bath
of sulfuric acid of suitable concentration [27]. Spatial control of the acidic group
distribution and increasing the charge density in the polymer electrolyte membrane not
only reduces the film thickness but also enables significant enhancements in the proton
conductivity. Susai et al. indicated back diffusion will be promoted by a thinner
membrane, thus producing a greater concentration gradient of water when enhanced
dehydration rate presents at higher temperatures. It is also indicated that the fuel cell can
run at lower humidity using a thinner membrane [28].
Membrane materials can be classified into the following category:
• perfluorinated ionomers,
28
• partially fluorinated polymers,
• non-fluorinated membranes with aromatic backbone,
• non-fluorinated hydrocarbons,
• acid–base blends.
Table 1.4 Classification of membrane materials (reprinted from reference 9)
Perfluorinated Partially
fluorinated
Non
fluorinated
Acid-base blends Others
-PFSA
-PFCA
-PFSI
-Gore-select
-PTFE-g-TFS
-PVDF-g--
PSSA
-NPI
-BAM3G
-SPEEK
-SPPBP
-MBS-PBI
-SPEEK/PBI/PVP
-SPEEK/PEI
-SPEEK/PSU(NH
2
)
2
-SPSU/PBI/PVP
-SPSU/PBI
-SPSU/PSU(NH
2
)
2
-PVA/H
3
PO
4
-Poly-AMPS
-Supported
composite
membrane
1.4.1 Perfluorinated membranes
The perfluorinated membranes currently used in portable fuel cell application,
such as Nafion, have perfluorinated structures with sulfonic acid functional groups
attached to the backbone. There are two research areas of interest on Nafion. One is the
29
transport phenomena within the membrane. The other is to modify the membrane
structure to increase its performance and water retention capacity. By varying the
operating parameters, such as temperature, water content and membrane thickness,
proton transport behavior in a Nafion membrane can be changed.
Hydration level of a Nafion
membrane is an important factor that has to be
maintained in order to retain the performance. Protons and the sulfonic acid groups are in
the solvated form in the presence of water, and this will facilitate the proton transport.
Some modeling studies have been conducted to understand the transport of water
in perfluorosulfonic acid membranes. The optimization of operating conditions and the
composition of the membrane will result in higher efficiencies and power densities. In the
early 1980s, Gierke and Hsu proposed the description of the microscopic structure of the
polymer, which established the basis for all these models [29]. By studying the
experimental data, they revealed the correlation between geometric and
phenomenological swelling of the polymer and the uptake of water and its effect on the
water diffusion coefficient in the membrane pores. A proposal of a widely accepted
description of the polymeric membrane was based on the correlation evolved by
analyzing the data taken under different operating conditions. Inverted micellar structures
are present in membranes. Being separated from the fluorocarbon backbone, the ion
30
exchange sites form spherical clusters (pores), which are connected by short narrow
ionomeric channels. Therefore the model was called cluster network model. When the
membrane is at dry state, clusters have average radius of about 1.8 nm and about 26 SO
3
−
groups are distributed on the inner pore surface. In the wet state of the membrane, the
number of fixed SO
3
−
groups increase to 70 while the cluster diameter increases to about
4 nm. At these conditions, each pore contains about 1000 water molecules and the
diameter of the connecting channels is about 1 nm.
Percolation theory was proposed for the correlation of proton conductivity with
the water content of the membrane. According to this theory, a critical amount of water is
needed in the membrane to retain the performance. Ion transport is very difficult due to
the absence of extended pathways when the water content is below certain value. Okada
et al reported some interesting results on the modeling of microstructure. [30]. By
interpolation of experimental data, they demonstrated that around 50% of the water
molecules in the membrane were associated with the protons or the SO
3
−
sites. The
remaining 50% of the water were semi-free in the pores like bulk water.
Recently, a “random network model” was proposed when studies found an
interpretation of the percolation properties of proton conductivity as a function of water
content [31]. This model is a modification of the “cluster network model”. As described
31
in this model, an intermediate region contains the side chains ending with pendant
sulfonic acid groups. These acid groups are ionically bonded to the perfluorinated
backbone and tend to cluster within the whole membrane structure, thus forming
hydrated regions. Different from the “cluster network model”, the hydrated regions in
membrane structure are distributed randomly in the polymer matrix. This phenomenon
renders quicker protons transportation because of the rotating ability of the side chains.
The traverse motion of protons through the membrane is possible even if the hydrated
regions drift apart. Through small angle X-ray scattering experiments (SAXS), Haubold
et al. verified that the “random network model” could be applied to Nafion [32]. Using
the atomic force microscopy (AFM) technique in the testing of Nafion117, James et al.
indicated that the “random network model” can be generally acceptable when the
membrane is in the region of 9–34% relative humidity [33].
Several efforts have been made to ascertain water retention of Nafion at higher
temperatures. It was found that addition of silica to Nafion can improve the retention of
water in the membrane and enable the operation of the fuel cell above 130 °C. Used in a
DMFC at 145 °C, this kind of membrane could obtain a power density of 240 mW/cm
2
.
Such membrane exhibited a significant improvement in proton conductivity. However,
methanol crossover hadn’t been reduced through this method.
32
Wasmus et al. [34] tried to improve the performance of a DMFC by equilibrating
Nafion117 with phosphoric acid moiety. High conductivity at 200 °C and lower methanol
crossover were observed accompanied by an improvement in the reaction kinetics at both
anode and cathode side. Mex and Muller [35] tried to reduce methanol crossover by
coating a thin layer of plasma polymerized tetrafluoroethylene (with vinyl phosphoric
acid) on Nafion membrane. Finsterwalder and Hambitzer adopted a similar method of
depositing a PTFE source with chlorosulfonic acid on Nafion to reduce methanol
crossover [36]. A significant decrease in methanol crossover was achieved, while a
decrease in conductivity was detected.
To improve the conductivity of Nafion, significant efforts have been made by
incorporating perfluorinated ionomers in Nafion matrix and by doping it with
heteropolyacids in Nafion membranes [37]. Tazi and Savadogo made a similar attempt to
improve proton conductivity and high power density with the incorporation of thiophene
(TH) and silicotungstic acid (SA) in Nafion 117 membrane [38]. Compared to the normal
value of 27% for Nafion 117, water uptake of the SA-incorporated Nafion was 60%,
while with both SA and TH it was 40%. With a substantial improvement in proton
conductivity, TH incorporated Nafion117 showed a high current density of 810 mA/cm
2
33
at 600 mV compared with a value of 640 mA/cm
2
for Nafion 117 under similar
conditions.
With the complex fuel cell requirements of high proton conductivity and high
chemical stability, Nafion and related polymers are still being studied to achieve the goal
to obtain a lifetime of 60,000 hrs at 80 °C. However, the major disadvantages of these
perfluorinated sulfonic acid materials are:
• Lack of safety during the membrane manufacture process.
• High cost of membrane material (about US$ 700 per square meter).
• Needing additional supporting equipment such as humidification system.
• Membrane decomposition related to temperature limitations.
Above 150 °C, Nafion and related membranes cause safety concerns because of
the toxic intermediates and corrosive gases liberated at high temperatures. During the
manufacturing process or vehicle accidents, decomposition products could be a concern.
There are extensive requirement for supporting equipment when PFSA
membranes are used. The hydration system such as a humidifier adds considerable cost
and complexity to the transportation applications.
Another issue to be concerned is the degradation of PFSA membranes at elevated
temperatures. Conductivity of membrane at 80 °C is reduced by as much as 10 fold
34
compared to that at 60 °C. At temperatures above 80 °C, the phenomena related to
membrane dehydration, ionic conductivity reduction, decreased affinity for water, loss of
mechanical strength due to softening of polymer backbone and increased parasitic losses
due to high fuel permeation rate are observed.
In direct methanol fuel cells (DMFC), Nafion exhibits high methanol permeability
at 80 °C, which dramatically reduces the DMFC performance and makes it a unsuitable
PEM material for DMFCs [39]. Significant efforts are made to eliminate the
disadvantages such as crossover problems and loss of hydration at high temperatures.
Nafion is still the choice for many PEFC and DMFC applications despite all the
shortcomings. Efforts had been made so that Nafion will be replaced by an alternative
membrane in the future. Research work is being carried out to identify promising
alternatives in order to overcome the disadvantages of the PFSA membranes. If the cost
factors are commercially realistic, some sacrifice in material lifetime and mechanical
stability may be acceptable to produce materials that are less expensive than Nafion.
Recently, the use of hydrocarbon polymers for the membrane materials has attracted
renewed interest, even though they had been previously abandoned because of their low
thermal and chemical stability.
35
1.4.2 Hydrocarbon membranes
CH
2
CH
CH
2
CH
3
CH
2
CH
2
CH
2
CH CH
2
CH
SO
3
H
x y
m
n
Figure 1.5 Structure of poly(butadiene styrene) block copolymer
CF
2
CF
2
CF
2
CF
CH
2
C H
CH
3
SO
3
H
m
n
x
(a)
(b)
Figure1.6 Structure of grafted membranes: (a) FEP main; (b) Sulfonated polystyrene side
chain
Hydrocarbon membranes have some advantages over PFSA membranes. They are
less expensive, commercially available and their structure permits the introduction of
polar sites as pendant groups in order to increase the water uptake. Gilpa and Hogarth
identified more than 50 alternatives to PFSA membranes. Among these, about 15
36
membranes showed potential for replacing Nafion membranes. The structures of
prominent ones are given in Figure 1.5 and Figure 1.6.
1.4.3 Aromatic polymers
To increase membrane stability at high temperatures, aromatic hydrocarbons can
be incorporated directly into the backbone of a hydrocarbon polymer. It can also be add
to polymers modified with bulky groups in the backbone to enhance the proton
conductivity. Polyarylenes are high temperature rigid polymers with glass transition
temperatures higher than 200 °C due to the inflexible and bulky aromatic groups in the
polymers [40]. The aromatic rings make electrophilic and nucleophilic substitution
possible. Polyether ketones (PEK) with different number of ketone and ether
functionalities (such as PEEK, PEKK, PEKEKK, etc.), Polyethersulfones (PESF),
poly(arylene ethers), polyesters and polyimides (PI) are examples of main chain
polyarylenes [41].
Research shows that polyesters are not a good choice for the membrane material
because the ester group imparts instability in aqueous acids while polyaromatics are good
PEM candidates due to their excellent thermal stability. Polymers on suitable
modification are not only thermally stable but also show stability in oxidizing, reducing
37
and acidic environments. Efforts have been made in this direction and design information
of membranes considered as alternatives to PFSA are reported [42-44]. Several research
groups have put great efforts on working with different sulfonated polymers containing
diarylsulfone units [45-47].
O
CH
3
C H
3
S
O O
O
SO
3
H
n
C O
O
SO
3
H
O
n
(a)
(b)
O
O
SO
3
H
n
(c)
Figure 1.7 Structures of (a) SPSU, (b) SPEEK and (c) SPPBP
38
Direct sulfonated polyimides have attracted some interest recently [48-50].
Sulfonated polybenzimidazoles have been studied by Asensio et al. and it was revealed
that they exhibit superior performance to Nafion at higher temperatures [51-52].
Poly(aryloxyphosphazene)s functionalized with sulfonimide units or phenyl phosphonic
units [53] have also found potential in membranes materials for fuel cells. As methods to
reduce water swelling and methanol permeation, blending and radiation crosslinking of
polyphosphazenes have also been studied [54]. Schuster et al. proved that imidazole-
terminated ethylene oxide oligomers can reach conductivities as high as 10
−5
S cm
−1
at
120 °C [55]. Some examples of prominent polymers are shown in Figure 1.7.
1.4.4 Acid-base complex
Acid–base complexes can maintain relatively high conductivity at high
temperatures without suffering from dehydration effects. They have been considered as a
possible alternative for fuel cell membranes. Incorporation of an acid component into an
alkaline polymer base renders the acid–base complexes the required proton conductivity
to be considered for fuel cell membranes.
39
The poly(2,2
1
-(m-phenylene)-5,5
1
-bibenzimidazole)/phosphoric acid (PBI/H
3
PO
4
)
complex is very promising as a new PEM material. Many attempts were made to
optimize this particular system because it has shown a great deal of potential for medium
temperature fuel cell applications. PBI complexes have been studied extensively at Case
Western Reserve University. The structure of the reactants and the corresponding
products are shown in Figure 1.8 [56].
N H
2
NH
2
NH
2
N H
2
+
OPh
O O
PhO
(a)
(b)
N
H
N
H
N N
n
(c)
Figure 1.8 Structure of (a) tetraaminobiphenyl, (b) diphenylisopthalate and (c) poly[2,2
’
-
(m-phenylene)-5,5
’
bibenzimidazole]
Qinfeng et al. have characterized the phosphoric acid doped PBI with various
separating unit doping levels. Doping level is defined as the molar percentage of acid per
40
repeating unit of the polymer. The conductivity of doped PBI does not depend on
humidity compared to Nafion
®
. However, the conductivity of such complexes is sensitive
to the doping level and temperature. With 450% doping at a temperature of 165 °C, the
conductivity of PBI membrane was about 4.6 × 10
−2
S/cm, while with very high levels of
doping (around 1600%), the proton conductivity could reach as high as 0.13 S/cm. A
practical fuel cell was made with a doped PBI/H
3
PO
4
membrane operating at 190 °C.
Under atmospheric pressure, the cell yielded a power density of 0.55 W/cm
2
at a current
density of 1.2 A/cm
2
[57]. It was interesting to note the poison tolerance of the electrode
catalysts is significantly improved at such elevated temperatures.
Reported by Qingfeng et al, another remarkable property of acid doped PBI is its
relatively low electro-osmotic drag coefficient (EODC). Nafion117 showed an EODC of
3.2 while the EODC was zero for acid doped PBI. Based on the electro-osmotic drag data
and the dependence of ionic conductivity on extent of acid doping, Grotthus mechanism
was used to explain the proton transport behavior in doped PBI. The distance between the
clusters of acid sites decreases when the doping content increases, and the anion moieties
will help the proton hopping between imidazole sites. Bouchet et al. also reported data
supporting a Grotthus mechanism [58]. The relatively high change in entropy ( ∆S, which
could be due to the molecular rearrangements necessary for the Grotthus mechanism) and
41
conductivity data of doped PBI at temperatures below the glass transition temperature
(Tg) show proof of such a mechanism.
Operating with a 500% doped PBI, a PEFC produced a power density of
25 W/cm
2
with a current density of 700 mA/cm
2
at 150 °C. The membrane used in this
case had excellent chemical and thermal stability, low gas crossover and good flexibility
at high temperature [59]. The water produced by the reaction was sufficient to maintain
the proton conductivity and the system can sustain an experimental period of 200 hrs.
Conclusion can be made that PEFC employing the doped PBI membranes can operate
efficiently at low humidity compared to Nafion membranes.
To achieve an improvement in proton conductivity, Samms et al. made efforts to
complex PBI with acids such as sulfuric acid, hydrochloric and phosphoric acid. A proton
conductivity of 9 × 10
−3
S/cm for PBI acidified with H
3
PO
4
was recorded for H
2
/O
2
fuel
cell. This is higher than the value of 8 × 10
−4
S/cm obtained for pure PBI membranes [60].
Doped PBI membranes are very promising in fuel cells operating at moderate
temperatures. Because acid doped PBI complexes are known to have lower methanol
permeability compared to Nafion, membranes in this category may be made suitable
alternatives to Nafion for DMFC applications in the near future.
42
O
CH
3
C H
3
S
O O
O
NH
2
NH
2
n
(a)
N
n
(b)
NH
NH
n
(c)
N
H
N
N
H
N
n
(d)
O
CH
3
C H
3
S
O O
O
SO
3
H
n
C O
SO
3
H
O
O
n
(e)
(f)
Figure 1.9 Structure of basic polymers (a–d) and acidic polymers (e, f)
Besides PBI, many polymers have also been evaluated for use in these types of
membranes [61]. Bozkurt and Meyer have studied poly(4-vinylimidazole)-H
3
PO
4
complexes and found its stability to be about 150 °C using thermogravimetric method.
43
Complexes of amorphous polyamide with H
3
PO
4
are reported by Lassegues to have high
conductivity but poor mechanical property and chemical stability at temperatures above
90 °C [62]. Blends of sulfonated polysulfones and PBI which were doped with H
3
PO
4
have been prepared [63] by Hasiotis. At 80% relative humidity, these membranes showed
improved mechanical properties and higher conductivities compared to acid-doped PBI
membranes under the same conditions.
Investigation of the long-term stability of doped PBI membranes has been carried
out. There are other acid-base blends reported [64]. The structures of some acidic and
basic polymers and their complexes are shown in Figure 1.9.
1.5 Conclusions
Based on the previous discussions, it can be concluded that despite the
advancement in the membrane development, some restrictions still need to be addressed.
There are still lots of room left to improve the desired membrane properties, identify
appropriate membrane candidates and accelerate the commercialization of fuel cell
technology.
Various solid polymer electrolytes tested for fuel cell applications as proton
conducting membranes are under investigation. Fuel cell membranes could be divided
44
into four categories as described above: perfluorinated ionomers, non-fluorinated
hydrocarbons, sulfonated polyarylenes and acid–base complexes.
Nafion is a prominent perfluorinated polymer material. The traditional Nafion
membranes for DMFC applications can not satisfy all the requirements. Thinner
membrane materials are preferred in hydrogen/oxygen fuel cell applications because of
increased electrical performance and reduced ionic resistance. However, thin membranes
(such as Nafion 112) result in a high methanol permeation rate in DMFC application [65].
As these disadvantages often exceed the advantage of low ionic resistance, thicker
membranes like Nafion 117 are more suitable in DMFC application.
It was shown that s-PPBP exhibits great potential for the sulfonated polyarylenes
and phosphoric acid doped PBI membranes are promising in acid–base complexes. These
polymers have good ionic conductivity, appropriate chemical and thermal stability to
meet the requirements of fuel cell membranes. Much efforts has been conducted to
understand the proton transport and to find new methods of improving PEM properties
but development of an alternative to Nafion is yet to be materialized. To develop a
polymer electrolyte membrane with high proton conductivity and low methanol cross-
over, significant effort has been made in our research group.
45
1.6 Chapter 1 References
1. Appleby, A.J., Foulkes, F.R., Fuel Cell Handbook, Krieger Publishing Co.,
Malabar, Florida, 1993.
2. Grubb, W.T., J. Electrochem. Soc. 106 (1959) 275-278.
3. Surampudi, S., Narayanan, S.R., Vamos, E., Frank, H., Hapert, G., Laconti, A.,
Kosek, J., Prakash, G.K.S., Olah, G.A., J. Power Sources 47 (1994) 377.
4. Narayanan, S.R., Kindler, A., Jeffries-Nakamura, B., Wang, C., Frank, H., Smart,
M., Surampudi, S., Halpert, G., “performance of PEM liquid Feed Direct
Methanol-Air Fuel Cells”, Electrochem. Soc. Proceedings volume 95-23 (1995)
61.
5. Othman, R., Yahaya, A.H., Arof, J.K., J. appl. Electrochem. 32 (2002) 1347.
6. Yang, C.C., Lin, S.J., Fauvarque, J.F. J. Power Sources 101 (2001) 267.
7. Service, R.F. Science 296 (2002) 1222.
8. Paulson, L.D. Compute 36 (2003) 10.
9. Smitha, B., Sridha, S., Kahn, A.A. J. Membrane Science 259 (2005) 10-26.
10. Larminie, J., Dicks, A. Fuel Cell Systems Explained, 2nd ed., Wiley: New York,
2003.
11. Lu, G.Q., Wang, C.Y., Yen, T.J., Zang, X., Electrochem. Acta 49 (2004) 821.
12. Blum, A., Duvdevani, T., Philosoph, M., Rudoy, N., Peled, E. J. Power Sources
117 (2003) 22.
13. Kelley, S.C., Deluga, G.A., Smyrl, W.H., Electrochem. Solid State Lett. 3 (2000)
407.
46
14. Hamnett, A. Catal. Today 38 (1997) 445.
15. Wasmus S., Kuver, A. J. Electroanal. Chem. 461 (1999) 14.
16. Ravikumar, M.K., Shukla, A.K. J. Electrochem. Soc. 143 (1996) 2601.
17. Scott, K., Taama, W., Cruickshank, J., J. Appl. Electrochem. 28 (1998) 289.
18. Ge, J., Liu, H., J. Power Sources 142 (2005) 56.
19. Magnet H.J.R., In: Handbook of Fuel Cell Technology, Berger, C., Editor,
Prentice-Hall, Englewood Cliffs, NJ, USA (1968) 425.
20. Bockris J.O’M.,and Srinivasan, S., Fuel Cells: Their Electrochemistry, McGraw-
Hill, New York (1969).
21. Grot, W.G., Discovery and development of Nafion perfluorinated membranes,
Chem. Ind. 19 (1985) 647.
22. Wakizoe, M., Velev, O.A., Srinivasan, S., Electrochim. Acta 40 (1995) 335.
23. Costamagna, P., Srinivasan, S., Quantum jumps in the PEMFC science and
technology from the 1960s to the year 2000 Part I. Fundamental scientific aspects,
J. Power Sources 102 (2001) 242–252.
24. Lipnizki, F., Hausmanns, S., Ten, P.K., Field, R.W. and Laufenberg, G.,
Organophilic pervaporation: prospects and performance, Chem. Eng. J. 73 (1999)
113–129.
25. Zadowzinski, T., Davey, J., Valerio, J., Gottesfeld, S. The water content
dependence of electro-osmotic drag in proton conducting polymer electrolytes,
Electrochim. Acta 40 (1995) 297–302.
47
26. Meier-Haack, J., Rieser, T., Lenk, W., Berwald, S., Lehmann, D., Build-up of
polyelectrolyte multilayer assemblies: a useful tool for controlled modification of
microfiltration and pervaporation membranes. In: Grassie, K., Tenckhoff, E.,
Wegner, G., Haußelt, J., Hanselka, H., Editors, Functional Materials Euromat ’99
vol. 13, Wiley-VCH (2000) 316–322.
27. Soontrapa, K., Srinapawong, N., J. Sci. Res. Chula. Univ. 26 (2) (2001) 59–70.
28. Susai, T., Kaneko, M., Nakatoa, K., Isono, T., Hamada, A., Miyake, Y.,
Optimization of proton exchange membranes and the humidifying conditions to
improve cell performance for polymer electrolyte membranes, Int. J. Hyd. Energy
26 (2001) 631–637.
29. Gierke, T.D., Hsu, W.Y., In: Eisenberg, A., Yeager, H.L., Editors, Perfluorinated
Ionomer Membranes, ACS Symposium Series No. 180, American Chemical
Society, Washington, DC (1982) 283.
30. Okada, T., Xie, G., Gorseth, O., Kjelstrup, S., Nakamura, N., Arimura, T., Effect
of water uptake and relative humidity on Nafion, Electrochem. Acta 43 (1998)
3741–3747.
31. Eikerling, M., Kornyshev, A.A., Stimming, U., J. Phys. Chem. B 101 (1997)
10807.
32. Haubold, H.G., Vad, T., Jungbluth, H., Hiller, P., Nano structure of Nafion: a
SAXS study, Electrochim. Acta 46 (2001) 1559–1563.
33. James, P.J., McMaster, T.J., Newton, J.M., Miles, M.J., In situ rehydration of
perfluorosulfonate ion-exchange membrane studied by AFM, Polymer 41 (2000)
4223–4231.
34. Wasmus, S., Valeriu, A., Mateescu, G.D., Tryk, D.A. and Savinell, R.F.,
Characterization of H
3
PO
4
-equlibriated Nafion 117 membranes using
1
H and
31
P
NMR spectroscopy, J. Membr. Sci. 185 (2000) 78–85.
48
35. Mex, L. and Muller, J., Plasma-polymerized electrolyte membrane for
miniaturized DMFC, Membr. Technol. 115 (1999) 5–9.
36. Finsterwalder, F. and Hambitzer, G., Proton conductive thin films prepared by
plasma polymerization, J. Membr. Sci. 185 (2001) 105–124.
37. Bahar, B., Hobson, A.R., Kolde, J.A., Zuckerbrod, D., Ultrathin integral
composite membrane, US Patent, 5,547,551 (1996).
38. Tazi, B., Savadago, O., New cation exchange membranes based on Nafion,
Silicotungstic acid and thiophene, J. New Mater. Electrochem. Syst., in press (cf.
JMS 185 (2001) 3–27).
39. Sakari, T., Takenaka, H., Wakabayashi, N., Kawami, Y. and Tori, K., Gas
permeation properties of SPE membranes, J. Electrochem. Soc. 132 (1985) 1328.
