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 polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)
(USC Thesis Other)
Development of polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DEVELOPMENT OF POLYSTYRENE SULFONIC ACID-POLYVINYLIDENE
FLUORIDE (PSSA-PVDF) BLENDS FOR DIRECT METHANOL FUEL CELLS
(DMFCS)
By
Ming Li
______________________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)
December 2014
Copyright 2014 Ming Li
ii
DEDICATION
To my wife, my son and my parents
iii
ACKNOWLEDGEMENTS
First and foremost, I would like to thank my advisor Professor Thieo Hogen-Esch for his
inexhaustible guidance and support throughout my studies at USC. It has been an honor to be his
Ph.D. student. He has inspired me, both professionally and personally, how research, good
experiment and hardworking is done. I appreciate all his contributions of time, ideas, guidance
and funding to make my Ph.D. experience productive and stimulating. The love, patience and
enthusiasm he has for research was motivational for me, leading me through the journey of my
Ph.D. pursuit. I would also like to express my great thanks for his guidance at key moments in
my work while also allowing me to work independently the majority of the time.
I am also deeply thankful for Professor Prakash, for his generosity, advice and strong
support throughout my research discovery in Direct Methanol Fuel Cells. Thank you for making
a lot of research opportunities possible and accessible for me.
The members of the research groups around me have contributed a lot to my professional
and personal time at USC. They have been a source of friendship as well as good advice and
collaboration. I am especially thankful for my group members: Dr. Janet Olsen, Ms. Merve
Yurdacan, Mr. Sergey N. Mukhin, Dr. Jingguo Shen, Dr. Victoria Punoiva, Dr. Dongqin Zhang.
I would also like to acknowledge the members from Dr. Prakash group: Dr. Bo Yang, who
helped set up the methanol fuel cell and prepare membrane electrolyte assembly using hot-press
during the early stage of this project; Dr. Alain Goeppert, for borrowing me the carbon dioxide
analyzer to measure membrane methanol crossovers; Mr. Marc Iuliucci and Mr. Dean Glass, for
thoughtful discussion of optimization of methanol fuel cell performances. I have also had the
pleasure working with numerous postdocs, summer and rotation students who have come
through the labs during my study at USC.
iv
I would also like to thank all the LHI colleagues and staff, for their continuous support in
every area of my academic and daily life. I would specially like to acknowledge the following
people: Dr. William Wilson, for helping me with thermogravimetric analysis experiments; Mr.
Ralph Pan, who helped to set up the Gas Chromatography for methanol permeability
measurements; Mrs. Jessy May, Mrs. Carole Philips, Mr. Robert Aniszfeld, and Mr. David
Hunter for all their help.
For this dissertation I would like to thank my screening and defense committee members:
Dr. Gupta, Dr. Benderskii, and Dr. Haiges for their time, support, encouragement and dedication
to academic excellence. I would also like to thank you for letting my screening and defense be
enjoyable moments, and for your brilliant comments and suggestions.
Last but not the least, I would like to thank my family for all their love and
encouragement. For my parents, Xiaoan and Maoyu, who raised me with a love of science and
supported me in all my pursuits. For my loving wife Ellen, who is always by my side and is my
strongest support throughout my research and study. And for my beloved son, thank you for
throwing me with Big Smiles everyday and always cheering me up. Thank You!
v
ABSTRACT
As direct methanol fuel cells (DMFCs) hold great promise for wide applications in
transportation, stationary and portable power sources, the study of fuel cell materials and their
contribution to the overall cost and performance are of considerable practical importance. One of
the key components that receives much research attention is the proton electrolyte membrane
(PEM). Perfluorosulfonic acid (PFSA) membranes such as Nafion
®
are considered to be the
current benchmark membranes for DMFCs. Despite of the merit of high proton conductivity and
stability, PFSA membranes suffer from high cost and excessive methanol crossover. Therefore,
alternative cost-effective and high performance materials are strongly desired. In this work, high
performance and cost-effective polymer electrolyte membranes (PEMs) based on tetra-
butylammonium poly(styrene sulfonate) (PTBASS)/PVDF polymer blends are developed. The
PTBASS/PVDF holds promise as a substitute membrane in DMFCs, and the methodology to
fabricate membranes in the present study also provides a tool to elucidate the complexities of the
structure-performance relationship in PEMs.
In Chapter I, the background of fuel cells, the working principle of DMFCs and the
polymer electrolyte membranes for DMFCs are reviewed and discussed. Various strategies
including sulfonation of aromatic polymer, impregnation of polyelectrolytes in inert polymers,
organic/inorganic composite, and polymer blending have been employed to develop alternative
membranes for DMFCs. In spite of some progress have been made in terms of reduction in
methanol crossover, development of a sturdy inexpensive substitute to perfluorosulfonic acid
membranes is yet to materialize.
In chapter II, a seemingly immiscible polystyrene sulfonic acid/polyvinylidene fluoride
(PVDF-PSSA) nano-composite homogeneous composite blend was synthesized by a “polymer
vi
camouflage” approach. Terpolymerization of an ionic liquid monomer, tetra-butylammonium
styrene sulfonate (BASS), with styrene and 4-chloromethyl styrene (CMS) is carried out with
free radical polymerization to give a terpolymer poly(BASS-S-CMS). Polymer blend membranes
consisting of poly(BASS-S-CMS) and PVDF was fabricated through solution-cast technique
followed by annealing, crosslinking and ion exchange, giving highly flexible, transparent and
tough PSSA/PVDF Nafion
®
-like membranes. Fuel cell relevant properties of PSSA/PVDF
membranes, such as the methanol permeability, water uptake, ion exchange capacity (IEC),
proton conductivity, membrane morphology and microstructure, and ultrafiltration property were
investigated. The relationship between membrane composition and property is systematically
studied. The experimental techniques including Gas Chromatography (GC), Transmission
Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), Energy Dispersive X-ray
spectroscopy (EDS), Thermogravimetric Analysis (TGA), UV-Visible Spectroscopy, and
Electrochemical Impedance Spectroscopy (EIS) are included.
In Chapter III, the electrochemical performance, methanol crossover, fuel utilization, and
fuel cell efficiency of PSSA/PVDF modular membranes in DMFCs were extensively measured
and analyzed. The effect of PSSA content on the fuel cell performance, methanol crossover and
fuel cell efficiency are addressed. The impact of various operation parameters including
temperature, methanol concentration, methanol and oxygen flow rate on the performance of
PSSA/PVDF in DMFCs were tested in both ambient oxygen and air environment. An
understanding of membrane property-performance relationship is examined and discussed. The
present PSSA/PVDF membranes exhibit comparable or superior electrochemical performance
compared to Nafion
®
-117 due to lower methanol crossover and higher proton conductivity. As a
vii
result, high fuel utilizations and overall fuel cell efficiency were achieved using PSSA/PVDF
membranes in DMFCs.
The polymer electrolyte membranes based on the polymer blend of PSSA/PVDF shows a
promising potential in DMFC applications with impressive economic advantages. The
methodology employed in membrane fabrication provide a general method for exploring many
other monomer, including ammonium salts of 4-vinyl sulfonic acid (VSA), 2-acrylamido-2-
methylpropane sulfonic acid) (AMPS), and 3-sulfopropyl methacrylate (SPM) as alternative
polymers electrolyte materials for DMFCs applications. The conclusions and outlook are
discussed in the last chapter.
viii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ........................................................................................................ iii
ABSTRACT ................................................................................................................................... v
LIST OF FIGURES ................................................................................................................... xiii
LIST OF SCHEMES ................................................................................................................... xx
LIST OF TABLES ...................................................................................................................... xxi
CHAPTER 1: Introduction ....................................................................................................... 1
1.1 Fuel cell backgrounds ............................................................................................................ 1
1.2 Direct Methanol Fuel Cells (DMFCs) ................................................................................... 4
1.2.1 Operation principle of the DMFC .................................................................................. 5
1.2.2 Thermodynamic background of DMFCs ........................................................................ 6
1.2.3 Polarization behavior of DMFCs .................................................................................... 7
1.3 Polymer electrolyte membranes .......................................................................................... 10
1.3.1 Polyperfluorosulfonic acid membranes ........................................................................ 11
1.3.2 Alternative proton exchange membranes ..................................................................... 13
1.3.2.1 PEMs prepared by post-polymerization sulfonation ............................................. 14
1.3.2.2 PEMs prepared by impregnation of polyelectrolytes. ........................................... 19
1.3.2.3 PEMs prepared by organic/inorganic composites ................................................. 21
1.3.2.4 PEMs based on polymer blends ............................................................................ 23
1.4 References ........................................................................................................................... 25
ix
CHAPTER 2: Membrane Fabrication and Characterization ............................................. 31
2.1 Sulfonated polystyrene (PS) and polyvinylidene fluoride (PVDF) polymer blend
membranes ................................................................................................................................. 31
2.1.1 Preparation of PS/PVDF polymer blend membranes ................................................... 32
2.1.2 Sulfonation of PS/PVDF membranes ........................................................................... 33
2.1.3 Ion exchange capacity (IEC) measurement .................................................................. 34
2.2 Poly(sodium 4-styrenesulfonate) (PSSNa)/PVDF polymer blend membrane .................... 36
2.2.1 Preparation of PSSNa/PVDF membranes by solution casting ..................................... 37
2.2.2 Preparation of PSSNa/PVDF membranes by in-situ polymerization of SSNa in PVDF
matrix ..................................................................................................................................... 38
2.3 Synthesis of polystyrene sulfonate (PSS) copolymer .......................................................... 40
2.3.1 Synthesis of ionic liquid monomer tetra-butylammonium styrene sulfonate (BASS) . 42
2.3.2 NMR analysis of monomer tetra-butylammonium styrene sulfonate (BASS) ............ 43
2.3.3 Terpolymerization of BASS, styrene, and 4-chloromethyl styrene (CMS) ................. 44
2.3.4 NMR analysis of terpolymer P(BASS-S-CMS) ........................................................... 46
2.3.5 Solubility of P(BASS-S-CMS) ..................................................................................... 47
2.4 P(BASS-S-CMS)/PVDF membranes fabrication through solution casting ........................ 48
2.4.1 P(BASS-S-CMS)/PVDF membranes cast from acetone at 30 °C (Method I) ............. 48
2.4.2 P(BASS-S-CMS)/PVDF membranes cast from DMF at 60 °C (Method II) ................ 51
2.4.3 P(BASS-S-CMS)/PVDF membranes cast from DMF at 160 °C (Method III) ............ 54
2.5 Membrane Characterization ................................................................................................ 57
2.5.1 Optical clarity measurements ....................................................................................... 57
2.5.2 Water uptake measurements ......................................................................................... 59
x
2.5.3 Ion exchange capacity (IEC) ........................................................................................ 61
2.5.4 Proton conductivity measurements ............................................................................... 62
2.5.5 Methanol permeability measurements .......................................................................... 65
2.5.6 Transmission Electron Microscopy (TEM) .................................................................. 70
2.5.7 Thermal stability measurements ................................................................................... 73
2.5.8 Scanning Electron Microscope (SEM) of PSSA/PVDF membranes ........................... 75
2.5.9 Energy Dispersive X-ray Electronic Spectroscopy (EDS) ........................................... 77
2.5.10 Chemical stability measurements ............................................................................... 79
2.5.11 Mechanical properties ................................................................................................ 80
2.5.12 Ultrafiltration properties of PSSA/PVDF membranes ............................................... 82
2.5.12.1 Water diffusion of PSSA/PVDF membranes ...................................................... 83
2.5.12.2 Ultrafiltration experiments of PSSA/PVDF membranes ..................................... 84
2.5.13 Reproducibility measurements ................................................................................... 88
2.6 References ........................................................................................................................... 90
CHAPTER 3: Electrical Performance ................................................................................... 95
3.1 Fuel cell performance of PSSA/PVDF membranes cast from acetone ............................... 95
3.1.1 Membrane electrolyte assembly fabrication ................................................................. 95
3.1.2 Electrical performance of MEA 12-20 ......................................................................... 97
3.1.3 Temperature effect on the electrical performance of MEA 12-20 ............................... 99
3.1.4 Methanol concentration effect on the electrical performance of MEA 12-20 ............ 100
3.1.5 Oxygen flow effect on the electrical performance of MEA 12-20 ............................. 101
3.2 Fuel cell performance of membranes cast from DMF with 20 wt% PSSA loading .......... 103
3.2.1 Membrane electrode assembly optimization .............................................................. 104
xi
3.2.2 Fuel cell performance of MEA 13-42 ........................................................................ 106
3.2.3 Temperature effect on electrical performance of MEA 13-42 ................................... 108
3.2.4 Methanol concentration effect on the electrical performance of MEA 13-42 ............ 110
3.2.5 Oxygen flow effect on the electrical performance of MEA 13-42 ............................. 112
3.2.6 Reproducibility of electrical performance .................................................................. 114
3.2.7 Fuel cell performance of MEA 13-42 in ambient air ................................................. 116
3.3 Fuel cell performance of PSSA/PVDF membranes with 25 wt% PSSA loading ............. 119
3.3.1 Effect of PSSA content on electrical performance of PSSA/PVDF membranes ....... 120
3.3.2 Electrical performance comparison with Nafion
®
-117 .............................................. 121
3.3.3 Effect of temperature on the performance of MEA 14-127 ....................................... 126
3.3.4 Impact of methanol flow rate on electrical performance of MEA 14-127 ................. 127
3.3.5 Electrical performance of MEA 14-127 in air ............................................................ 129
3.4 Methanol crossover analysis of PSSA/PVDF membranes ................................................ 132
3.4.1 Effect of temperature on methanol crossover ............................................................. 133
3.4.2 Effect of methanol concentration on methanol crossover .......................................... 135
3.4.3 Effect of PSSA uptake on the methanol crossover ..................................................... 138
3.5 Fuel utilization of PSSA/PVDF polymer blend membranes ............................................. 139
3.5.1 Temperature effect on fuel utilization ........................................................................ 141
3.5.2 Effect of methanol concentration on fuel utilization .................................................. 143
3.6 Fuel cell efficiency of PSSA/PVDF polymer blend membranes ...................................... 145
3.6.1 Comparison of fuel cell efficiency between Nafion
®
-117 and PSSA/PVDF membranes
............................................................................................................................................. 146
3.6.2 Effect of methanol concentration on fuel cell efficiency ........................................... 147
xii
3.6.3 Effect of temperature on fuel cell efficiency of PSSA/PVDF membranes ................ 151
3.7 Conclusions ....................................................................................................................... 155
3.8 References ......................................................................................................................... 159
Bibliography ............................................................................................................................... 160
xiii
LIST OF FIGURES
Figure 1.1 Schematic representation of a single fuel cell. .............................................................. 2
Figure 1.2 Summary of fuel cell types (reprinted from ref.
16
). ...................................................... 4
Figure 1.3 Schematic representation of the operating principle of a single direct methanol fuel
cell (DMFC) (adapted from ref.
32
). ................................................................................................. 6
Figure 1.4 Polarization curve illustrating contributions of different overpotentials in a DMFC. 10
Figure 1.5 Chemical structure of Nafion
®
and other polyperfluorosulfonic acid PEMs. ............. 12
Figure 1.6 Structures of commonly studied PEMs based on post-polymerization sulfonation. .. 18
Figure 2.1 Photograph of PS/PVDF polymer blend membranes prepared by solvent casting. .... 33
Figure 2.2 Photograph of sulfonated PS/PVDF membranes. ....................................................... 34
Figure 2.3 Effect of molecular weight on ion exchange capacity of SPS/PVDF membranes. ..... 36
Figure 2.4 Photograph of PSSNa/PVDF membranes prepared by solvent casting. ..................... 38
Figure 2.5 Photograph of PSSNa/PVDF membranes prepared by in-situ polymerization of SSNa
in PVDF. ........................................................................................................................................ 40
Figure 2.6
1
H NMR spectrum of TBASS. .................................................................................... 44
Figure 2.7
1
H NMR spectrum of terpolymer P(BASS-S-CMS). .................................................. 47
Figure 2.8 Photograph of a PSSA/PVDF membrane cast from acetone at room temperature. .... 51
Figure 2.9 Photograph of a PSSA/PVDF membrane cast from DMF at 60 °C. ........................... 53
Figure 2.10 Photographs of PSSA/PVDF membranes cast from DMF at 160 °C placed upon
USC logos. ..................................................................................................................................... 57
Figure 2.11 Optical transmittance measurement of a PSSA/PVDF membrane with 30 wt%
PSSA. ............................................................................................................................................. 58
xiv
Figure 2.12 Water uptake values of various PSSA/PVDF membranes with different PSSA
uptakes. .......................................................................................................................................... 61
Figure 2.13 PSSA loading effect on proton conductivity and IEC of PSSA/PVDF membranes. 65
Figure 2.14 Diagram of the apparatus used to evaluate methanol permeability. ......................... 67
Figure 2.15 Methanol concentration vs. time of through membranes with varying PSSA
contents. ......................................................................................................................................... 68
Figure 2.16 Dependence of methanol permeability on PSSA content for PSSA/PVDF
membranes. .................................................................................................................................... 69
Figure 2.17 Cluster-network model for the morphology of hydrated Nafion
®
. (adapted from
ref.
46
) .............................................................................................................................................. 71
Figure 2.18 Transmission Electron Microscopy (TEM) of membrane M20 stained with
Pb(NO
3
)
2
. ....................................................................................................................................... 72
Figure 2.19 TEM image of membrane M30 stained with Pb(NO
3
)
2
. ........................................... 73
Figure 2.20 Thermal degradation study of PSSA/PVDF polymer blend membranes. ................. 74
Figure 2.21 Scanning electron micrograph of the sulfonated polystyrene/PVDF blend membrane
compatibilized with PS-b-PMMA block copolymer (adapted from ref.
103
). ................................. 76
Figure 2.22 SEM image of membrane M30. ................................................................................ 77
Figure 2.23 EDS element mapping of sulfur on the surface of a PSSA/PVDF membrane. ......... 78
Figure 2.24 Element compositions on the surface of a PSSA/PVDF membrane coated with a thin
layer of gold. .................................................................................................................................. 79
Figure 2.25 Tensile tests of PSSA/PVDF polymer blend membranes with different PSSA
loading contents. ............................................................................................................................ 82
Figure 2.26 Stirred Cell used for ultrafiltration experiments. ...................................................... 84
xv
Figure 2.27 Photographs of dyed polystyrene sphere solution (left) and the filtrate (right) after
ultrafiltration using PSSA/PVDF membrane with 40 wt% PSSA. ............................................... 86
Figure 2.28 Comparison of transmittances of solutions before (blue) and after (red)
ultrafiltration. ................................................................................................................................. 86
Figure 2.29 Photographs of feed solutions and ultrafiltrates of NaCl titrated with AgNO
3
. ........ 87
Figure 2.30 Reproducibility demonstration of proton conductivity and methanol permeability of
samples fabricated with different precursors on different days. .................................................... 89
Figure 3.1 Schematic illustration of the structure of a membrane electrolyte assembly (MEA)
used in DMFCs. ............................................................................................................................. 97
Figure 3.2 Performance of MEA 12-20 in a direct methanol fuel cell at 90 °C with 2.0 M
methanol utilizing ambient oxygen at the cathode. ....................................................................... 99
Figure 3.3 Temperature effect on the polarization behavior of MEA 12-20. ............................. 100
Figure 3.4 Methanol concentration effect on the performance of MEA 12-20. ......................... 101
Figure 3.5 Oxygen flow effect on the electrical performance of MEA 12-20. .......................... 103
Figure 3.6 Performance of MEA 14-42 in a DMFC at 90 °C with 2.0 M methanol utilizing 0.02
L/min ambient oxygen at the cathode. ........................................................................................ 107
Figure 3.7 Temperature effect on the electrical performance of MEA 13-42 with 2.0 M methanol
and 0.02 L/min ambient oxygen at the cathode. .......................................................................... 109
Figure 3.8 Effect of temperature on the performance of Nafion
®
-117 in a DMFC with 2.0 M
methanol utilizing 0.02 L/min ambient oxygen at the cathode. .................................................. 110
Figure 3.9 Methanol concentration effect on electrical performance of Nafion
®
-117 at 90 °C
with 0.02 L/min ambient oxygen at the cathode. ........................................................................ 111
xvi
Figure 3.10 Methanol concentration effect on the electrical performance of MEA 13-42 at 90 °C
with 0.02 L/min ambient oxygen at the cathode. ........................................................................ 112
Figure 3.11 Oxygen flow effect on the electrical performance of Nafion
®
-117 at 90 °C with 2.0
M methanol utilizing ambient oxygen at the cathode. ................................................................ 113
Figure 3.12 Oxygen flow effect on the electrical performance of MEA 13-42 at 90 °C with 2.0
M methanol utilizing ambient oxygen at the cathode. ................................................................ 114
Figure 3.13 Reproducibility measurement of MEAs at 90 °C with the same PSSA uptake but
fabricated on different days with different polymer precursors. ................................................. 116
Figure 3.14 Performance of MEA 13-42 at various temperatures with 0.5 M methanol utilizing
0.10 L/min ambient air at the cathode. ........................................................................................ 117
Figure 3.15 Methanol concentration effect on electrical performance of MEA 13-42 at 30 °C
utilizing 0.10 L/min ambient air at the cathode. .......................................................................... 118
Figure 3.16 Methanol concentration effect on electrical performance of Nafion
®
-117 at 30 °C
utilizing 0.10 L/min ambient air at the cathode. .......................................................................... 118
Figure 3.17 Fuel cell performances of Nafion
®
-117 and MEA 13-42 at 30 °C with 2.0 M
methanol utilizing 0.10 L/min ambient air at the cathode. .......................................................... 119
Figure 3.18 Performance of MEAs 14-127 and 13-42 in DMFCs at 90 °C with 1.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 121
Figure 3.19 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 0.5 M methanol using
0.10 L/min ambient oxygen flow at the cathode. ........................................................................ 122
Figure 3.20 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 1.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode. ........................................................................ 123
xvii
Figure 3.21 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 1.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode. ........................................................................ 124
Figure 3.22 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 3.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode. ........................................................................ 125
Figure 3.23 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 4.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode. ........................................................................ 125
Figure 3.24 Effect of temperature on the performance of MEA 14-127 with 1.0 M methanol
utilizing ambient oxygen at the cathode. ..................................................................................... 126
Figure 3.25 Effect of temperature on the cell performance of MEA 14-127 with 4.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 127
Figure 3.26 Effect of methanol flow rate on the performance of MEA 14-127 at 90 °C with 1.0
M methanol utilizing 0.10 L/min ambient oxygen at the cathode. .............................................. 128
Figure 3.27 Performance of MEA 14-127 and Nafion
®
-117 at 60 °C with 0.5 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 129
Figure 3.28 Performance of MEA 14-127 and Nafion
®
-117 at 60 °C with 1.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 130
Figure 3.29 Performance of MEA 14-127 and Nafion
®
-117 at 60 °C with 2.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 131
Figure 3.30 Performance of MEA 14-127 and Nafion
®
-117 at 30 °C with 1.0 M methanol
utilizing 0.10 L/min ambient air at the cathode. .......................................................................... 132
Figure 3.31 Methanol crossover current density of MEA 13-42 and Nafion
®
-117 measured with
2.0 M methanol in the temperature range of 30-90 °C. ............................................................... 135
xviii
Figure 3.32 Effect of methanol concentration on the methanol crossover current density of
Nafion
®
-117. ................................................................................................................................ 136
Figure 3.33 Effect of methanol concentration on crossover current density of MEA 13-42. .... 137
Figure 3.34 Effect of methanol concentration on crossover current density of MEA 14-127. .. 137
Figure 3.35 Crossover current density vs. methanol concentration for MEAs 13-42, 14-127, and
Nafion
®
-117. ................................................................................................................................ 138
Figure 3.36 Effect of PSSA uptake on methanol crossover of PSSA/PVDF polymer blend
membranes. .................................................................................................................................. 139
Figure 3.37 Effect of PSSA content on the fuel utilization of PSSA/PVDF polymer blend
membranes. .................................................................................................................................. 141
Figure 3.38 Effect of temperature on the fuel utilization of MEA 13-42 and Nafion
®
-117 in a
DMFC with 1.0 M methanol. ...................................................................................................... 142
Figure 3.39 Fuel utilization of MEA 14-127 in a DMFC at various temperatures with 1.0 M
methanol. ..................................................................................................................................... 142
Figure 3.40 Effect of methanol concentration on the fuel utilization of Nafion
®
-117 at 60 °C. 143
Figure 3.41 Effect of methanol concentration on the fuel utilization of MEA 13-42 with 20 wt%
PSSA loading at 60 °C. ............................................................................................................... 144
Figure 3.42 Effect of methanol concentration on the fuel utilization of MEA 14-127 with 25
wt% PSSA loading at 60 °C. ....................................................................................................... 144
Figure 3.43 Fuel cell efficiencies of MEA 13-42, 14-127 and Nafion
®
-117 at 90 °C with 0.5 M
methanol utilizing 0.10 L/min ambient oxygen at the cathode. .................................................. 147
Figure 3.44 Effect of methanol concentration on the fuel cell efficiency of Nafion
®
-117 at 60 °C
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 149
xix
Figure 3.45 Effect of methanol concentration on the fuel cell efficiency of MEA 13-42 at 60 °C
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 150
Figure 3.46 Effect of methanol concentration on the fuel cell efficiency of MEA 14-127 at 60 °C
utilizing 0.10 L/min ambient oxygen at the cathode. .................................................................. 150
Figure 3.47 Effect of temperature on the fuel cell efficiency of Nafion
®
-117 in a DMFC with 0.5
M methanol utilizing 0.10 L/min ambient oxygen at the cathode. .............................................. 152
Figure 3.48 Temperature effect on the fuel cell efficiency of MEAs with 20 wt% PSSA. ........ 152
Figure 3.49 Effect of temperature on the fuel cell efficiency of MEA 14-127 with 0.5 M
methanol utilizing 0.1 L/min ambient oxygen at the cathode. .................................................... 153
Figure 3.50 Effect of temperature on the fuel cell efficiency of MEA 13-42 with 3.0 M methanol
utilizing 0.1 L/min ambient oxygen at the cathode. .................................................................... 154
Figure 3.51 Fuel cell efficiency of MEAs with 20 wt% PSSA and Nafion
®
-117 with 2.0 M
methanol utilizing ambient oxygen at the cathode. ..................................................................... 155
xx
LIST OF SCHEMES
Scheme 1.1 Electrochemical reactions in a direct methanol fuel cell. ............................................ 6
Scheme 1.2 Schematic illustration of fabricating semi-IPN membranes (adapted from ref.
75
). ... 21
Scheme 1.3 Schematic representation of different routes for the preparation of organic-inorganic
PEMs (adapted from ref.
79
) ........................................................................................................... 23
Scheme 2.1 Synthesis hydrophobic ionic liquid monomer tetra-butylammonium styrene
sulfonate. ....................................................................................................................................... 43
Scheme 2.2 Cross-linking of polystyrene copolymer through Friedel-Crafts reaction. ............... 45
Scheme 2.3 Schematic representation of terpolymerization of BASS, S, and CMS. ................... 46
Scheme 2.4 Schematic representation of crosslinking and protonation of terpolymer P(BASS-S-
CMS). ............................................................................................................................................ 50
xxi
LIST OF TABLES
Table 2.1 Water uptake and appearance of SPS/PVDF membranes. ........................................... 34
Table 2.2 Ion exchange capacity (IEC) of sulfonated PS/PVDF membranes. ............................. 35
Table 2.3 Characteristics of PSSA/PVDF membranes as function of PSSA content. .................. 60
Table 2.4 Oxidative stability of the blend membranes. ................................................................ 80
Table 2.5 Mechanical property of membranes with various PSSA loading. ................................ 82
Table 2.6 Ultrafiltration property of PSSA/PVDF membranes and Nafion
®
-117. ....................... 84
Table 3.1 Anode and cathode catalyst ink compositions for 25 cm
2
MEAs.
b
.............................. 97
Table 3.2 PSSA uptake and related properties of membrane 12-20 cast from acetone. ............... 98
Table 3.3 Improved composition of catalyst inks. ...................................................................... 105
Table 3.4 Properties of MEAs based on membranes cast from DMF with a 20 wt% PSSA
loading. ........................................................................................................................................ 106
1
CHAPTER 1: Introduction
1.1 Fuel cell backgrounds
Despite their modern high-tech aura, the concept of a fuel cell had been demonstrated in
the early 19th century by Humphrey Davy.
1
This was followed by pioneering work by the
scientist William Grove, who is generally credited with inventing the fuel cell in 1839.
2,3
Generally, fuel cells are electrochemical devices that directly convert chemical energy of fuels
through electrochemical reactions to electrical energy. As shown in Figure 1.1, fuel cells consist
of an electrolyte medium sandwiched between two electrodes: anode and cathode. The anode
facilitates electrochemical oxidation of fuels, while the cathode promotes electrochemical
reduction of oxidants. Ions generated during oxidation or reduction are transported from one
electrode to the other through the ionically conductive but electronically insulating electrolyte,
while the electrons generated at the anode during oxidation pass through the external circuit, thus
generating electricity, to the cathode and complete the reduction reaction.
A
2
Figure 1.1 Schematic representation of a single fuel cell.
In the past decade, the interest in the fuel cell technology has increased dramatically.
4-15
Earth environmental issues related with air pollution, green house effects and global warming are
the major driving forces. In contrast to the environmental issues and efficiency limitations
associated with thermal processes that are commonly used for producing energy from fossil
fuels, fuel cells have some major advantages, including higher efficiencies (not limited by Carnot
efficiency), absence of gaseous pollutants, such as nitrogen oxides and sulfur dioxide, as well as
striking simplicity and compactness. Fuel cells have potentially comprehensive applications
range from stationary power generation and power for transportation, down to portable power
sources to supply electronic devices, such as notebooks, cell phones and cameras.
11,15
Fuel cells are primarily classified according to the electrolyte material they used. After
many years of research and development, today there are various types of fuel cells under active
development including: 1. alkaline fuel cell (AFC); 2. polymeric-electrolyte-membrane fuel cell
3
(PEMFC); 3. phosphoric-acid fuel cell (PAFC); 4. molten-carbonate fuel cell (MCFC); and 5.
solid-oxide fuel cell (SOFC). The general operation principles of these fuel cells along with
electrolyte used, operating temperature, and electrode reaction are summarized in Figure 1.2.
16
Most fuel cell systems operate with hydrogen as a fuel. However, the production, storage
and distribution of hydrogen are challenging.
4
The fundamental disadvantage is the low storage
density, and hence high cost, of commercially available hydrogen storage means. No efficient
and practical method of storing hydrogen for fuel cell applications currently exists. State-of-the-
art hydrogen storage canisters store hydrogen as metal hydrides, however, the reversible storage
of hydrogen in metal hydrides, has been limited to relatively low achievable specific energy.
17
While liquefaction leads to a form of hydrogen that is potentially attractive for use in large fuel
cell system, the liquefying process itself requires considerable amounts of energy, and the
volumetric power density of liquid hydrogen is still low and accounts for only one-third of that
of gasoline.
4
Although less reactive compared to hydrogen, methanol is considered to be an alternative
fuel due to its high energy density, being easy to store, transport and distribute. As a convenient
and safe liquid (b.p.: 64.7 °C) at room temperature, methanol can be produced from other natural
resources (plants, natural gas, coal) or from the reduction of carbon dioxide with hydrogen.
4
In
comparison to other carbonaceous or alcoholic fuels, methanol also has advantages in terms of
both energy density and rate of electro-oxidation.
4
Figure 1.2 Summary of fuel cell types (reprinted from ref.
16
).
1.2 Direct Methanol Fuel Cells (DMFCs)
As a consequence of the issues and challenges associated with the use of hydrogen, in
recent years interest in other fuels for low temperature fuel cells has emerged, such as direct
methanol fuel cells (DMFCs). The direct methanol fuel cell is a fuel cell that runs directly on
liquid methanol without having to first convert it into hydrogen gas. Developed by USC in a
collaborative effort with Jet Propulsion Laboratory (JPL) of Caltech in 1990, DMFCs have draw
tremendous attentions for the past two decades due to their potential applications in
transportation and portable power sources.
18-31
Compared to other types of fuel cells, DMFCs
offer a number of advantages. First, DMFCs can be operated at low temperatures and use liquid
methanol as fuel, which eliminating the need for a fuel reformer. In addition, methanol as a
liquid fuel is easier to store, transport and distribute using the current infrastructure. Further
more, methanol as a fuel has a higher energy density compared to pressurized hydrogen gas, and
does not have many of the fuel storage issues typical of hydrogen fuel cells. Finally, DMFC
5
represents a new era of methanol economy, which may fundamentally alleviate our dependence
on fossil fuels in the future.
4
1.2.1 Operation principle of the DMFC
The general principle of DMFCs was shown in Figure 1.3. In the DMFC, the chemical
energy of methanol is directly converted to electricity through electrochemical reactions. The
center of a DMFC is a solid polymer electrolyte membrane (PEM), which serves as a proton
conductor as well as a fuel barrier and separator for the anode and the cathode. The catalyst
layers, typically containing platinum and ruthenium (Pt-Ru) (anode) or platinum (Pt) (cathode)
deposited on carbon paper, are bonded onto both sides of the polymer electrolyte membrane to
form so-called membrane electrolyte assemblies (MEAs). At the anode of a DMFC, liquid
methanol is oxidized and produces protons and electrons. The resulted protons cross the polymer
electrolyte membrane to the cathode while the electrons pass through the external circuit, thus
generating an electric current, and combine with protons and oxygen in the cathode to produce
water as byproducts. The anode, cathode, and overall reactions of a DMFC are summarized in
Scheme 1.1.
6
Figure 1.3 Schematic representation of the operating principle of a single direct methanol fuel
cell (DMFC) (adapted from ref.
32
).
Anode reaction: CH
3
OH + H
2
O ⟶ CO2 + 6 H
+
+ 6 e
-
Cathode reaction: 3/2 O
2
+ 6 H
+
+ 6 e
-
⟶ 3 H
2
O
Overall reaction: CH
3
OH + 3/2 O
2
⟶ CO
2
+ 2H
2
O
Scheme 1.1 Electrochemical reactions in a direct methanol fuel cell.
1.2.2 Thermodynamic background of DMFCs
In a fuel cell, operating at isothermal conditions, if the enthalpy energy of both anode and
cathode reactions could be fully converted into electric work, the enthalpic cell voltage U
H
,
obtained would be:
15,33-35
where z is the number of electrons involved in the electrochemical reaction, F is the Faraday
constant (96484.6 C/mol) and ΔH
R
is the overall reaction enthalpy at standard conditions.
7
However, according to the second law of thermodynamic, if an electrochemical cell
operates reversibly, there will be a variation of the system entropy (released heat). Thus, the
maximum electric work of an electrochemical cell is obtained from the Gibb’s free energy
variation, ΔG
R
,
and the maximum (reversible) fuel cell voltage, U
rev
, is obtained as follows:
16,34
Where T is the system absolute temperature and ΔS
R
is the variation of the system entropy at
standard conditions.