40. Soczka-Guth, T., et al., International Patent WO99/29763 (1999).
41. Gowariker, V.R., Vishwanathan, N.V. and Sridhar, J., Polym. Sci. New Age
International, New Delhi (1999).
42. Carretta, N., Tricoli, V. and Picchioni, F., Ionomeric membranes based on
partially sulfonated poly(styrene): synthesis, proton conduction and methanol
permeation, J. Membr. Sci. 166 (2000) 189–197.
43. Bashir, H., Linares, A. and Acosta, J.L., Heterogeneous sulfonation of blend
systems based on hydrogenated poly butadiene-styrene block copolymer.
Electrical and structural characterization, Solid State Ionics 139 (2001) 189–197.
44. Buchi, F., Gupta, B., Haas, O. and Scherer, G., Study of radiation grafted FEP-g-
polystyrene membranes as polymer electrolytes in fuel cells, Electrochim. Acta 40
(1995) 345–353.
49
45. Wang, F., Hickner, M., Kim, Y.S., Zawodzinski, T.A. and McGrath, J.E., Direct
polymerization of sulfonated poly(arylene ether sulfone) random (statistical)
copolymers: candidates for new proton exchange membranes, J. Membr. Sci. 197
(2002) 231–242.
46. Poppe, D., Frey, H., Kreuer, K.D., Heinzel, A. and Mulhaupt, R., Carboxylated
and sulfonated poly(arylene-co-arylene sulfone)s: thermostable polyelectrolytes
for fuel cell applications, Macromolecules 35 (2002) 7936–7941.
47. Lafitte, B., Karlsson, L.E. and Jannasch, P., Sulfophenylation of polysulfones for
proton conducting fuel cell membranes, Rapid Macromol. Commun. 23 (2002)
896–900.
48. Guo, X., Fang, J., Watari, T., Tanaka, K., Kita, H. and Okamoto, K., Novel
sulfonated polyimides as polyelectrolytes for fuel cell application. 2. Synthesis
and proton conductivity of polyimides from 9, 9-bis (4-aminophenyl)fluorene-2,7-
disulfonic acid, Macromolecules 35 (2002) 6707–6713.
49. Genies, C., Mercier, R., Sillion, B., Cornet, N., Gebel, G. and Pineri, M., Soluble
sulfonated naphthalenic polyimides as materials for proton exchange membranes,
Polymer 42 (2001) 359–373.
50. Genies, C., Mercier, R. and Sillion, B., et al., Stability study of sulfonated
phthalic and naphthalenic polyimide structures in aqueous medium, Polymer 42
(2001) 5097–5105.
51. Asensio, J.A., Borros, S. and Gomez, R., Proton conducting polymers based on
benzimidazole and sulfonated benzimidazoles, J. Polym. Sci. Part A: Polym.
Chem. 40 (2002) 3703–3710.
52. Bae, J.M., Honma, I., Murata, M., Yamamoto, T., Rikukawa, M. and Ogata,
N.,Properties of selected sulfonated polymers as proton-conducting electrolytes
for polymer electrolyte fuel cells, Solid State Ionics 147 (2002) 189–194.
50
53. Hofmann, M.A., Ambler, C.M. and Maher, A.E., et al., Synthesis of
polyphosphazenes with sulfonimide side groups, Macromolecules 35 (2002)
6490–6493.
54. Carter, R., Wycisk, R., Yoo, H., and Pintauro, P.N., Blended
polyphosphazeneypolyacrylonitrile membranes for direct methanol fuel cells,
Electrochem. Solid State Lett. 5 (2002) A195–A197.
55. Schuster, M., Meyer, W.H. and Wegner, G., et al., Proton mobility in oligomer-
bound proton solvents: imidazole immobilization via flexible spacers, Solid State
Ionics 145 (2001) 85–92.
56. Qinfeng, L., Hjirker, H.A. and Bjerrum, N.J., Phosphoric acid doped
polybenzimidazole membranes, J. Appl. Electrochem. 31 (2001) 773–779.
57. Steiner, P. and Sandor, R., Polybenzimidazole prepreg: improved elevated
temperature properties with autoclave processability, High Perform. Polym. 3
(1991) 139–150.
58. Bouchet, R., Miller, S., Deulot, M. and Sonquet, J.L., A thermodynamic approach
to proton conductance in acid-doped polybenzimidazole, Solid State Ionics 145
(2001) 69–78.
59. Wang, J.J., Savinell, R.F., Wainright, J., Litt, M. and Yu., H., A H
2
/O
2
fuel cell
using acid doped polybenzimidazole as polymer electrolyte, Electrochim. Acta 41
(1996) 193–197.
60. Samms, S.R., Wasmus, S. and Savinell, R.F., Thermal stability of protons
conducting acid doped PBI in simulated fuel cell environments, J. Electrochem.
Soc. 143 (1996), 1225.
61. Bozkurt, A. and Meyer, W.H., Proton conducting blends of poly(4-vinylimidazole)
with phosphoric acid, Solid State Ionics 138 (2001), 259–265.
51
62. Lassegues, J.C., Grondin, J., Hernandez, M. and Maree, B., Proton conducting
polymer blends and hybrid organic inorganic materials, Solid State Ionics 145
(2001) 37–45.
63. Hasiotis, C., Deimede, V. and Kontoyannis, C., New polymer electrolytes based
on blends of sulfonated polysulfones with polybenzimidazole, Electrochim. Acta
46 (2001) 2401–2406.
64. Kerres, J., Ullrich, A., Meier, F. and Haring, T., Synthesis and characterization of
novel acid–base polymer blends for application in membrane fuel cells, Solid
State Ionics 125 (1999) 243–249.
65. Neburchilov, V., Martin, J., Wang, H., Zhang, J., J. power sources 169 (2007)
221-238.
52
Chapter 2
Development of PVDF-PSSA-PMMA composite membranes
for direct methanol fuel cells
2.1 PVDF-PSSA-PMMA membrane preparation
Polymer electrolyte membrane (PEM) is one of the key components that
determine the electronic performance of a PEM fuel cell. Our research group has
developed a method to prepare ion exchange membranes consisting of sulfonated
polystyrene-divinylbenzene polymerized in a poly(vinylidenefluoride) (PVDF) matrix
similar to the one described by Hodgdon and Boyack [1]. This method was further
modified to fabricate semi-interpenetrating polymer network (sIPN) composite
membranes consisting of poly(vinylidenefluoride) (PVDF), poly(styrenesuflonic acid)
(PSSA) and poly(methyl methacrylate) (PMMA). The present research work
demonstrated improved performance characteristics of these composite membranes in
direct methanol fuel cells. These membranes exhibit lower methanol crossover rates
compared with Nafion
®
-117, display comparable proton conductivity at ambient
temperatures, and their performance has been demonstrated in direct methanol fuel cell
cells being developed for portable power systems.
53
2.1.1 PVDF precursor study
The first step involved in the process of membrane development is the preparation
of a PVDF precursor that be used as the inert polymer matrix host. The PVDF powder is
commercially available. The polymer is obtained by free radical polymerization of 1,1-
difluoroethylene (vinylidene fluoride) as a high molecular weight, partially crystalline
material. Emulsion, suspension and solution polymerization can be used for the PVDF
synthesis and chain transfer reagents are used to moderate the molecular weights and
their distribution. The polymerization condition has strong influence on PVDF properties
such as crystallinity and mechanical strength. The spatial symmetrical disposition of the
hydrogen and fluorine atoms on the polymer chain produces special polarity property that
affects dielectric properties, solubility and crystal morphology. PVDF is melt processed
by either molding or extrusion methods whereas porous membranes can be prepared from
solution casting using polar solvents.
The physical stability, chemical resistance and methanol rejecting properties of
PVDF make it an attractive precursor material. A PVDF precursor was achieved either by
hot pressing PVDF powder in a mold at 160-180
o
C or by casting PVDF membranes
from solutions of polar solvents, such as acetone, DMSO or THF, and then subsequently
hot pressing to form PVDF sheets. Films of various thicknesses (125 to 350 μm) could be
54
reproduced by this method. The precursor polymer matrix controls the membrane
morphology by determining membrane domain size and the distribution of the
impregnated polymer. Surface morphology is a very important factor in selecting
membranes for MEA fabrication and fuel cell testing. Membranes with irregular surface
morphology will lead to poor interfacial contact at the electrolyte-electrode interface, thus
resulting in a high contact resistance and reduced utilization of the catalyst hence
adversely affecting the overall electrical performance.
The most profound factor influencing the morphology of the membrane after
impregnation is the precursor crystallinity. The crystalline forms of PVDF include
lamellar and spherulitic structures, whose size and distribution strongly affect precursor
morphology and IPN properties. The precursor crystallinity is also a crucial factor in
determining physical properties such as toughness, mechanical strength and the high
impact resistance of the polymer.
PVDF has four crystal forms with different chain formations: α, β, γ and δ. The
polymorphs are decided by different processing conditions such as pressure, electric
fields, controlled melt crystallization and seeding crystallization. The α form is
thermodynamically stable and usually appears during normal melt processing. The β
form has an all trans chain conformation and usually appears during thermal extrusion
55
process or compression molding of melted polymers. Mechanical deformation of PVDF
with the β form results in highly oriented films with strong mechanical strength. It can be
also obtained through mechanical drawing of the α form membranes or slow heating of a
DMF solution of PVDF and then annealing at a certain temperature [2-4]. The γ form is
obtained from annealing of high molecular weight PVDF while the δ form is obtained by
a distortion of the other forms under a high electrical field.
The study of the crystallization of PVDF is very important to find a suitable
precursor in IPN synthesis. The degree and order of crystalline domains have great
influence on the distribution of the impregnated polymer thus controlling the whole
membrane morphology. PVDF obtained from the melt results mainly in the α form.
Mechanical stretching is needed to induce the formation of the β form. Studies show that
rapid quenching of the melt at low temperature result primarily in the β form [5]. The use
of different solvents to cast PVDF films can also induce changes of polymer crystallinity.
2.1.2 Synthesize PVDF-PSSA using IPN methodology
As discussed before, Nafion
®
is not a very effective membrane in DMFC. There
has been some interest in PSSA grafted membranes since 1960s. Typical graft
copolymers consist of long segment of one monomer with branches consisting of long
56
segments of a second monomer. These PSSA grafted membranes have been used
extensively on some separation processes. However, these grafted membranes can swell
by 90 vol% in water, two times the value of Nafion117. The increase in volume and
dimension of this membrane resulted in poor physical properties. Flint studied the
incorporation of polystyrene grafted into a PVDF matrix and found that subsequent
sulfonation and hydrolysis create a two-phase hydrophobic, hydrophilic polymer [6].
MEA prepared from these membranes had low open circuit voltage (OCV) due to poor
interfacial contact at the electrode-electrolyte surface. This was caused by phase
separation of the grafted membrane and is a significant concern in catalyst utilization in
application in fuel cells. Some analysis of PVDF-g-PSSA system showed that the
polystyrene components are incompatible with the PVDF matrix. Phase separated
domains were found within the amorphous regions [7]. With increasing PSSA
components, bulk like water content will increase. This raises concerns regarding
membrane stability. Furthermore, the agglomerated pockets of PSSA grafts would cause
high water permeability, which will inhibit electrical performance.
Our research group has used IPN methodology [8-9] to prepare novel polymer
electrolyte membranes for direct methanol fuel cell applications. An interpenetrating
polymer network is a blend of two different polymer networks without covalent bonds
57
connecting them. IPNs have been used to modify the physical properties of many kinds
of plastics. The two individual polymer network form continuous phases and the phase of
the first formed network control the physical properties of the whole IPN.
One crosslinked polymer precursor is reacted with the other monomer and
crosslinking agents producing a “fully” interpenetrating sequential-IPN. If one of the
polymer networks is not crosslinked, the system will be called a “semi” interpenetrating
sequential-IPN. Our PVDF-PSSA-PMMA system belongs to the latter category. Most
semi-IPNs utilize a crosslinked polymer matrix with a linear additive polymer. Our
system uses the reverse process whereby the polymer additive is crosslinked within the
polymer matrix. So the crystalline regions within the PVDF matrix act as pseudo-
crosslinks controlling domain size and subsequent distribution of impregnated styrene.
The processing conditions, the PVDF molecular weight and its distribution will affect the
degree of crystallinity and styrene distribution. So the IPN methodology ensures uniform
distribution of PSSA throughout the PVDF matrix.
Our research group has used the IPN methodology to prepare PVDF-PSSA
membranes and resulting MEAs show excellent electrical performance and very low
methanol crossover compared to Nafion117 [9]. As discussed before, styrene grafted
membranes often show irregular surface morphology and the uneven polystyrene
58
distribution can lead to poor interfacial contact at the electrolyte-electrode surface. Phase
separation in membranes can cause severe membrane wrinkling and brittleness during the
process of styrene polymerization. The development of PVDF-PSSA membranes in our
group has greatly improved the membrane surface morphology and physical properties.
However, PVDF-PSSA membranes still show brittleness at dry state, which makes the
MEA fabrication difficult. Thus the addition of MMA component to the PVDF-PSSA
semi-IPN has been anticipated to alleviate the problem.
2.1.3 Introduction of PMMA into PVDF-PSSA membranes
It is well known that the presence of certain polymeric species, such as suitably
chosen block polymers or graft polymers, can make immiscible polymer systems
compatible by reducing the interfacial tension between the two. The segments of these
copolymers can be chemically identical with those polymers on the respective phases, or
it can be miscible with one of the phases. Each block is partially or totally miscible with
one of the blend components and the copolymer is located at the interfaces between the
immiscible phases, increasing the adhesion between them.
It has been found that the inclusion of 20 wt% PMMA to the PVDF melt resulted
in films with maximum β phase content. The inclusion of PMMA greatly reduced the rate
59
and temperature of crystallization within the polymer systems. This is mainly due to the
increase in the glass transition temperature and is recommended as one of the best ways
to enhance the efficiency of quenching resulting in elongation enhancement and better
membrane physical property.
Some research work showed that the domain size of PVDF/Polystyrene blends
decreased with increasing PMMA content. This was because the relatively low
interaction energy between polystyrene and PMMA. So the PMMA is an effective
compatibilizer that reduces interfacial tension between the two relatively immiscible
polymer systems.
PVDF is among the few partially crystalline polymers that exhibit thermodynamic
compatibility with other polymers such as methacrylic resins [10]. The addition of
compatibilizer such as PMMA was necessary to avoid macroscopic phase separation
between the PVDF and Polystyrene phases. Thus PMMA is widely considered to be
compatible with PVDF due to the favorable hydrogen bonding between the two polymers.
The inclusion of PMMA into PVDF-Polystyrene membrane results in a more flexible
membrane, which is further sulfonated into a promising PVDF-PSSA-PMMA membrane
for use in direct methanol fuel cells. PMMA component helped to reduced interfacial
tension between the immiscible PVDF and polystyrene, insured fine distribution of PSSA
60
in membranes after sulfonation. PVDF-PSSA-PMMA membranes show great
improvement of the mechanical properties as well as electrical performance.
2.2 Experimental
The procedure to make the PVDF-PSSA-PMMA membrane is described below:
a) Making 10 mil precursors
1. Weigh 12 g PVDF powder purchased from Aldrich and sandwich it between two
plates.
2. Put it under press machine at a force of 25,000 lbs.
3. Heat at 240 °C for one hour.
4. Heat at 240 °C under force for 15 minutes.
5. Cool to 25 °C under force in 45 minutes.
6. Open the press machine and obtain a PVDF precursor membrane (10 mils).
b) Impregnation to get 15% PVDF-PS-PMMA membranes
1. Weigh the PVDF precursor membrane (12 g, 10 mils).
2. Put it into acetone bath, at 35 °C for at least 12 hours.
3. Weigh 0.5 g AIBN, 56 mL divinylbenzene (DVB), 63 mL methyl methacrylate
(MMA) and 450 mL styrene monomer and mix them uniformly.
61
4. Get the precursor out of acetone bath and put it into the styrene bath made in step
3 for 1.5 hours.
5. Get the precursor out of the styrene bath and sandwich it between two aluminum
foils and two titanium plates.
6. Put it into press machine, 60 °C for 1 hour and 90 °C for 1 hour, using low force
of 1000 lbs.
7. Weigh the PVDF-PS-PMMA membrane and calculate the PS uptake.
8. Repeat step 1 to step 7.
9. Put the membrane under vacuum at 90 °C for 12 hrs.
10. Weigh the membrane and calculate final PS uptake.
c) Sulfonation process.
1. Mix chlorosulfonic acid and chloroform together, volume ratio: 30:70.
2. Place the PVDF-PSSA-PMMA membrane in, keep in the chlorosulfonic acid
solution at 40-50 °C for 3days.
3. Cool to R. T. and keep in chlorosulfonic acid solution for 2 days.
d) Hydrolysis process
1. Put the membrane into 500 mL deionized water.
2. Heat it up slowly to 50-60 °C and keep it in water for 3 days.
62
3. Heat it up to 90 °C and keep it for 2 days.
The whole fabrication path is shown in Scheme 2.1.
F
2
CCH
2
m
PVDF matrix
Styrene/MMA
DVB/AIBN
1.ClSO
3
H/CHCl
3
2. H
2
O
F
2
CCH
2
m
x
y
COOCH
3
PVDF-PS-PMMA
F
2
CCH
2
m
x
y
COOCH
3
SO
3
H
PVDF-PSSA-PMMA
Scheme 2.1 Synthetic scheme of PVDF-PSSA-PMMA membrane
Membrane electrode assemblies (MEAs) (23 mils) were fabricated by hot
pressing the PVDF-PSSA-PMMA membrane sample in its hydrated state with catalyzed
porous carbon electrodes. The prepared MEA with 6 or 25 cm
2
active area were used for
fuel cell testing.
The overall cost to make the PVDF-PSSA-PMMA membrane is very low
compared to Nafion ($700/m
2
). The partially fluorinated polymer PVDF was chosen as
the precursor membrane because it is known as a low-cost material (Kynar membranes)
and is used as films for water/alcohol separation purposes. Prakash et al. have carried out
significant advancement in the area of membrane development using the semi-IPN
technology [8, 9].
63
2.3 Membrane characterization
2.3.1 Thermal analysis
Thermal analysis was used to study the properties of the membranes relating to
the degree of crystallinity, polymer miscibility as well as membrane stability.
Differential Scanning Calorimetry (DSC) is one method used to understand
thermal transition properties related with polymer mobility (T
g
) and crystallinity (T
m
).
DSC study was conducted on both commercially available Kynar PVDF membranes and
a compression molded PVDF precursor prepared in our group. The PVDF precursor
membrane prepared at USC has a melt temperature of 160
o
C with a corresponding
endothermic heat capacity about 45.05 J/g. This is comparable with the reported value of
a phase crystals with a degree of crystallinity about 43% [11]. DSC analysis on Kynar
460 and 740 membranes was also conducted and the results are summarized in table 2.1.
As shown in Table 2.1, molecular weight has a strong influence on the degree of
crystallinity of PVDF precursors. Higher Molecular weight results in lower degree of
crystallinity in PVDF precursors because polymer chain entanglements will affect the
crystallization process. The USC - PVDF membranes have a lower degree of crystallinity
and will have better membrane morphology after styrene impregnation.
64
Table 2.1 DSC analysis of the various PVDF precursors
PVDF
precursor
Processing
method
M
w
Tm
o
C Crystallinity
USC (Aldrich) Compression
molding
540, 000 160 43%
Elf Atochem
Kynar460
Compression
molding
540, 000 160 46%
Weslake
Kynar 740
Thermal
extrusion
370, 000 166 55%
Elf Atochem
Kynar740
Thermal
extrusion
370, 000 166 55%
Results from DSC analysis show that PVDF-PSSA membranes have good
polymer miscibility. The PVDF-PS membrane sample has a melting endotherm at 160
o
C,
heat capacity of 46.4 J/g and a degree of crystallinity of 43%, which are similar to the
values of PVDF precursor membrane. This indicates that the incorporation of
polystyrene/divinylbenzene doesn’t interfere with the crystallization process in PVDF
itself. The T
g
of PVDF in PVDF-PS membrane is about -40.9
o
C, which is also similar to
that value of PVDF precursor membrane. This indicates the incorporation of polystyrene
doesn’t affect the T
g
of PVDF matrix.
The T
g
of PVDF in the PVDF-PSSA membrane is -42
o
C, suggesting a lack of
influence from styrenesulfonic acid on the segmental motion of PVDF matrix. Based on
these DSC data, it can be concluded that the styrene polymerizes mainly in the
65
amorphous region in the precursor matrix. Thus the thermal transitions of PVDF will not
be affected by the incorporation of polystyrene in the amorphous regions.
TGA analysis has been used to study the stability and miscibility of styrene
grafted PEMs. TGA analysis of PVDF-PS membranes shows membrane decomposition
starts at 325
o
C and a sharp drop in membrane weight occurs at 360
o
C. This indicates
improved miscibility of PVDF-PS using IPN technology compared to styrene grafted
membranes. TGA analysis of PVDF-PSSA shows initial mass loss of 20% at 75-100
o
C,
indicating membrane water loss. Desulfonation happens between 200
o
C and 300
o
C,
while membrane degradation takes place around 350
o
C. TGA analysis suggests that the
incorporation of PSSA into PVDF using IPN methodology results in uniform PEMs. The
lack of multi-step degradative mechanism indicates improved polymer miscibility in
PVDF-PSSA membrane.
2.3.2 Energy dispersive X-ray analysis (EDAX)
Energy dispersive X-ray Analysis is used to measure the homogeneity of
polystyrenesulfonic acid distribution along the cross-section of a PVDF-PSSA membrane.
The detector of EDAX can be tuned to measure certain elemental X-ray emissions such
as carbon, fluorine and sulfur. Sulfur X-rays come from the emissions from
66
styrenesulfonic groups on polystyrene backbone. So the measurement of sulfur content
can reveal polystyrene sulfonic acid distribution in the membrane.
During the impregnation process, the membrane swelling time in acetone bath is
important for the distribution of polystyrene in the subsequent hot-press polymerization.
Lack of ample swelling time will cause styrene to polymerize at the edges of the PVDF
host without sufficient diffusion into the bulk. EDAX analysis shows that with swelling
time of 12 hours at 35
o
C, uniform styrene adsorption was obtained and the homogeneous
polystyrene distribution can be achieved in the next step. EDAX analysis is an effective
tool used to characterize PSSA distribution in PVDF-PSSA membranes. It helps to screen
membranes for uniform sulfonic acid distribution.
2.3.3 Membrane water uptake
Membrane water content can affect the fuel cell performance in several different
ways. Sufficient membrane water content is needed to maintain proton conductivity of
the membrane to sustain the cathode oxygen reduction. However, high water
permeability will result in cathode flooding. High methanol permeability often comes
together with high water permeability, resulting in mixed cathode potentials at the
cathode that reduce overall fuel cell performance.
67
The sulfonic acid content of the membrane will determine membrane water
management properties. PSSA grafted membranes with high degree of grafting will prove
high proton conductivity at the expense of mechanical stability due to unrestricted
membrane swelling. Crosslinking will partially solve the problem but the low water
content will result in low proton conductivity. The sulfonic acid groups are distributed
uniformly through the membrane fabricated using IPN methodology, resulting in uniform
water sorption within the stable PVDF matrix. The water uptake of a membrane is
determined by the following formula:
The water uptake test is carried out this way:
1. Dry a membrane sample and get the weight.
2. Put the membrane sample in water at 90 °C for 6 hrs.
3. Take the sample out of water and wipe off surface water and weight it quickly.
4. Calculate water uptake according to the formula above.
Water uptake =
Weight of dry membrane
x 100%
Weight of wet membrane- Weight of dry membrane
68
water uptake vs. PSSA content
0
5
10
15
20
25
30
35
40
45
0 5 10 15 20 25 30
PSSA Content %
water uptake %
10% DVB
5% DVB
Figure 2.1 Membrane water content of PVDF-PSSA membrane vs. PSSA uptakes
The influence of crosslinking density and the PSSA uptake on water content of a
membrane has been studied. It is interesting to note that the water content increases
linearly with PSSA uptake (Figure 2.1). When the PSSA uptake is over 17% with 10%
DVB crosslinking, the water content is even higher than of Nafion117 (30%).
The data in Figure 2.1 suggest that crosslinking density of polystyrene does not
have a significant impact on membrane water content. At the same level of PSSA uptake,
membranes with 5% DVB crosslinking have only slightly higher water content than
water content of membranes with 10% DVB crosslinking. So the water content of
69
membranes with 5% DVB crosslinking and 15% PSSA uptake is comparable to
Nafion117. Since Nafion is reported to exhibit adequate proton conductivity with
sufficient water content, the proton conductivity of PVDF-PSSA with certain water
content should be similar.
When a current scan experiment is run on an actual direct methanol fuel cell,
methanol will be transported at the cathode in addition to the permeation of water from
anode side (Figure 2.2). The corresponding water produced at the cathode (contained by
oxygen reduction as well as crossover methanol oxidation) is collected since the
beginning of the scan experiment at different experimental conditions.
It was interesting to note that the water content at the cathode side with oxygen is
higher than that with air as oxidant, which is predictable since the reduction kinetics is
faster with oxygen. When the methanol concentration and fuel cell running temperature
increase, the amount of water collected at the cathode will increase correspondingly.
With fuel cell running at 90
o
C, the water collected at the cathode increased significantly
with either oxygen or air as the cathode oxidant. This information will be useful in
solving cathode flooding problems.
70
0
2
4
6
8
10
12
14
16
12 34 56
run
water (g)
air
oxygen
Figure 2.2 Water collected at the cathode during a scan current experiment
1) 0.25 M methanol, 23
o
C; 2) 0.5 M methanol, 55
o
C; 3)0.5 M methanol, 60
o
C;
4) 1 M methanol, 55
o
C; 5) 1 M methanol, 60
o
C; 6)1 M methanol, 90
o
C
2.3.4 Proton conductivity measurement
Our research group has developed a 4-probe method to measure the specific
proton conductivity of membranes. The membranes have to be fully hydrated before
proceeding to proton conductivity measurement. Nafion117 was reported to have a
specific conductivity 70-75 mS/cm when fully hydrated at room temperature. Using a
similar method, PVDF-PSSA membranes have proton conductivity values range from 30
mS/cm to 90 mS/cm depending on PSSA uptake and crosslinking densities (Table 2.2).
71
Table 2.2 Specific proton conductivity values for Nafion117 and PVDF-PSSA
membranes
Membrane Nafion 97-33 98-35 98-36 01-58 01-60 01-72 01-82
Conductivity
(ms/cm)
73 60 49 30 57 88 73 75
It was observed that the membrane conductivity increases as the PSSA uptake
increases while the degree of polystyrene crosslinking didn’t have a significant impact on
the conductivity. Influence of membrane water content on proton conductivity was also
investigated. It was also noted that the conductivity of a dry membrane is very low. The
conductivity will increase linearly with the membrane water content and will reach a
steady state when the membrane is fully hydrated.
The Nafion 117 contains some sulfonic acid sodium salt content. It has to be
treated with sulfuric acid before the conductivity measurement. An interesting
phenomenon has been observed when Nafion membrane was put into 0.01 M NaCl
solution. The conductivity of Nafion will decrease to 0.04 mS/cm when it was left in
NaCl solution for 10 hrs (Figure 2.3).
72
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0 10 2030 4050 6070 80
Time (hrs)
σ
(s/cm)
Figure 2.3 NaCl sodium exchange effect on the conductivity of Nafion117
We also built a humidifier chamber to observe the relation between proton
conductivity and temperature. The membrane sample in the humidifier chamber will
establish equilibrium with water vapor at different temperatures before proton
conductivity was measured.
Using this method, proton conductivity of different PVDF-PSSA-PMMA
membrane samples were compared to Nafion117 at different temperatures. It been
observed from Figure 2.4, the conductivity of PVDF-PSSA-PMMA membranes is
comparable to Nafion at temperature ranging from 25-85
o
C.
73
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 10 2030 4050 6070 8090 100
Temp (
o
C)
σ
(s/cm)
Nafion117
MMA072806
MMA112408
Figure 2.4 Proton conductivity vs. temperature under humidified conditions
2.3.5 Ion Exchange Capacity Test
Ion exchange capacity test is used to determine the content of acidic proton in a
membrane. Equivalent weight is calculated as a result to show how many grams of
material is needed to have one mole of acidic proton. The procedure is showed below:
1. Weigh a dry membrane sample and put it into deionized water at 90 °C for 6
hours.
2. Put the sample into 3 M aqueous KBr solution, at 70 °C for 6 hours.
74
3. The resultant solution in step 2 is titrated using 0.005 M Na
2
CO
3
solution to reach
equivalent point.
4. Using the result, the equivalent weight is calculated.
Table 2.3 Equivalent weight of various membranes
Membrane Thickness (mil) PS% Eq. Wt.
00-61 10 23.8 550
00-62 10 23.1 683
00-72 10 17.2 1078
MMA01-1 10 14.1 1507
MMA01-2 10 16.6 1255
MMA01-3 10 17.1 1022
ky460.3 16 12.2 2736
ky460.4 16 11.9 2213
Table 2.3 shows that our PVDF-PSSA-PMMA membranes, with 14-17% PSSA
uptake, have similar ion exchange capacity compared to Nafion117 (eq. weight = 1100).
The PVDF-PSSA membrane 00-72 has a similar eq. weight compared to Nafion117,
while the eq. weights of the other two PVDF-PSSA membranes 00-61 and 00-62 are only
75
half of that of Nafion. This is not surprising since the polystyrene uptake of these two
membranes is as high as 23%. Some commercially available PVDF precursors (Kynar
460) are also impregnated with polystyrene and sulfonated to form PVDF-PSSA
membranes. The equivalent weights of these two Kynar 460 membranes are so high that
they display very low proton conductivities. This is due to the high membrane thickness
as well as the precursor physical properties, which caused the unfavorable polystyrene
distribution.
2.3.6 Methanol permeability test using a GC tracking method
The methanol-repellent property of PVDF precursor combined with the semi-IPN
method to avoid the distinct two phase character of PSSA grafted membranes helps to
reduce the methanol permeability. The inclusion of compatibilizer (or called plasticizer)
PMMA insured the uniform distribution of PSSA, further reducing methanol crossover.
A Gas Chromatograph (GC) method was established to monitor the methanol
crossover. The method was designed to monitor methanol diffusion through membrane
samples along a certain concentration gradient. Different kinds of membranes were
sandwiched between methanol solution and de-ionized water. The water samples were
monitored using GC to analyze the methanol diffusion through the membrane.
76
As shown in Figure 2.5, two 250 mL volume vessels were connected together
with a membrane clamped in between them. The inner diameter of the joint is 2.4 cm
(area 4.52 cm
2
). 250 mL of a 3.0 M methanol solution was added to the first vessel and
250 mL of de-ionized water was added to the second vessel. Both solutions were stirred
to have the uniform concentration.
Figure 2.5 Apparatus to evaluate methanol permeability
Over certain periods of time, water samples were taken from the de-ionized water
side and injected into a Varian 3300 GC equipped with a carbowax column to analyze the
methanol diffused through the membrane. An internal standard solution of isopropanol
was used as a reference to calculate the concentration of each water sample. The
calculated methanol concentration was plotted vs. time to obtain the mass transfer
coefficient from the slope of the plot (Figure 2.6).
PEM
3 M methanol
De-ionized water
77
y = 2.37E-10x - 2.39E-07
0.00E+00
2.00E-07
4.00E-07
6.00E-07
8.00E-07
1.00E-06
1.20E-06
1.40E-06
1.60E-06
1.80E-06
2.00E-06
0 2000 4000 6000 8000 10000
time (s)
conc. (mol/mL)
Figure 2.6 Methanol concentrations of GC samples vs. time
The methanol diffusion coefficient was determined according to the following
equation:
m = slope = C*/ γ = DA / (tV) (2.1)
C* = concentration of the standard methanol solution
γ = mass transfer coefficient
A = area of joint
t = membrane thickness
V = volume of the vessel with de-ionized water
78
Methanol Permeability
3 M Methanol Standard, Room Temp.
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0 20 40 60 80 100 120 140 160
Time (minutes)
Methanol (Moles/liter)
Nafion-117
MMA01-3
MMA01-1
MMA02-6
MMA01-1, PS14.6% MMA 0.5%
MMA02-6, PS16.5% MMA 2.5%
MMA01-3, PS17.1% MMA 12.5%
Figure 2.7 Methanol cross-over of Nafion117 and USC membranes using GC tracking
method
From Figure 2.7, it was observed that the methanol diffusion rates of PVDF-
PSSA-PMMA membranes are almost one seventh of Nafion117. As PSSA uptake
increased, the methanol crossover rate increased also. This is because the membrane
water content increases linearly with PSSA uptake and will provide a more aqueous
environment for methanol crossover. It was found that methanol would be completely
impermeable through a PVDF precursor before impregnation. It can be concluded that
methanol permeability was solely attributed to the PSSA content of the membrane.
79
It was interesting to note that some PVDF-PSSA-PMMA membranes had similar
water uptake and proton conductivity compared to Nafion117 while they had a significant
reduction in methanol crossover rate. This phenomenon suggests that proton transport
and methanol crossover may depend on different issues.
Aqueous channels are the proton transport path in PEMs. It is believed that the
formation of aqueous channels depends on the agglomeration of ionic clusters formed
during membrane hydration. The water content and the state of the bound water within
the membrane combined to determine proton transport. Kreur et al. found that proton
migration was assisted by the translational dynamics of bigger species referred to as the
vehicle mechanism because protons diffuse together with vehicular species such as water,
in the form of H
3
O
+
[12-13]. The counterdiffusion of unprotonated water allows for the
net transport of protons. Another mechanism called Grotthuss mechanism proposed that
the protons transfer from one vehicle to another takes place within hydrogen bonds.
Uninterrupted trajectory for proton migration was formed when proton environment
reorganized [14].
Zawodzinski used H
1
NMR to calculate H
+
and H
2
O diffusion coefficients in
Nafion117 at different hydration state [15]. It was concluded that H
+
and H
2
O will diffuse
according to the vehicle mechanism driven by a concentration gradient, when the water
80
content of membrane is low. When the water content of membrane is high, the increasing
of more bulk-like water will allow for effective reorganization of the solvent environment.
In this case, the proton diffusion will follow the Grotthus mechanism.
It was reported that significant amounts of water within non-porous PVDF-g-
PSSA membrane are confined to the PVDF backbone [16]. There were about 20 water
molecules per sulfonate group in the non-porous PVDF precursor, while the sulfonate
group in the porous PVDF precursor had about 60 water molecules. Increasing the degree
of grafting ultimately resulted in two phase morphology with more bulk-like water.
Similar results were observed in Nafion, which would contain more bulk-like water with
decreasing equivalent weight.
The transport mechanism is quite unique compared to Nafion and PSSA grafted
membranes. With the IPN methodology, the PVDF-PSSA-PMMA membranes will
reduce water and methanol permeability because of the reduced formation of aqueous
channels inside the membranes. The non-porous PVDF precursor will confine a
significant amount of water to the PVDF backbone, which further explains why the
PVDF-PSSA-PMMA membranes had reduced methanol crossover with an equivalent
weight similar to Nafion117.
81
2.3.7 Methanol permeability test using CO
2
analyzer
Another method to measure methanol crossover is using the use of CO
2
analyzer
to measure the CO
2
volume percent in the cathode exit stream. The CO
2
volume percent
generated from methanol crossover was then converted to current density according to
the following equations:
ICO
2
= nFe*1000/A = mA/cm
2
(2.2)
n = Moles of CO
2
, F = 96487 C, e = 6 e
-
, A = the area of carbon paper in cm
2
n = PV / RT (2.3)
P = 1 atm, V = flow rate of CO
2
in the cathode exit stream,
R= 0.0821 L.atm.K
-1
, T = 298 K
In DMFC testing, methanol will permeate through the membrane and will be
oxidized at the cathode to produce CO
2
and a mixed potential. This mixed potential can
be converted to methanol
crossover current density compared to the cell operating current
density to understand how much the methanol crossover costs the fuel cell performance.
82
Nafion03-4, 1 M methanol
0.00
20.00
40.00
60.00
80.00
100.00
120.00
0 100 200 300 400 500 600 700
I (mA/cm
2
)
Crossover current density (mA/cm
2
)
25 ℃
55 ℃
60 ℃
90 ℃
Figure 2.8 Methanol cross-over of Nafion117 using the CO
2
analyzer tracking method.
During DMFC testing, the methanol crossover of Nafion117 at different operating
temperatures was measured using CO
2
analyzer. As the cell operating temperature
increased, methanol crossover also increased due to the increased water content in the
membrane (Figure 2.8). When temperature rose from 25
o
C to 90
o
C, at the operating
current density of 100 mA/cm
2
, the crossover current density increased from 24.5
mA/cm
2
to 91.4 mA/cm
2
. When the operating current density increased, the crossover
current density decreased because of greater consumption of methanol at the anode side
due to the faster anode oxidation kinetics.
83
methanol crossover, 25
o
C, 1 M methanol
0.00
5.00
10.00
15.00
20.00
25.00
30.00
0 50 100 150 200 250
I (mA/cm
2
)
crossover current density (mA/cm
2
)
MMA01-3
MMA01-1
MMA02-6
Nafion03-4
00-72b
Figure 2.9 Methanol cross-over of Nafion117 and USC membranes using the CO
2
analyzer tracking method at 25
o
C
methanol crossover, 55
o
C, 1 M methanol
0
10
20
30
40
50
60
0 50 100 150 200 250 300 350 400 450
I(mA/cm
2
)
crossover current density (mA/cm
2
)
MMA01-3
MMA01-1
MMA02-6
Nafion03-4
00-72b
Figure 2.10 Methanol cross-over of Nafion117 and USC membranes using the CO
2
analyzer tracking method at 55
o
C
84
methanol crossover, 60
o
C, 1 M methanol
0
10
20
30
40
50
60
70
0 50 100 150 200 250 300 350 400 450
I (mA/cm
2
)
crossover current density (mA/cm
2
)
MMA01-3
MMA01-1
MMA02-6
Nafion03-4
00-72b
Figure 2.11 Methanol cross-over of Nafion117 and USC membranes using the CO
2
analyzer tracking method at 60
o
C
methanol crossover, 90
o
C, 1 M methanol
0
20
40
60
80
100
120
0 100 200 300 400 500 600 700
I (mA/cm
2
)
crossover current density (mA/cm
2
)
MMA01-3
MMA01-1
MMA02-6
Nafion03-4
00-72b
Figure 2.12 Methanol cross-over of Nafion117 and USC membranes using the CO
2
analyzer tracking method at 90
o
C
85
Using this method, methanol cross-over of Nafion117 was compared to different
USC membranes at different operating temperatures. Nafion 03-4 is a MEA prepared
from a Nafion117 film. The 00-72b used a PVDF-PSSA film with a PSSA uptake of 16%.
MMA01-1, MMA01-3 and MMA02-6 are all PVDF-PSSA-PMMA membranes with
PSSA uptake of 14.6%, 17.1% and 16.5%, respectively. The results are shown in Figure
2.9-2.12. At all these different operating temperatures, all the MEAs containing USC
made membranes had a significantly lower methanol crossover compared to Nafion117
because of the IPN morphology discussed before. PVDF-PSSA-PMMA membranes had
the best result in reducing methanol crossover because the compatibilizer PMMA
reduced the phase separation and insured more uniform PSSA distribution.
2.4 Electrical performance of PVDF-PSSA-PMMA membranes
2.4.1 Membrane electrode assembly fabrication
Membrane electrode assembly is the key part of a PEM fuel cell. The fabrication
methods generally focus on reduced catalyst loading and strong binding force between
catalyst layer and membrane surface. At the membrane electrode interface, electrical
conductivity, proton conductivity as well as facile catalytic activity are necessary to
assure adequate fuel cell performance.
86
The catalysts used for anode and cathode in PEM fuel cell have platinum black
with normally high surface areas. However, in a DMFC fuel cell, incomplete oxidation of
methanol at the anode side will produce CO, which will bind to platinum strongly and
block the active catalyst sites. So far, the solution has been to use binary or ternary
catalysts based on platinum, all containing ruthenium as the activity promoting
component. According to the bifunctional mechanism, the CO poisoned platinum can be
reactivated via a surface reaction between CO and O-type species associated with
ruthenium to yield CO
2
. According to the ligand model, the change in Pt electronic
properties induced by Ru makes Pt more susceptible for OH adsorption or even for
dissociative adsorption of methanol. Thus Pt-Ru is widely used as the anode catalyst in
DMFC applications.
Early development of Nafion based MEAs for H
2
/O
2
fuel cells involved hot
pressing a mixture of Pt powder and PTFE on both sides of MEA [17]. Nafion-H ionomer
was then introduced into catalyst to replace PTFE, resulting in significant improvements
in MEA fabrication and enhanced fuel cell performance with reduced Pt loading [18].
Early work in our research group has developed methods to fabricate MEA
involving hot-pressing of Teflon-impregnated porous carbon electrodes to a catalyzed
membrane at 112
o
C with a 1000-3000 psi pressure for 15 minutes. The preparation of
87
catalyzed substrates involved preparing an ink consisting of catalyst (180 mg of Pt
powder or 140 mg of Pt-Ru), Nafion-H ionomer (720 mg of a 5% wt solution dispersed
in lower alcohols) and an aqueous solution of PTFE (405 mg). The catalyst ink was then
painted onto teflon-coated carbon paper and membrane surface to prepare 1” x 1” (6.45
cm
2
) and 2” x 2” (25 cm
2
) MEAs. Initial testing of these membranes showed promising
electrical performance, which motivated our interest in modifying MEA fabrication
procedure.
The two most important factors in MEA fabrication are processing temperature
and pressing force used. The pressing force has a limit since the electrode paper can only
sustain a predetermined pressure. A suitable pressing force helps the binding between
carbon electrode and the membrane. It was also found that increasing hot-press
temperatures would improve interfacial bonding at the catalyst-membrane and electrode-
electrolyte interfaces.
Another factor in MEA fabrication is the use of a co-solvent capable of improving
the melt flow characteristics of the membrane. PVDF is known to swell and even
dissolve in a number of polar solvents, such as dimethyl acetamide (DMA). DMA is a
solvent used in solution-cast PVDF films and is highly soluble in water. Thus DMA can
be easily added into catalyst ink mixture.
88
Table 2.4 Anode and cathode catalyst ink composition I
Anode Coat Cathode Coat
0.2 g Pt-Ru 0.2 g Pt
0.2 g Nafion 0.2 g Nafion
0.8 g H
2
O 0.8 g H
2
O
Based on discussion above, we modified the MEA fabrication method to prepare
PVDF-PSSA-PMMA MEAs for testing. The components of a typical catalyst ink are
shown in Table 2.4 and 2.5.
Table 2.5 Anode and cathode catalyst ink composition II
Anode Coat Cathode Coat
0.2 g Pt-Ru 0.2 g Pt
0.2 g Nafion 0.2 g Nafion
0.8 g H
2
O 0.8 g H
2
O
0.2 g DMA
89
The membrane was sanded to roughen the surface to improve contact between
catalyst ink and the membrane. Prepared catalyst ink was applied to the electrode papers
and membrane surface. The two carbon papers painted with catalyst ink and the catalyst-
coated membrane is hot-pressed at 180
o
C and 2000 psi for 20 minutes. The fabricated
MEAs were assembled into the hardware and evaluated for electrical performance.
Table 2.6 MEA fabrication pressure vs. MEA properties
Membrane PSSA Water MEA MEA MEA Cell Voltage Cross-Over
uptake uptake pressure thickness resistance at 100 mA/cm
2
at 100 mA/cm
2
lbs mils 55
o
C, 0.50 M 55
o
C, 0.50 M
00-02 15% 30% 2000 23 N/A 0.407 V 11.6 mA/cm
2
00-27A 16% 33% 1500 25 N/A 0.401 V 11.6 mA/cm
2
00-27B 16% 33% 1000 25 15 mOhms 0.436 V 8.3 mA/cm
2
00-27C 16% 33% 500 26 14 mOhms 0.454 V 10.4 mA/cm
2
00-40 15.60% 35% 500 27 13.5 mOhms 0.450 V 12.6 mA/cm
2
00-42 16% 35% 0 28.5 N/A 0.415 V 27 mA/cm
2
Several PVDF-PSSA MEAs were prepared to study the pressure effect on the
bonding between catalyst layer and membrane surface (Table 2.6). The PVDF-PSSA
90
membranes used had PSSA uptake 15-16%. Water uptake of these membranes ranged
from 30% to 35% and the physical properties of these membranes were very similar.
These membranes were made into MEAs and tested at 55
o
C with 0.5 M methanol
solution and 0.1 L/min ambient air (Figure 2.13).
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 50 100 150 200 250
Current Density (mA/cm
2
)
Cell Voltage (Volts)
MEA 00-27C, 500 lbs
MEA 00-27B, 1000lbs
MEA 00-27A, 1500 lbs.
MEA 00-02, 2000 lbs.
MEA 00-40, 500 lbs.
55
o
C, 0.50 M Methanol
0.10 L/min Ambient Air
Figure 2.13 MEA fabrication pressure vs. electrical performance
Upon MEA fabrication pressure elevation, the thickness of the corresponding
MEA would decrease while the cell resistance was similar. We also studied the effect of
pressing force on the methanol crossover of the corresponding MEAs using CO
2
analyzer
described before. It was interesting to note that when the MEA fabrication force
91
increased from 500 lbs to 2000 lbs, the membrane thickness decreased from 28.5 mil to
23 mil, methanol crossover current density decreased from 27 mA/cm
2
to 11.6 mA/cm
2
.
This is because better bonding between catalyst layer and membrane was created when a
higher pressure was used in MEA fabrication. It was observed that beyond certain
pressing force of 500 lbs, there was no significant decrease of methanol crossover upon
pressure elevation. The electrical performance of non-pressurized MEA 00-42 at 55
o
C
with 0.5 M methanol solution and 0.1 L/min ambient air showed a 0.415 V at 100
mA/cm
2
, while MEA00-27C showed a great improvement of 0.454V at the same
operating condition. It was concluded that the MEA fabrication pressure is necessary for
the improvement of bonding between catalyst layer and membrane surface. However,
beyond a certain pressure threshold, the pressure increase was not helpful.
2.4.2 Ionomer studies with PVDF-PSSA beads
One idea to refine the MEA fabrication process is to add hydrophobic agents to
the cathode paper to increase oxygen solubility and remove accumulated water. This
technique has been used in the preparation of Nafion-117 (by our JPL team) and has
resulted in favorable performance, especially at low temperatures. Two kinds of PVDF-
PSSA beads were prepared. The ion exchange capacities of these beads are shown below:
92
0.5% Crosslinked Beads: 0.50 mE/g
1.0% Crosslinked Beads: 0.44 mE/g
Nafion-117: 0.90 mE/g
We prepared several MEAs using the ionomer and it was readily apparent that the
use of the beads when incorporated with catalyst and water was not convenient to paint
due to the formation of a suspension.
Glycerol: Addition of glycerol into the catalyst ink improved the paintability of the
mixture but would not dry. The catalyst formed a gel that squeezed out of the MEA upon
pressing.
DMAc: We normally included DMAc in our anode composition to improve the
melt flow properties of the PVDF thereby improving bonding. However with inclusion at
both anode and cathode the viscosity was reduced and the MEA was destroyed due to the
some “wrinkling” experienced earlier.
Carbon Black: Investigation on ionomer testing with carbon black that looked very
promising. Further studies are on-going.
93
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00
Current Density (mA/cm
2
)
Volts (V)
00-41B: 2X PVDF
00-41C: 1X PVDF
00-41D: No PVDF
MEA 00-41B, 00-41C & 00-41D
25
o
C, Ambient Air 0.10 L/min
0.25 M Methanol
0.523 V at 16 mA/cm
2
Figure 2.14 The effect of PVDF-PSSA beads on cell performance I
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00
Current Density (mA/cm
2
)
Volts (V)
00-41B: 2X PVDF
00-41C: 1X PVDF
00-41D: No PVDF
MEA 00-41B, 00-41C & 00-41D
25
o
C, Ambient Air 0.10 L/min
0.50 M Methanol
0.500 V at 16 mA/cm
2
Figure 2.15 The effect of PVDF-PSSA beads on cell performance II
94
The MEAs with PVDF-PSSA beads added to the cathode side were tested at 25
o
C with 0.25 M or 0.5 M methanol solution and 0.1 L/min ambient air. The results are
shown at Figure 2.14 and Figure 2.15.
00-41B: PVDF added to both the anode and cathode catalyst inks and electrode
papers resulted in reduced cell performance at all operating conditions as compared to
that with the absence of additive at the cathode.
00-41C: PVDF was added to the electrode (cathode) paper only but it was not
immediately clear if it helped the performance. The performance was very similar to that
of a standard MEA, slightly better at certain conditions.
00-41D: This was the control MEA that had no PVDF additive at all. The
performance was consistent with our newly established baseline of 0.45 V at 100 mA/cm
2
.
It was very difficult to determine if PVDF had a beneficial impact on fuel cell
performance. Our data suggests that at moderate temperatures the addition of PVDF does
not help cathode performance. Since the PVDF-PSSA membranes have reduced rates of
both water and methanol permeability, the increased hydrophobicity of the additive may
hinder cathode kinetics. However, at lower temperatures, the addition of the
homopolymer seemed to help with performance, but it did not have an overwhelming
impact.