Since not all the fuel chemical energy in a DMFC is converted into electric work, the
thermodynamic fuel cell efficiency is limited by the fuel intrinsic properties. Therefore, the
maximum thermodynamic efficiency that can be achieved by a DMFC can be calculated from
the following equation:
15,16
1.2.3 Polarization behavior of DMFCs
In an ideal fuel cell, the cell voltage is independent on the current drawn.
36,37
In practice,
the cell voltage departs from the open circuit voltage (OCV) and depends strongly on the current
density. The difference between actual cell voltage at a given current density and the reversible
cell voltage can be attributed to the following primary sources: mixed potential at electrodes,
activation losses, ohmic losses and mass transport losses, as demonstrated in Figure 1.4.
33-38
The difference between measured OCV and the reversible voltage is associated with
mixed potential at electrodes. The mixed potential is caused by parasitic reactions, where
oxidation and reduction reactions are taking place simultaneously at the same electrode with no
8
net current.
37
This can be seen as an internal closed circuit at the electrode surface. One
particularly important cause of mixed potentials is the crossover of fuels through the electrolyte
from anode to cathode or vice versa, especially in DMFCs, in which methanol crossover is
known to be high.
The typically sharp initial voltage drop is associated to the activation of reactions
occurring at each electrode in the low current density region (< 100 mA/cm²). At the
electrode/electrolyte interface, a buildup of charge occurs forming the double-layer structure.
39,40
The voltage difference required to drive a given electrochemical reaction rate across the double-
layer interface is the source of the activation overpotential, which leads to a potential drop in the
polarization curve of a fuel cell. The activation overpotential at each electrode can be expressed
with the Tafel equation:
34
where ΔV
Activation
is the overpotential, i and i
0
are the current density and the exchange current
density, respectively and A is the Tafel slope, which provides information about the reaction
kinetics. According to Tafel equation, the loss of voltage increase logarithmically with increasing
current density. Activation losses can be alleviated by increasing of temperature, the surface area
of catalyst, the operation pressure, or by using a more effective catalyst.
The voltage loss in the moderate current density region is dominated by ohmic losses due
to the internal electronic and ionic resistances of the cell components, e.g., electrodes and
electrolyte. The extent of the ohmic losses depends on the current draw from the cell giving that
the resistance of a DMFC is a constant under a given condition. Therefore, a nearly linear
behavior of the polarization curve at intermediate current densities is reflected by Ohm’ law as
the following equation:
38
9
where R is the area specific resistance in Ω/cm². To minimize ohmic losses and increase the fuel
cell efficiency, membranes with high proton conductivity are desired. In addition, to facilitate
proton transfer, the catalyst layer should to be well bonded to the electrolyte to minimize the
interfacial resistance between electrode and electrolyte.
Mass transport losses are the primary source of overpotential at high current densities. At
higher current densities, the reactant consumption rates increases, causing depletion of reactants
at the catalyst surface, which leads to a reduction of reaction kinetics. The factors that limit
reactant availability include fuel diffusion limitation, inefficient removal of CO
2
from the anode,
and low supply of oxygen when air is used. In addition, for DMFCs, water flooding at the
cathode is also a common issue, which also restricts the gas access to the catalyst surface. The
limiting current (i
lim
) at high current densities is expressed:
35
in which z is the number of charges; F is Faraday constant; D is the diffusion coefficient; c
r
is the
reactant concentration in the bulk and δ is the diffusion distance. From the limiting current, the
mass transport losses (ΔV
mass
) can be calculated using the Nernst equation:
33
10
Figure 1.4 Polarization curve illustrating contributions of different overpotentials in a DMFC.
1.3 Polymer electrolyte membranes
As indicated by in the previous session, the centerpiece of DMFCs is a solid, polymeric
membrane that separates the anode from the cathode, prevents the fuel and the oxidant from
mixing, as well as serves as a proton conductor. The requirements for high performance polymer
electrolyte membranes (PEMs) include:
41-45
• High proton conductivity, but low or no electronic conductivity
• Low permeability to the fuel and the oxidant
• High oxidative and hydrolytic stability
• Necessary flexibility for fabrication into MEAs
• Good mechanical properties in both dry and hydrated states to ensure long term durability
• Reasonable cost compatible with intended applications
11
1.3.1 Polyperfluorosulfonic acid membranes
Perfluorinated sulfonic acid copolymers are the most widely studied and applied proton
exchange membrane for proton exchange membrane fuel cells (PEMFCs). In particular, Nafion
®
membranes, produced and marketed by DuPont, have served as benchmarks in the field of
polymer electrolyte membranes as they satisfies an array of above mentioned requirements for
good performance in fuel cells.
46
The proposed structure of Nafion
®
is shown in Figure 1.5.
47
Other perfluorosulfonate proton exchange membranes with similar structure have also been
developed, such as Dow XUS (Dow Chemical, USA), Flemion (Asahi Glass Company, Japan),
and Aciplex (The Asahi Kasei Chemicals, Japan).
47-49
A general drawback with all perfluorinated
proton exchange membrane is their cost due to the associated high costs of the perfluorinated co-
monomers.
41-44,48,50
The combination of super hydrophobic polymer backbones and super hydrophilic side
chains give Nafion
®
its inherent properties. The Teflon-like polymer backbone have
demonstrated high thermal and chemical stability under fuel cell operation conditions due to the
strong carbon-fluorine bonds, which are approximately 4 kcal/mol stronger than aliphatic
carbon-hydrogen bonds. The presence of perfluorinated sulfonic acid group at the side chains
endows Nafion
®
membranes with desirable proton conductivity when fully hydrated.
The microstructure of Nafion
®
membranes has been extensively investigated.
46,51-54
Among these are the influence and importance of phase separation or ionic cluster formation
within the Nafion
®
and similar perfluorosulfonic acid copolymers. These ionic aggregates within
Nafion
®
have been proposed to contribute to enhanced proton conductivity.
46
The flexible,
hydrophobic fluorinated backbone of the polymer is believed to promote aggregation of the
12
hydrophilic side chains containing the sulfonic acid groups during processing and/or conversion
to its sulfonic acid form.
46
Figure 1.5 Chemical structure of Nafion
®
and other polyperfluorosulfonic acid PEMs.
An additional important feature of Nafion
®
-like membranes is the presence of semi-
crystallinity.
55-57
Semi-crystalline domains in these presumed random copolymer are, in part,
responsible for the mechanical strength, water insolubility and relatively modest water swelling.
The amount of crystallinity is dependent on equivalent weight where the lower equivalent weight
(EW) polymers will necessarily have low crystallinity and, at some EW values, the crystallinity
will disappear entirely.
58
Despite their advantages, Nafion
®
and other polyperfluorosulfonic acid membranes suffer
from several shortcomings, which limit their utility and performance in DMFC appellations.
41-
45,47,48,50,59-62
One of the most pronounced problems of Nafion
®
membranes in DMFC
applications is their high methanol crossover rates under fuel cell operation conditions
41-45,47,59,60
as it is prone to both spontaneous and electro-osmotic drag of methanol crossover causing as
13
much as 40 percent methanol loss. The inherently high cost of Nafion
®
membranes (20-30% of a
DMFC cost) is an additional limitation, making their use in cost critical applications, such as fuel
cells for electrical vehicles, unlikely.
59
Some studies also indicated that Nafion
®
membranes
suffers from high ruthenium crossover, in which ruthenium crosses the membrane and re-
deposits on the cathode, resulting in decreasing the fuel cell performance.
63,64
As a consequence,
in recent years enormous effort has been put for the search of new polymeric materials with high
proton conductivity and low methanol permeability for the application in DMFCs.
45,49,65-68
1.3.2 Alternative proton exchange membranes
Due to the shortcomings of Nafion
®
and other polyperfluorosulfonic acid membranes,
there have been extensive research efforts, both academically and industrially, to develop
alternative low cost materials as proton exchange membranes, and to understand the complex
relationship between chemical structure, morphology, physicochemical properties and fuel cell
performance of membranes. Different strategies have been explored in order to obtain low-cost
hydrocarbon based polymer electrolyte membranes with properties of high proton conductivity
and low methanol crossover.
Polymer electrolyte membranes for DMFCs typically have a large hydrophobic content
(60-90 wt. percent) but also require the hydrophilic ionic component to provide proton
conductivity. This could be achieved by a number of synthesis methods including but not limited
to: (a) direct sulfonation of existing polymers, such as sulfonated poly(arylene ether)s,
69
polyimides,
70
polybenzimidazoles.
71,72
(b) impregnation of polyelectrolytes through the
polymerization of a precursor monomer in the bulk phase of another giving an interpenetrating
network (IPN) or semi-IPN followed by the functionalization (i.e. sulfonation).
73-75
(c) composite
14
membranes consisting of organic/inorganic components.
76-79
(d) Blends of carefully chosen
hydrophobic and hydrophilic polymers followed by later crosslinking if required.
1.3.2.1 PEMs prepared by post-polymerization sulfonation
Post-modification of hydrocarbon polymers are thought to be one of the promising routes
to alternative high performance PEMs because of their availability, process ability, wide variety
of chemical compositions, and anticipated stability in the fuel cell environment.
47
Although
sulfonated polyaryls have been demonstrated to suffer from hydroxyl radical initiated
degradation, it is found that sulfonated polyacryls to be durable under fuel cell conditions over
several thousand hours.
44,59,80
. There are several major drawbacks with the post-sulfonation
method including the lack of precise control over the degree and location of functionalization,
the inability to adjust molecular weight to enhance durability and optimize fuel cell performance,
the possibility of side reactions, or degradation of the polymer backbone. Regardless, this area of
PEM synthesis has received much attention and may be the source of emerging new PEM
products.
60
Sulfonated polymers including sulfonated polystyrene or derivatives, poly(arylene
ether)s, polyimides, polybenzimidazoles, etc. have been intensively investigated and
reviewed.
65,68
These aromatic polymers are easily sulfonated by concentrated sulfuric acid,
chlorosulfonic acid, pure or complex sulfur trioxide, and acetyl sulfate, with the degree of
sulfonation controlled by the reaction time, temperature and the concentration of sulfonation
agents. The structures of common sulfonated polymers are shown in Figure 1.6.
Sulfonated polystyrene (SPS) and derivatives are the basic materials for polymer
electrolyte membrane in fuel cells. These polymers are readily sulfonated by various sulfonation
15
agents mentioned above. Tricoli et al.
69
prepared ionomeric membranes based on partially
sulfonated polystyrene, and found out that SPS with 20 mol% of sulfonation exhibits proton
conductivity equal to that of Nafion
®
membranes and relatively lower methanol crossover. Elabd
et al.
81
characterized sulfonated poly(styrene-tributylene-styrene) triblock copolymer and
achieved high ion exchange capacities. However, these membranes deteriorated quickly in
hydrogen peroxide solution, indicating insufficient oxidative stability for use in fuel cells. And
the sulfonated triblock copolymer exhibits excessive water swelling (up to 351 wt%) and
mechanically instable as PEMs for fuel cell.
Among various aromatic polymers, poly(arylene ether) materials such as poly(ether ether
ketone) (PEEK), polysulfone and their derivatives are the focus of many recent investigations.
This family of copolymers is attractive for use in PEMs because of their well-known oxidative
and hydrolytic stability under fuel cell operation conditions, their readily availability and low
cost. The synthesis of sulfonated poly(arylene ether)s and their performance in a DMFC have
been widely reviewed by Kreuer, Kerres, Bauer, Roziere, McGrath and their co-worker and
others. Wang et al.
69
prepared sulfonated poly(ether ether ketone) (SPEEK) with various degrees
of sulfonation and claimed that the proton conductivity of SPEEK was exceeded 10
-2
S/cm above
80 °C and the methanol permeability was about an order of magnitude lower than that of
Nafion
®
-115. Guiver et al.
82
sulfonated commercial Victrex
®
and Gatone
®
PEEK. The
conductivity and the water swelling behavior were shown to be increased with increasing degree
of sulfonation and temperature. In view of the problems of SPEEK membranes, such as
brittleness at elevated temperatures, excessive water swelling, and relatively high methanol
crossover with a high degree of sulfonation all lead to insufficient durability. The further
16
modification of SPEEK, such as cross-linking or blending may provide space for the adaptation
of these polymers to particular fuel cell applications.
59
Sulfonated polybenzimidazoles (SPBI) also received a great amount of attention for both
as a proton exchange membrane candidate and also as a host for phosphoric acid, especially
given the success of unmodified PBI/phosphoric acid membranes in high temperature fuel cell
applications.
71
However, the post-sulfonation may induce scission or crosslinking of polymer
backbones, as evidenced by the insolubility and brittleness of sulfonated PBIs.
60
Polyimides, especially the six-membered rings of the naphthalenic polyimides exhibit
hydrolytic stability suitable for PEM fuel cells.
83,84
Preliminary study of sulfonated polyimides
suggested that they had some promise as PEM, however, their poor solubility limits membrane
formation and subsequent use in fuel cells. The introduction of unsulfonated diamines units
containing ether linkages or bulky substituents was shown to improve solubility.
60
Polyphosphazene-based polymer electrolyte membranes are potentially attractive
materials for both hydrogen and methanol fuel cells because of their reported chemical and
thermal stability, and the ease of various side chain attachment.
85-87
Studies by Pintauro and co-
workers
88
have shown that poly[(3-methylphenoxy)(phenoxy)phosphazene] and poly[bis(3-
methylphenoxy)phosphazene] can be sulfonated using SO
3
solution in dichloroethane. High ion
exchange capacity (> 2.0 mequiv/g) materials were reported with no detectable polymer
degradation.
PEMs prepared by direct sulfonation of exiting polymers in general possess a lower
methanol crossover than Nafion
®
, but their proton conductivities often depend on the sulfonation
degree.
43
In other words, the PEMs with high degrees of sulfonation exhibit high proton
conductivities which are comparable to or even higher than Nafion
®
membranes, but he
17
mechanical properties of these PEMs tend to deteriorate progressively with high degrees of
sulfonation due to excessive swelling in the hydrated state and brittleness in the dry state.
89
Therefore, further modification of sulfonated aromatic polymers is required, such as
crosslinking, blending with other inert polymers, reinforcement with porous supporting
materials, in order to obtain the desirable properties for DMFC applications.
18
Figure 1.6 Structures of commonly studied PEMs based on post-polymerization sulfonation.
19
1.3.2.2 PEMs prepared by impregnation of polyelectrolytes.
As polymer electrolyte membranes for DMFCs call for high ionic conductivity, low
methanol permeability as well as good mechanical stability, this multiple demand upon PEMs
limits the full exploitation of either ionic conductivity or membrane strength.
90
In order to
address this dilemma, the impregnation of polyelectrolytes (polystyrene sulfonic acid in most
cases) into swollen and porous hydrophobic inert polymers, such as polytetrafluoroethylene
(PTFE), fluorinated ethylene propylene (FEP), ethylene tetrafluoroethylene (ETFE),
polyvinylidene fluoride (PVDF), polyvinyl fluoride (PVF), perfluoroalkoxy alkane (PFA) and
others has been extensively investigated to reinforce polymer electrolyte membranes while
maintaining desirable proton conductivity.
75,91
Polymer electrolyte membranes with an interpenetrating polymer network (IPN) or semi-
IPN have been prepared by the impregnation of electrolyte monomers in porous substrates
followed by polymerization, crosslinking and subsequent sulfonation (Scheme 1.2). Prakash et
al.
92
studied a semi-IPN membrane by impregnation of styrene (monomer), divinylbenzene
(DVB) (crosslinker) and AIBN (initiator) in PVDF matrix followed polymerization and
sulfonation. Kalita et al.
93
prepared a different composite membrane utilizing a similar procedure
except the inert matrix was changed to polyvinylidene fluoride-co-hexafluoropropylene) (PVDF-
co-HFP). Kim et al.
74
prepared sulfonated polystyrene/PTFE composite membranes by
immersing a porous PTFE film into a mixture solution containing styrene, divinylbenzene
(DVB) and AIBN followed polymerization and sulfonation. The ion conductivity and methanol
permeability values of these interpenetrating polymer network membranes were reported to be
comparable or better than that of Nafion
®
-117, however, brittleness and lack of reproducibility
limit their use in DMFC applications.
20
In the contrast to the post-polymerization sulfonation method, membranes prepared by in-
situ polymerization of ionic monomers containing sulfonate groups within the polymeric matrix
have also been reported. Kundu et al.
94
prepared a polymer electrolyte membrane by in-situ
polymerization of sodium styrene sulfonate (SSNa) within the polymeric blend of PVDF-co-
HFP/Nafion
®
. The fuel cell performance of resulting membranes in a DMFC was recorded to
have a current density of 120 mA/cm² and a power density of 24 mW/cm² at 0.2 V and 60 °C.
Zhang et al.
95
also synthesized a proton exchange membrane by in-situ reaction and grafting of
SSNa to hydrogenated nitrile butadiene rubber (HNBR) during peroxide curing. The resulted
membrane exhibits a proton conductivity on the order of 0.01 S/cm and slightly lower methanol
crossover rates compared to Nafion
®
membranes. However, no membrane morphology and
electrical performance was reported.
Polymer electrolyte membranes with pore-filling architecture consisting of a porous
polyolefin substrate having pore sizes of one micron or less and a polyelectrolyte that fills in the
pores of the substrate were also reported. Yamaguchi et al.