95
The PVDF-PSSA-PMMA membrane electrode assembly was inserted into the
necessary cell hardware and evaluated for electrical performance. There are many
elements that affect the MEA performance. Among them, we studied the methanol
concentration, temperature effect, different flow rate effects, PSSA content effect and the
catalyst loading effect.
2.4.3 Methanol concentration effect on electrical performance
The cell performance of a PVDF-PSSA-PMMA MEA at methanol concentration
of 0.5 M, 1 M and 2 M was measured at 55
o
C using 0.7 L/min air or oxygen flow at
ambient pressure, as illustrated in Figure 2.16 and Figure 2.17.
Increasing methanol concentration can benefit anode performance at high current
densities because methanol mass transfer limitations are alleviated. However, the
methanol crossover will also increase when methanol concentration increases, which will
reduce the cathode performance.
At low current densities, when fuel is not consumed, the mass transfer limitation
is not an issue while methanol crossover has a great impact on cell performance, and
hence low methanol concentration is preferred. At high current densities, the methanol
crossover is reduced because of the high fuel consumption rate at the anode and mass
96
transfer plays major role in cell performance, making higher methanol concentration
better in this case.
55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 50 100 150 200 250 300
I (mA/cm
2
)
E (V)
0.5 M methanol, 0.7 L/min air
1 M methanol, 0.7 L/min air
2 M methanol, 0.7 L/min air
Figure 2.16 Methanol concentration vs. cell performance I
Fuel cell operation using air further compounds cathode performance because
mass transfer of oxygen becomes an issue. At high current densities, cell with 1 M
methanol performs better than cell with 2 M methanol with air as oxidant, while the
performance of 2 M methanol is better when oxygen is used. This suggests that the
cathode using air is highly sensitive to methanol crossover and cathode polarization
97
losses outweigh anode potential gain due to the mass transfer limitations of oxygen in air
coupled with the poisoning effect of methanol crossover.
55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.00 100.00 200.00 300.00 400.00 500.00
I (mA/cm
2
)
E (V)
0.5 M methanol, 0.7 L/min oxygen
1 M methanol, 0.7 L/min oxygen
2 M methanol, 0.7 L/min oxygen
Figure 2.17 Methanol concentration vs. cell performance II
2.4.4 Impact of PSSA uptake on electrical performance
PSSA uptake is related with proton conductivity, water content and methanol
crossover in DMFC. Membranes with 15% PSSA uptake has shown promising electrical
performance compared to Nafion. MEAs with different PSSA uptake were made to
further study the impact on fuel cell related properties.
98
0
5
10
15
20
25
30
8 1012 1416 18 2022 24
PSSA uptake %
Cell resistance mOhms
60 ℃
90 ℃
Figure 2.18 Effect of PSSA uptake on OCV resistance
0
5
10
15
20
25
30
35
40
45
8 1012 1416 1820 2224
PSSA uptake %
OCV methanol crossover mA/cm
2
60 ℃
90 ℃
Figure 2.19 Effect of PSSA uptake on methanol crossover
99
PVDF-PSSA-PMMA membranes with different PSSA uptakes were assembled
into test hardware and the steady state OCV resistance values were measured at 60
o
C and
90
o
C (Figure 2.18). It was indicated that there is a linear relationship between cell
resistance and PSSA uptake. The results were in accordance with the fact that increasing
membrane water content associated with PSSA uptake will lower the grain-boundary
resistance.
PVDF-PSSA-PMMA MEAs were tested at 60
o
C and 90
o
C (Figure 2.19). The
methanol crossover rate was measured through the CO
2
analyzer method described
before. We also found that the methanol crossover increased linearly with increasing
PSSA uptake. This is because the membrane water content and associated methanol
diffusion coefficient increase with PSSA uptake.
It is interesting to notice that increasing PSSA uptake will have a mixed effect on
fuel cell performance. Increasing PSSA uptake leads to low cell resistance and high
proton conductivity. However, methanol crossover rate will increase with higher PSSA
uptake. The electrical performance of the cell will depend on finding a delicate balance
between increasing proton conductivity and decreasing methanol permeation rate. The
electrical performances of MMA01-1 (PS 14.6%, PMMA 0.5%) and MMA01-3 (PS
17.1%, PMMA12.5%) are compared at different temperatures (Figure 2.20-2.23).
100
55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 50 100 150 200 250 300
I (mA/cm
2
)
E (V)
MMA01-1, 0.1 L/min Air
MMA01-3, 0.1 L/min Air
MMA01-1, PS 14.6% MMA 0.5%
MMA01-3, PS 17.1% MMA 12.5%
Figure 2.20 PSSA uptake on MEA performance at 55
o
C with air
55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 50 100 150 200 250 300 350 400
I (mA/cm
2
)
E (V)
MMA01-1, 0.1 L/min Oxygen
MMA01-3, 0.1 L/min Oxygen
MMA01-1, PS 14.6% MMA 0.5%
MMA01-3, PS 17.1% MMA 12.5%
Figure 2.21 PSSA uptake on MEA performance at 55
o
C with oxygen
101
90
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 50 100 150 200 250 300
I (mA/cm
2
)
E (V)
MMA01-1, 0.1 L/min Air
MMA01-3, 0.1 L/min Air
MMA01-1, PS 14.6% MMA 0.5%
MMA01-3, PS 17.1% MMA 12.5%
Figure 2.22 PSSA uptake on MEA performance at 90
o
C with air
90
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400 500 600 700
I (mA/cm
2
)
E (V)
MMA01-1, 0.1 L/min Oxygen
MMA01-3, 0.1 L/min Oxygen
MMA01-1, PS 14.6% MMA 0.5%
MMA01-3, PS 17.1% MMA 12.5%
Figure 2.23 PSSA uptake on MEA performance at 90
o
C with oxygen
102
With a lower PSSA uptake, MMA01-1 performed better at 55
o
C with low flow
rate of ambient air or oxygen at the cathode. This is because the methanol crossover is a
major factor in determining cell performance. Lower PSSA uptake with adequate proton
conductivity will enhance cell performance. While at high temperature, when the
electrode reaction rate is high, the consumption rate of methanol will be fast and the mass
transport of proton is a major factor in cell performance, MMA01-3 performed better at
90
o
C with the higher proton conductivity.
OCV usually doesn’t reach the theoretical value because of the activation
polarization loss as well as the mixed potential at the cathode side caused by methanol
crossover. However, we have recorded OCV values as high as 0.9 V compared to an
OCV value about 0.7 V of Nafion117 at the same test condition. This is another
confirmation of the low methanol crossover property related to the ability of PVDF-
PSSA-PMMA membranes to reduce the formation of aqueous channels and confine a
significant amount of water to the PVDF backbone.
The results also show the PVDF-PSSA-PMMA membranes with appropriate
PSSA uptake display excellent electrical performance at different temperatures, even
with a very low flow rate of air or oxygen at ambient pressure. At a same condition,
Nafion can obtain a similar performance only with very high flow rate of air/oxygen or
103
with a pressured system. This comparison gives a great advantage of our PVDF-PSSA-
PMMA membranes over Nafion considering practical uses. It would be very convenient
and economical for the fuel cell in portable electronic devices to use ambient air instead
of a complex pressured oxygen supply system.
2.4.5 Impact of membrane thickness on electrical performance
Table 2.7 Physical properties of membranes and related MEA properties
Membrane PSSA
uptake
Thickness Proton
conductivity
MEA
thickness
Methanol
diffusion coefficient
Nafion117 N/A 7 mils 60 mS/cm 20 mils 3.5E-07 cm
2
/s
MMA02-8 15.2% 10.6 mils 56 mS/cm 23 mil 5.8E-07 cm
2
/s
MMA02-9 14.9% 13.2 mils 48 mS/cm 26 mil 5.1E-07 cm
2
/s
The membrane thickness has a significant effect on the membrane resistivity
values. Normally, the cell resistivity will reduce when the membrane thickness decreases.
Two PVDF-PSSA-PMMA membranes with different thicknesses (Table 2.7) were made
into MEAs and tested at 90
o
C (Figure2.24). MMA02-8 and MMA02-9 have similar
PSSA uptakes. It was observed that the methanol diffusion coefficient decreased with
104
increasing membrane thickness. At 90
o
C, the performance of MMA02-8 was much better
than MMA02-9 because of the reduced cell resistance related to membrane thickness.
Despite slight increase of methanol crossover, the cathode kinetics of MMA02-8
benefited from the improved proton flux and availability of oxygen. The cell performance
is better with decreased membrane thickness due to the decreased ohmic losses
throughout the bulk.
90
o
C,1 M methanol, 0.1 L/min oxygen
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.00 100.00 200.00 300.00 400.00 500.00
I (mA/cm
2
)
E (V)
MMA02-8
MMA02-9
Figure2.24 Effect of membrane thickness on cell performance
105
2.4.6 Temperature effect on cell performance
Increasing the operating temperature can help the MEA performances, because
the catalyst activity and utilization improve significantly at higher temperature. As can be
seen from Figure 2.25, with 2 M methanol solution on the anode and 0.1 L/min oxygen at
cathode side, the cell performance at 90 °C is the best compared with that of other lower
temperatures. At 100 mA/cm
2
, the cell voltage is only 0.38 V at 25 °C, while at 90 °C the
cell voltage can reach up to 0.54 V.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 450.00
I (mA/cm
2
)
E (V)
25 ℃, 2 M methanol, 0.1 L/min oxygen
55 ℃, 2 M methanol, 0.1 L/min oxygen
60 ℃, 2 M methanol, 0.1 L/min oxygen
90 ℃, 2 M methanol, 0.1 L/min oxygen
Figure 2.25 The temperature effect on a 25 cm
2
MEA performance
106
methanol crossover rate at 90
o
C
0
50
100
150
200
250
300
350
0 200 400 600 800 1000
I (mA/cm
2
)
crossover current density
(mA/cm2)
00-72a, 2 M methanol
Nafion117, 2 M methanol
Figure 2.26 Comparison between USC PVDF-PSSA membrane and Nafion117
The ability of the cell to operate at high temperature is very unique considering
the inability of Nafion117 based systems to operate under similar conditions. In the case
of Nafion117, high water permeability rates will flood the cathode. Therefore, with
Nafion, high pressure oxygen or air has to be employed with high flow rates. Also, the
methanol crossover rate of our membrane is only 1/3 of that of Nafion117 under the same
conditions (Figure 2.26). The methanol crossover rate can be monitored at different
current densities using CO
2
analyzer. The methanol crossover rate decreased as current
density increased because the catalyst utilization at anode increased at higher current
107
density, thus decreasing the rate of methanol to permeating through the polymer
electrolyte.
2.4.7 Gas flow rate effect
Different gas flow rate can also have profound effect on the MEA performance.
Under similar conditions, when the flow rate of air is increased, the cell performance gets
better.
MEA0072b, 55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 50 100 150 200 250 300
I (mA/cm
2
)
E (V)
1 M methanol, 0.1 L/min air
1 M methanol, 0.3 L/min air
1 M methanol, 0.5 L/min air
Figure 2.27 The performance of MEA00-72b at different flow rates of air
108
As can be seen from Figure 2.27, the air flow rate increased from 0.1 L/min to 0.5
L/min, the cell voltage increased from 0.40 V to 0.43 V. It means that the cathode is
“starving” for the air supply at lower flow rates.
MEA0072b, 55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 50 100 150 200 250 300 350 400
I (mA/cm
2
)
E (V)
1 M methanol, 0.1 L/min oxygen
1 M methanol, 0.3 L/min oxygen
1 M methanol, 0.5 L/min oxygen
Figure 2.28 The performance of MEA00-72b at different flow rates of oxygen
For the oxygen, there is almost no big difference in the cell performance with the
increase of oxygen flow rate. There is enough oxygen pressure at the cathode even at low
flow rates (Figure 2.28). Actually at very high flow rate of oxygen, MEA performance
109
will decease because high flow rate will drive out the moisture content necessary for the
cathode reaction and the cathode will dry out.
2.4.8 Catalyst loading effect
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.00 50.00 100.00 150.00 200.00 250.00
Current Density (mA/cm
2
)
Volts (V)
USC 00-27C
USC 00-72A
USC 00-72C
55
o
C, 0.50 M Methanol
0.12 L/min Ambient Air
USC 00-27C
Anode = 15 mg/cm
2
Cathode = 17 mg/cm
2
USC 00-72A
Anode = 7.5 mg/cm
2
Cathode = 8.5 mg/cm
2
USC 00-72C
Anode = 4 mg/cm
2
Cathode = 4 mg/cm
2
Figure 2.29 Cell performance vs. catalyst loading
Platinum is a relatively expensive noble metal catalyst for PEM fuel cell use.
Reducing the amount of catalyst used and still maintain a decent cell performance will
greatly reduce the cost of practical use. Currently, we still use relatively large amounts of
110
catalyst on both anode and cathode sides. Efforts have been made to decrease catalyst
loading and still maintain a good performance. Seen from Figure 2.29, we have
successfully cut the catalyst loading to half and get even better cell performance. But if
we only use ¼ of catalyst, there is a large decrease in the cell performance.
2.5 PVDF-PSSA-PMMA performance at room temperature
DMFC has a potential in the application for portable devices because of its liquid-
feed character, system simplicity and ability for start-up at low temperature. It would be
interesting to examine the performance of PVDF-PSSA-PMMA MEAs at room
temperature (Figure 2.30, 2.31).
With 0.1 L/min air flow at the cathode, MMA01-1 had the best performance at
low current densities (0-40 mA/cm
2
). This is due to the relatively low methanol crossover
rate related to the low PSSA uptake of 14.6%. Beyond 40 mA/cm
2
, MMA02-6 with
PSSA uptake of 16.5% led in performance due to lower cell resistivity. Although
MMA01-3 had the highest PSSA uptake, it couldn’t match the performance of other
MEAs because of its lower cell resistivity couldn’t compensate the higher methanol
crossover rate. It’s not surprising that MMA02-6 produced the highest power density of
22 mW/cm
2
at a current density of 60 mA/cm
2
with ambient air at room temperature.
111
25
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 20 40 60 80 100 120 140
I (mA/cm
2
)
E (V)
0
5
10
15
20
25
Power density(mW/cm
2
)
MMA01-3, 0.1 L/min Air
MMA02-6, 0.1 L/min Air
MMA01-1, 0.1 L/min Air
MMA01-1, PS14.6% MMA 0.5%
MMA02-6, PS16.5% MMA 2.5%
MMA01-3, PS17.1% MMA 12.5%
Figure 2.30 PVDF-PSSA-PMMA MEA performance at room temperature with air
25
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 50 100 150 200 250
I (mA/cm
2
)
E (V)
0
5
10
15
20
25
30
35
40
45
Power density (mW/cm
2
)
MMA02-6, 0.1 L/min Oxygen
MMA01-3, 0.1 L/min Oxygen
MMA01-1, 0.1 L/min Oxygen
MMA01-1, PS14.6% MMA 0.5%
MMA02-6, PS16.5% MMA 2.5%
MMA01-3, PS17.1% MMA 12.5%
Figure 2.31 PVDF-PSSA-PMMA MEA performance at r. t. with oxygen
112
With 0.1 L/min oxygen at the cathode, the performance of all these PVDF-PSSA-
PMMA MEAs improved substantially. The maximum current density can reach 200
mA/cm
2
, compared to 100 mA/cm
2
with air. Power density of 41 mW/cm
2
was obtained
at a current density of 140 mA/cm
2
during the scan current test of MMA02-6. we can
conclude that using pure oxygen will improve the cathode reaction kinetics to obtain a
better cell performance. PVDF-PSSA-PMMA membranes with the appropriate PSSA
uptake would have a great potential for fuel cell applications in portable devices. With
the development of MEA stack, higher current and power densities can be realized for
high power application purposes.
2.6 Comparison between PVDF-PSSA-PMMA and Nafion117
The electrical performance of PVDF-PSSA-PMMA membranes was compared
with Nafion117 and PVDF-PSSA membranes at different temperatures. Both USC-made
membranes have PSSA uptake around 15% and a proton conductivity similar to
Nafion117.
The specific IPN morphology gave PVDF-PSSA-PMMA membranes advantage
in terms of methanol crossover. At 25
o
C, OCV crossover density of MMA01-1, 0072b,
Nafion117 MEAs are 2.2, 6.8, 26.2 mA/cm
2
, respectively (Figure 2.9). The methanol
113
crossover effect was significant in this case. As can be seen from Figure 2.32, MMA01-1
had an OCV value of 0.88 V, while Nafion’s OCV was only 0.77 V. MEA 0072b and
MMA01-1 performed superior to Nafion at all current densities with maximum current
density of 200 mA/cm
2
.
25
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.00 50.00 100.00 150.00 200.00
I (mA/cm
2
)
E (V)
0072b, 1 M methanol, 0.1 L/min oxygen
Nafion, 1 M methanol, 1 L/min oxygen
MMA01-1, 1 M methanol, 0.1 L/min oxygen
MMA01-1, PS 14.6% MMA 0.5%
0072b, PS 16%
Figure 2.32 Comparison between USC MEAs with Nafion117 at 25
o
C
As discussed before, the specific water management properties of IPN membranes
alleviate the cathode flooding problem. Nafion117 will need a high flow rate of air or
oxygen to take away the excess of water permeating through the membrane to avoid
114
cathode flooding. At 55
o
C (Figure 2.33), MMA01-1 performed better than Nafion at
low current densities (0-120 mA/cm
2
). When the methanol crossover current density was
reduced from 52 to 27 mA/cm
2
at the high current densities, the low cell resistivity factor
took a major role and the electrical performance of Nafion117 improved.
55
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 50 100 150 200 250 300 350 400
I (mA/cm
2
)
E (V)
0072b, 1 M methanol, 0.1 L/min oxygen
Nafion, 1 M methanol, 1 L/min oxygen
MMA01-1, 1 M methanol, 0.1 L/min oxygen
MMA01-1, PS14.6% MMA 0.5%
0072b, PS16%
Figure 2.33 Comparison between USC MEAs with Nafion117 at 55
o
C
The same trend can be observed when the MEA performance was examined at
90
o
C (Figure 2.34). MMA01-1 and 00-72b performed better than Nafion at low current
densities. Nafion117 still had advantage at higher operating current densities. However,
115
the performance of Nafion dropped dramatically at a current density of 450 mA/cm
2
. It
was probably due to excessive water accumulation at the cathode side. It was exciting to
see that the OCV value of MMA01-1 still reached 0.88 V despite the increased methanol
crossover rate at high temperature.
90
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400 500 600 700
I (mA/cm
2
)
E (V)
0072b, 1 M methanol, 0.1 L/min oxygen
Nafion, 1 M methanol, 1 L/min oxygen
MMA01-1, 1 M methanol, 0.1 L/min oxygen
MMA01-1, PS14.6% MMA 0.5%
0072b, PS16%
Figure 2.34 Comparison between USC MEAs with Nafion117 at 90
o
C
2.7 Fuel efficiency and fuel cell efficiency
The fuel efficiency is used to calculate how much fuel is used in the anode
reaction compared to the total fuel supplied. The most immediate consequence of
116
methanol permeation is the reduction in fuel utilization at the cathode. PVDF-PSSA-
PMMA membranes have a low methanol crossover rate, which will increase the fuel
efficiency. The fuel efficiency can be calculated from the following equation:
Fuel efficiency % = Iload
/ (Iload + Icrossover) (2.4)
Iload is the operating current density and Icrossover is the calculated crossover current
density to measure how much fuel has permeated through the membrane on to the
cathode side.
0
20
40
60
80
100
120
0 20 40 60 80 100 120 140 160
I (mA/cm
2
)
Fuel efficiency %
MMA01-1
MMA02-6
MMA01-3
00-72b
Nafion117
Fuel efficiency % = Iload / (Iload+Icrossover)
25
o
C, 1 M Methanol
Figure 2.35 Fuel efficiency at 25
o
C
117
The fuel efficiencies of various PVDF-PSSA-PMMA membranes were compared
with Nafion at 25
o
C and 90
o
C (Figure 2.35, Figure2.36). It’s known that methanol
crossover rates decrease with increase of current density as a result of increased anode
utilization, thus improving fuel cell efficiency. Because of reduced methanol crossover,
the fuel efficiency of PVDF-PMMA-PSSA membranes can reach 90% at 25
o
C, much
higher than Nafion117 tested under similar conditions.
Fuel efficiency % = Iload / (Iload+Icrossover)
0
20
40
60
80
100
120
0 100 200 300 400 500 600 700
I (mA/cm
2
)
Fuel efficiency %
MMA01-3
MMA01-1
MMA02-6
Nafion117
90
o
C, 1 M methanol
Figure 2.36 Fuel efficiency at 90
o
C
118
When temperature rose to 90
o
C, fuel efficiency also improved. Because at higher
temperature, the reaction kinetics at the anode improved and more methanol would be
consumed, thus reducing the overall crossover rate. The fuel efficiency of PVDF-PSSA-
PMMA MEAs can reach as high as 98% at high current density.
0
5
10
15
20
25
30
35
40
45
0 20 40 60 80 100 120 140 160
I (mA/cm
2
)
Fuel Cell efficiency
MMA01-1
MMA02-6
MMA01-3
00-72b
Nafion117
Fuel cell efficiency % = (Operating voltage / 1.21 V) Fuel efficiency
25
o
C, 1 M methanol
Figure 2.37 Fuel cell efficiency at 25
o
C
Fuel utilization is only part of the overall picture when evaluating fuel cell
efficiencies. The overall cell performance as compared to the theoretical potential of 1.21
V must be factored into the equation to truly understand the fuel cell operation. The fuel
119
cell efficiency of an operating system factors in both fuel efficiency and voltage
efficiency according to the following equation:
Fuel cell Efficiency = (Operating voltage/1.21 V) x Fuel efficiency% (2.5)
Fuel cell efficiency % = (Operating voltage / 1.21 V) Fuel efficiency
0
5
10
15
20
25
30
35
40
45
0 100 200 300 400 500 600 700
I (mA/cm
2
)
Fuel cell efficiency %
MMA01-3
MMA01-1
MMA02-6
Nafion117
00-72b
90
o
C, 1 M methanol
Figure 2.38 Fuel cell efficiency at 90
o
C
Fuel cell efficiency was measured at different temperatures (Figures 2.37-2.38).
The fuel cell efficiency of PVDF-PSSA-PMMA MEAs can reach as high as 40%, while
Nafion117 can only reach 25% at the most optimized condition.
120
2.8 PVDF-PSSA-PMMA performance on a H
2
/O
2
fuel cell
0
0.2
0.4
0.6
0.8
1
1.2
0 200 400 600 800 1000 1200
Current Density (mA/cm
2
)
Voltage (Volts)
MMA02-8, 7 mils
Nafion117, 7 mils
Tcell = 80
o
C
TH
2
= 78
o
C, Flow Rate = 6.8 ml/s
TO
2
= 75
o
C, Flow Rate = 6.4 ml/s
Figure 2.39 Comparison between PVDF-PSSA-PMMA and Nafion117 on H
2
/O
2
fuel cell
It was interesting to see how the PVDF-PSSA-PMMA membranes perform in
H
2
/O
2
fuel cell compared to Nafion117. We reduced the PVDF-PSSA-PMMA membrane
thickness to 7 mils to match Nafion117 and got a comparable electrical performance of
PVDF-PSSA membranes (Figure 2.39). The PS uptake of this membrane was 15% and
this membrane had a similar equivalent compared to Nafion117. A humidifier was built
to obtain the desired anode and cathode operating temperature, and MEAs using both
Nafion117 and PVDF-PSSA-PMMA membranes can sustain a high current density of
121
1000 mA/cm
2
. This is quite exciting since PVDF-PSSA-PMMA MEA is designed for
direct methanol full cell application.
2.9 Conclusions
PVDF-PSSA-PMMA membranes have been demonstrated to exhibit substantially
low methanol crossover rates compared to Nafion and high electrical performance in
direct methanol fuel cell applications.
PVDF is chosen to precursor membrane material since it is known as a low-cost
material to be used as films for water/alcohol separation purposes. The crystallinity of
PVDF precursor has great impact on the polystyrene distribution in the IPN and
ultimately polystyrene sulfonic acid distribution. Appropriate PVDF precursor selection
resulted in favorable surface morphology suitable for membrane electrode fabrication.
The concept of IPN was applied in the synthesis of PVDF-PSSA-PMMA
composite membranes. IPN methodology retained the excellent physical and chemical
properties of PVDF and insured a more uniformed distribution of PSSA moiety in the
membranes. PMMA was introduced into IPN network as compatibilizer to reduce the
interfacial tension between PVDF and polystyrene, thus alleviating phase separation
problem often encountered in styrene-grafted membranes. Subsequent sulfonation led to
122
PVDF-PSSA-PMMA membranes with great improvement of the mechanical properties
as well as electrical performance.
Thermal analysis was used to study the nature of the membranes relating to the
degree of crystallinity, polymer miscibility as well as membrane stability. Energy
dispersive X-ray analysis is used to insure the homogeneity of polystyrenesulfonic acid
distribution along the cross-section of a PVDF-PSSA membrane. These techniques
helped to screen MEA candidates for DMFC.
Membrane water uptake studies showed that the both water content and proton
conductivity increased linearly with PSSA uptake. The degree of polystyrene
crosslinking doesn’t have a significant impact on proton conductivity and membrane
water content. Ion exchange capacity test is used to determine the content of acidic
protons in the membrane. It was found that membranes having around 15% PSSA uptake
had proton conductivity similar to Nafion117.
Methanol permeability of these membranes was studied using both GC tracking
method and CO
2
analyzer monitoring of operating fuel cell systems. The transport
mechanism of PVDF-PSSA-PMMA membranes is quite unique compared to Nafion and
PSSA grafted membranes. With IPN methodology, the PVDF-PSSA-PMMA membranes
have reduced water and methanol permeability because of the reduced formation of
123
aqueous channels. The non-porous PVDF precursor will confine a significant amount of
water to the PVDF backbone, further helped to explain why the PVDF-PSSA-PMMA
membranes had reduced methanol crossover with an equivalent weight comparable to
Nafion117.
Membrane electrode assembly fabrication method was modified to optimize the
electrical performances of MEAs. The PVDF-PSSA-PMMA membrane electrode
assembly was inserted into the necessary cell hardware and evaluated for electrical
performance. We studied the impact of methanol concentration, temperature effect,
different flow rate effect, PSSA content effect and the catalyst loading effect on cell
performance.
Results showed that PVDF-PSSA-PMMA membranes had a dramatically lower
methanol cross-over rate (up to 90%) and much higher fuel cell efficiency compared to
commercially available Nafion-117. The specific physical and chemical properties of the
membrane made it a good PEM candidate for direct methanol fuel cell applications.
124
2.10 Chapter 2 References
1. Hodgdon, R. B. and Boyack, J. R., J. Polym. Sci.: Part A 3 (1965) 1463.
2. Yang, D., Thomas, E. L., J. Mater. Sci. Lett. 3 (1984) 929.
3. Petermann, J., Gohill, R.M., J. Mater. Sci. 14 (1979) 2620.
4. Gregorio, R., Cestari, M., J. Polym. Sci.: Part B: Polym. Phys. 32 (1994) 859-870.
5. Kim, K.J., Chao, Y.J., Kim, Y.H., Vib. Spectrosc. 9 (1995) 147-159.
6. Flint, S.D., Slade, R.C.T., Solid State Ionics 97 (1997) 299-307.
7. Hietala, S., Koel, M., Skou, E., Elomaa, M., Sundholm, F., J. Mater. Chem. 8(5)
(1998) 1127-1132.
8. Smart, M., “Chemical and Electrochemical Oxidation of small organic
Molecules”, UMI Ann Arbor, MI, (1998).
9. Atti, A.R., Ph. D. Dissertation, University of Southern California (2000) 109-112.
10. Mijovic, J., Luo, H.L., Han, C. D., Polym. Eng. Sci. 22(4) (1982) 234.
11. Lee, J.H., Kim, S.C., Macromolecules 19 (1986) 644-648.
12. Kreuer, K.D., Chem. Mater. 8 (1996) 610-641.
13. Kreuer, K.D., Weppner, W., Rabenau, A., Angew. Chem. Int. Ed. Engl. 21 (1982)
208.
14. Grottuss, S.G., Work in Progress.
15. Zawodzinski, T.A., Derouin, C., Redzinski, S., Sherman, R.J., Smith, V.T.,
Springer, T.E., Gottesfeld, S., J. Electrochem. Soc. 140 (1993) 1041.
125
16. Ostrovskii, D.I., Torell, L.M., Paronen, M., Hierala, S., Sundholm, F., Solid state
ionics 97 (1997) 315-321.
17. Watkins, D., Dircks, K., Epp, D.A., in “Proceedings of 32
nd
Power Sources
Conference”, Cherry Hill, NJ, June9-12, 1986, The Electrochem. Soc., Inc., P.590.
18. Raistrick, I.D., in “Diaphragms, Separators, and Ion exchange Membranes”, Zee,
J.W., White, R.E., Kinoshita, K., Burney, H.S., Editors, P.172, The Electrochem.
Soc. Softbound Proceeding Series, Pennington, New Jersy, (1986).
126
Chapter 3
New polymer electrolyte membrane materials
In order to increase the performance of the MEA and reduce the cost, various
attempts have been made in our laboratories to develop several different polymer
electrolyte materials. One method we tried is to introduce a copolymer with polystyrene,
which can increase the mechanical stability and enhance the PSSA distribution in an IPN
system. The other method we probed is to identify new monomers with functional groups
which can be polymerized and then hydrolyzed to sulfonic acid in order to avoid the
corrosive sulfonation step. We also used the commercially available PVDF precursors
(Kynar materials) to synthesize IPN for comparison. Several new polymer electrolyte
membranes were synthesized and tested in direct methanol fuel cell. Some of them
exhibited promising results.
3.1 PVDF-Poly(vinylsulfonyl chloride) membrane
Poly (vinylsulfonyl chloride) was considered to be an ideal candidate to add into
the IPN system because of the low cost of the material and no further need of sulfonation
step. Vinylsulfonyl chloride was readily synthesized from sodium vinylsulfonate (from
127
Aldrich) by adding PCl
5
in chloroform at 35
o
C [1]. After 4 hours, the reaction mixture
was vacuum distilled to get the expected vinyl sulfonic chloride monomer. Using PVDF
as a matrix, AIBN as initiator, divinylsulfone as cross linker, vinylsulfonic chloride
monomers went through impregnation step to obtain PVDF-Poly(vinylsulfonyl chloride)
semi-IPN membranes.
SO
3
Na
PCl
5
, CHCl
3
, 35
o
C
SO
2
Cl
Scheme 3.1 Synthesis of vinylsulfonyl chloride
Table 3.1 PVDF-Poly(vinylsulfonyl chloride) membranes
Membranes Poly(vinylsulfonic chloride) weight
percent
Water
uptake
Methanol
crossover
00-Cl1 3% 2.1% 0
00-Cl2 4% 6.4% 0
00-Cl3 5% 8.1% 0
Three membranes were fabricated with 3.5%, 4.7% and 5% poly (vinylsulfonyl
chloride) in weight, respectively. Then after hydrolysis in 1 M sulfuric acid, water
uptakes of these membranes were 2.1%, 6.4%, 8.1% respectively. One of them, with
128
8.1% water uptake, was fabricated into 5 cm
2
MEA and tested. But no current sustained
indicating inadequate monomer penetration into the membrane. Methanol crossover test
of these membranes further confirmed the inadequate monomer penetration. Appropriate
swelling condition in the impregnation step had to be found to improve the distribution of
poly (vinylsulfonyl chloride) in the IPN system.
3.2 PVDF - Poly trifluoro-1-(trifluoromethyl)ethyl vinylsulfonate
C H
2
CH
SO
2
Cl
+
CF
3
CF
3
O H
Et
2
O, NEt
3
, 25
o
C
C H
2
CH
SO
3
CH(CF
3
)
2 -Et
3
N
+
HCl
-
Scheme 3.2 Synthesis of trifluoro-1-(trifluoromethyl)ethyl vinylsulfonate
Since vinylsulfonyl chloride is immiscible with PVDF, we attempted to use the
corresponding vinylsulfonate in an PVDF based semi-IPN system.
Following the IPN synthetic procedure described before, a PVDF-Poly trifluoro-
1-(trifluoromethyl)ethyl vinylsulfonate membrane was made, with 3% ester in weight.
After hydrolysis, we found no water uptake took place after boiling in water for 6 hours.
129
Either higher ester uptake should be obtained before making it into MEA or the
hydrolysis condition need to be modified.
3.3 Poly(sodium vinylsulfonate) membrane
As mentioned, sodium vinylsulfonate is a low cost material, which is
commercially available. Simple free radical polymerization of the material would
produce a high molecular weight polymer and subsequent hydrolysis would give us a
possible membrane candidate [3]. Several methods to polymerize vinyl sulfonic acid
sodium salt were attempted:
a) Using AIBN as the initiator with DVB as the crosslinker.
b) Using K
2
S
2
O
8
as initiator, divinyl sulfone as the crosslinker.
c) Copolymerization with styrene.
Using all these methods, we were able to synthesize copolymers. But the melting
points of the polymers were too high to fabricate membranes under hot-press. However,
solvent casting methods were not investigated.
130
3.4 PVDF-poly(vinyl sulfonic acid tetrabutylammonium salt) membrane
SO
3
Na
-
+
+
(CH
3
CH
2
CH
2
CH
2
)
4
N Cl
+
-
CHCl
3
/H
2
O
35 C SO
3
TBA
+
-
Scheme 3.3 Synthesis of tetrabutylammonium vinylsulfonate
As discussed before, it was hard to make membranes with good mechanical
properties using sodium vinylsulfonate. Then, we consider introducing
tetrabutylammonium salt into the IPN system because of its property as a good phase
transfer reagent and can be quite soluble in organic solutions. Such properties of this salt
could help to vary the miscibility of the monomer in PVDF precursor.
Synthetic procedure of tetrabutylammonium vinylsulfonate is very simple [4].
One equivalent of tetrabutylammonium chloride and one equivalent of sodium sulfonate
are mixed together with chloroform/water as solvent. Stirring the mixture at 35
o
C for 24
hours gave you the desired TBA monomer.
Poly(tetrabutylammonium vinylsulfonate) was introduced into PVDF precursor
following the standard IPN synthetic procedure. Membrane with 10% TBA salt was
obtained. Then, after hydrolysis, it was made into MEA. However, the material could not
sustain any current during cell operation, which indicates that no adequate TBA
131
monomers were able to penetrate into the inner part of the membrane. Better solvent need
to be found for the impregnation step.
3.5 PVDF-PSSA-poly(trifluoroethyl acrylate)
C H
2
CH
SO
2
Cl
+
C H
2
CH
SO
3
CH
2
CF
3
CH
2
CF
3
O H
x y
COOCH
2
CF
3
Styrene
DVB/AIBN
Et
2
O
NEt
3
x y
COOCH
2
CF
3
x y
COOCH
2
CF
3
SO
3
H
1.ClSO
3
H/CHCl
3
2. H
2
O
Scheme 3.4 Synthesis of PVDF-PSSA-poly(trifluoroethyl acrylate)
We have had tremendous success in using PMMA as copolymer of PSSA in
DMFC. Therefore it was of interest to investigate other similar copolymers involving
poly(trifluoroethyl acrylate) [5]. Bearing –CF
3
group on the side chain, poly
(trifluoroethyl acrylate) was believed to have better compatibility with PVDF matrix. It
could serve as a better compatibilizer role compared to PMMA.
132
FEA01-1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.00 50.00 100.00 150.00 200.00 250.00 300.00
I (mA/cm
2
)
E (V)
90 ℃, 0.5 M methanol, 0.1 L/min oxygen
90 ℃, 0.5 M methanol, 0.1 L/min air
55 ℃, 0.5 M methanol, 0.1 L/min oxygen
25 ℃, 0.5 M methanol, 0.1 L/min oxygen
25 ℃, 0.5 M methanol, 0.1 L/min air
55 ℃, 0.5 M methanol, 0.1 L/min air
Figure 3.1 Cell performance of PVDF-PSSA-poly(trifluoroethyl acrylate)
Using the IPN methodology, PVDF-PSSA-poly(trifluoroethyl acrylate)
membrane with 15% PSSA and 2.5% poly(trifluoroethyl acrylate) was made into MEA.
The electrical performance at different temperatures is shown in Figure 3.1. Decent
performance was realized at different temperatures. This is the first MEA of this kind that
we made. There is still ample room left for improving the performance of such MEAs.
133
3.6 PVDF-PSSA-poly(heptafluorobutyl acrylate)
While studying the PVDF-PSSA-poly(trifluoroethyl acrylate) membrane, we were
also interested in fabrication of the corresponding heptafluorobutyl acrylate [6]. With
more fluorine moiety on the side chain, poly(heptafluorobutyl acrylate) might be a better
compatibilizer than PMMA and even poly(trifluoroethyl acrylate). A membrane with
15% PSSA uptake and 2.5% acrylate was fabricated. However, the MEA fabrication with
the material has not been carried out.
x
y
COOCH
2
CF
2
CF
2
CF
3
x
y
COOCH
2
CF
2
CF
2
CF
3
SO
3
H
1.ClSO
3
H/CHCl
3
2. H
2
O
Scheme 3.5 Synthesis of PVDF-PSSA-poly(heptafluorobutyl acrylate)
3.7 PVDF-PVP- CF
3
SO
3
H membrane
Polyvinylpyridine (PVP) is known to complex many kinds of Bronsted acids [7].
PVP membranes doped with strong acid like triflic acid would provide the PVP
membrane with proton conducitivity. PVDF-PVP membrane material was fabricated
134
readily using the IPN methodology. Then PVDF-PVP was treated with triflic acid to
obtain PVDF-PVP- CF
3
SO
3
H membrane.
n
N
CF
3
SO
3
-
+
H
Figure 3.2 Polyvinylpyridine doped with triflic acid
However the MEA made out of this membrane could not sustain any current
during cell testing. It could be due to the penetration problem of the triflic acid into the
bulk membrane. Further research is needed to solve the problem.
3.8 PVDF-PVP-HF membrane
n
N
+
H (HF)
n
F
-
Figure 3.3 Polyvinylpyridine doped with HF
135
A different superacid HF was used to study the PVP-superacid complex.
Vinylpyridine monomer was quite readily polymerized using our developed methodology.
The membrane was doped in HF for a period of time and washed with DI water. It was
confirmed that despite continued washing, a significant portion of the acid was retained
within the membrane. Methanol permeability studies on the membrane have been carried
out. The MEA evaluation has not yet been performed.
3.9 PVDF-PSSA-PVP membrane
Like PVDF-PSSA-PMMA membranes, PVP can copolymerize with polystyrene
to form an IPN in the PVDF matrix. Subsequent sulfonation will give us a new PVDF-
PSSA-PVP membrane which can be utilized in fuel cell studies.
x
y
SO
3
H
N
Figure 3.4 Structure of PSSA-PVP
136
PVDF-PS-PVP 25
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 20 40 60 80 100 120 140
I (mA/cm
2
)
E (V)
1 M methanol, 0.1 L/min air
1 M methanol, 0.1 L/min oxygen
Figure 3.5 Cell performance of PVDF-PSSA-PVP at 25
o
C
PVDF-PS-PVP 60
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400
I (mA/cm
2
)
E (V)
1 M methanol, 0.1 L/min air
1 M methanol, 0.1 L/min oxygen
Figure 3.6 Cell performance of PVDF-PSSA-PVP at 60
o
C
137
PVDF-PS-PVP 90
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400 500 600
I (mA/cm
2
)
E (V)
1 M methanol, 0.1 L/min air
1 M methanol, 0.1 L/min oxygen
Figure 3.7 Cell performance of PVDF-PSSA-PVP at 90
o
C
PVDF-PSSA-PVP membranes were fabricated into MEAs and then tested at
different temperatures in DMFC (Figure 3.5-3.7). The electrical performance of this type
of MEA at 25
o
C is a little bit lower to that of PVDF-PSSA-PMMA MEA but still greater
than Nafion under similar conditions.
Cell performance at 60
o
C is comparable to the performance of PVDF-PSSA-
PMMA MEA, reaching a current density as high as 380 mA/cm
2
with 0.1 L/min flow rate
of oxygen. Similar trend can be observed at 90
o
C, the maximum current density of 500
mA/cm
2
was easily obtained with 0.1 L/min flow rate of oxygen. This is a very promising
PEM material, which could exceed the performance of PVDF-PSSA-PMMA with further
modifications.
138
3.10 PVDF-PSSA-Poly( α,α-bis(trifluoromethyl)-β-4-ethenylbenzeneethanol)
x
y
SO
3
H
CF
3
OH
CF
3
Figure 3.8 PSSA- Poly( α,α-bis(trifluoromethyl)-β-4-ethenylbenzeneethanol)
To increase the membrane flexibility and further modify PVDF-PSSA IPN
morphology, poly4-ethenyl- α, α-bis(trifluoromethyl)benzeneethanol (PEFBE) was
introduced into polystyrene backbone as a copolymer. It was believed that
benzeneethanol component will help the membrane to retain more water content in the
dry state, thus greatly improving the membrane mechanical property.
The synthetic route of the copolymer is shown above. PVDF-polystyrene
membrane is reacted with bis(trifluoromethyl) oxirane at 0-20
o
C in DCM with triflic
acid as a catalyst [8]. Subsequent sulfonation is expected to give desired PVDF-PSSA-
PEFBE material, which will be used in DMFC testing.
139
n
x
y
CF
3
OH
CF
3
O
CF
3
CF
3
+
CH
2
Cl
2
CF
3
SO
3
H
Scheme 3.6 Synthesis of PSSA-PEFBE
3.11 Kynar membranes
A 4 mil commercial Kynar460 PVDF precursor was impregnated with 15% PSSA
and fabricated into an MEA. The MEA exhibited promising electrical performance
reaching our desired baseline of 0.430 V at 100 mA/cm
2
using roughly 1.0 L/min of air
flow at 55
o
C. The higher flow rate would be expected due to the thin nature of the
membrane. However, no significant increase in methanol crossover was observed. This
suggests that the 'ordered' distribution of crystalline domains, relative to the amorphous
areas precludes methanol's ability to readily diffuse. This may also result in a decrease in
catalyst utilization but the performance data so far is quite positive.
Efforts to prepare MEAs using the 15 mil precursor were unsuccessful as it
appears that none of the sulfonic acid moieties were able to penetrate the inner portion of
the material. Kynar460 films have been used to fabricate 15% PVDF-PSSA MEAs. This
is a promising work because the crossover is much lower than the corresponding 4 mil
140
membranes. The best result is 0.42 V at 100 mA/cm
2
in 0.5 M methanol aqueous solution,
as is shown in Figure 3.9.
55
o
C, Kynar460
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 20 40 60 80 100 120 140 160 180 200
I (mA/cm
2
)
E (V)
0.5 M methanol, 0.1 L/min air
0.5 M methanol, 0.3 L/min air
0.5 M methanol, 0.5 L/min air
0.5 M methanol, 1 L/min air
Figure 3.9 Electrical performance of Kynar460 4 mil MEA
As the cell operating temperature rises to 90
o
C, Kynar460 MEA performs well
with different flow rates of oxygen at the cathode (Figure 3.10). It was observed that the
operating current density can reach as high as 560 mA/cm
2
at an oxygen flow rate of 0.3
L/min. In the hydrogen-oxygen fuel cell, this 4 mil Kynar membrane is even better than
Nafion117 at low current density as shown in Figure 3.11.
141
90
o
C, Kynar460
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 100 200 300 400 500 600
I (mA/cm
2
)
E (V)
1 M methanol, 0.1 L/min oxygen
1 M methanol, 0.3 L/min oxygen
1 M methanol, 0.5 L/min oxygen
Figure 3.10 Kynar460 4 mil MEA performance at high temperature
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
1.1
0 50 100 150 200 250 300
I (mA/cm
2
)
E (V)
Kynar460
Nafion117
Tanode 80
o
C, Tcathode 78
o
C, Tcell 80
o
C,
1.8 L/min H
2
, 1.5 L/min O
2
Figure 3.11 Kynar460 4 mil MEA performance compared with Nafion 117
142
3.12 Norbornene based Polymer electrolyte membrane
Ring-opening metathesis polymerization (ROMP) is a type of polyethylene
metathesis polymerization that produces important products in industry. The driving
force of the reaction is relief of ring strain in cyclic olefins such as norbornene or
cyclopentene. Research has shown that the molecular weight of the polymer produced
can be changed by the addition of substituents to the monomer and different solvents.
The most famous catalyst used in the ROMP reaction is the Grubbs' catalysts [9-
11]. Because the driving force of the reaction is relief of ring strain, a strained cyclic
structure is required in a ROMP catalytic cycle. After formation of the metal-carbene
species, the double bond in the ring is attacked by the carbene, thus forming a highly
strained metallacyclobutane intermediate. The ring then opens up giving the beginning of
the polymer: a linear chain double bonded to the metal with a terminal double bond. The
new carbene reacts with the double bond on the next monomer, thus propagating the
reaction to form a polymer product.
Ring-opening metathesis polymerization of cycloalkenes can produce many
important materials for industry. Polynorbornene is also an important ROMP product on
the market. A side reaction of the polymerization of norbornene will produce
polydicyclopentadiene products such as Telene and Metton. The ROMP process is very
143
useful because polymers synthesized have a regular amount of double bonds. The
resulting polymer product can be subjected to hydrogenation or can be functionalized into
more useful polymer materials.
We have investigated the addition of sulfonic ester group to the ring of
norbornene. The subsequent sulfonation will produce a new PEM material for DMFC.
The synthetic procedure is shown in Scheme 3.7.
SO
3
Na
PCl
5
, CHCl
3
, 35
o
C
SO
2
Cl
C H
2
CH
SO
2
Cl
+
R
R
O H
Et
2
O, NEt
3
, 25
o
C
C H
2
CH
SO
3
CHR
2
C H
2
CH
SO
3
CHR
2
+
SO
3
CHR
2
Toluene, 150
o
C, 14 hrs
SO
3
CHR
2
ROMP
SO
3
CHR
2
n
SO
3
H
n
Hydrolysis
R = -CH
3
, -CF
3
Scheme 3.7 Scheme of synthesis of norbornene-based PEM
144
Ru
Ph
Cl
P(Cy)
3
P(Cy)
3
Cl
Grubbs' catalyst
Dicyclopentadiene
Norbornene
Figure 3.12 Structures of Grubb’s catalyst, dicyclopentadiene and Norbornene
Poly(vinylsulfonyl chloride) was synthesized following the procedures described
before. Isopropanol or hexafluoropropanol was used to prepare the corresponding
vinylsulfonic ester.
S
O
O O
CF
3
F
3
C
S
O
O O
CH
3
C H
3
(a) (b)
Figure 3.13 Structures of (a) FPHS and (b) PHS
Diels-Alder reaction was carried out with toluene as the solvent, when
cyclopentadiene and vinylsulfonic ester was mixed in a pressure tube. Cyclopentadiene
normally exist as dimer at room temperature. Dicyclopentadiene has to be cleaved in a
retro-Diels-Alder reaction at 160
o
C to obtain cyclopentadiene. With isopropyl and
145
hexafluoro isopropyl vinylsulfonic esters, we got two corresponding norbornene-based
esters: (a) 2,2,2-Trifluoro-1-(trifluoromethyl)ethyl bicyclo[2.2.1]hept-5-ene-2-sulfonate
(FPHS) and (b) Isopropyl bicyclo[2.2.1]hept-5-ene-2-sulfonate (PHS).
Two methods were used for the hydrolysis of poly(FPHS). In the first method
methanol/HCl (volume ratio 1/1) solution was used. After 14 hours reflux, only 10% of
the sulfonate groups were hydrolyzed. The second method involved reflux in
methanol/KOH (volume ratio 1/1) solution for 4 hours followed by methanol/HCl
(volume ratio 1/1) solution reflux for another 4 hours. The hydrolysis of poly (FBHS)
was complete using the second method.
The hydrolysis of poly (PHS) was much easier using the first method described
above. Ion exchange capacity test of this type of polymer resulted in an equivalent weight
of 1088, which is very close to the value of Nafion117. However, the physical properties
of these membranes are not satisfactory. Membranes became deformed in water at 80
o
C
due to the low melting point of the polymer. More work is needed to modify the
polymerization condition to increase the molecular weight of the polymer. At the same
time, increasing the crosslinking density could be another way to improve the physical
properties of the membranes.
146
3.13 Experimental
The synthetic procedure of the trifluoro-1-(trifluoromethyl)ethyl vinylsulfonate is
described below [2]:
1) 3 g (0.02 mol) of vinylsulfonyl chloride and 3.98 g (0.02 mol) hexafluoropropanol are
mixed together into 50 mL ethyl ether in a ice bath.
2) After 2.4 g (0.02 mol) triethylamine had been added drop wise into the mixture, the
ice bath was removed and reaction mixed was kept for 4 hours under vigorous stirring.
3) The precipitated white solid Et
3
NH
+
Cl
-
was filtered.
4) The organic ether layer was washed several times and evaporated to obtain the liquid
vinylsulfonate product.