96
developed a pore-filling membrane
with porous cross-linked high-density polyethylene (CLPE) as the substrate and poly(acrylamide
tert-butyl sulfonate sodium salt) (ATBS) as the filling polyelectrolyte. The resulting membrane
exhibited low methanol crossover and can be operated with high concentration of methanol. Choi
et al.
97
prepared a PEM by filling a porous substrate with a mixture of electrolyte monomer
(ethylene-sulfonic acid), crosslinker (N,N’-ethylenebisacrylamide), and photoinitiator (2-
hydroxy-2-methyl-1-phenylpropan-1-one), followed by UV light initiated polymerization and
crosslinking. The resulting membrane exhibited proton conductivity of 0.09 S/cm at room
temperature with stable fuel cell operation for 1000 h.
21
The reinforcement of polymer electrolyte membranes by impregnation of polyelectrolytes
into inert hydrophobic substrates offers a simple and efficient way to improve the dimensional
stability, lower methanol permeability. However, in-situ polymerization also has some major
drawbacks including poor control of molecular weight of polyelectrolyte in substrates
60
, adverse
change of the substrate morphology and mechanical property due to loss of crystallinity,
43
formation of voids and bubbles in membranes as a result of distinct difference of hydrophilicity
between substrates and polyelectrolytes, and finally poor control of reproducibility.
44
Scheme 1.2 Schematic illustration of fabricating semi-IPN membranes (adapted from ref.
75
).
1.3.2.3 PEMs prepared by organic/inorganic composites
Another strategy to reduce crossover is to use composite membranes. Inorganic fillers act
as blockers to methanol while maintaining the necessary proton conductivity. In addition, the
incorporation of the inorganic phase also enhances thermal and mechanical stability, which is of
great interest for high temperature fuel cells.
78,79
Several polymer electrolyte membranes have been prepared containing inorganic fillers,
for example, graphene/graphene oxide, silica, zirconium phosphate, phosphotungstic acid,
molybdophosphoric acid, silicon dioxide, silane-based filters, titanium oxide, hydroxyapatite,
22
laponite, montmorillonite, zeolites, and palladium.
77,78,98
Organic/inorganic composite system
can be prepared in various ways, such as blending, sol-gel, and in-situ polymerization, as
represented in Scheme 1.3.
79
Kim et al.
76
reported a hybrid membrane prepared by solution blending of sulfonated
poly(arylene ether sulfone) and 30 wt% phosphotungstic acid. The resulting membrane exhibited
a proton conductivity value of 0.15 S/cm at 130 °C at 100% relative humidity. Scott et al.
99
prepared a composite membrane consisting of poly(ethylene oxide) and 0.5 wt% graphene oxide
without any chemical modification. Polarization curves in a DMFC with this membrane gave a
maximum power density of 53 mW/cm² at 60 °C. However, the composite membrane was shown
to be brittle as the fractured elongation was less than 5%. Bjerrum et al.
100
studied phosphoric
acid doped polybenzimidazole (PBI) composite membranes containing various inorganic proton
conductors including zirconium phosphate (ZrP), (Zr(HPO
4
)
2
·nH
2
O) and phosphotungstic acid.
A higher conductivity was observed for the acid doped membranes containing 15-20 wt.% ZrP in
PBI than for pure PBI, exhibiting a proton conductivity of 7.9 x 10
-2
S/cm at 200 °C and 5%
relative humidity.
Hybrid composite membranes shows a great potential for applications in high
temperature fuel cells as they exhibits high proton conductivity and mechanical stability even at
high temperatures and low humidity. However, some challenges associated to composite
membranes have to be overcome prior to the application in DMFCs, such as poor water
management and brittleness due to high inorganic contents,
43,44
the aggregation or segregation of
particles due to heterogeneous particle dispersion, the leaching/bleeding of heteropolyacid due to
its dissolution in water.
47
23
Scheme 1.3 Schematic representation of different routes for the preparation of organic-inorganic
PEMs (adapted from ref.
79
)
1.3.2.4 PEMs based on polymer blends
Another strategy to obtain new materials for polymer electrolyte membranes in DMFC
applications is through polymer blending. Physical polymer blending has been a long-standing
tool to modify the properties of polymeric materials. Polymer blending can potentially combine
the desired properties from different polymers and alter certain properties.
101
In the field of
proton exchange membranes, polymer blending, in general, has been investigated to improve or
modify properties such as mechanical strength, water swelling behavior, methanol permeability,
and proton conductivity.
102
The choice of polymer blends for PEMs is based on the premise that
the hydrophobic polymer is chemically inert under the acid condition of use and acts as a
scaffold with desired thermo-mechanical strength and stability, while the polymer electrolyte
functions as a proton exchange material.
The wide selection of methods and materials and low processing costs seem to make
blending an attractive but challenging option. PEMs based on polymer blends of a hydrophobic
and a hydrophilic component typically produce materials with poor and non-reproducible
24
properties.
102
The inherent incompatibilities between between polymer components causes
micro- and macroscopic phase separation with the formation of large separated domains that, in
turn, give rise to poor mechanical properties as well as poorly controlled morphologies. For
instance, Wootthikanokkhan et al.
103
studied sulfonated polystyrene/PVDF polymer blend
membranes compatibilized with polystyrene-b-poly(methyl methacrylate) (PS-co-PMMA). The
resulting membranes exhibited pronounced coarse morphology with phase separation at the
micron level or larger, resulting in poor mechanical properties under a strain of less than 5
percent elongation. Seeponkai et al.
104
prepared a PEM based on polymer blends of sulfonated
poly[styrene-(ethylene-butylene)-styrene] triblock copolymer and PVDF compatibilized with
PS-co-PMMA. Again, the polymer blend yielded highly heterogeneous and hence inferior
membranes with weak mechanical strengths (tensile strength < 5.96 MPa) due, in part, to the
poor compatibility between SPS and PVDF. Wootthikanokkhan et al.
105
also studied the
methanol permeability and proton conductivity of DMFC membranes based on sulfonated
PEEK/PVDF blends and found that methanol permeability of membranes decreased along with
proton conductivity with increasing of PVDF content. No membrane morphology and electrical
performance of such polymer blend systems was reported.
It is well known that polymer blends with poor miscibility often end up with poor and
non-reproducible mechanical properties due to multiple factors including the absence of
polymer-polymer interaction or adhesion.
101
Although blending of low cost hydrocarbon
polymer acids and chemically inert polymers appear to be an effective method of fabricating
PEMs, highly compatible blend membranes with desirable properties is a major challenge.
102
25
1.4 References
(1) Stambouli, A. B.; Traversa, E. Renew Sust Energ Rev 2002, 6, 295.
(2) Appleby, A. J. Journal of Power Sources 1990, 29, 3.
(3) Shah, R. K. In Recent Trends in Fuel Cell Science and Technology; Basu, S., Ed.;
Springer New York: 2007, p 1.
(4) Olah, G. A. Angew Chem Int Ed Engl 2005, 44, 2636.
(5) Hoffert, M. I.; Caldeira, K.; Benford, G.; Criswell, D. R.; Green, C.; Herzog, H.;
Jain, A. K.; Kheshgi, H. S.; Lackner, K. S.; Lewis, J. S.; Lightfoot, H. D.; Manheimer, W.;
Mankins, J. C.; Mauel, M. E.; Perkins, L. J.; Schlesinger, M. E.; Volk, T.; Wigley, T. M. Science
2002, 298, 981.
(6) O'Brian, N.; Grübler, A.; Nakicenovic, N.; Obersteiner, M.; Riahi, K.;
Schrattenholzer, L.; Toth, F. Science 2003, 300, 581.
(7) Watt, G. D. Renewable Energy 2014, 72, 99.
(8) Edwards, P. P.; Kuznetsov, V. L.; David, W. I. F.; Brandon, N. P. Energy Policy
2008, 36, 4356.
(9) Turner, J. A. Science 1999, 285, 687.
(10) Pehnt, M. Int J LCA 2003, 8, 283.
(11) Dresselhaus, M. S.; Thomas, I. L. Nature 2001, 414, 332.
(12) Ashley, S. Sci Am 2005, 292, 50.
(13) Demirdoven, N.; Deutch, J. Science 2004, 305, 974.
(14) Chu, S.; Majumdar, A. Nature 2012, 488, 294.
(15) Carrette, L.; Friedrich, K. A.; Stimming, U. Chemphyschem 2000, 1, 162.
(16) Steele, B. C.; Heinzel, A. Nature 2001, 414, 345.
(17) Sakintuna, B.; Lamari-Darkrim, F.; Hirscher, M. International Journal of
Hydrogen Energy 2007, 32, 1121.
(18) Schultz, T.; Zhou, S.; Sundmacher, K. Chemical Engineering & Technology
2001, 24, 1223.
(19) Ren, X.; Zelenay, P.; Thomas, S.; Davey, J.; Gottesfeld, S. Journal of Power
Sources 2000, 86, 111.
26
(20) Dillon, R.; Srinivasan, S.; Aricò, A. S.; Antonucci, V. Journal of Power Sources
2004, 127, 112.
(21) Kamarudin, S. K.; Achmad, F.; Daud, W. R. W. International Journal of
Hydrogen Energy 2009, 34, 6902.
(22) Kamarudin, S. K.; Daud, W. R. W.; Ho, S. L.; Hasran, U. A. Journal of Power
Sources 2007, 163, 743.
(23) Wasmus, S.; Küver, A. Journal of Electroanalytical Chemistry 1999, 461, 14.
(24) McGrath, K. M.; Prakash, G. K. S.; Olah, G. A. J. Ind. Eng. Chem. 2004, 10,
1063.
(25) Narayanan, S. R.; Kindler, A.; Jeffries-Nakamura, B.; Chun, W.; Frank, H.;
Smart, M.; Valdez, T. I.; Surampudi, S.; Halpert, G.; Kosek, J.; Cropley, C. In Battery
Conference on Applications and Advances, 1996., Eleventh Annual 1996, p 113.
(26) Mitzel, J.; Arena, F.; Walter, T.; Stefener, M.; Hempelmann, R. Zeitschrift für
Physikalische Chemie International journal of research in physical chemistry and chemical
physics 2014, 227, 497.
(27) Yuan, W.; Zhou, B.; Deng, J.; Tang, Y.; Zhang, Z. C.; Li, Z. T. International
Journal of Hydrogen Energy 2014, 39, 6689.
(28) Dillon, R.; Srinivasan, S.; Arico, A. S.; Antonucci, V. Journal of Power Sources
2004, 127, 112.
(29) Garc’ia, B. L.; Weidner, J. W. Modern Aspects of Electrochemistry 2007, 40, 229.
(30) Li, X. L.; Faghri, A. Journal of Power Sources 2013, 226, 223.
(31) Hacquard, A. Worcester Polytechnic Institute 2005.
(32) Silva, V. S.; Mendes, A. M.; Madeira, L. M.; Nunes, S. P. In Advances in fuel
cells; Zhang, X., Ed.; Research Signpost: Kerala, India, 2005.
(33) Carrette, L.; Friedrich, K. A.; Stimming, U. Fuel Cells 2001, 1, 5.
(34) Winkler, W.; Nehter, P. In Modeling Solid Oxide Fuel Cells; Bove, R., Ubertini,
S., Eds.; Springer Netherlands: 2008, p 13.
(35) Sharaf, O. Z.; Orhan, M. F. Renewable and Sustainable Energy Reviews 2014, 32,
810.
(36) Pilatowsky, I.; Romero, R. J.; Isaza, C. A.; Gamboa, S. A.; Sebastian, P. J.;
Rivera, W. In Cogeneration Fuel Cell-Sorption Air Conditioning Systems; Springer London:
2011, p 25.
27
(37) Ota, K.-i. In Encyclopedia of Applied Electrochemistry; Kreysa, G., Ota, K.-i.,
Savinell, R., Eds.; Springer New York: 2014, p 918.
(38) Struchtrup, H. In Thermodynamics and Energy Conversion; Springer Berlin
Heidelberg: 2014, p 563.
(39) Mench, M. M. In Fuel Cell Engines; John Wiley & Sons, Inc.: 2008, p 121.
(40) Revankar, S. T.; Majumdar, P. Fuel Cells: Principles, Design, and Analysis;
Taylor & Francis, 2014.
(41) Peighambardoust, S. J.; Rowshanzamir, S.; Amjadi, M. International Journal of
Hydrogen Energy 2010, 35, 9349.
(42) Bose, S.; Kuila, T.; Thi, X. L. N.; Kim, N. H.; Lau, K. T.; Lee, J. H. Progress in
Polymer Science 2011, 36, 813.
(43) Zhang, H.; Shen, P. K. Chem Soc Rev 2012, 41, 2382.
(44) Zhang, H.; Shen, P. K. Chem Rev 2012, 112, 2780.
(45) Chandan, A.; Hattenberger, M.; El-Kharouf, A.; Du, S. F.; Dhir, A.; Self, V.;
Pollet, B. G.; Ingram, A.; Bujalski, W. Journal of Power Sources 2013, 231, 264.
(46) Mauritz, K. A.; Moore, R. B. Chemical Reviews 2004, 104, 4535.
(47) Rikukawa, M.; Sanui, K. Progress in Polymer Science 2000, 25, 1463.
(48) Kirt, A. P.; Brandon, W. R. In Polymers for Energy Storage and Delivery:
Polyelectrolytes for Batteries and Fuel Cells; American Chemical Society: 2012; Vol. 1096, p
147.
(49) Rana, D.; Matsuura, T.; Zaidi, S. M. J. In Polymer Membranes for Fuel Cells;
Zaidi, S. M. J., Matsuura, T., Eds.; Springer US: 2009, p 401.
(50) Dupuis, A. C. Progress in Materials Science 2011, 56, 289.
(51) Savage, J.; Tse, Y.-L. S.; Voth, G. A. The Journal of Physical Chemistry C 2014.
(52) Ban, S.; Huang, C.; Yuan, X.-Z.; Wang, H. The Journal of Physical Chemistry C
2012, 116, 17424.
(53) Daly, K. B.; Benziger, J. B.; Panagiotopoulos, A. Z.; Debenedetti, P. G. The
Journal of Physical Chemistry B 2014, 118, 8798.
(54) Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.; Garland, N.;
Myers, D.; Wilson, M.; Garzon, F.; Wood, D.; Zelenay, P.; More, K.; Stroh, K.; Zawodzinski,
T.; Boncella, J.; McGrath, J. E.; Inaba, M.; Miyatake, K.; Hori, M.; Ota, K.; Ogumi, Z.; Miyata,
28
S.; Nishikata, A.; Siroma, Z.; Uchimoto, Y.; Yasuda, K.; Kimijima, K.-i.; Iwashita, N. Chemical
Reviews 2007, 107, 3904.
(55) Ludvigsson, M.; Lindgren, J.; Tegenfeldt, J. Journal of The Electrochemical
Society 2000, 147, 1303.
(56) Porat, Z. e.; Fryer, J. R.; Huxham, M.; Rubinstein, I. The Journal of Physical
Chemistry 1995, 99, 4667.
(57) Zhang, W.; Yue, P.-L.; Gao, P. Langmuir 2011, 27, 9520.
(58) Michael, A. Y.; Steven, J. H. In Fuel Cell Chemistry and Operation; American
Chemical Society: 2010; Vol. 1040, p 15.
(59) Kreuer, K. D. Journal of Membrane Science 2001, 185, 29.
(60) Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; McGrath, J. E. Chem
Rev 2004, 104, 4587.
(61) Neburchilov, V.; Martin, J.; Wang, H. J.; Zhang, J. J. Journal of Power Sources
2007, 169, 221.
(62) Rozière, J.; Jones, D. J. Annual Review of Materials Research 2003, 33, 503.
(63) Piela, P.; Eickes, C.; Brosha, E.; Garzon, F.; Zelenay, P. Journal of the
Electrochemical Society 2004, 151, A2053.
(64) Zelenay, P.; Kim, Y. S.; Bashyam, R.; Choi, J.-H. ECS Transactions 2006, 1, 437.
(65) Maier, G.; Meier-Haack, J. In Fuel Cells II; Scherer, G., Ed.; Springer Berlin
Heidelberg: 2008; Vol. 216, p 1.
(66) Kerres, J. A. Journal of Membrane Science 2001, 185, 3.
(67) Jannasch, P. Current Opinion in Colloid & Interface Science 2003, 8, 96.
(68) Higashihara, T.; Matsumoto, K.; Ueda, M. Polymer 2009, 50, 5341.
(69) Li, L.; Zhang, J.; Wang, Y. X. Journal of Membrane Science 2003, 226, 159.
(70) Catherine, M.; Gérard, G.; Olivier, D.; Régis, M. Sulfonated Polyimides.
(71) Berber, M.; Fujigaya, T.; Sasaki, K.; Nakashima, N. Scientific reports 2013, 3,
1764.
(72) Subianto, S. Polymer International 2014, 63, 1134.
(73) Chen, N.; Hong, L. Polymer 2004, 45, 2403.
29
(74) Shin, J.; Chang, B.; Kim, J.; Lee, S.; Suh, D. Journal of Membrane Science 2005,
251, 247.
(75) Chikh, L.; Delhorbe, V.; Fichet, O. Journal of Membrane Science 2011, 368, 1.
(76) Kim, Y. S.; Wang, F.; Hickner, M.; Zawodzinski, T. A.; McGrath, J. E. Journal of
Membrane Science 2003, 212, 263.
(77) Ruffmann, B.; Silva, H.; Schulte, B.; Nunes, S. P. Solid State Ionics 2003, 162–
163, 269.
(78) Gaowen, Z.; Zhentao, Z. Journal of Membrane Science 2005, 261, 107.