Ring opening metathesis polymerization of norbornene-based sulfonate was
carried out using Grubbs’ catalyst [12]. The catalyst used was the first generation
Grubbs’ catalyst RuCl
2
(==CH-p-C
6
H
5
)(PCy
3
)
2
. Dicyclopentadiene (10%) was used as
crosslinker. After 14 hrs at room temperature, the reaction mixture was filtered through a
short column of silica gel followed by precipitation in methanol with vigorous stirring.
The resulting white polymer was washed several times with methanol and dried under
vacuum. The polymer obtained was dissolved in THF and made into a membrane using
147
solvent casting methodology. The subsequent hydrolysis of the membrane would give us
the new norbornene-based PEM for fuel cell studies.
3.14 Conclusions
Several new PEMs were synthesized and tested as candidates for DMFC
membranes. PVDF-poly(vinylsulfonyl chloride) membrane and PVDF-Poly trifluoro-1-
(trifluoromethyl)ethyl vinylsulfonate membrane showed some promising result in water
uptake test. However, the low proton conductivities of these membranes indicated that
insufficient monomer penetration in the inner parts of the membrane. Poly(sodium
vinylsulfonate) membrane is very difficult to prepare because of the high melting point of
the polymer. Kynar membranes exhibited promising electrical performance reaching our
desired baseline of 0.43 V at 100 mA/cm
2
. We are currently investigating Kynar
membranes of different thickness to get a better picture.
PVDF-poly(vinyl sulfonic acid tetrabutylammonium salt) membrane was easily
made but we still have to find the appropriate swelling agents to improve the PSSA
distribution in this kind of membrane.
PVDF-PSSA-poly(trifluoroethyl acrylate) membranes showed promising
performance at all different temperatures. PVDF-PSSA-poly(heptafluorobutyl acrylate)
148
membrane is still under investigation. However, with more fluorine moiety on side chain,
poly(heptafluorobutyl acrylate) might be a better compatibilizer than PMMA and
poly(trifluoroethyl acrylate).
The acid-doped PVDF-PVP membranes have great mechanical properties. With
modified methods to improve the acid penetration into the membrane, we could find a
PEM synthetic method without sulfonation step.
PVDF-PSSA-Poly4-ethenyl- α, α-bis(trifluoromethyl)benzeneethanol film has the
same synthetic advantage while showing the potential to have better water management
property. PVDF-PSSA-PVP membrane is a very promising PEM material. With a
comparable performance to PVDF-PSSA-PMMA membrane, better results are expected
with further modifications.
Norbornene based polymer electrolyte membrane is synthesized based on a
synthetic strategy coupled with the other membrane fabrication. The ring-opening
metathesis polymerization method could give novel PEM materials for use as DMFC
membranes.
149
3.15 Chapter 3 References
1. Blumstein, A., Kakivaya, S.R., Blumstein, R., Suzuki, T., Macromolecules 8 (4)
(1975) 435–437.
2. Mellor, J.M., Sagheera, A.H., Tamanyb, H. and Metwally, R.N., Tetrahedron 56
(51) (2000) 10067-10074.
3. Cascone, M.G., Lazzeri, L., Barbani, B., Cristallini, C., Polacco, G., Polym. Int.
41 (1) (1999) 17-21.
4. Roush, W.R., Gwaltney, S.L., Cheng, J., Scheidt, K.A., McKerrow, J.H. and
Hansell, E., J. Am. Chem. Soc. 120 (1998) 10994-10995.
5. Narita, T., Hagiwara, T., Hamana, H., Miyasaka, T., Wakayama, A., Hotta, T.,
Die Makromolekulare Chemie 188 (2) (2003) 273 – 279.
6. Skaat, H., Belfort, G. and Margel, S., Nanotechnology 20 (2009) 225106.
7. Ikkala, O., Pietilä, L.O., J. Chem. Phys. 116 (2002) 2417.
8. Itoh, T., Shiromoto, M., Inoue, H., Hamada, H., Nakamura, K., Tetrahedron Lett.
37 (28) (1996) 5001-5002.
9. Trnka, T.M., Grubbs, R.H., "The Development of L2X2Ru=CHR Olefin
Metathesis Catalysts: An Organometallic Success Story". Acc. Chem. Res. 34 (1)
(2001) 18–29.
10. Huang, J., Stevens, E.D., Nolan, S.P., Petersen, J.L., "Olefin Metathesis-Active
Ruthenium Complexes Bearing a Nucleophilic Carbene Ligand". J. Am. Chem.
Soc.121 (12) (1999) 2674-2678.
150
11. Hong S.H., and Grubbs R.H., "Highly Active Water-Soluble Olefin Metathesis
Catalyst". J. Am. Chem. Soc. 128 (11) (2006) 3508–3509.
12. Schwab, P., Grubbs, R., Ziller, J., J. Am. Chem. Soc. 118 (1) (1996) 100–110.
151
Chapter 4
Testing of alternative liquid fuels in fuel cells
Electro-oxidation of small organic molecules takes place on the anode of fuel
cells [1]. The energy released during this process depends on the type of the organic
molecules involved. The oxidation of methane involves a release of 8 electrons per
carbon atom while the oxidation of methanol is a 6-electron transfer process. Due to the
obvious ease of handling of methanol and ease of anodic oxidation, compared to that of
methane, there has been extensive research in the development of direct methanol fuel
cells (DMFCs) using noble metal catalysis [2-4]. Thus, methanol is the simplest organic
molecule that has been widely investigated for fuel cell applications. However, methanol
is toxic when ingested and therefore alternate nontoxic fuels have been considered for
fuel cell applications (Figure 4.1).
4.1 Crossover test of different fuels
As described before, Gas Chromatograph (GC) method was established to
monitor the methanol crossover. To screen the possible fuels, the same method was used
to check the permeation rate of different fuels through Nafion117 membranes.
152
Dimethyl carbonate (DMC) is a flammable clear liquid with a boiling point of 90
°C. It has been recently used as a methylating reagent. The main advantage over other
methylating reagents is its lesser toxicity and its biodegradability. Also, it is now
prepared from catalytic oxidative carbonylation of methanol with carbon monoxide and
oxygen, instead of phosgene, making its production non-toxic and more environmentally
friendly [5]. This property allows dimethyl carbonate to be considered a green fuel. It can
also be used as a cleaner-burning diesel substitute.
O
OO
CH
3
C H
3
Dimethyl carbonate
Dimethoxymethane
OO
CH
3
C H
3
Dimethyl oxalate
O
O
CH
3
O
C H
3
O
Trimethyl orthoformate
H
3
CO OCH
3
OCH
3
Dioxolane
OO
Glyoxal
O
O O
O O
O
O H
O H
OH
OH
Glyoxal trimer dihydrate
Pentaerythritol
OH
OH
O H
O H
Acetaldehyde
C H
3
O
Figure 4.1 Alternative organic fuels in fuel cell
153
The crossover test result is shown in table 4.1. Dimethyl carbonate has the
crossover coefficient value almost identical to that of methanol in crossover tests using
Nafion117 or PVDF-PSSA membranes.
Table 4.1 Crossover coefficient of DMC
Fuel Crossover coefficient (cm
2
/second)
Nafion 117 PVDF-PSSA
DMC 2.67E-07 5.73E-08
Methanol 2.94E-07 4.68E-08
Dimethoxymethane (methylal) is a colorless flammable liquid with a low boiling
point, low viscosity and an excellent dissolving power [6]. It has chloroform-like odor
and a pungent taste. Dimethoxymethane is the dimethyl acetal of formaldehyde.
Dimethoxymethane is miscible with most common organic solvents and soluble
in three parts water. It was produced by oxidation of methanol or by the reaction of
formaldehyde with methanol. It is hydrolyzed back to formaldehyde and methanol in
aqueous acid. Primarily used as a solvent, it was also used in the manufacture of
adhesives, perfumes, resins, paint strippers and protective coatings. As shown in Table
154
4.2, dimethoxymethane did not show superior crossover reduction compared to methanol
at similar testing conditions.
Table 4.2 Crossover coefficient of dimethoxymethane
Fuel Crossover coefficient (cm
2
/second)
Nafion 117 PVDF-PSSA
Dimethoxymethane 3.00E-07 4.05E-08
Methanol 2.94E-07 4.68E-08
Dimethyl oxalate (C
4
H
6
O
4
) is available as white crystals. It has good solubility in
alcohol and ether and decomposes in hot water. Dimethyl oxalate was mainly used to
make pure methanol, medicine and agriculture products. It is also a plasticizer [7].
Dimethyl oxalate shows greatly reduced crossover rate than methanol (Table 4.3). It
would be of interest to probe to the electrical performance of this particular fuel.
Table 4.3 Crossover coefficient of dimethyl oxalate
Fuel Crossover coefficient (cm
2
/second)
Nafion 117 PVDF-PSSA
Dimethyl oxalate 2.76E-08 1.47E-08
Methanol 2.94E-07 4.68E-08
155
Trimethylorthoformate [CH(OCH
3
)
3
] may be hydrolyzed (under acidic conditions)
to methylformate and methanol, and may be further hydrolyzed to formic acid and
methanol [8]. As expected, trimethylorthoformate did not show reduced crossover rate
compared to methanol as it is readily hydrolyzed to methanol in aqueous environments.
Table 4.4 Crossover coefficient of trimethyl orthoformate
Fuel Crossover coefficient cm
2
/second
Nafion 117 PVDF-PSSA
Trimethyl orthoformate 1.70E-07 3.05E-07
Methanol 2.94E-07 4.68E-08
4.2 Tracking crossover rates of certain fuels in fuel cells using CO
2
analyzer
CO
2
analyzer was used to track the crossover current density of glyoxal, glyoxal
trimetric dihydrate and 1,3-dioxolane. The methanol crossover current density is shown
in Figure 4.2 using Nafion117 as PEM.
Glyoxal is an organic compound with the formula CH
2
O
2,
which is the smallest
dialdehyde. Commercial glyoxal is prepared either by the gas phase oxidation of ethylene
156
glycol or by the liquid phase oxidation of acetaldehyde in nitric acid. Glyoxal is prepared
in the laboratory by oxidation of acetaldehyde with Selenious acid (H
2
SeO
3
).
Nafion03-4, 0.5 M methanol
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
0 50 100 150 200 250 300 350 400 450
I (mA/cm
2
)
Crossover Current Density (mA/cm
2
)
25 ℃
55 ℃
60 ℃
90 ℃
Figure 4.2 Methanol crossover density
Glyoxal was used as a crosslinker for starch-based polymers and also as a starting
material with urea for wrinkle-resistant chemical treatments. It is generally used as a
cross-linking agent in polymer chemistry. It is also a valuable building block in organic
synthesis, especially in the synthesis of heterocycles. It is evident from Figure 4.3 that
glyoxal has a little bit lower crossover rate than methanol (Figure 4.2) at different testing
temperatures.
157
Nafion03-4, 0.5 M Glyoxal
0.00
10.00
20.00
30.00
40.00
50.00
60.00
0 20 40 60 80 100 120 140 160 180 200
I (mA/cm
2
)
Crossover Current Density (mA/cm
2
)
25 ℃
55 ℃
60 ℃
90 ℃
Figure 4.3 Glyoxal crossover density
Nafion03-4, 0.1 M Glyoxal Trimer Dihydrate
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
0 5 10 15 20 25 30 35 40 45 50
I (mA/cm
2
)
Crossover Current Density (mA/cm
2
)
25 ℃
55 ℃
60 ℃
90 ℃
Figure 4.4 Glyoxal trimer dihydrate crossover density
158
Glyoxal is supplied as a 40% aqueous solution [9]. Glyoxal forms hydrates like
other small aldehydes. Furthermore, the hydrates can condense to give a series of
oligomers, the structures of which remain uncertain. However, the exact nature of the
species in solution is inconsequential for most applications. Glyoxal trimer dihydrate:
[(CHO)
2
]
3
(H
2
O)
4
can precipitate, due to lower solubility, from its solutions below 4
o
C
[10]. The crossover rate of glyoxal trimeric dihydrate is very low (Figure 4.4). However,
it can only sustain very low current density even at high operating temperatures.
Nafion03-4, 0.5 M 1,3-Dioxolane
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
0 50 100 150 200 250 300 350 400 450
I (mA/cm
2
)
Crossover Current Density (mA/cm
2
)
25 ℃
55 ℃
60 ℃
90 ℃
Figure 4.5 Dioxolane crossover density
159
Dioxolane or 1,3-dioxolane is an heterocyclic acetal with the chemical formula
C
3
H
6
O
2
. It is an analog of tetrahydrofuran with an additional ring oxygen atom and an
analog of the 6 member ring 1,3-dioxane [11]. The isomeric 1,2-dioxolane is classified as
a peroxide. Dioxolane can be used as a solvent and as a co-monomer in polyacetals.
When dioxolane is used in fuel cell, the crossover rate of dioxolane is even higher than
methanol at 90
o
C.
4.3 Electrical performance of Nafion117 using different fuels
Nafion117 was made into a standard MEA ( 8 mg/cm
2
Pt-Ru on anode and 8
mg/cm
2
Pt on cathode, 25 cm
2
active area) and tested with different fuels at different
temperatures.
First we tried pentaerythritol. Pentaerythritol is a white, crystalline polyol with the
formula C(CH
2
OH)
4.
It is a very useful building block for the preparation of
polyfunctionalized compounds such as the explosive PETN and pentaerythritol triacrylate.
Derivatives of pentaerythritol are generally components of varnishes, alkyl resins, PVC
stabilizers, oil esters and olefin antioxidants. However, when tested in fuel cell, the
electrical performance of pentaerythritol is not satisfactory. It can only sustain a current
density of 40 mA/cm
2
at 90
o
C.
160
With a very low concentration of glyoxal trimer dihydrate, we were able to obtain
some electrical performance at different temperatures (Figure 4.6). The OCV of the cell
was quite good, which was as expected considering the low crossover rate of this fuel.
Nafion03-4, 0.1 M Glyoxal Trimer Dihydrate
0
0.2
0.4
0.6
0.8
1
020 40 60 80
I (mA/cm
2
)
E (V)
25 ℃, 0.1 L/min oxygen
55 ℃, 0.1 L/min oxygen
90 ℃, 0.1 L/min oxygen
Figure 4.6 Cell performance of Nafion117 using glyoxal trimer dihydrate
When the fuel was changed to 0.3 M glyoxal, we saw some improvement at
different operating temperatures. However, due to the difficult oxidation at the anodic
side, the electrical performance of cells using glyoxal or its dimer as fuels was not very
good.
161
Nafion03-4, 0.3 M Glyoxal
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 20 40 60 80 100 120 140
I (mA/cm
2
)
E (V)
25 ℃, 0.1 L/min oxygen
55 ℃, 0.1 L/min oxygen
90 ℃, 0.1 L/min oxygen
Figure 4.7 Cell performance of Nafion117 using glyoxal as fuel
Nafion03-4, 0.5 M 1,3-Dioxolane
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 50 100 150 200 250 300
I (mA/cm
2
)
E (V)
25 ℃, 0.1 L/min oxygen
55 ℃, 0.1 L/min oxygen
90 ℃, 0.1 L/min oxygen
Figure 4.8 Cell performance of Nafion117 using 1,3-dioxolane as fuel
162
Dioxolane was a better fuel than glyoxal, as can be seen from Figure 4.8.
However, its relatively high crossover rate limited the further improvement in electrical
performance. Trimethylorthoformate gave the best result among the alternative fuels we
tested (Figure 4.9). At 55
o
C, a current density of 250 mA/cm
2
can be obtained using 1
L/min air under ambient pressure. The performance of trimethylorthoformate was
comparable to methanol probably because it can hydrolyze into methanol in aqueous
solution. We also explored the possibility of using acetaldehyde as a fuel [12]. Some
initial test result is shown in Figure 4.10.
55
o
C, 1 L/min O
2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400
I (mA/cm
2
)
E (V)
1 M Trimethylorthoformate
1 M Methanol
Figure 4.9 Cell performance of Nafion117 using trimethyl orthoformate
163
55
o
C, 1M acetaldehyde
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 20 406080
c. d. / mA cm
-2
Voltage / V
1 L/min O
2
0.1L/min O
2
Figure 4.10 Cell performance of Nafion117 using acetaldehyde
4.4 Performance evaluation of fuel cells with ethanol as a fuel
Ethanol is a two-carbon alcohol often produced from corn or sugar cane, and has
a higher octane rating than gasoline. Ethanol can be used as a fuel in gasoline engines
like other alcohols. With little or no modification of the engine, ethanol is blended with
gasoline and can be used in cars. As a mixture of about 10-percent ethanol and 90-percent
gasoline, gasohol will run an unmodified car.
Ethanol appears to be an attractive internal combustion fuel from various points of
view; because of its low toxicity, ease of synthesis and low cost. However, its production
on massive scale has resulted in increase in food prices. Although the electrooxidation of
164
ethanol has been investigated to some extent, studies of direct ethanol fuel cell (DEFC)
are sparse [13-15]. Since the performance of DMFC has been well understood and
documented, the performance of an alternate alcohol fuel such as ethanol is of some
significance. A comparison between the performance of DMFC and DEFC under similar
conditions was made. Oxygen reduction is the cathode reaction of DMFC and DEFC:
O
2
+ 4H
+
+ 4e
-
→ 2H
2
O (4.1)
The anode reaction of the DMFC and DEFC are shown in Eq.(4.2) and Eq.(4.3),
respectively.
CH
3
OH + H
2
O → CO
2
+ 6H
+
+6e
-
(4.2)
C
2
H
5
OH + 3H
2
O → 2CO
2
+ 12H
+
+12e
-
(4.3)
Although the complete oxidation of alcohol molecules is assumed in Eq. (4.2) and
(4.3), formation of several reaction intermediates is possible. Furthermore, the reactions
may terminate with the formation of an intermediate without proceeding to completion.
This possibility is more likely in the case of ethanol due to the presence of a C-C bond.
Cyclic voltammograms of a Pt foil electrode were recorded for oxidation of
methanol and ethanol in 0.5 M H
2
SO
4
as the supporting electrolyte at different
temperatures. The nature of the voltammograms of CH
3
OH oxidation is similar to the
data reported in the literature. A broad anodic peak appearing in the forward sweep
165
direction is due to oxidation of CH
3
OH and reaction intermediates, since the oxidation of
CH
3
OH to CO
2
proceeds through the formation of several intermediates adsorbed on the
electrode. There is an anodic peak present in the reverse direction of the potential sweep
also, which is attributed to the oxidation of adsorbed intermediate species.
Similar to the voltammograms of CH
3
OH, the voltammograms of C
2
H
5
OH also
contains oxidation current peaks during forward and reverse potential sweeps. However,
the peaks are broader and the current values are lower for C
2
H
5
OH compared to the
values of CH
3
OH. The voltammograms result suggests that the oxidation of C
2
H
5
OH
takes place on Pt with adsorbed reaction intermediates.
The polarization curves at different temperatures of DMFC are shown in Figure
4.11. The open-circuit voltage of the DMFC was in the range between 0.60 and 0.86 V,
and its values increases with temperature. On loading the cell initially with a current
density (c.d.) of 20 mA/cm
2
, there is a large voltage drop. At 90 °C, for instance, the
DMFC experiences a voltage drop of about 150 mV. At 25 °C, a voltage drop about 280
mV was observed. This voltage drop is attributed to the impact of the ohmic resistance of
the cell and polarization resistances of the electrodes. Since these resistances decrease
with an increase of temperature, the voltage drop decreases at higher temperatures.
166
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
1.00 1.50 2.00 2.50 3.00
log c. d. / mA cm
-2
Voltage / V
20
o
C
55
o
C 90
o
C
Figure4.11 Polarization curves at different temperatures of DMFC
The linearity between log c.d. and voltage suggests that the cell reactions occur
under activation control, thus obeying Tafel behavior. The effect of temperature is
evident in the polarization data. At a cell voltage of 0.2 V, the c.d. values are 174, 450
and 680 mA/cm
2
, respectively, at 25, 55 and 90 °C. These data suggest that the cell
reactions are faster by about 4 times at 90 °C, and 2.5 times at 55 °C compared to the data
of 25 °C.
The polarization data of the DEFC (figure 4.12) shows that the initial voltage drop
values are about 350, 300 and 200 mV at 25, 55 and 90 °C, respectively. The non-
linearity of log c.d. versus cell voltage suggests that the occurrence of the cell reaction is
167
under diffusion control throughout the current range. At a cell voltage of 0.2 V, the c.d.
values are 50, 125 and 230 mA/cm
2
respectively. These data indicate increasing reaction
rate with the operating temperature of DEFC also.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
1.00 1.50 2.00 2.50 3.00
log c. d. / mA cm
-2
Voltage / V
20
o
C
55
o
C
90
o
C
Figure4.12 Polarization curves at different temperatures of DEFC
A comparison of the performance of DMFC and DEFC was also made. In both of
the cells, the reaction conditions of the cathode are the same whereas the anode fuel is
changed. Therefore, the variation of the polarization data from DMFC to DEFC is
attributed to the change of fuel from methanol to ethanol.
168
0
5
10
15
20
25
30
35
40
45
50
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Voltage / V
η / %
20
o
C
55
o
C
90
o
C
Figure 4.13 Anode efficiency of DEFC compared to DMFC
In the temperature range 25-90 °C, the maximum current values are much smaller
for the DEFC than the DMFC. The anode efficiency (η) of DEFC taking DMFC as the
reference is defined as:
η= 100I
DEFC
/I
DMFC
(4.4)
A plot of η versus the cell voltage is plotted at several temperatures. At high cell voltage
values when the cell was loaded with low values of current, the value of η is lower.
However it increases with the decrease of the cell voltage.
The mechanism of oxidation of an organic molecule is very complex because of
formation of several reaction intermediates. The complete oxidations of methanol and
169
ethanol to CO
2
involve a transfer of 6 electrons per carbon atom and it is believed that the
transfer of electrons occurs in consecutive steps. A generally accepted mechanism of
methanol is as follows:
M + (CH
3
OH)
sol
→ M-(CH
3
OH)
ads
(4.5)
M-(CH
3
OH)
ads
→ M-(⋅CH
2
OH)
ads
[ or M-(CH
3
O⋅)
ads
] + H
+
+e
-
(4.6)
M-(⋅CH
2
OH)
ads
[M-(CH
3
O⋅)
ads
] → M-(⋅CHOH) [M-(CH
2
O⋅)
ads
] + H
+
+ e
-
(s)
(4.7)
M-(⋅CHOH) [M-(CH
2
O⋅)
ads
] → M-(⋅CHO)
ads
+ H
+
+ e
-
(4.8)
M-(⋅CHO)
ads
→ M-(⋅CO)
ads
+ H
+
+ e
-
(4.9)
M + H
2
O → M-(OH)
ads
+ H
+
+ e
-
(4.10)
M-(⋅CO)
ads
+ M-(OH)
ads
→ 2 M + CO
2
+ H
+
+ e
-
(4.11)
In the above reaction steps, M stands for the catalyst and free radicals formed are
in adsorbed state. Different species formed by steps (4.6) to (4.8) have been detected by
in situ infrared reflectance spectroscopy. The consecutive dissociative steps are widely
accepted, although the exact nature of the species formed in step (4.8) is still under
discussion. After step (4.8), (⋅CHO)
ads
species is spontaneously dissociated as in step (4.9)
producing (⋅CO)
ads
species on the electrode.
170
The strongly adsorbed CO species is identified as the main poisoning species
blocking the electrode active sites from further adsorption of intermediates formed during
methanol oxidation. The oxidation of water molecule producing (OH)
ads
species is shown
in step (4.10), and the formation of CO
2
by an interactive oxidation of (⋅CO)
ads
and
(OH)
ads
of adjacent sites on the electrode is shown in step (4.11). It is now believed that
the Pt-Ru alloy is a good catalyst for the electrooxidation of CH
3
OH, as the adsorption of
(⋅CO)
ads
occurs on the Pt sites and adsorption of (OH)
ads
on the Ru site making step (4.11)
occur fast.
Compared to many investigations on the methanol oxidation and its mechanism as
detailed above, only a few studies are reported on ethanol oxidation and its mechanism.
The spectroscopic studies have indicated the presence of acetaldehyde, acetic acid and
CO
2
during oxidation of ethanol. A strongly adsorbed reaction intermediate has been
considered to undergo cleavage of C-C bond producing CO
2
. Thus, there is uncertainty
on the nature of the intermediates for the complete electrooxidation of ethanol.
The DMFC and DEFC were subjected to constant current loading and the
voltages were monitored for 24 hours. The data are presented in Figures 4.14 and 4.15,
respectively, for DMFC and DEFC. The cell voltages were kept constant over 24hours at
different constant current densities.