(79) Tripathi, B. P.; Shahi, V. K. Progress in Polymer Science 2011, 36, 945.
(80) Soczka-Guth, T.; Baurmeister, J.; Frank, G.; Knauf, R.; International Patent:
1999.
(81) Elabd, Y. A.; Napadensky, E. Polymer 2004, 45, 3037.
(82) Xing, P.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.; Wang, K.;
Kaliaguine, S. Journal of Membrane Science 2004, 229, 95.
(83) Asano, N.; Aoki, M.; Suzuki, S.; Miyatake, K.; Uchida, H.; Watanabe, M. Journal
of the American Chemical Society 2006, 128, 1762.
(84) Fang, J.; Zhai, F.; Guo, X.; Xu, H.; Okamoto, K.-i. Journal of Materials
Chemistry 2007, 17, 1102.
(85) Zhou, X.; Weston, J.; Chalkova, E.; Hofmann, M. A.; Ambler, C. M.; Allcock, H.
R.; Lvov, S. N. Electrochimica Acta 2003, 48, 2173.
(86) Carter, R.; Wycisk, R.; Yoo, H.; Pintauro, P. N. Electrochemical and Solid-State
Letters 2002, 5, A195.
(87) Allcock, H. R.; Wood, R. M. Journal of Polymer Science Part B: Polymer
Physics 2006, 44, 2358.
(88) Wycisk, R.; Ballengee, J.; Pintauro, P. N.; Hoek, E. M. V.; Tarabara, V. V. In
Encyclopedia of Membrane Science and Technology; John Wiley & Sons, Inc.: 2013.
(89) Schmittinger, W.; Vahidi, A. Journal of Power Sources 2008, 180, 1.
(90) Gryte, C. C.; Gregor, H. P. Journal of Polymer Science: Polymer Physics Edition
1976, 14, 1855.
(91) Chen, J. H.; Asano, M.; Maekawa, Y.; Yoshida, M. Journal of Membrane Science
2006, 277, 249.
30
(92) Prakash, G. K. S.; Smart, M. C.; Wang, Q.-J.; Atti, A.; Pleynet, V.; Yang, B.;
McGrath, K.; Olah, G. A.; Narayanan, S. R.; Chun, W.; Valdez, T.; Surampudi, S. Journal of
Fluorine Chemistry 2004, 125, 1217.
(93) Moszczyński, P.; Kalita, M.; Parzuchowski, P.; Siekierski, M.; Wieczorek, W.
Journal of Power Sources 2007, 173, 648.
(94) Kumar, P.; Dutta, K.; Das, S.; Kundu, P. P. Applied Energy 2014, 123, 66.
(95) Wei, Z.; He, S.; Liu, X.; Qiao, J.; Lin, J.; Zhang, L. Polymer 2013, 54, 1243.
(96) Yamaguchi, T.; Kuroki, H.; Miyata, F. Electrochemistry Communications 2005,
7, 730.
(97) Park, J.-S.; Choi, Y.-W. Chemistry Letters 2013, 42, 998.
(98) Asensio, J. A.; Sanchez, E. M.; Gomez-Romero, P. Chemical Society Reviews
2010, 39, 3210.
(99) Cao, Y.-C.; Xu, C.; Wu, X.; Wang, X.; Xing, L.; Scott, K. Journal of Power
Sources 2011, 196, 8377.
(100) He, R.; Li, Q.; Xiao, G.; Bjerrum, N. J. Journal of Membrane Science 2003, 226,
169.
(101) Ajji, A.; Utracki, L. A. Polymer Engineering & Science 1996, 36, 1574.
(102) Kerres, J. In Polymer Membranes for Fuel Cells; Javaid, Z. S. M., Matsuura, T.,
Eds.; Springer US: 2009, p 1.
(103) Piboonsatsanasakul, P.; Wootthikanokkhan, J.; Thanawan, S. Journal of Applied
Polymer Science 2008, 107, 1325.
(104) Seeponkai, N.; Wootthikanokkhan, J. Journal of Applied Polymer Science 2010,
117, 393.
(105) Wootthikanokkhan, J.; Seeponkai, N. Journal of Applied Polymer Science 2006,
102, 5941.
31
CHAPTER 2: Membrane Fabrication and Characterization
2.1 Sulfonated polystyrene (PS) and polyvinylidene fluoride (PVDF) polymer blend membranes
Polymer blending is a potentially effective method to design materials with the
combination of desirable polymer properties. Most polymer electrode membrane materials, such
as Nafion
®
, partially sulfonated poly(ether ether ketone) (SPEEK), consist of a single polymeric
component. In this case, the single component PEM must satisfy the following two aspects for
DMFC applications: first, it must exhibit some degree of hydrophilicity so that the desirable
water uptake and high proton conductivity can be obtained; but at the same time it has to possess
substantial hydrophobicity to ensure the membrane integrity as well as low methanol
permeability. The technological challenge here is to produce an inexpensive material that can
satisfy this seemly contradictory requirement.
In comparison to sulfonated single polymers, the incorporation of two distinct polymeric
components into the polymer electrolyte membrane has certain advantages as one component can
provide a strong membrane scaffold while the other provides for proton transport.
106,107
The two-
component composite structure is essential for designing low cost and highly efficient polymer
electrolyte membranes because it seems unlikely that any homopolymer alone can combine such
distinct properties.
102,108,109
Poly(styrene sulfonic acid) (PSSA) is a basic material in the field of polymer electrolyte
membranes. And polyvinylidene fluoride (PVDF) exhibits high thermal and aging resistance and
superior inertness to various organic solvents, oils, and acids.
109,110
It is also more
environmentally friendly than perfluoropolymers and shows low permeabilities to many gases
and liquids, including a low methanol crossover rate.
103-105,111
The unique properties of PVDF
32
have prompted its use as a polymer matrix for polymer electrolyte membranes. The initial
attempt was to create a polymer blend of polystyrene (PS) and PVDF followed by the
sulfonation with concentrated sulfuric acid or chlorosulfuric acid. However, PS and PVDF have
very distinct cohesive energy density (PS: 17.52 MPa
1/2
; PVDF: 23.2 MPa
1/2
)
112
implying these
two polymers have poor compatibility. Nevertheless, PS/PVDF polymer blend was explored in
order to lay a foundation for study the compatibility of polystyrene sulfonate copolymer and
PVDF in the further.
2.1.1 Preparation of PS/PVDF polymer blend membranes
PS and PVDF polymer blend membranes were prepared by dissolved both polymers
according to appropriate weight ratios in a cosolvent, such as tetrahydrofuran (THF),
dimethylformamide (DMF), dimethylacetamide (DMAC), or dimethyl sulfoxide (DMSO) to
obtain a 20 wt% polymer solution. The polymer mixture was poured into glass molds and placed
in the fume hood (THF as casting solvent) at room temperature or an oven (DMF as casting
solvent) at 60 °C to evaporate the casting solvent. The resulted PS/PVDF membranes appear to
be opaque (Figure 2.1) regardless which solvent was used due to the poor compatibility between
these two polymers. In the case of using low boiling point solvents, like THF, PS/PVDF
membranes are wrinkled and brittle presumably due to the repaid evaporation of THF, which
aggravates the phase separation of the polymer blend. The wrinkled morphology issue was
alleviated by using high boiling point solvents (DMF, DMSO), however, the PS/PVDF
membrane still exhibit macro-phase separation resulting in poor mechanical strength due to the
immiscibility between PS and PVDF.
104,108
33
Figure 2.1 Photograph of PS/PVDF polymer blend membranes prepared by solvent casting.
2.1.2 Sulfonation of PS/PVDF membranes
The sulfonation of PS/PVDF was carried out with concentrated sulfuric acid or
chlorosulfuric acid according to published procedures.
69,113
The degree of sulfonation can be
adjusted by the sulfonation time and the concentration of strong acids. As shown in Figure 2.2,
the opaque appearance of membranes becomes yellow to dark brown upon sulfonation indicating
the presence of some undesired side reactions. In order to minimize the leaching problem of
sulfonated PS from the membrane, all membranes were sulfonated in concentrated sulfuric acid
at room temperature for 24 h followed by dialysis against with deionized water for another 48 h.
The surface roughness of sulfonated polystyrene (SPS)/PVDF membrane increased as compared
to unsulfonated PS/PVDF precursor. The water uptake of the sulfonated membrane is
summarized in Table 2.1, which is unusually high presumably due to the highly porous structure
of SPS/PVDF as a result of incompatibility. It is worth to note that the SPS/PVDF membranes
34
were extremely brittle due to the side reactions and crosslinking brought by sulfonation and the
phase separation,
114,115
making them unsuitable for DMFC applications.
Table 2.1 Water uptake and appearance of SPS/PVDF membranes.
SPS wt% in the membranes Water uptake Appearance
15% 39% Yellow
20% 45.5% Brown
30% 61% Dark brown
A
Figure 2.2 Photograph of sulfonated PS/PVDF membranes.
2.1.3 Ion exchange capacity (IEC) measurement
Even though the mechanical strength of sulfonated PS/PVDF membrane is unfavorable in
terms of membrane electrolyte assembly (MEA), the ion exchange capacity measurement of
35
sulfonated PS/PVDF membranes was conducted in order to evaluate the feasibility of preparing
polymer electrolyte membrane via polymer blending.
The sulfonated PS/PVDF membranes were immersed in brine solution for 24 h prior to
IEC measurement. A portion of brine solution was titrated with a known concentration (6.2
mmol/L) of NaOH aqueous solution using phenolphthalein as a PH indicator. The proton
concentration in the membrane was determined from the following equation.
where E is the ion exchange capacity of the membrane (mmol/g); V
NaOH
and C
NaOH
are the
volume (ml) and molar concentration, respectively, of NaOH solution; and W
dry
the mass of dry
membrane. And the characteristics of various SPS/PVDF membranes were summarized in Table
2.2.
Table 2.2 Ion exchange capacity (IEC) of sulfonated PS/PVDF membranes.
Membrane PS%
Mw of PS
(KDa)
Thickness
(mm)
IEC (mmol/g)
12-23 15 35 0.18 0.48
12-24 15 192 0.17 0.80
12-25 15 280 0.18 0.79
12-26 15 350 0.18 0.63
12-27 20 35 0.19 0.48
12-28 20 192 0.18 0.79
12-29 20 280 0.17 0.98
12-30 20 350 0.18 0.74
12-31 25 35 0.17 0.37
12-32 25 192 0.19 1.06
12-33 25 280 0.18 1.15
12-34 25 350 0.18 1.01
36
The effect of molecular weight (MW) on the IECs of SPS/PVDF membranes was shown in
Figure 2.3. The data indicated that membranes prepared with a low MW (35K) PS suffered from
significant leaching of polystyrene sulfonic acid (PSSA) after dialysis, explaining the extremely
low ion exchange capacities even with 25% polystyrene loading the membrane. The IEC
increased with the increase of molecular weight (MW) of polystyrene up to 280K, beyond which
the MW of PS has negative influence on the IEC of membranes, presumably due to a lower
degree of sulfonation when the MW is above 280 K. All in all, SPS/PVDF membranes prepared
by post-sulfonation are characterized with macro-phase separation, highly porous structure,
excessive water swelling, and poor mechanical property, suggesting they have limited
applications in DMFC.
Figure 2.3 Effect of molecular weight on ion exchange capacity of SPS/PVDF membranes.
2.2 Poly(sodium 4-styrenesulfonate) (PSSNa)/PVDF polymer blend membrane
In order to avoid the inevitable side reactions caused by the sulfonation process, polymers
with sulfonic acid group such as poly(sodium 4-styrene sulfonate) (PSSNa) had been explored to
blend with the inert polymer PVDF. PSSNa is commercially available and relatively
37
inexpensive, however without crosslinking, it is highly water-soluble. Therefore, the main
purpose of making PSSNa/PVDF polymer blend membranes here was to testify whether PSSNa
is compatible with PVDF to create a homogeneous polymer electrolyte membrane that is suitable
for DMFC applications.
2.2.1 Preparation of PSSNa/PVDF membranes by solution casting
N,N-Dimethylformamide (DMF) was used as a cosolvent for casting PSSNa/PVDF
membranes as both polymers are readily dissolved in it at room temperature. Membranes with
different composition (PSSNa : PVDF ratio) were prepared as the following typical procedure: to
a 50 ml round bottom flask, 3.0 g PVDF was dissolved in 20 ml DMF at room temperature. To
which, 10 g 10 wt% PSSNa solution in DMF was added slowly with vigorously stirring for 30
min. The polymer mixture was poured into a petri dish (diameter = 13.8 mm) and transferred into
an oven preheated at 60 °C for 48 h. The membrane was dried in the oven for 48 h. After the
solvent was evaporated, the membrane was immersed in deionized water for 3 h prior to retrieve
it from the glass substrate.
As shown in Figure 2.4, PSSNa/PVDF polymer blend membranes were opaque with
yellow spots (attributed to the PSSNa domains) dispersed in translucent PVDF matrix. It was
clear that PSSNa and PVDF have a poor compatibility as evidenced by a macro-phase separation
in the membrane, which again caused the brittleness and poor mechanical strength.
38
Figure 2.4 Photograph of PSSNa/PVDF membranes prepared by solvent casting.
2.2.2 Preparation of PSSNa/PVDF membranes by in-situ polymerization of SSNa in PVDF
matrix
As membranes prepared by direct blending of PSSNa and PVDF exhibits heterogeneous
morphology and unfavorable mechanical properties for DMFC applications. The in-situ
polymerization of monomer sodium styrene sulfonate (SSNa) in PVDF matrix was conducted
with the expectation of creating a more homogenous polymer blend membrane with desired
properties.
The procedure to prepare PSSNa/PVDF membranes with in-situ polymerization of SSNa
is as follows: to a 50 ml round bottom flask, 2.26 g PVDF was dissolved in 20 ml DMF at room
temperature with vigorously stirring. After that, the PVDF solution was degassed and refilled
with argon three times before injecting 1.73 g monomer SSNa and 0.021 g initiator AIBN
dissolved in 10 ml DMF. The mixture was stirred for another 30 min while purging with argon.
39
The mixture was then poured into a glass mold and transferred to an oven degassed with argon.
The polymerization of SSNa in PVDF matrix was started by setting the temperature of oven to
80 °C and maintaining at that temperature for 24 h. After evaporation of DMF, PSSNa/PVDF
membrane was retrieved after immersed in water for 3 h.
The morphology of PSSNa/PVDF membranes prepared by in-situ polymerization was
shown in Figure 2.5. The in-situ polymerization method shows some improvement in
homogeneity, to some degree, as compared to PSSNa/PVDF membranes prepared by direct
mixing polymer PSSNa and PVDF. For example the phase separation become less pronounced
and the domain sizes for each polymer become smaller. However, membranes prepared by in-
situ polymerization still exhibit poor mechanical strength and are brittle in the dry state due to
the poor miscibility between polymer PSSNa and PVDF, which prevent them to be used as
polymer electrolyte membrane for DMFC applications.
It is important to note that although polymer blend membranes based on in-situ
polymerization of SSNa had been reported.
93,94,116
However, the homogeneity and mechanical
properties are rarely mentioned. Kundu et al.
94
prepared a polymer electrolyte membrane by in-
situ polymerization of sodium salt of sulfonated styrene in the polymeric blend of PVDF-co-
HFP/Nafion
®
. The maximum current density was recorded to be 120 mA/cm² at 0.2 V with a
powder output of 24 mW/cm² at 60 °C in air. However, no morphology and mechanical property
was reported in the article. Kalita et al.
93
prepared a composite electrolyte by soaking PVDF
sponge with water solution of ionomer SSNa followed by in-situ polymerization. SEM images
reveal that those membranes had a coarse and irregular morphology, in addition, no mechanical
strength and electrical performance was reported in the article. Clearly developing a better
methodology is needed for prepare polymer electrolyte membranes based on polymer blends
40
with desired properties, such as homogeneity, toughness, flexibility in dry state, and so on, for
direct methanol fuel cell applications.
A
Figure 2.5 Photograph of PSSNa/PVDF membranes prepared by in-situ polymerization of SSNa
in PVDF.
2.3 Synthesis of polystyrene sulfonate (PSS) copolymer
In the previous attempt, the direct blending of PS and PVDF followed by sulfonation
with strong acids did not generate membranes with favorable properties (chemical/thermal
stable, high mechanical strength) for DMFC applications. Therefore, no further electrochemical
performance was conducted in a DMFC. Nevertheless, the initial attempts of exploring
sulfonated PS/PVDF membranes did provide some valuable insights for preparing usable
membrane in DMFC applications utilizing the polymer blend technique. In order to generate
41
robust and low cost polymer electrolyte membranes for DMFC applications utilizing polymer-
blending technique, the polymer system must satisfy the following criteria:
1. Considerable compatibility between polymer components;
2. Crosslinking of hydrophilic polymer component;
3. Direct incorporation of sulfonic acid groups in polymerization rather than post-sulfonation;
4. Low cost and readily available materials;
5. High mechanical strength of polymer blend membranes.
There are much more requirements in terms of designing novel polymer electrolyte membranes
for DMFC applications, however, the above mentioned criteria are more prominent when
developing DMFC membranes based polymer blends.
The post-sulfonation process during polymer electrolyte membrane fabrication inherits
undesirable side reactions evident by the dark brown color of membranes. The inevitable side
reactions motivated us to utilize other polymers preferably those prepared with direct
polymerization of monomers bearing sulfonic acid groups so that any post-sulfonation would be
unnecessary. Polystyrene sulfonic acid is considered as the basic component for designing
alternative hydrocarbon polymer electrolyte membranes for DMFCs.
47,59,60
The simplest vinylic
monomer with sulfonate functional group which is also commercially available is sodium styrene
sulfonate (SSNa).
116
In principle, polymer blend membranes for DMFCs can be made by simply
mixing an inert polymer matrix (such as PVDF) with poly(styrenesulfonic acid sodium salt)
(PSSNa) to offer the proton transpiring property to the resulting membrane. However, PSSNa is
highly hydrophobic and barely mixable with other hydrophobic polymer matrix, which imposes
an insurmountable obstacle for designing PEMs utilizing polymer blending.
103,104
Furthermore,
the absence of crosslinking or curing make the hydrophilic PSSNa vulnerable to leach out from
42
the polymer blend membrane as discussed before. In order to verify our hypothesis, membranes
consisting of only PSSNa and PVDF were prepared by solution casting, and the resulting
membranes were extremely brittle and are subject to leaching issue after dialysis in demonized
water. Therefore, it is clear that the proton transporting/hydrophilic component in the membrane
must be crosslinked somehow prior to or post membrane preparation in order to make it feasible
for DMFC applications.
With these thoughts in mind, we decided to explore the possibility of creating a polymer
blend membrane consisting of cross-linkable polystyrene sulfonate (PSS) and PVDF. Preferably,
the crosslinking process should occur after the membrane preparation to simplify the membrane
fabrication procedures. As mentioned before, the monomer SSNa is relatively inexpensive and
commercial available, which make it an ideal raw material to start with.
2.3.1 Synthesis of ionic liquid monomer tetra-butylammonium styrene sulfonate (BASS)
Although sodium styrene sulfonate appear to be a versatile monomer to prepare
polyelectrolyte, it is difficult to directly copolymerize SSNa with other hydrophobic monomers,
such as styrene and 4-vinylbenzyl chloride.
116,117
The challenge in the direct copolymerization of
SSNa-containing copolymers is the strong immiscibility and distinct polarities of the hydrophilic
SSNa and most hydrophobic monomers and organic solvent.
118
To enhance the compatibility of
SSNa with other hydrophobic monomers, SSNa was modified to a more hydrophobic form,
which is amenable to the conventional radical polymerization with other monomers.
Herein, tetra-butylammonium styrene sulfonate (BASS) was synthesized by ion
exchange between SSNa and tetra-alkylammonium salt,
116
such as tetra-butylammonium
hydroxide, bromide or chloride, tetra-octylammonium bromide or chloride, dodecyl-
43
dimethylammonium bromide or chloride, in aqueous media followed by extraction with organic
solvent (CH
2
Cl
2
), as shown in Scheme 2.1.
Scheme 2.1 Synthesis hydrophobic ionic liquid monomer tetra-butylammonium styrene
sulfonate.
In a typical synthesis of BASS, 5 grams (24.2 mmoles) of SSNa dissolved in 50 ml
deionized water was placed in a 250 ml round bottom flask, and 16.1 ml of 40 wt. percent
Bu
4
NOH (24.2 mmols) or Bu
4
NBr was added dropwise with stirring at room temperature. The
solution became cloudy and was stirred for additional 10 min. The BASS was extracted with 100
ml of CH
2
Cl
2
(DCM) and the solution was dried overnight over anhydrous MgSO
4
. Evaporation
of the organic solvent CH
2
Cl
2
with a rotary evaporator under vacuum gave a light yellowish
viscous liquid (10.1 g) in 98% yield.
2.3.2 NMR analysis of monomer tetra-butylammonium styrene sulfonate (BASS)
The resulting ionic liquid monomer BASS are readily soluble in most common organic
solvents, such as acetone, THF, DMF, DMSO, DCM, chloroform, methanol, ethanol, and 1,2-
dichloroethane (DCE), and it is also miscible with many hydrophobic vinyl monomers, such as
styrene, 4-chloromethyl styrene (CMS), and 4-vinylpyridine (4-VP), which allows for the
44
copolymerization with a variety of other monomers utilizing the conventional free radical
polymerization. Proton NMR was used to verify the stoichiometric amount of the respective
protons in BASS as shown in Figure 2.6. The vinyl protons appeared at 5.24 ppm, 5.75 ppm
(vinyl, =CH
2
) and 6.65 ppm (vinyl, -CH=CH
2
), respectively. The peaks at 7.36 ppm and 7.84
ppm (aromatic, 4H) are attributed to protons on the aromatic ring. And the peaks appear between
3.27 ppm and 0.98 ppm correspond to protons in butyl groups.
Figure 2.6
1
H NMR spectrum of TBASS.
2.3.3 Terpolymerization of BASS, styrene, and 4-chloromethyl styrene (CMS)
The homopolymerization of BASS gives water-soluble polymer PBASS, which is subject
to the leaching issue if no crosslinking is introduced later. In order to introduce the crosslinking
in the membrane, a small amount (10 mol%) of styrene (S) and 4-chloromethyl styrene (CMS)
45
were copolymerized with BASS to give the terpolymer P(BASS-S-CMS). A small amount of
monomer CMS is usually used to copolymerized with styrene giving crosslinked polystyrene
with enhanced thermal stability and flame retardancy via Friedel-Crafts chemistry (Scheme
2.2).
119-124
The Friedel-Crafts reaction is particularly favorable in terms of optimizing the
membrane preparation procedure, because the crosslinking process can be introduced by thermal
annealing the terpolymer P(BASS-S-CMS) after blending it with inert polymer matrix PVDF.
Scheme 2.2 Cross-linking of polystyrene copolymer through Friedel-Crafts reaction.
The terpolymerization was carried out through the free radical polymerization with 2,2´-
azobis-(isobutyronitrile) (AIBN) as the initiator. In the typical polymerization, 5 g (11.7 mmols)
of BASS, 0.153 g (1.47 mmols) of styrene, 0.224 g (1.47 mmols) of CMS, 21 mg (0.13 mmols)
of AIBN and 5 ml 1,2-dichloroethane were charged in a dry 50 ml round bottom flask purged
with argon. The system was then degassed and refilled with argon three times. The
polymerization was started by heating at 65 °C while stirring. After 10 h, the reaction mixture
was diluted with 20 ml dichloromethane and precipitated into 500 ml anhydrous THF at room
46
temperature. Poly(BASS-S-CMS) (91% yield) was obtained by drying the polymer in vacuum at
50 °C for 24 h.
Scheme 2.3 Schematic representation of terpolymerization of BASS, S, and CMS.
2.3.4 NMR analysis of terpolymer P(BASS-S-CMS)
The composition of copolymer P(BASS-S-CMS) was analyzed with proton NMR in
CDCl
3
. As shown in Figure 2.7, the spectrum of the polymer was characterized with the shape
peaks attributed to the butyl groups. On the other hand, those peaks attributed the aromatic and
polymer backbone protons are barely shown in the spectrum. The absence of peaks attributed to
aromatic protons presumably due to the self-assembled structures of polyelectrolyte P(BASS-S-
CMS) in CDCl
3
, in which the polyanionic polymer backbones are well-surrounded by the
counterions tetra-butylammonium cations.
125-127
Therefore, the proton NMR spectrum of
polymer P(BASS-S-CMS) was dominated by the shape peaks attributed to the small tetra-
butylammonium cations and the peaks corresponding to the polymer backbones are barely shown
in the NMR spectrum.
47
Figure 2.7
1
H NMR spectrum of terpolymer P(BASS-S-CMS).
2.3.5 Solubility of P(BASS-S-CMS)
The terpolymer of ammonium substituted styrene sulfonate (BASS), styrene and 4-
chloromethyl styrene turned out to be very soluble in most polar organic solvents, such as
acetone, DCM, DMF, DMAc and DMSO. It is also soluble in water, methanol and ethanol,
validating our previous conclusion that the introduction of crosslinking of the hydrophilic
polymer in polymer blend membranes is necessary for applications in direct methanol fuel cells.
Interestingly, the terpolymer has a poor solubility in THF and diethyl ether, which makes
purification of the polymer by precipitation feasible. On the contrast, the purification of
poly(sodium styrene sulfonate) (PSSNa) by precipitation is problematic as few solvents can
precipitate the polymer PSSNa without precipitating the unreactive monomer SSNa.
117,118,128
The
48
good solubility of P(BASS-S-CMS) also provides a facial approach to fabricate the
corresponding PBASS/PVDF membranes by solution-cast techniques.
2.4 P(BASS-S-CMS)/PVDF membranes fabrication through solution casting
Unlike commercial available sodium salt of polystyrene sulfonic acid which has poor
solubility in most common organic solvents and is barely mixable with other hydrophobic
polymers, the terpolymer P(BASS-S-CMS) is readily soluble in polar solvents, which facilitates
membrane preparation via solution casting. Finding an optimal methodology to prepare P(BASS-
S-CMS)/PVDF membranes required trial and error as a lot of external factors including casting
solvent, solution viscosity, evaporation rate, anneal temperature and quench temperature etc.
affect the overall property of resulting membranes. Nevertheless, a lot of experiments were
carried out in our lab for finding an optimal method to prepare P(BASS-S-CMS)/PVDF
membranes through solvent casting.
2.4.1 P(BASS-S-CMS)/PVDF membranes cast from acetone at 30 °C (Method I)
As the tetraalkyl ammonium salt of polystyrene sulfonate copolymer is soluble in various
organic solvents, it was plausible find appropriate solvents to give membranes with desirable
properties. The commercially available polymer matrix PVDF can be cast from acetone, THF,
DMF, DMAc, DMSO and N-methyl-2-pyrrolidone (NMP).
129
In the early stage of developing
membrane fabrication techniques, low boiling point (BP) solvents, like acetone, were first
selected for the membrane fabrication as the removal of acetone from membranes is easier
compared the above higher boiling point solvents.
130
49
Even though PVDF is not readily dissolved in acetone at room temperature, it can be
dissolved in acetone when refluxed at 60 °C for a short period of time. A typical preparation of
P(BASS-S-CMS)/PVDF membrane is as follows: to a 100 ml round bottom flask charged with a
Teflon stir bar, 1.25 g polystyrene sulfonate copolymer P(BASS-S-CMS) was dissolved in 25 ml
of acetone at room temperature. After that 2.75 g PVDF powder was added to the solution and
stirred at room temperature for 1 h. The temperature of the mixture was brought to 60 °C and
refluxed for another 1 h until the solution appeared to be homogeneous. The mixture was
allowed to cool down at room temperature while stirring and 1 ml 1.0 wt% acetone solution of
ZnCl
2
as the crosslinking catalyst was slowly added. The mixture was poured into a glass petri
dish covered with a lid in order to reduce the solvent evaporation rate and placed in a fume hood
for 24 h allowing acetone to be removed slowly at room temperature. After acetone was
evaporated, the membrane was transferred to an oven followed by incremental heating to 100 °C
over a period of 1 h and heating at that temperature for another 1 hour in order to induce the
ZnCl
2
-mediated crosslinking of the benzyl chloride and phenyl groups (Scheme 2.4).
120,121,123,124
After that, the P(BASS-S-CMS)/PVDF membrane was soaked in 1.0 M sulfuric acid solution for
3 days at 90 °C in order to replace the tetra-butylammonium ions with protons providing
corresponding PSSA/PVDF membranes. The resulting membrane was washed with water
thoroughly to remove sulfuric acid and any residual catalyst ZnCl
2
.
50
Scheme 2.4 Schematic representation of crosslinking and protonation of terpolymer P(BASS-S-
CMS).
The resulting opalescent PSSA/PVDF membranes shown in Figure 2.8 have a striking
improved homogeneity compared to PSSA/PVDF membranes prepared by post-polymerization
sulfonation. The improved homogeneity of PSSA/PVDF membrane was attributed to the
improved compatibility between terpolymer P(BASS-S-CMS) and PVDF due to the enhanced
hydrophobicity of tetra-butylammonium styrene sulfonate. It is important to note that although
PSSA/PVDF membranes cast from acetone achieved a significantly improved homogeneity, the
use of acetone as a casting solvent has some issues. For instance membranes cast from acetone
often show irregular and wrinkled surface due to the rapid evaporation of acetone. Upon closed
observation this was seen to be due to some areas with higher evaporation rates being detached
SO
3
-
x
y
z
y
z
x
q
Chains
Crosslinks
+
Cl
SO
3
-
+
SO
3
-
+
HO
3
S
SO
3
H
SO
3
H
SO
3
H
N(Bu)
4
N(Bu)
4
N(Bu)
4
+ PVDF
PVDF
Matrix
i. ZnCl
2
; Annealing
ii. 1.0M H
2
SO
4
51
from the glass substrate. In addition, membranes prepared by acetone with PSSA contents above
15 wt. percent appear to be brittle when being fabricated into membrane electrolyte assembles.
Further more, these membranes appeared to be opaque due to microphase separation of P(BASS-
S-CMS) and PVDF, presumably due to marginal solubility of PVDF in acetone. Despite this,
several membranes were fabricated into MEAs the electrical performance of which will be
discussed in the next chapter.
Figure 2.8 Photograph of a PSSA/PVDF membrane cast from acetone at room temperature.
2.4.2 P(BASS-S-CMS)/PVDF membranes cast from DMF at 60 °C (Method II)
Apparently, a better solvent was needed in order to fabricate homogeneous polymer
blend membranes. N,N-Dimethylformamide (DMF) and N,N-Dimethylacetamide (DMAc) are
listed as active solvents for the film-forming of PVDF, meaning PVDF can be readily dissolved
in those solvents at room temperature.
109,131
Meanwhile, DMF and DMAc are less volatile and
52
have much higher boiling points (153 °C and 165 °C, respectively) compared to acetone.
Therefore, DMF and DMAc were chosen as the solvent for solution cast of P(BASS-S-
CMS)/PVDF membranes.
The fabrication of P(BASS-S-CMS)/PVDF membranes using DMF as the solvent is
similar to that using acetone, except that both P(BASS-S-CMS) and PVDF polymers were fully
dissolved in DMF at room temperature prior to pouring into glass molds. In a typical example,
1.58 g P(BASS-S-CMS) terpolymer and 2.42 g PVDF powder were added to a 100 ml round-
bottom flask charged with 25 ml DMF and a Teflon stirrer. The polymer mixture was stirred at
room temperature for 12 h until both polymers were completely dissolved and the solution
became optically clear. After that 1 ml ZnCl
2
solution (1 wt% in DMF) was added slowly and
stirred for another 1 h. The mixture was then poured into glass molds and transferred to an oven
located in a fume hood. The oven was adjusted to 60 °C and the membrane was allowed to dry
for 48 h followed by annealed at 100 °C for another 1 h. After annealing, the membrane was
allowed to cool down at room temperature before treating with a sulfuric acid solution.
The membrane was strongly stuck to the glass substrate after complete removal of DMF
and the annealing process. Therefore, it was necessary to soften the membrane by immersing it
in deionized water over night prior to peeling it off from the glass substrate. After the membrane
was retrieved from the petri dish, the resulting PSSA-PVDF membrane was soaked in 1 L 1.0 M
H
2
SO
4
solution and heated to 90 °C for 72 h in order to obtain the corresponding polystyrene
sulfonic acid composite (PSSA-PVDF) through ion exchange. After acidification, the membrane
was further dialyzed with 500ml deionized water at 80 °C for 3 days, with deionized water being
changed every 12 hours. Finally the PSSA-PVFD membranes were dried at 60 °C for 24 h under
vacuum.
53
The absence of chloride was confirmed by boiling a small membrane sample in 10 ml of
water for 2 h and titration of the aqueous phase with AgNO
3
with no precipitate being observed.
The incorporation of P(SSA-S-CMS) in the PVDF matrix was confirmed by boiling in
dichloromethane for 2 h after which no loss of mass was observed. In the absence of
crosslinking, a significant loss (> 15%) of PSSA was seen.
As shown in Figure 2.9, the resulting membrane cast from DMF or DMAc at 60 °C had
much better morphology as compared to those cast from acetone. The membranes appear to be
opalescent rather than opaque, indicating the improved homogeneity. The membrane surfaces
were also flatter and much smoother compared to the membranes processed in acetone.
Figure 2.9 Photograph of a PSSA/PVDF membrane cast from DMF at 60 °C.
Although membranes cast from DMF appear to be much favorable toward MEA
fabrication, they still suffers from brittleness in the completed dry state especially when the
54
PSSA content is above 20 wt%, presumably due to the low crystallinity and phase separation of
polymer blends. One of the important factors that has strong influence on the morphology and
mechanical property of PVDF membranes by solvent casting is the evaporation temperature.
132-
135
Therefore, different casting temperatures were explored in order to further optimize the
membrane fabrication procedure, providing desired mechanical properties for applications in
DMFCs.
2.4.3 P(BASS-S-CMS)/PVDF membranes cast from DMF at 160 °C (Method III)
PVDF exhibits a complex crystalline polymorphism absent in most other synthetic
polymers. There are at least three distinct crystal forms with different corresponding chain
conformation: α (TGTGT’), β (TTTT’), γ (TTTGTTTG’).
136,137
For membranes prepared by
solution casting, the polymorphs are present in different proportions in samples and depend on
the casting solvent and evaporation temperature. Buonomenna et al.
132
studied the role of casting
temperature in the crystalline structure of PVDF and found out that PVDF films cast below an
annealing temperature of 70 °C primarily consist of a β crystalline phase, and those cast above
110 °C primary consist of α crystalline phase.
138
The evaporation conditions are also essential factors to be considered when fabricating
membrane through solution cast. Membranes cast from low temperatures take longer to
evaporate, which increases the chance of phase separation due to polymers immiscibility. The
quality of cast membranes depends in large part on the temperature and convection conditions
prevailing in the oven during the drying and crosslinking.