171
0.45
0.5
0.55
0.6
0.65
0.7
0 5 10 15 20 25 30
Time / h
Voltage / V
20 mA cm
-2
40 mA cm
-2
60 mA cm
-2
80 mA cm
-2
100 mA cm
-2
(c) 90
o
C
Figure 4.14 Voltage vs. constant current of DMFC at 90
o
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0 5 10 15 20 25 30
Time / h
Voltage / V
20 mA cm
-2
60 mA cm
-2
100 mA cm
-2
(c) 90
o
C
Figure 4.15 Voltage vs. constant current of DEFC at 90
o
C
172
Using the data of Figures 4.14 and 4.15, specific energy (SE) values were
calculated on the basis of geometrical area of the membrane electrode assembly. The
variation of SE is shown in Figures 4.16 and 4.17 for DMFC and DEFC, respectively.
It was noticed that the specific energy produced with both DMFC and DEFC will
increase as operating temperatures increase. At low current densities, specific energy will
increase as operating currents increase. However, when the operating current was high
then the voltage will drop dramatically, resulting in a decrease of specific energy
produced at high operating currents.
0
200
400
600
800
1000
1200
1400
0 20 406080 100 120
c. d. / mA cm
-2
S. E. / mWh cm
-2
90
o
C
55
o
C
20
o
C
Figure 4.16 Specific energy vs. current density of DMFC
173
0
50
100
150
200
250
300
350
400
450
500
0 20 40 60 80 100 120
c. d. / mA cm
-2
S. E. / mWh cm
-2
90
o
C
55
o
C
Figure 4.17 Specific energy vs. current density of DEFC
4.5 Experimental section of DEFC evaluation
Pt-Ru alloy powder and Pt powder catalysts were purchased from Alfa Aesor,.
Methanol (Aldrich) and ethanol were of high purity. A solution of methanol or ethanol at
a concentration of 1M in de-ionized water was used as the fuel. High purity oxygen was
used as the oxidant.
Carbon paper (Troy) was cut into 5 cm × 5 cm size pieces for coating of the
catalysts. The catalyst powder (10%), Nafion suspension (10%) and water were taken in a
glass vial, and subjected to sonication for about 15 minutes to make a homogeneous
catalyst ink. The catalyst ink was brushed on the carbon paper and dried under a stream
174
of air at ambient temperature. Brushing and drying were repeated twice to get a loading
level of 8 mg/cm
2
of the catalyst. The Pt-Ru coated carbon paper electrode was used for
the anode, and the Pt coated electrode for the cathode. The Nafion 117 membrane was
treated in H
2
O
2
, and diluted H
2
SO
4
sequentially.
The membrane electrode assembly (MEA) was made by hot-pressing at 140
o
C. A
fuel cell was assembled by using carbon black current collectors, which had provisions
for circulation of fuel at the anode and oxygen gas at the cathode. The aqueous fuel
solution was held in a glass container of 2 L capacity surrounded by a heating coil. The
fuel was heated to the desired temperature using a temperature controller. The cell was
conditioned initially by passing currents of different magnitudes for a long duration till
the open-circuit voltage of the cell became a stable value around 0.7 V. For evaluation of
the two fuels, a single cell was employed and the measurements were carried out with
one fuel at a time. The cell was washed thoroughly by circulating hot de-ionized water
whenever the fuel was changed.
Polarization and constant current loading experiments were carried out by using a
fuel cell workstation. A. C. impedance spectra were recorded by using a Solartron
electrochemical system models 1260 and 1280 in combination. A Pt foil electrode was
175
used to record cyclic voltammograms of oxidation of CH
3
OH and C
2
H
5
OH in 0.5 M
H
2
SO
4
. A saturated calomel electrode (SCE) was used as the reference electrode.
176
4.6 Chapter 4 References
1. Scheijen, F.J.E., Beltramo, G. L.,
Hoeppener, S., Housmans, T.H.M., and Koper,
M.T.M., J. Solid State Electrochem. 12 (2008) 482-495.
2. Heinzel, A. and Barragán, V.M., J. Power Sources 84 (1) (1999) 70-74.
3. Liua, H., Songa, C., Zhanga, L., Zhanga, J., J. Power Sources 155 (2) (2006) 95.
4. Wasmusa, S. and Küve, A., J. Electroanal. Chem. 461 (29) (1999) 14-31.
5. Zhao, T., Han Y. and Sun, Y., Fuel Processing Technology 62 (2-3) (2000) 187.
6. Chetty, R., Scott, K., J. Power Sources 173(1) (2007) 166-171.
7. Ma, X., Gong, J., Wang, S., Gao, N., Wang, D., Yang X. and He, F., Catal. Comm.
5 (3) (2004) 101-106.
8. Bachman, G. B., Org. Synth. 2 (1943) 323.
9. Snyder, H.R., Handrick, R.G., Brooks, L. A. Org. Synth. 3 (1995) 471.
10. Whipple, E.B., J. Am. Chem. Soc. 90 (1970) 7183–7186.
11. Adams, R.D., Barnard, T.S. and Brosius, K., J. Organomet. Chem. 582, (2) (1999)
358-361.
12. Wittig, G., Hesse, A., Org. Synth. 6 (1988) 901.
13. Kordesch, K., Hackerb V. and Bachhiesl, U., J. Power Sources 96 (1) (2001) 200.
177
14. Lamy, C., Rousseau, S., Belgsir, E.M., Coutanceau, C., Leger, J.M., Electrochem.
Acta 49 (22-23) (2004) 3901-3908.
15. Vigier, F., Coutanceau, C., Hahn, F., Belgsir, E.M., Lamy, C., J. Electrochem.
563 (1) (2004) 81-89.
178
Chapter 5
Carbon-based solid acid: A catalyst for intramolecular
Friedel–Crafts acylation
Solid acids are useful for applications in various acid catalyzed reactions [1]. The
advantage of simplified processes using solid acids has attracted great deal of interest in
their development. A typical solid acid is an ion-exchange resin which is polymer having
sulfonic groups. For example, Nafion-H
®
, a perfluoroalkyl sulfonic acid polymer, has
been developed. However, its utility in synthetic reactions becomes limited when the
reaction temperature exceeds 200
o
C. We have now investigated a carbon-based solid
acid that can be readily obtained by carbonization and sulfonation of an aromatic
compound. Such carbon-based solid acids have attracted some attention due to their
inexpensiveness and high performance. Our research group has widely used solid acid
catalysts in many synthetic organic transformations. We have identified a carbon-based
acid and explored its potential as a proton conductive material in fuel cells.
Intramolecular Friedel-Crafts ring closure reactions of benzoic acids with various
substituents bearing phenyl group in the ortho position were catalyzed by a new carbon-
based solid acid we prepared. Cyclic ketones were prepared through intramolecular
179
Friedel-Crafts acylation in good to excellent yields (80-95%). No intermolecular
condensation and other side reactions were observed under the reaction conditions.
5.1 Preparation of cyclic ketones catalyzed by carbon-based solid acid
Cyclic ketones have considerable importance in the preparation of biologically
active target molecules as well as synthetically useful intermediates. For example,
cyclohexanone, a symmetrical cyclic ketone, is an important intermediate in the
production of nylon. Isophorone, derived from acetone, is an unsaturated, unsymmetrical
ketone which acts as the precursor to other polymers. Muscone, 3-methylpentadecanone,
is an animal pheromone.
There are many methods to prepare ketones. Among them, the most important
method probably involves oxidation of hydrocarbons in industry. For an example, billion
kilograms of cyclohexanone are produced annually by aerobic oxidation of cyclohexane.
Normally, such oxidations employ air or oxygen for the industrial processes. Such
reactions rely on a strong oxidant such as potassium permanganate or Cr (VI) compound
for specialized applications. Milder conditions such as use of the Dess-Martin reagent or
the Moffatt-Swern oxidation are commonly adopted in organic synthesis, with alcohols as
substrates.
180
Intramolecular ring closure reactions of benzoic acids with phenyl substituents
catalyzed by conventional acid systems are typically associated with problems due to
decomposition at high temperature, long reaction times, excess amounts of catalysts, and
difficulty in separation and recovery. Solid acids have served as efficient catalysts for the
petroleum refinery industry as well as many acid–catalyzed reactions, such as
esterification, hydration, hydrolysis, and Friedel-Crafts reactions [2-6]. Because many
solid acids are less corrosive, less hazardous, and easy to be separated and recovered,
they are more advantageous than conventional liquid acids as catalyst for Friedel-Crafts
acylation.
Industrial alkylation reactions have traditionally used either hydrofluoric acid or
sulfuric acid as a catalyst. However, both these processes have environmental and safety
concerns associated with acid spillage and waste disposal. Major oil companies have
attempted to develop solid acid catalysts as an alternative to the hazardous liquid acids.
Although some zeolites catalyze the alkylation reaction, they deactivate rapidly due to the
formation of low-volatile hydrocarbon substance that plug up the micropores and block
the active sites. With enhanced catalyst activity and longevity while retaining product
quality and process reliability, a solid acid catalyzed alkylation technology will clearly be
181
an environmentally superior alternative to conventional technology that employs liquid
acids.
Further studies have shown that Nafion-H is an efficient solid acid catalyst in
many organic synthetic reactions [7-11]. Nafion-H has been proven to be an effective
catalyst in the ring closure reaction of 2, 2’-diamino or 2, 2’-dihydroxy biphenyls
(Scheme 5.1) [8, 9] and intramolecular acylation of benzoic acids with phenyl
substituents [12]. However, this perfluoroalkanesulfonic acid resin is expensive and the
reaction temperature cannot exceed 200
o
C due to its decomposition.
Y
X X
Nafion-H
X = -NH
2
, -OH Y = -NH-, -O-
Scheme 5.1 Nafion-H catalyzed intramolecular cyclization
Recently, the synthetic procedure for a new carbon-based solid acid was reported
[13]. This new type of acid has a high density of sulfonic acid groups and we have found
that it is a highly active catalyst in the preparation of cyclic ketones. The synthetic
approach is simple and the resulting carbon-based solid acid is thermally stable and
182
insoluble in water and many organic solvents (ether, chloroform, benzene, hexane, and
THF).
HO
3
S
SO
3
H
SO
3
H
HO
3
S
SO
3
H
SO
3
H
SO
3
H
Figure 5.1 Proposed schematic structure of the carbon-based solid acid
The titration results showed that 4.9 mmol/g SO
3
H was attached to the polycyclic
aromatic carbons, which indicated that the active acid sites are 5 times higher than that of
commercial perfluorosulfonated ionomer Nafion117 (0.9 mmol/g SO
3
H).
The thermal stability of the carbon material was checked by thermogravimetric
analysis (TGA) under an air flow. The sample weight decreased slightly at the initial
heating stage because of the losing of water content, and then reached a plateau at 150 -
183
300
o
C. Further heating the sample resulted in a decrease in weight (Figure 5.2). This
indicates that the carbon-based solid acid is reasonably stable and can function at
temperatures as high as 300
o
C, which is much higher compared to
perfluoroalkanesulfonic acids.
0
20
40
60
80
100
120
0 200 400 600 800 1000
Temp
o
C
Weight (%)
Figure 5.2 Thermogravimetric analysis (TGA) of the carbon-based solid acid
This new carbon-based solid acid has been proven to be highly active in
catalyzing the esterification of acetic acid, hydrolysis of cyclohexyl acetate, and
hydration of 2,3-dimethyl-2-butene [11]. In all these reactions, the carbon-based solid
184
acid showed high catalytic activity compared to Nafion, niobic acid, or using the same
amount of concentrated sulfuric acid as a catalyst.
X
O
OH
X
O
Carbon-based solid acid
Dichlorobenzene
1a-g 2a-g (80%-95%)
X = -O-, -CO-, -CH
2
-,
-CH
2
-CH
2
-, etc.
Scheme 5.2 Carbon-based solid acid catalyzed intramolecular cyclization
Using the carbon-based solid acid, the intramolecular acylation reactions of the
benzoic acid derivatives were carried out by heating the mixture of the carboxylic acid
derivatives (1a-g) and the solid acid (40%) at 180
o
C for 3-6 hours (Scheme 5.2). 1,3-
dichlorobenzene was chosen as reaction solvent because it is a suitably deactivated high
boiling solvent, which lead to good reaction completion without any side reactions from
solvent interaction [12-15]. The reactions afforded the products in high yields (Table 5.1)
and no intermolecular acylation reaction was observed. The amount of catalyst was 40-
50% by weight compared to the substrate.
185
Table 5.1 Carbon-based solid acid catalyzed intramolecular acylation
Entry Starting Material (1a-g) Time (hrs) Yield (%) Product (2a-g)
a
O
O
OH
6 88
O
O
b
O
OH
3 91
O
c
O
OH
3 88
O
d
O
OH Cl
O
6 80
O
Cl
O
e
O
O
OH
3 92
O
O
f
O
OH
HOOC
3 95
O
HOOC
g
O
OH
6 91
O
186
5.2 Experimental section of cyclic ketones synthesis
All starting materials (Aldrich) were used directly without further purification.
The 1, 2-dichlorobenzene was dried overnight using molecular sieves. The process of the
reactions was followed by GC/MS (Thermo Finigan). NMR spectra data were recorded
on a 300 MHz Varian Unity 300 NMR spectrometer with CDCl
3
as solvent.
In a typical synthesis, 50 g naphthalene was heated in concentrated sulfuric acid
(200 mL) at 260
o
C under a flow of N
2
for 16 hrs. Excess sulfuric acid was removed by
vacuum distillation at 260
o
C. The resulting black solid carbon material was ground to
powder and was first washed with dichloromethane and then washed repeatedly in
boiling deionized water and filtered until sulfate ions were no longer detected in the
filtrate. In this reaction, naphthalene was found to undergo sulfonation then followed by
partial carbonization to yield a solid with a composition of CH
0.35
O
0.35
S
0.14
(Figure 5.1).
The general Procedure for preparation of cyclic ketones is described below:
2-Benzoyl benzoic acid 1a (500 mg, 2.4 mmol) and carbon-based solid acid (250 mg,
50%) were mixed together and kept under reflux in 50 mL 1,3-dichlorobenzene for 3
hours. The completion of the reaction was determined by GC/MS analysis. The resulting
solution was filtered to remove the solid acid, and the filtrate was evaporated under
187
vacuum. The residue was recrystallized from dichloromethane and pentane mixture to get
yellow crystals of anthraquinone (m. p. 284-286
o
C).
5.3 Potential of carbon-based solid acid in fuel cell applications
At the beginning of this decade, there were predictions that the manufacture of
fuel cell vehicles would be well underway by mid decade signaling the inception of a
hydrogen era. However, cost issue and many unsolved technical problems hindered the
realization of these dreams. Fuel cells encounter special problems which have no proven
easy solutions. For example, the PEM type of fuel cell, generally seen as the most
promising technology, is still too expensive in designated markets and suffers from
various problems associated with manufacture, lifespan and reliability.
Solid acids are a recently described as chemicals intermediate between salts and
acids in their characteristics. At somewhat elevated temperatures in the range of 300 to
400
o
C, such substances become “superprotonic” permitting proton migration to cathode.
Nafion is used in PEM fuel cells with similar properties. But Nafion tends to
degrade at temperatures over 160
o
C. Nafion is also vulnerable to carbon monoxide
poisoning and requires ultra high purity hydrogen gas in fuel cell applications.
Furthermore, Nafion must be continually hydrated with deionized water and this will add
188
the cost of practical use. Nafion is used in most PEM fuel cells applications today.
Despite all the limitations, it is still considered to be the best available material by most
manufacturers.
The recently developed carbon-based acid is a solid acid material which has
shown some potential in fuel cell applications. We have already shown that the active
acid sites inside the polycyclic aromatic sulfonic acid are 5 times that of commercial
perfluorosulfonated ionomer Nafion117. Specific efforts were made to produce a thin
carbon-based acid membrane for fuel cell testing. This solid acid membrane possesses all
of the desirable characteristics long sought in a fuel cell electrolyte. It operates at medium
to high temperatures and is highly economic. Furthermore, it is physically and chemically
stable and doesn’t need complex hydration system. With all these advantages, we force a
promising future in the development of the carbon-based acid fuel cell material.
189
5.4 Chapter 5 References
1. Barton, D.G., Shtein, M., Ryan D., Wilson, R.D., Stuart, L., Soled, S.L. and
Iglesia, E., J. Phys. Chem. B, 103 (1999) 630-640.
2. Okuhara, T., Chem Rev. 102 (2002) 3641.
3. Misono, M., Nojiri, N., Appl. Catal. 64 (1990) 1.
4. Armor, J.N., Appl. Catal. 78 (1991) 141.
5. Tanabe, K., Hoelderich, W.F., Appl. Catal. A: Gen. 181(1999) 399.
6. Grieco, P.A., Ed.; Organic Synthesis in Water; Blackie Academic and
Professional: Glasgow, Scotland, 1998.
7. Olah, G.A., Arvanaghi, M., Kirshnamurthy, V.V., J. Org. Chem. 48 (1983) 3359.
8. Yamato, T., Hidershima, C., Prakash, G.K.S., Olah, G.A., J. Org. Chem. 56 (1991)
3192.
9. Yamato, T., Hidershima, C., Prakash, G.K.S., Olah, G.A. J. Org. Chem. 56 (1991)
6248.
10. Olah, G.A., Shamma, T., Prakash, G.K.S., Catal. Lett. 46 (1997) 1.
11. Hachoumy, M., Mathew. T., Tongo, E.C., Vankar, Y.D., Prakash, G.K.S., Olah,
G. A., Synlett (1999) 363.
12. Olah, G.A., Mathew, T., Farnia, M., Prakash, G.K.S. Synlett 7 (1999) 1067.
13. Hara, M., Yoshida, T., Takagaki, A, Takata, T., Kondo, J.N., Hayashi, S., Doman,
K., Angew. Chem. Int. Ed. 43 (2004) 2955.
190
14. Bakkeren, F.J.A.D., Schroer, F., Klunder, A.J.H., Zwannenburg, B., Tet. Lett. 39
(1998) 9531.
15. Mathew, T., Keller, M., Hunker, D., Prinzbach, H., Tet. Lett. 37 (1996) 4491.
191
Bibliography
1. Adams, R.D., Barnard, T.S. and Brosius, K., J. Organomet. Chem. 582, (2) (1999)
358-361.
2. Alberti G., Casciola M., Annu. Rev. Mater. Res. 33 (2003) 129–54.
3. Alberti G., Casciola M., Solid State Ionics 145 (2001) 3–16.
4. Appleby, A.J., “The Electrochemical Engine for Vehicles.” Scientific American
(1999) 74.
5. Appleby, A.J., Foulkes, F.R., Fuel Cell Handbook, Krieger Publishing Co.,
Malabar, Florida, 1993.
6. Armor, J.N., Appl. Catal. 78 (1991) 141.
7. Asensio, J.A., Borros, S. and Gomez, R., Proton conducting polymers based on
benzimidazole and sulfonated benzimidazoles, J. Polym. Sci. Part A: Polym.
Chem. 40 (2002) 3703–3710.
8. Atti, A.R., Ph. D. Dissertation, University of Southern California (2000) 109-112.
9. Bachman, G. B., Org. Synth. 2 (1943) 323.
10. Bacon, F.T. and Fry, T.M., "Review Lecture: The Development and Practical
Application of Fuel Cells." Proceedings of the Royal Society of London, Series A,
Mathematical and Physical Sciences 334 (1973) 431.
11. Bae, J.M., Honma, I., Murata, M., Yamamoto, T., Rikukawa, M. and Ogata,
N.,Properties of selected sulfonated polymers as proton-conducting electrolytes
for polymer electrolyte fuel cells, Solid State Ionics 147 (2002) 189–194.
12. Bahar, B., Hobson, A.R., Kolde, J.A., Zuckerbrod, D., Ultrathin integral
composite membrane, US Patent, 5,547,551 (1996).
192
13. Bakkeren, F.J.A.D., Schroer, F., Klunder, A.J.H., Zwannenburg, B., Tet. Lett. 39
(1998) 9531.
14. Barbir F., Gómez T., Int. J. Hydrogen Energy 22 (1997) 1027–37.
15. Barton, D.G., Shtein, M., Ryan D., Wilson, R.D., Stuart, L., Soled, S.L. and
Iglesia, E., J. Phys. Chem. B, 103 (1999) 630-640.
16. Bashir, H., Linares, A. and Acosta, J.L., Heterogeneous sulfonation of blend
systems based on hydrogenated poly butadiene-styrene block copolymer.
Electrical and structural characterization, Solid State Ionics 139 (2001) 189–197.
17. Baur, E. and Preis, H., Zeitschrift für Elektrochemie. 43 (1937) 727.
18. Baur, E. and Tobler, J., Zeitschrift für Elektrochemie 39 (1933)169.
19. Borchers, W., "Direct production of electricity from coal and combustible
gasses." Translation, Electrical Review (London), 35 (1894) 887.
20. Blomen, L., and Mugerwa, M., Fuel Cell Systems. New York: Plenum Press,
(1993).
21. Blum, A., Duvdevani, T., Philosoph, M., Rudoy, N., Peled, E. J. Power Sources
117 (2003) 22.
22. Blumstein, A., Kakivaya, S.R., Blumstein, R., Suzuki, T., Macromolecules, 8 (4)
(1975) 435–437.
23. Bockris J.O’M.,and Srinivasan, S., Fuel Cells: Their Electrochemistry, McGraw-
Hill, New York (1969).
24. Bouchet, R., Miller, S., Deulot, M. and Sonquet, J.L., A thermodynamic approach
to proton conductance in acid-doped polybenzimidazole, Solid State Ionics 145
(2001) 69–78.
193
25. Bozkurt, A. and Meyer, W.H., Proton conducting blends of poly(4-vinylimidazole)
with phosphoric acid, Solid State Ionics 138 (2001), 259–265.
26. Breeze, P., Power Generation Technologies: Evaluating the Cost of Electricity.
London: Financial Times Energy, (1998).
27. Buchi, F., Gupta, B., Haas, O. and Scherer, G., Study of radiation grafted FEP-g-
polystyrene membranes as polymer electrolytes in fuel cells, Electrochim. Acta 40
(1995) 345–353.
28. Carretta, N., Tricoli, V. and Picchioni, F., Ionomeric membranes based on
partially sulfonated poly(styrene): synthesis, proton conduction and methanol
permeation, J. Membr. Sci. 166 (2000) 189–197.
29. Carrette L, Friedrich K.A., Stimming U., Fuel Cells 1 (2001) 1–34.
30. Carter, R., Wycisk, R., Yoo, H., and Pintauro, P.N., Blended
polyphosphazeneypolyacrylonitrile membranes for direct methanol fuel cells,
Electrochem. Solid State Lett. 5 (2002) A195–A197.
31. Cascone, M.G., Lazzeri, L., Barbani, B., Cristallini, C., Polacco, G., Polym. Int.
41 (1) (1999) 17-21.
32. Chetty, R., Scott, K., J. Power Sources 173(1) (2007) 166-171.
33. Costamagna, P., Srinivasan, S., Quantum jumps in the PEMFC science and
technology from the 1960s to the year 2000 Part I. Fundamental scientific aspects,
J. Power Sources 102 (2001) 242–252.
34. Dornheim, M.A., "Helios breakup review." Aviation Week & Space Technology
(2004) 59.
35. Douglas, D.L., “Advances in Basic Sciences: 1. Fuel Cells.” Electrical
Engineering (1959).
194
36. Douglas, D.L. and Liebhafsky, H.A., “Fuel Cells: History, Operation and
Applications.” Physics Today. (1960) 26.
37. Dyer, C.K., “Replacing the Battery in Portable Electronics.” Scientific American
(1999) 88.
38. Editor. "The Jacques Carbon Generator." The Electrical Review 38 (1896) 826.
39. Eikerling, M., Kornyshev, A.A., Stimming, U., J. Phys. Chem. B 101 (1997)
10807.
40. Elmore, G.V. and Tanner, H.A., "Intermediate Temperature Fuel Cells." J.
Electrochem. Soc. (1961).
41. Fairley, P. and Scott, A., "Competing to Fuel the Fuel Cells." Chemical Week.
New York (1999).
42. Fellows, L., "British Fuel Cell Runs a Lift Truck" The New York Times (1959) 6.
43. Finsterwalder, F. and Hambitzer, G., Proton conductive thin films prepared by
plasma polymerization, J. Membr. Sci. 185 (2001) 105–124.
44. Flint, S.D., Slade, R.C.T., Solid State Ionics 97 (1997) 299-307.
45. Glipa X., Bonnet, B., Mula B., Jones, D.J., Rozière, J., J. Mater. Chem. 9 (1999)
3045–49.