133,139
Subsequent studies have shown
that rapid solvent evaporation typically leads inferior films while slow evaporation can lead to
films showing partial phase separation. This can result in the formation of a PVDF-rich
55
layer.
90,140,141
As PVDF has a melting point around 165 °C, an annealing temperature around 165
°C could be helpful as polymers become miscible above the upper critical solution temperature
(UCST). Martin et al.
133,139
reported that perfluorosulfonate ionomer (PFSI) films cast from at
low temperatures were soluble in a variety of polar organic solvents at ambient temperature, as
well as brittle and noncohesive. However, solution-cast PFSI films prepared by using high
temperature and suitable solvents had the desirable physical, mechanical, chemical
characteristics of the as-received commercial films. The morphological differences of films cast
from different temperatures are believed to account for the distinct mechanical properties.
Therefore, a higher casting temperature was necessary to prepare P(BASS-S-
CMS)/PVDF membranes through solution casting in order to obtain better homogeneity and
mechanical properties. The procedure of Method III was similar to Method II except that the
membrane was cast at a much higher temperature, 100 °C and then annealed at 150-160 °C. In
order to improve the homogeneity of membranes, special attention was paid on the morphology
change of membranes during the evaporation process. It was found that membranes were totally
optically clear at higher temperatures (> 150 °C), however, they became translucent when
allowed to cool down slowly, indicating the formation of different polymer- or crystalline
domains. Therefore it was necessary to rapidly quench membranes after annealing in order to
retain the homogeneity. After exploring different casting and annealing temperatures on the
morphology and mechanical property of P(BASS-S-CMS)/PVDF, an optimized membrane
fabrication procedure was summarized as follows: 1. P(BASS-S-CMS) and PVDF precursors
with appropriate weigh ratios are dissolve in DMF or DMAc and stirred at room temperature for
24 h in order to give a homogeneous solution. 2. A small amount (~1 ml) of ZnCl
2
(1 wt% in
DMF) was added the polymer solution and stirred for another hour. 3. The polymer solution was
56
poured into glass or Teflon molds and transferred into an oven heating from 100~180 °C with a
15 °C increment of temperature for every one hour interval. 4. After annealing, the dried
membrane was quenched quickly by immersing it in a water bath (room temperature). 5. After
soaking in water for 12 h, the membrane was carefully peeled off from the mold. 6. Membrane
was soaked in 1.0 M sulfuric acid solution for 72 h at 90 °C. 7. Membrane was soaked in
deionized water for 72 h at 90 °C. 8. Membranes was dried at 60 °C for 24 h in vacuum.
The impact of optimized procedure on the morphology and properties of membranes was
observed immediately as the resulting membrane was highly transparent, flexible and tough even
in the dry state. As shown in Figure 2.10, the transparency of membranes indicated a
considerably increased homogeneity rarely reported for PEMs prepared by polymer blending,
such as sulfonated polystyrene/PVDF, sulfonate PEEK/PVDF. The flexibility and toughness of
membranes guarantee the integrity of MEAs fabricated by hot-press and also ensure the long-
term durability for DMFCs. Those favorable properties of membranes motivate us to fully
characterize the membrane and explore its potential as polymer electrolyte membranes for
DMFC applications.
57
Figure 2.10 Photographs of PSSA/PVDF membranes cast from DMF at 160 °C placed upon
USC logos.
2.5 Membrane Characterization
As mentioned in the introduction, a potential polymer electrolyte membrane for DMFCs
should have at least the following properties: a. high proton conductivity. b. low methanol
crossover rate. c. high mechanical strength. d. optimal degree of aqueous swelling. e. high
thermal stability. Most reported PEMs based on polymer blending met some of these criteria, but
fail in other requirements. In an effect to develop low-cost alternative PEMs for DMFCs through
polymer blending, membranes fabricated with the optimized procedure were extensively
characterized and compared with the properties of commercial Nafion
®
-117.
2.5.1 Optical clarity measurements
The optimized polymer blend membranes exhibit high optically transparency as
compared to membranes prepared by mixing sulfonated PS and PVDF. The PSSA is successfully
blended with PVDF into membranes by the “polymer camouflage” approach, in which the
58
P(BASS-S-CMS)/PVDF blend is then transformed into a PSSA terpolymer/PVDF blend by ion
exchange. The transparency of PSSA/PVDF indicates a high homogeneity between crosslinked
PSSA terpolymer and PVDF is obtained, which is difficult to achieve by directly mixing PS and
PVDF followed by post-sulfonation.
The transmittance of the membranes was further confirmed by ultraviolet–visible
spectroscopy (UV-Vis) measurement. As shown in Figure 2.11, PSSA/PVDF membranes
prepared using the optimized procedure exhibits transmittances above 90 percent in the range of
wavelengths between 350 and 1000 nm. The transmittance of membranes prepared by polymer
blending is always the first indication of miscibility between each polymer component. And such
a high transparency was rarely documented in PEMs based on polymer blending.
102,109,142,143
Figure 2.11 Optical transmittance measurement of a PSSA/PVDF membrane with 30 wt%
PSSA.
59
2.5.2 Water uptake measurements
The water swelling of PEMs is closely related to their mechanical stability and plays an
essential role on their proton conducting behavior. For sulfonic acid based PEMs, proton
conductivity depends to a large extent on the amount of absorbed water in membranes.
46,47,144
On
the other hand, the methanol crossover is also associated to water concentration in polymer
electrolyte membranes. Apart from this, the absorbed water also influences the ionomer
microstructure, cluster and channel size as well as modifies the membrane mechanical
properties.
32
The water uptake of the membrane was measured by soaking the samples overnight in
water at 25 °C. The samples were removed from the water, and any excess surface water was
wiped off using dry filter papers prior to weighing swollen membranes (W
wet
). The water uptake,
Φ
w
, defined by
where W
wet
and W
dry
are the weights of swollen and of dry membranes, respectively.
In order to study the effect of PSSA content on the water swelling, various membrane
with different content of PSSA were fabricated and their major characteristics was list in Table
2.3. For PSSA/PVDF polymer blend membranes, the inner polymer matrix PVDF is highly
hydrophobic and its water absorption is negligible, therefore, the water uptake of PSSA/PVDF
membranes depends to a large extent on the amount of PSSA in membranes. As shown in Figure
2.12, the water uptake of PSSA/PVDF membranes increases nearly linearly with PSSA content.
The hydration number, although not identical are quite similar and correspond roughly to about
12 water molecular per PSSA unit, explaining the more or less linear correlation between the
water uptake of membranes and the amount of PSSA.
60
It is worth noting that the hydration number is relatively small as compared to that of
Nafion
®
-117 (hydration number = 19)
145
despite some higher IEC and proton conductivity
values, implying the water swelling in PSSA/PVDF is more restricted so that a reduced methanol
crossover can be achieved. The distinct correlation between proton conductivity and water
swelling also indicates that PSSA/PVDF membranes has a very different microstructure from
Nafion
®
-117. For PSSA/PVDF membranes, the water swelling is more regulated by the
hydrophobic PVDF matrix and that of Nafion
®
-117 is primary controlled by the hydrophobic
tetrafluoroethylene backbone. Membranes with PSSA contents above 40 percent show excessive
swelling that adversely affects their mechanical properties and increases methanol permeability,
therefore, no characterization was conducted.
Table 2.3 Characteristics of PSSA/PVDF membranes as function of PSSA content.
Membrane
Calc.
PSSA
(wt.%)
(IEC)
(mmol/g)
Proton
conductivity
(mS/cm)
Methanol
Permeability
10
-7
(cm
2
·sec
-1
)
Water
(wt.%)
[H
2
O]
c
[PSSA]
M10 10 0.51 18.2 0.3 13.9 13.5
M15 15 0.80 49.6 1.0 19.7 12.9
M20 20 0.96 78.6 5.8 22.6 11.7
M25 25 1.10 115.6 9.9 28.6 12.1
M30 30 1.22 132.8 12.8 32.2 10.9
M35 35 1.36 173.3 15.0 39.5 11.5
Nafion
®
117 -- 0.95 76.3 15.9 38 19.4
d
(a) IEC was measured in fully hydrated state at 25°C using the “four probe” method.
71
(b) Methanol permeability was measured at 25°C at an initial concentration of 2.0 M
MeOH.
(c) Number of water molecules per unit of PSSA.
(d) Estimated value based on IEC and conductance data.
61
Figure 2.12 Water uptake values of various PSSA/PVDF membranes with different PSSA
uptakes.
2.5.3 Ion exchange capacity (IEC)
The IEC values of polymer electrolyte membranes depend directly on the amount of
sulfonic acid groups in membranes and are the indication of the actual ion exchange sites
available for proton conduction.
146
Generally higher values of the IEC are desirable to achieve
higher proton conductivities as well as higher fuel cell performances. However, the methanol
permeability of membranes also has a monotonic correlation with the content of PSSA, which
reduces the fuel cell performance and the overall efficiency. Therefore, it is critical to balance
the proton conductivity and the methanol permeability of membranes in order to achieve high
fuel cell performances.
62
PSSA/PVDF membranes were placed in a saturated NaCl solution for 24 h membranes
prior to measuring their ion exchange capacity (IEC). The proton concentration in the NaCl
solution soaked by membranes was then determined by titrating with 0.01 M NaOH solution
using phenolphthalein as the indicator. The IEC (mmol/g) of the membrane was determined
from:
where V
NaOH
and C
NaOH
are the volume (ml) and molar concentration, respectively, of the sodium
hydroxide solution and W
dry
is the mass of dry membranes.
The values of ion exchange capacity of typical membranes are listed in Table 2.3. As
predicted, the membrane IEC increased approximately linearly with PSSA content as shown in
Figure 2.13. One of great advantages of PEMs based on polymer blends, in which one
component endows a strong membrane scaffold while the other provides proton transport, is that
the IEC and proton conductivity can be easily adjusted by varying the amount of PSSA content
in membranes without redesigning the entire membrane system.
2.5.4 Proton conductivity measurements
The proton conductivity of a PEM is strictly related with the ohmic losses associate to the
membrane during DMFC operation. One of the keys for PEM research is to develop membranes
with improved proton transport properties with restricted methanol crossover rates. The former
reduces ohmic losses of fuel cell polarization behaviors, while the later increases the fuel
utilization and the overall efficacy.
63
The specific proton conductivity of membrane samples was measured using a D.C., 4-
probe apparatus similar to that described by Cahan and Wainright.
147
The proton conductivity of
membrane samples was determined by an electrochemical impedance spectroscopy (EIS) using
an impedance/gain-phase analyzer (Solartron 1260) in combination with an electrochemical
interface (Solartron 1287). A homemade “four probe” apparatus was employed to measure the
proton conductivity. The technique was standardized using fully hydrated samples of Nafion
®
-
117 since it has been previously reported that conductivity values increase and reach a steady
state value when the membrane is fully hydrated.
148
This test resulted in specific proton
conductivity values of 70-75 mS/cm for Nafion
®
-117, in accordance with published data.
149,150
The technique was then used to determine the specific proton conductivity for fully hydrated
PSSA/PVDF polymer blend membranes. The proton conductivity (σ) was calculated using a
relationship:
where L, A and R represents the distance between the two inner pt-probes, the cross-sectional
area of membranes, and the resistance, respectively.
The specific proton conductivity of membrane samples was measured in the fully
hydrated state at room temperature. The correlation between the proton conductivity and the
PSSA loading in membranes is shown in Figure 2.13. It can be seen that the proton conductivity
of membranes increases nearly linearly with the increasing amount of PSSA as the proton
conductivity of pure PVDF matrix is negligible. Such as phenomenon is quite common for
polymer electrolyte membranes, as the increase of membrane IEC usually accompanies with the
increase of water swelling and proton conductivity.
151,152
For example, Chen et al.
91
studies
64
various perfluorinated and partially fluorinated polymers as base materials for radiation grafted
PSSA polymer electrolyte membranes, it was shown that the proton conductivity linearly
increases with the increase in the ion exchange capacity, and is quite dependent on the nature of
the base films. For PSSA/PVDF membranes with PSSA weight fractions above 20%, the IEC
and the proton conductivity were comparable or exceeded that of Nafion
®
-117, implying the
potential applications for PEMFCs. Therefore, it is clear that the membrane PSSA content is a
major parameter in the optimization of the membrane electrochemical performance.
The proton conductivity of PSSA/PVDF though direct copolymerization of BASS was
shown to be much higher than that of sulfonated PS and PVDF blends with the same level of
PSSA content. For instance, Chen et al.
73
reported that the proton conductivity values of
sulfonated poly(styrene-methyl methacrylate) [P(SSA-MMA)] are ranged from 2 to 41 mS/cm
with 11 to 36 wt% SSA content in the polymer. This is clearly due to the high concentration of
sulfonic acid groups in P(BASS-S-CMS) prepared by the direct copolymerization of BASS,
which ensures one aromatic ring has one sulfonic acid group. The direct copolymerization of
ionic liquid monomer BASS offers a great advantage to synthesize high conductive membranes
as compared to the relatively inefficient and sometimes poorly reproducible post-sulfonations
that often require elevated temperatures or harsh conditions and thus lead to undesired side
reactions.
65
Figure 2.13 PSSA loading effect on proton conductivity and IEC of PSSA/PVDF membranes.
2.5.5 Methanol permeability measurements
DMFC is a promising energy conversion device for the future. However, at present, two
major technical problems must be overcome before DMFCs can validly be proposed for practical
power sources. One issue is slow methanol oxidation kinetics on the anode catalyst. The second
is methanol crossover across the polymer electrolyte membrane from the anode to the cathode
153-
155
. Methanol crossover from anode to the cathode is a very challenging problem that severely
reduces the cell voltage, the fuel utilization and hence the cell performance.
156-158
Since methanol
can be dissolved into water to any degree and the commonly used solid polymer electrolyte,
Nafion
®
, readily absorbs water as well as methanol, it has been found that over 40% of the
methanol can be wasted in DMFCs across Nafion
®
membranes.
25
The study of the methanol
66
permeability is essential for DMFC applications due to its detrimental effect on the DMFC
performance and efficiency as discussed before.
The methanol permeability characteristics of the PSSA/PVDF polymer blend membrane
and Nafion
®
-117 samples were measured using the two vessels apparatus joined together by a
clamp with a membrane sample in between as shown in the schematic diagram of Figure 2.14.
One compartment of the diffusion cell (V
A
=200ml) was filled with an aqueous methanol solution
of (2.0 M) and a trace of 1-butanol (0.1 mM as internal standard), while the other compartment
(V
B
=200ml) contained a 1-butanol (0.1 mM) solution in deionized water. A rectangular
membrane with a diffusion area of 4.52 cm
2
was clamped between the two compartments, and
the methanol permeation cross the membrane from one vessel to the other was monitored by a
Gas Chromatography (GC, Thermo Trace 2000 GC) equipped with a carbowax column.
159,160
The reservoirs were sufficiently large so that pseudo steady-state condition prevailed in which
the methanol concentration in the aqueous compartment was always negligible compared that of
the 2.0 M methanol solution. Assuming that methanol diffusivity inside the membrane is
constant and its partitioning into the aqueous solution does not depend on concentration, the
concentration of methanol in the receiving compartment as a function of time can be given
by:
153,161
where c is the concentration (mol/L). A and L the membrane area and membrane thickness (cm).
D and K are the menthol diffusivity and partition coefficient between the membrane and adjacent
solution respectively. Equation above can be solved to give:
67
where P (= D·K) is the membrane methanol permeability; the parameter, t
0
, also termed time lag,
is explicitly related to the diffusivity: t
0
= L
2
/6D.
The concentration of methanol (C
B
) in the reservoir was measured several times during
an experiment. The internal standard 1-butanol was used as a reference to calculate the
concentration of methanol in reservoir through methanol diffusion. The calculated methanol
concentration was plotted as a function of time and the permeability is calculated from the slope
of resulting straight lines.
Figure 2.14 Diagram of the apparatus used to evaluate methanol permeability.
As shown in Figure 2.15 The plot of methanol concentration into the aqueous
compartment versus time was shown to be linear consistent with a simple diffusion process.
161
It
is important to stress that the rates of methanol diffusion are highly reproducible. The
corresponding methanol diffusion coefficients are plotted vs. PSSA content as shown in Figure
2.16. PSSA/ PVDF membranes with PSSA uptakes below 10 wt% inhibited very low methanol
crossover rates, which suggests that hydrophilic channels in membranes are less connected due
to low PSSA content resulting in a lack of percolation of methanol through membranes. Above
68
this limit the methanol permeability increases sharply between 10 and 15 percent (~ 3 fold),
presumably due to rapidly increasing of well-connected hydrophilic channels thus larger degrees
of percolation.
59
After that the range methanol diffusion appears to increase more or less linearly
with PSSA content. As pure PVDF has a negligible methanol crossover rate, it is clear then that
methanol permeability is mediated by the hydrophilic PSSA domains only. It is worth pointing
out that methanol permeability of PSSA/PVDF membranes fabricated in this study is
significantly lower than that of Nafion
®
-117. For instance, the methanol permeability of typical
membranes with 20 wt% PSSA has a permeability coefficient 5.8 x 10
-7
(cm²·sec
-1
), which only
accounts for 36% of that of Nafion
®
-117 (15.9x 10
-7
(cm²·sec
-1
)). In fact, even membranes with
PSSA contents as high as 35 wt% still give lower methanol diffusion coefficients than that of
Nafion
®
-117.
Figure 2.15 Methanol concentration vs. time of through membranes with varying PSSA
contents.