46. Ge, J., Liu, H., J. Power Sources 142 (2005) 56.
47. Genies, C., Mercier, R. and Sillion, B., et al., Stability study of sulfonated
phthalic and naphthalenic polyimide structures in aqueous medium, Polymer 42
(2001) 5097–5105.
48. Genies, C., Mercier, R., Sillion, B., Cornet, N., Gebel, G. and Pineri, M., Soluble
sulfonated naphthalenic polyimides as materials for proton exchange membranes,
Polymer 42 (2001) 359–373.
195
49. Gierke, T.D., Hsu, W.Y., In: Eisenberg, A., Yeager, H.L., Editors, Perfluorinated
Ionomer Membranes, ACS Symposium Series No. 180, American Chemical
Society, Washington, DC (1982) 283.
50. Gowariker, V.R., Vishwanathan, N.V. and Sridhar, J. Polm. Sci. New Age
International, New Delhi (1999).
51. Gregorio, R., Cestari, M., J. Polym. Sci.: Part B: Polymer Phisics, 32 (1994) 859-
870.
52. Grieco, P.A., Ed.; Organic Synthesis in Water; Blackie Academic and
Professional: Glasgow, Scotland, 1998.
53. Grot, W.G., Discovery and development of Nafion perfluorinated membranes,
Chem. Ind. 19 (1985) 647.
54. Grottuss, S.G., Work in Progress.
55. Grove, W.R., “On a Gas Voltaic Battery.” Proceedings of the Royal Society (1843)
268-78, 346-54, 422-432.
56. Grove W.R., “On a Gaseous Voltaic Battery.” Philos. Mag. And J. Sci. 21 (1842)
417-20.
57. Grove, W.R., “On Voltaic Series and the Combination of Gases by Platinum.”
Philos. Mag. And J. Sci. 14 (1839) 127.
58. Grove, W.R., “On a new Voltaic Combination.” Philos. Mag. And J. Sci. 13 (1838)
430.
59. Grubb, W. T., J. Electrochem. Soc. 106 (1959) 275-278.
60. Guo, X., Fang, J., Watari, T., Tanaka, K., Kita, H. and Okamoto, K., Novel
sulfonated polyimides as polyelectrolytes for fuel cell application. 2. Synthesis
and proton conductivity of polyimides from 9, 9-bis (4-aminophenyl)fluorene-2,7-
disulfonic acid, Macromolecules 35 (2002) 6707–6713.
196
61. Hachoumy, M., Mathew. T., Tongo, E.C., Vankar, Y.D., Prakash, G.K.S., Olah,
G. A., Synlett (1999) 363.
62. Hacker, B.C. and Grimwood, G.M., On the Shoulders of Titans: a history of
Project Gemini. NASA: Washington, DC, (1977).
63. Hamnett, A. Catal. Today 38 (1997) 445.
64. Hara, M., Yoshida, T., Takagaki, A, Takata, T., Kondo, J.N., Hayashi, S., Doman,
K., Angew. Chem. Int. Ed. 43 (2004) 2955.
65. Hasiotis, C., Deimede, V. and Kontoyannis, C., New polymer electrolytes based
on blends of sulfonated polysulfones with polybenzimidazole, Electrochim. Acta
46 (2001) 2401–2406.
66. Haubold, H.G., Vad, T., Jungbluth, H., Hiller, P., Nano structure of Nafion: a
SAXS study, Electrochim. Acta 46 (2001) 1559–1563.
67. Heinzel, A. and Barragán, V.M., J. Power Sources 84 (1) (1999) 70-74.
68. Hietala, S., Koel, M., Skou, E., Elomaa, M., Sundholm, F., J. Mater.Chem. 8(5)
(1998) 1127-1132.
69. Hodgdon, R.B. and Boyack, J.R., J. Polymer Sci.: Part A 3 (1965) 1463.
70. Hong S.H., and Grubbs R.H., "Highly Active Water-Soluble Olefin Metathesis
Catalyst". J. Am. Chem. Soc. 128 (11) (2006) 3508–3509.
71. Hofmann, M.A., Ambler, C.M. and Maher, A.E., et al., Synthesis of
polyphosphazenes with sulfonimide side groups, Macromolecules 35 (2002)
6490–6493.
72. Huang, J., Stevens, E.D., Nolan, S.P., Petersen, J.L., "Olefin Metathesis-Active
Ruthenium Complexes Bearing a Nucleophilic Carbene Ligand". J. Am. Chem.
Soc.121 (12) (1999) 2674-2678.
197
73. Hunt, L.B., “The first fuel cell.” Platinum Met. Rev. (1978) 22-43.
74. Ikkala, O., Pietilä, L.O., J. Chem. Phys. 116 (2002) 2417.
75. Itoh, T., Shiromoto, M., Inoue, H., Hamada, H., Nakamura, K., Tetrahedron Lett.
37 (28) (1996) 5001-5002.
76. Jacoby, M., "New Fuel Cells Run Directly on Methane." Chemical & Engineering
News Washington; (1999).
77. James, P.J., McMaster, T.J., Newton, J.M., Miles, M.J., In situ rehydration of
perfluorosulfonate ion-exchange membrane studied by AFM, Polymer 41 (2000)
4223–4231.
78. Kelley, S.C., Deluga, G.A., Smyrl, W.H. Electrochem Solid State Lett. 3 (2000)
407.
79. Kerres, J., Ullrich, A., Meier, F. and Haring, T., Synthesis and characterization of
novel acid–base polymer blends for application in membrane fuel cells, Solid
State Ionics 125 (1999) 243–249.
80. Ketelaar, J.A.A., "History." Leo J.M., Blomen, J. and Mugerwa, M.N., ed. Fuel
Cell Systems. New York: Plenum Press, (1993) 24.
81. Kim, K.J., Chao, Y.J., Kim, Y.H., Vib. Spectrosc. 9 (1995) 147-159.
82. Kordesch, K.V., "25 Years of Fuel Cell Development (1951-1976)." J.
Electrochem.l Soc. (1978) 77.
83. Kordesch, K., Hackerb V. and Bachhiesl, U., J. Power Sources 96 (1) (2001) 200-
203.
84. Kordesch, K., and Simader, G., Fuel Cells and Their Applications. New York:
VCH, (1996).
198
85. Kragh, H., "Confusion and Controversy: Nineteenth-Century Theories of the
Voltaic Pile." Bevilacqua F. and Fregonese, L., eds. Nuova Voltiana: Studies on
Volta and his Times 1 (2000) 133.
86. Kreuer, K.D., Chem. Mater. 8 (1996) 610-641.
87. Kreuer, K.D., Weppner, W., Rabenau, A., Angew. Chem. Int. Ed. Engl. 21 (1982)
208.
88. Lafitte, B., Karlsson, L.E. and Jannasch, P., Sulfophenylation of polysulfones for
proton conducting fuel cell membranes, Rapid Macromol. Commun. 23 (2002)
896–900.
89. Lamy, C., Rousseau, S., Belgsir, E.M., Coutanceau, C., Leger, J.M., Electrochem.
Acta, 49 (22-23) (2004) 3901-3908.
90. Larminie, J., Dicks, A. Fuel Cell Systems Explained, 2nd ed., Wiley: New York,
2003.
91. Lassegues, J.C., Grondin, J., Hernandez, M. and Maree, B., Proton conducting
polymer blends and hybrid organic inorganic materials, Solid State Ionics 145
(2001) 37–45.
92. Lee, J.H., Kim, S.C., Macromolecules 19 (1986) 644-648.
93. Liebhafsky, H.A., “Fuel Cells.” International Science and Technology (1962) 54.
94. Liebhafsky, H.A. and Cairns, E.J., “Hydrocarbon Fuel Cells: A Survey.”
Presentation to the American Institute of Electrical Engineers 1962 Pacific Energy
Conversion Conference (1962) 12-16. San Francisco (GE reprint DE-34).
95. Linden, D., Handbook of Batteries and Fuel Cells. New York: McGraw-Hill,
(1984).
199
96. Lipnizki, F., Hausmanns, S., Ten, P.K., Field, R.W. and Laufenberg, G.,
Organophilic pervaporation: prospects and performance, Chem. Eng. J. 73 (1999)
113–129.
97. Lischka, J.R. Ludwig Mond and the British alkali industry. New York: Garland,
(1985).
98. Liua, H., Songa, C., Zhanga, L., Zhanga, J., J. Power Sources 155 (2) (2006) 95-
110.
99. Lloyd, A.C., “The Power Plant in Your Basement.” Scientific American (1999) 80.
100. Lu, G.Q., Wang, C.Y., Yen, T.J., Zang, X. Electrochem. Acta 49 (2004) 821.
101. Ma, X., Gong, J., Wang, S., Gao, N., Wang, D., Yang X. and He, F.,
Catal.Commun. 5 (3) (2004) 101-106.
102. Magnet H.J.R., In: Handbook of Fuel Cell Technology, Berger, C., Editor,
Prentice-Hall, Englewood Cliffs, NJ, USA (1968) 425.
103. Maio, P., "Driving Force." Wall Street Journal (Eastern Edition). 234 (51) (1999)
8.
104. Mathew, T., Keller, M., Hunker, D., Prinzbach, H., Tet. Lett. 37 (1996) 4491.
105. Meier-Haack, J., Rieser, T., Lenk, W., Berwald, S., Lehmann, D., Build-up of
polyelectrolyte multilayer assemblies: a useful tool for controlled modification of
microfiltration and pervaporation membranes. In: Grassie, K., Tenckhoff, E.,
Wegner, G., Haußelt, J., Hanselka, H., Editors, Functional Materials Euromat ’99
vol. 13, Wiley-VCH (2000) 316–322.
106. Mellor, J.M., Sagheera, A.H., Tamanyb, H. and Metwally, R.N., Tetrahedron 56
(51) (2000) 10067-10074.
107. Mex, L. and Muller, J., Plasma-polymerized electrolyte membrane for
miniaturized DMFC, Membr. Technol. 115 (1999) 5–9.
200
108. Mijovic, J., Luo, H.L., Han, C. D., Polym. Eng. Sci. 22(4) (1982) 234.
109. Misono, M., Nojiri, N., Appl. Catal. 64 (1990) 1.
110. Mond, L and Langer, C., Proc. R. Soc. 46 (1889) 296.
111. Narayanan, S.R., Kindler, A., Jeffries-Nakamura, B., Chun. W., Frank, H., Smart,
M., Surampudi, S., Halpert, G., “performance of PEM liquid Feed Direct
Methanol-Air Fuel Cells”, Electrochem. Soc. Proceedings volume 95-23 1995) 61.
112. Narita, T., Hagiwara, T., Hamana, H., Miyasaka, T., Wakayama, A., Hotta, T.,
Die Makromolekulare Chemie 188 (2) (2003) 273 – 279.
113. Neburchilov, V., Martin, J., Wang, H., Zhang, J., J. Power Sources 169 (2007)
221-238.
114. Nixon, W., "Back to the future." Amicus Journal 21 (2) (1999) 17.
115. Norbeck, J., Hydrogen Fuel for Surface Transportation. Warrendale, P.A.,
Society of Automotive Engineers, (1996).
116. Okada, T., Xie, G., Gorseth, O., Kjelstrup, S., Nakamura, N., Arimura, T., Effect
of water uptake and relative humidity on Nafion, Electrochem. Acta 43 (1998)
3741–3747.
117. Okuhara, T., Chem. Rev. 102 (2002) 3641.
118. Olah, G.A., Arvanaghi, M., Kirshnamurthy, V.V., J. Org. Chem. 48 (1983) 3359.
119. Olah, G.A., Mathew, T., Farnia, M., Prakash, G.K.S. Synlett 7 (1999) 1067.
120. Olah, G.A., Shamma, T., Prakash, G.K.S., Catalysis Lett. 46 (1997) 1.
121. Ostrovskii, D.I., Torell, L.M., Paronen, M., Hierala, S., Sundholm, F., Solid state
ionics 97 (1997) 315-321.
201
122. Ostwald, W., Electrochemistry: History and Theory. Translated by N. P. Date.
New Delhi: Amerind for the Smithsonian Institution and the National Science
Foundation, (1980).
123. Ostwald, W., Zeitschrift für Elektrochemie 1 (1894) 122.
124. Othman, R., Yahaya, A. H., Arof, J. K., J. appl. Electrochem. 32 (2002) 1347.
125. Paulson, L. D. Compute 36 (2003) 10.
126. Petermann, J., Gohill, R.M., J. Mater. Sci. 14 (1979) 2620.
127. Poppe, D., Frey, H., Kreuer, K.D., Heinzel, A. and Mulhaupt, R., Carboxylated
and sulfonated poly(arylene-co-arylene sulfone)s: thermostable polyelectrolytes
for fuel cell applications, Macromolecules 35 (2002) 7936–7941.
128. Qinfeng, L., Hjirker, H.A. and Bjerrum, N.J., Phosphoric acid doped
polybenzimidazole membranes, J. Appl. Electrochem. 31 (2001) 773–779.
129. Raistrick, I.D., in “Diaphragms, Separators, and Ion exchange Membranes”, Zee,
J.W., White, R.E., Kinoshita, K., Burney, H.S., Editors, P.172, The Electrochem.
Soc. Softbound Proceeding Series, Pennington, New Jersy, (1986).
130. Ravikumar, M.K., Shukla, A. K. J. Electrochem. Soc. 143 (1996) 2601.
131. Reed, C.J., "Gas Batteries." Electrical World 25 (1895) 419 and (1895) 482.
132. Rhodin, J.G.A., "Can the heat of combustion of coal be turned directly into
electric energy?" The Engineer (London). 142 (1926) 80.
133. Rikukawa M., Sanui K., Prog. Polym. Sci. 25 (2000) 1463–502
134. Roush, W.R., Gwaltney, S.L., Cheng, J., Scheidt, K.A., McKerrow, J.H. and
Hansell, E., J. Am. Chem. Soc. 120 (1998) 10994-10995.
202
135. Sakari, T., Takenaka, H., Wakabayashi, N., Kawami, Y. and Tori, K., Gas
permeation properties of SPE membranes, J. Electrochem. Soc. 132 (1985) 1328.
136. Samms, S.R., Wasmus, S. and Savinell, R.F., Thermal stability of protons
conducting acid doped PBI in simulated fuel cell environments, J. Electrochem.
Soc. 143 (1996), 1225.
137. Savadogo O., J. New Mater. Electrochem. Syst. 1 (1998) 47–66.
138. Scheijen, F.J. E., Beltramo, G. L.,
Hoeppener, S., Housmans, T. H. M., and Koper,
M. T. M., J. Solid State Electrochem. 12 (2008) 482-495.
139. Schuster, M., Meyer, W.H. and Wegner, G., et al., Proton mobility in oligomer-
bound proton solvents: imidazole immobilization via flexible spacers, Solid State
Ionics 145 (2001) 85–92.
140. Schwab, P., Grubbs, R., Ziller, J., J. Am. Chem. Soc., 118 (1) (1996) 100–110.
141. Scott, K., Taama, W., Cruickshank, J., J Appl Electrochem 28 (1998) 289.
142. Service, R. F., Science 296 (2002) 1222.
143. Service, R.F., "New Tigers in the Fuel Cell Tank." Science 288 (2000) 1955.
144. Skaat, H., Belfort, G. and Margel, S., Nanotechnology 20 (2009) 225106.
145. Smart, M., “Chemical and Electrochemical Oxidation of small organic
Molecules”, UMI Ann Arbor, MI, (1998).
146. Smitha, B., Sridha, S., Kahn, A. A. J. Membrane Science 259 (2005) 10-26.
147. Snyder, H.R., Handrick, R.G., Brooks, L. A. Org. Synth. 3 (1995) 471.
148. Soczka-Guth, T., et al., International Patent WO99/29763 (1999).
149. Soontrapa, K., Srinapawong, N., J. Sci. Res. Chula. Univ. 26 (2) (2001) 59–70.
203
150. Srinivasan, S., "Fuel Cells for Extraterrestrial and Terrestrial Applications." J.
Electrochem. Soc. (1989) 41.
151. Staff. "Cleaner and Quieter Electrical Generator Seen in Five Years" Wall Street
Journal (1959) 8.
152. Steiner, P. and Sandor, R., Polybenzimidazole prepreg: improved elevated
temperature properties with autoclave processability, High Perform. Polym. 3
(1991) 139–150.
153. Surampudi, S., Narayanan, S.R., Vamos, E., Frank, H., Hapert, G., Laconti, A.,
Kosek, J., Prakash, G.K.S., Olah, G.A., J. Power Sources 47 (1994) 377.
154. Susai, T., Kaneko, M., Nakatoa, K., Isono, T., Hamada, A., Miyake, Y.,
Optimization of proton exchange membranes and the humidifying conditions to
improve cell performance for polymer electrolyte membranes, Int. J. Hyd. Energy
26 (2001) 631–637.
155. Tanabe, K., Hoelderich, W.F., Appl. Catal. A: Gen. 181(1999) 399.
156. Tazi, B., Savadago, O., New cation exchange membranes based on Nafion,
Silicotungstic acid and thiophene, J. New Mater. Electrochem. Syst., in press (cf.
JMS 185 (2001) 3–27).
157. Trnka, T.M., Grubbs, R.H., "The Development of L2X2Ru=CHR Olefin
Metathesis Catalysts: An Organometallic Success Story". Accounts of Chemical
Research 34 (1) (2001) 18–29.
158. U.S. Department of Energy. Fuel Cells for the 21st Century: Collaboration for a
Leap in Efficiency and Cost Reduction. Morgantown, WV: US DOE, (1999).
159. Vigier, F., Coutanceau, C., Hahn, F., Belgsir, E.M., Lamy, C., J. Electrochem.
563 (1) (2004) 81-89.
160. Wakizoe, M., Velev, O. A., Srinivasan, S., Electrochim. Acta 40 (1995) 335.
204
161. Wang, F., Hickner, M., Kim, Y.S., Zawodzinski, T.A. and McGrath, J.E., Direct
polymerization of sulfonated poly(arylene ether sulfone) random (statistical)
copolymers: candidates for new proton exchange membranes, J. Membr. Sci. 197
(2002) 231–242.
162. Wang, J.J., Savinell, R.F., Wainright, J., Litt, M. and Yu., H., A H
2
/O
2
fuel cell
using acid doped polybenzimidazole as polymer electrolyte, Electrochim. Acta 41
(1996) 193–197.
163. Warshay, M, and Paul R.P., The fuel cell in space: yesterday, today and tomorrow.
NASA-TM-102366 E-5084 NAS 1.15:102366.
164. Wasmusa, S. and Küve, A., J. Electroanal. Chem. 461 (29) (1999) 14-31.
165. Wasmus, S., Valeriu, A., Mateescu, G. D., Tryk, D.A. and Savinell, R.F.,
Characterization of H
3
PO
4
-equlibriated Nafion 117 membranes using
1
H and
31
P
NMR spectroscopy, J. Membr. Sci. 185 (2000) 78–85.
166. Watkins, D., Dircks, K., Epp, D.A., in “Proceedings of 32
nd
Power Sources
Conference”, Cherry Hill, NJ, June9-12, 1986, The Electrochemical Society, Inc.,
P.590.
167. Wendel, C.H., the Allis-Chalmers Story. Sarasota, Florida: Crestline Publishing
Co., (1988).
168. Whipple, E.B., J. Am. Chem. Soc. 90 (1970) 7183–7186.
169. Williams, K.R., “Francis Thomas Bacon, (21 December 1904) – (24 May 1992)”
Biographical Memoirs of the Fellows of the Royal Society 39 (1994) 3-18.
170. Wittig, G., Hesse, A., Org. Synth. 6 (1988) 901.
171. Wright, A., Charles R. and Thompson. C., "Note on the Development of Voltaic
Electricity by Atmospheric Oxidation of Combustible Gasses and other
Substances." Proc. R. Soc. London, 46 (1889).
205
172. W.T.C. “The Union Carbide Fuel Cell System: A Brief Description.” (1966)
Union Carbide Corporation.
173. Xing, B., Savadogo, O., Electrochem. Commun. 2 (2000) 697–702.
174. Yaeger, E. "Fuel Cells." Science New Series. 134 (1961) 1178.
175. Yamato, T., Hidershima, C., Prakash, G.K.S., Olah, G.A., J. Org. Chem. 56 (1991)
3192.
176. Yamato, T., Hidershima, C., Prakash, G.K.S., Olah, G.A. J. Org. Chem. 56 (1991)
6248.
177. Yang, C.C., Lin, S.J., Fauvarque, J.F. J. Power Sources 101 (2001) 267.
178. Yang, D., Thomas, E.L., J. Mater. Sci. Lett. 3 (1984) 929.
179. Young, G.J., ed. Fuel Cells, 2 volumes. New York: Reinhold Publishing Corp., 1
(1959), 2 (1961). Both volumes reprint papers given at symposia in 1959 and
1961, respectively.
180. Zadowzinski, T.A., Davey, J., Valerio, J., Gottesfeld, S. The water content
dependence of electro-osmotic drag in proton conducting polymer electrolytes,
Electrochim. Acta 40 (1995) 297–302.
181. Zawodzinski, T.A., Derouin, C., Redzinski, S., Sherman, R.J., Smith, V.T.,
Springer, T.E., Gottesfeld, S., J. Electrochem. Soc. 140 (1993) 1041.
182. Zhao, T., Han, Y. and Sun, Y., Fuel Process. Technol. 62 (2-3) (2000) 187-194.
Abstract (if available)
Abstract
Fuel cell has long been viewed as a promising scientific technology for power generation in many applications. Polymer electrolyte membrane (PEM) is one of the key components which determine the electronic performance of a PEM fuel cell. Significant effort has been made in our research group to develop a suitable polymer electrolyte membrane in direct methanol fuel cells (DMFC). The concept of interpenetrating polymer network (IPN) was applied in the synthesis of poly(vinylidenefluoride)-poly(styrenesuflonic acid)-poly(methyl methacrylate) (PVDF-PSSA-PMMA) composite membranes. IPN technology retained the excellent physical and chemical properties of PVDF and insured a more uniformed distribution of PSSA moiety in the membranes. PMMA was introduced into IPN network as compatibilizer to reduce the interfacial tension between PVDF and polystyrene, Optimization of experimental condition led to great improvement of the mechanical properties of PEM duringthus alleviating phase separation problem often encountered in styrene-grafted membranes. PVDF-PSSA-PMMA membranes have been demonstrated to exhibit substantially low methanol crossover rates compared to state of art materials and high electrical performance in direct methanol applications.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Development of polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)
PDF
Blends of polystyrene sulfonic acid copolymers and polyvinylidene fluoride as polyelectrolyte membranes
PDF
Dynamics of direct hydrocarbon polymer electrolyte membrane fuel cells
PDF
PSSA-PVDF semi-IPN blends for direct methanol fuel cells
PDF
Electrocatalysts for direct liquid-feed fuel cells and advanced electrolytes for lithium-ion batteries
PDF
Development of high-performance polymer electrolyte membranes for direct methanol fuel cells
PDF
Studies on direct oxidation formic acid fuel cells: advantages, limitations and potential
PDF
Applications of advanced electrochemical techniques in the study of microbial fuel cells and corrosion protection by polymer coatings
PDF
Direct C−H arylation for the synthesis of conjugated polymers
PDF
Studies on direct methanol, formic acid and related fuel cells in conjunction with the electrochemical reduction of carbon dioxide
PDF
Design, dynamics, and control of miniature catalytic combustion engines and direct propane PEM fuel cells
PDF
Development of a family of semi-random multichromophoric polymers for application in organic solar cells
PDF
Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction
PDF
Design of nanomaterials for electrochemical applications in fuel cells and beyond
PDF
The modification of catalysts and their supports for use in various fuel cells
PDF
Influence of polymer structure on electronic properties and performance as donor materials in bulk heterojunction solar cells
PDF
Electron transfer capability and metabolic processes of the genus Shewanella with applications to the optimization of microbial fuel cells
PDF
Electrokinetic transport of Cr(VI) and integration with zero-valent iron nanoparticle and microbial fuel cell technologies for aquifer remediation
Asset Metadata
Creator
Yang, Bo
(author)
Core Title
Development of polymer electrolyte membranes for fuel cell applications
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry
Publication Date
11/11/2009
Defense Date
10/28/2009
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
DMFC,fuel cell,OAI-PMH Harvest,PEM
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Prakash, G.K. Surya (
committee chair
), Olah, George A. (
committee member
), Yen, Teh Fu (
committee member
)
Creator Email
boyang@usc.edu,boyangboyang@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m2732
Unique identifier
UC1183389
Identifier
etd-Yang-3360 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-274501 (legacy record id),usctheses-m2732 (legacy record id)
Legacy Identifier
etd-Yang-3360.pdf
Dmrecord
274501
Document Type
Dissertation
Rights
Yang, Bo
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
DMFC
fuel cell
PEM