69
Figure 2.16 Dependence of methanol permeability on PSSA content for PSSA/PVDF
membranes.
As shown in Figure 2.16, PSSA/PVDF membranes achieve lower methanol diffusion
coefficients while exhibiting similar or higher proton conductivities than Nafion
®
-117, which
may originate from the very distinct transport properties and morphologies between hydrocarbon
and perfluorinated PEMs.
59
It has been reported that hydrocarbon based sulfonic acid PEMs have
narrower, less separated and more branched with dead-end “pockets” water channels compared
to Nafion
®
,
59,162
which is agreed with the finding that PSSA/PVDF membranes have smaller
water uptakes as compared to that of Nafion
®
-117 despite their higher proton conductivity
discussed in the water uptake section. The combined effect serves as an advantage in DMFCs
because it can reduce the loss of reactants and the mixed potential, increasing the fuel cell
efficiency and performance.
70
2.5.6 Transmission Electron Microscopy (TEM)
The microstructure of Nafion
®
has been widely investigated by a variety of techniques,
including transmission electron microscopy (TEM). Among the earliest concepts that were set
forth regarding microstructure of Nafion
®
, the cluster-network model proposed Gierke et
al.
163,164
have been widely endured for many years as a conceptual basis for rationalizing the
properties of Nafion
®
membranes, especially ion and water transport and ion permselectvity. As
illustrated in Figure 2.17, the cluster-network model has ~ 4 nm diameter cluster of sulfonate-
ended perfluoroalkyl ether groups that are organized as inverted micelles and arranged on a
lattice. These micelles are connected by proposed pores or channels with ~1 nm in size.
46
These -
SO
3
coated channel were proposed to account for inter cluster ion hopping of positive charge
species but rejection of negative ions. Later a three-phase model first suggested by Rodmacq and
co-workers and later improved by Yeager and Steck was introduced.
165,166
In this model, Nafion
®
consists of three regions: a perfluorocarbon hydrophobic backbone, an interfacial region, and
ionic clusters. As compared to the model of Gierke et al.,
164
the three-phase model does not have
a strict geometrical definition (spherical inverted micelles connected by cylindrical pores) and
their geometrical distribution has a lower degree of order. Most importantly, there’re transitional
interphases between hydrophobic and hydrophilic regions, a concept that is becoming
increasingly accepted.
46
Although these and other proposed models for the microstructure of Nafion
®
membranes
may be not able to accurately describe PEMs based on PSSA/PVDF polymer blends, the
recognition that the ionic groups aggregate in the hydrophobic polymer matrix to form a network
of clusters that allow for efficient ionic transport through these nanometer-scale domains is
generally accepted.
46,59,68
As water management and ionic transport are closed related to the
71
microstructure of polymer electrolyte membranes, a study of nanoscale morphology is important
to tune PSSA/PVDF membranes for optimum performances in DMFCs.
Figure 2.17 Cluster-network model for the morphology of hydrated Nafion
®
. (adapted from
ref.
46
)
TEM observation was performed on a JEOL JEM-2100F TEM at an accelerating voltage
of 200 kV. The membrane samples were immersed in saturated Pb(NO
3
)
2
aqueous solution for 2
days to stain the ionic clusters of membranes. The films were then washed with deionized water
and dried at room temperature. The stained membrane were embedded in an epoxy resin (Low
Viscosity Embedding Media Spurr's Kit, Spurr Co, PA) and cut into TEM specimens with a
Leica EM UC6 Cryo-Ultramicrotome.
The TEM morphology of 20% PSSA and 30% PSSA membranes were shown in Figure
2.18 and Figure 2.19, respectively. In the micrograph, the dark sites represent ionic domains,
which were stained with Pb(NO
3
)
2
solution. The degree of darkness of ionic domains in TEM
image depends on -SO
3
H group aggregation density and the deepness of the ionic cluster below
the film surface. As shown in Figure 2.18, the clusters of ionic aggregation have a size around
3~4 nm, which is smaller than the reported ionic cluster size of Nafion
®
(> 7 nm).
167,168
Perfluorosulfonic polymers like Nafion
®
combine the extremely high hydrophobicity of the
72
perfluorinated backbone (tetrafluoroethylene) with the extremely high hydrophilicity of the
sulfonic acid groups in one macromolecule, which gives rise to prominent
hydrophobic/hydrophilic nano-separation especially in the presence of water. Compared to
Nafion
®
membranes, PSSA/PVDF membranes have smaller hydrophilic/hydrophobic difference,
thus less pronounced separation of hydrophilic-hydrophobic domains.
46,59
The smaller ionic
cluster creates small proton transporting channels contributing to the distinct low methanol
permeabilities of PSSA/PVDF membranes, which is of significant importance for applications in
direct liquid feed methanol fuel cells. The even dispersion of ionic clusters also indicates the
excellent homogeneity of the hydrophilic component (PSSA) dispersed in the hydrophobic
matrix (PVDF).
Figure 2.18 Transmission Electron Microscopy (TEM) of membrane M20 stained with
Pb(NO
3
)
2
.
73
Figure 2.19 TEM image of membrane M30 stained with Pb(NO
3
)
2
.
2.5.7 Thermal stability measurements
A thermal stability analysis of PEMs is required to assure an adequate thermal stability
for their use in DMFCs. Several studies have investigated the thermal stability of Nafion
®
and
analogous polymers. Chu et al.
169
used IR spectroscopy to study the effect of heating Nafion
®
membranes with an equivalent weight of 1100 at temperatures of 22, 110, 160, 210, and 300 °C.
These samples were coated onto a platinum film and heated in air. The study concluded that
Nafion
®
loses its sulfonic acid groups after being heated at 300 °C for 15 min. Surowiec et al.
170
conducted a thermal study of Nafion
®
membranes using thermogravimetric analysis (TGA),
differential thermal analysis (DTA), and FT-IR spectroscopy. Their studies conclude that
Nafion
®
membranes was stable up to 280 °C. Another study was conducted by Samms et al.
171
in
simulated fuel cell environments. In all cases, Nafion
®
membranes were found to be stable up to
280 °C, at which temperature the sulfonic acid groups began to decompose.
74
The thermal stability of our PSSA/PSSA polymer blend membranes was studied with a
thermogravimetric analyzer (SHIMADZU TGA-50) under ambient air atmosphere. The sample
was placed in a platinum crucible and heated from 30 °C to 800 °C with a heating rate of 10
°C/min. During the heating the mass loss of membrane samples was recorded and plotted with
temperature.
Figure 2.20 Thermal degradation study of PSSA/PVDF polymer blend membranes.
The thermal properties of PSSA/PVDF membrane samples and corresponding precursors
terpolymer P(TBASS-S-CMS) and PVDF were obtained in Figure 2.20. The initial < 2% weight
loss of all samples are attributed the loss of water in the samples.
103,172
The weight loss of pure
P(TBASS-S-CMS) starting from 260 °C may be associated with the onset of desulfonation of
tetra-butylammonium sulfonate groups. The PSSA/PVDF membrane on the other hand shows
three-step decomposition with additional step of PVDF degradation. Compare to the precursor
75
P(BASS-S-CMS), the onset decomposition temperature of PSSA/PVDF membranes increased to
300 °C from 260 °C. The results from the TGA analysis suggest that PSSA/PVDF membranes is
suitable for applications in DMFCs which have a typical operating temperature below 90 °C. The
thermal analysis of PSSA/PVDF membranes also suggests that these membranes can also be
applied in hydrogen fuel cell applications, in which a higher operational temperature (>120 °C)
is generally preferred as it increases the kinetic rates for the fuel cell reactions and reduces issues
of catalyst poisoning by absorbed carbon monoxide.
42,45
2.5.8 Scanning Electron Microscope (SEM) of PSSA/PVDF membranes
Blending of two very different polymers in general is difficult especially if the polymer
properties are very different. In poorly miscible blends, the interfaces and morphology are gross,
thus greatly reducing the strength and toughness of such blends. Sulfonated PS were shown to be
immiscible with PVDF, as a results, membranes prepared by mixing sulfonated PS and PVDF
often lead to and weak mechanical strength or brittleness.
103,104,108
The preferred remedy for such
blends is to use compatibilizers such as copolymer of styrene and methyl methacrylate (MMA),
to bond the phases. However, compatibility enhancement by such compatibilizers was shown to
be very limited. For instance, Piboonsatsanasakul et al. prepared a PEM based on sulfonated
PS/PVDF blend compatibilized with PS-co-PMMA block copolymer. A typical SEM image of
the resulting membrane showed a very coarse membrane surface and large domain sizes due to
phase separation as shown in Figure 2.21.
103
Hong et al.
73
also observed a similar phase
separated morphology for membranes prepared by embedding PSSA copolymer P(MMA-SSA)
into PVDF matrix.
76
Figure 2.21 Scanning electron micrograph of the sulfonated polystyrene/PVDF blend membrane
compatibilized with PS-b-PMMA block copolymer (adapted from ref.
103
).
PSSA/PVDF membranes in the present study exhibit high optical clarity and toughness,
which is a clear indication that these membranes may have a distinct morphology as compared to
membrane based on mixing sulfonated PS and PVDF. The surface morphology of the modular
membranes was studied by scanning electron microscopy (SEM) (JEOL JSM-7001). Samples of
PSSA/PVDF polymer blend membranes were coated with a thin layer of gold using a sputter
coater (Cressington) prior to imaging with SEM.
As shown in Figure 2.22, PSSA/PVDF membranes exhibit very smooth surfaces with no
separated individual polymer domains except some occasionally irregular spots resulted from
sample preparation in contrast with the rough and phase-separated SEM morphology in Figure
2.21. This smooth surface indicates a substantial homogeneity of PSSA/PVDF polymer blends
was achieved by the novel camouflaged method described in the membrane preparation. For
polymer blends, the overall mechanical performance was directly related to by the interfacial
adhesion between different polymer phases. This apparent homogeneity of PSSA/PVDF ensures
77
the necessary mechanical strength for DMFC applications, which is discussed in the mechanical
properties section.
Figure 2.22 SEM image of membrane M30.
2.5.9 Energy Dispersive X-ray Electronic Spectroscopy (EDS)
The sulfonic acid group distribution is a major parameter in designing polymer
electrolyte membranes based on polymer blends. The nonuniform distribution of sulfonic acid
groups results in reduction in proton conductivity and poor reproducibility. The distribution of
PSSA on the membrane was investigated using energy-dispersive X-ray spectroscopy (EDS).
Samples of PSSA/PVDF membranes were coated with a thin layer of gold using a sputter coater
(Cressington) prior to imaging.
As shown in Figure 2.23, the element S in the membrane is found to be evenly
distributed on the entire surface rather than aggregated into separated domains from PVDF
domains. The homogeneity of PSSA distribution also indicates that PSSA/PVDF membranes
78
fabricated by our optimized membrane preparation technique could be highly reproducible as
membranes with uneven distribution of sulfonic acid groups may result in different electrical
performances depending on which area of membranes was utilized as electrodes. The density of
the –SO
3
H groups on the PSSA/PVDF modular membrane was also confirmed by quantitative
energy dispersive X-ray spectroscopy (EDS) mapping. Quantitative analysis (Figure 2.24)
shows that the element mass ratios of carbon, oxygen, fluorine and sulfur are 50.2 wt%, 3.8 wt%,
42.9 wt%, and 2.48 wt%, respectively. The density of –SO
3
H group on the surface of the
PSSA/PVDF membrane was calculated to be ~0.8 mmol g
-1
, which is lower than the ion
exchange capacity of the membrane 1.22 mmol g
-1
, probably due to the lower surface energy of
fluorine atoms which overwhelming the surface of the membrane.
Figure 2.23 EDS element mapping of sulfur on the surface of a PSSA/PVDF membrane.
79
Figure 2.24 Element compositions on the surface of a PSSA/PVDF membrane coated with a thin
layer of gold.
2.5.10 Chemical stability measurements
The accelerated test for oxidative stability of the blend membrane was also conducted by
soaking a small piece of well-dried membrane in Fenton’s reagent (3% H
2
O
2
+ 2 ppm FeSO
4
)
173-
175
at 80 °C and monitoring the crack formation. At the mean time, the weight loss (wt%) of
membrane after immersing in Fenton’s reagent for 24 h was recorded.
Table 2.4 lists the oxidative stability test, which is evaluated by the weight ratio of dry
membranes after and before Fenton test. The blend membranes are found to be highly stable
against chemical oxidation, as no crack was formed in membranes and nearly no weight loss is
detected after the oxidative test. The high oxidative stability of PSSA/PVDF may be attributed to
the high chemical stability of membrane matrix PVDF, which is known to have superior
inertness to many organic solvents, oils, and acids.
109,110
These results also validates our initial
design of polymer electrolyte membrane using polymer blends, in which one component
80
provides thermal, chemical and mechanical stability to the membrane while the other offers
proton transport capacity.
Table 2.4 Oxidative stability of the blend membranes.
Membrane
PSSA Content in
membrane (wt%)
Oxidative stability (%) (in Fenton's
reagent at 80 °C for 24h)
14-43 15% > 99.9%
14-33 20% > 99.9%
14-38 25% > 99.9%
14-48 30% > 99.8%
2.5.11 Mechanical properties
The mechanical integrity of membranes as mounted in cells, and under perturbation of
pressure gradients, swelling-dehydration cycles, mechanical creep, extreme temperatures, and
the onset of brittleness and tear resistance are important and must be taken into consideration
when designing a polymer electrolyte membrane.
60
In most cases, the literature of PEMs based
on polymer blends has largely been concerned with proton ductility and electrical performance
while the mechanical property was not always documented. Wootthikanokkhan et al.
105
studied
the methanol permeability and proton conductivity of sulfonated PEEK/PVDF blends, and found
out that methanol permeability was reduced by blending with PVDF at the expense of decreasing
proton conductivity. However, no mechanical property was mentioned. Hong et al.
73
prepared a
PEM by embedding PSSA into PVDF matrix and discrete domains aggregated by SSA segments
have been observed throughout the membrane. And micro-cracks were observed when SSA
content was higher than 40 wt%. Kim et al.
74
prepared SPS/PTFE composite membrane by
impregnating styrene/divinylbenzene (DVB) in PTFE substrate followed by polymerization and
81
crosslinking. The composite membranes were shown to have comparable and higher ion
conductivity and lower methanol permeability than Nafion
®
-117 membranes. Yin et al.
176
studied PEMs based on SPEEK and PVDF polymer blends, which exhibited comparable proton
conductivity but lower methanol permeability as compared to Nafion
®
-117 membranes.
However, no morphology and mechanical strength measurement was reported in those studies.
As MEA is the heart of a fuel cell, the endurance of polymer electrolyte membranes and
MEAs directly determines the life span of fuel cells.
49,177,178
The membranes need to be flexible
enough and thermally stabile to tolerate the high pressure and the high temperature during
membrane electrolyte assembly (MEA) fabrication. Also PEMs need to have high deformation
resistance to ensure the long-term operation of fuel cells. Tensile test was conducted to study
Young’s modulus, tensile strength and tensile elongation of PSSA/PVDF polymer blend
membranes. The impact of PSSA content on the mechanical properties of membranes was also
investigated. As shown in Table 2.5 and Figure 2.25, the Young’s moduli of the PSSA-PVDF
membranes were much higher (up to 10 fold) than that of the Nafion
®
-117. The maximum
strengths were also between 40 and 120% higher. It is important to note that the mechanical
properties of PSSA-PVDF was far better than SPS/PVDF membranes made by post-
polymerization-sulfonation, (Young’s moduli 6.85 MPa, Elongation at break < 5%)
103,104
giving
much higher Young’s moduli as well as greater strains at membrane failure. As the tensile tests
were performed on fully hydrated membrane, the strengths of membranes decreased with
increasing of PSSA content as water acts as a plasticizer. The superior mechanical strength and
flexibility of PSSA/PVDF membranes is attributed to the enhanced compatibility and
homogeneity between proton conductor component and polymer scaffold.
82
Table 2.5 Mechanical property of membranes with various PSSA loading.
Membranes Young’s
Modulus (GPa)
Maximum
strength (MPa)
Maximum
elongation (%)
M15 1.36 50.6 76
M20 0.95 38.2 96
M25 0.55 34.1 159
Nafion
®
-117 0.11 24 >180
Figure 2.25 Tensile tests of PSSA/PVDF polymer blend membranes with different PSSA
loading contents.
2.5.12 Ultrafiltration properties of PSSA/PVDF membranes
Water transport through polymer electrolyte membranes is essential for the proton
transport trough polymer electrolyte membrane in DMFCs.
179,180
Water transport trough PEMs
under fuel cell operation conditions occurs in two major ways: diffusion of water due to the
active concentration gradient, and electro-osmotic drag (EOD) of water with protons from the
anode to cathode
165,181,182
. To attain optimal fuel cell performance, it is critical to have a good
water balance so that the polymer electrolyte is hydrated for sufficient proton conductivity while
the cathode flooding and the methanol crossover are minimized.
83
2.5.12.1 Water diffusion of PSSA/PVDF membranes
PSSA/PVDF membranes with various PSSA compositions and Nafion
®
-117 were
punched into circles with a 1.0 inch diameter and was placed in a stirred cell (millipore, 8010)
(Figure 2.26), supported by a nitrocellulose filter membrane (millipore, pore size 0.65 µm, diam.
25 mm).
90
Deionized water (10ml) was introduced above the membrane and the stirred cell was
pressurized with 60 psi nitrogen. The effluent was collected and the water flux was calculated.
The water transport property of PSSA/PVDF membranes and Nafion
®
-117 under pressure is
shown in Table 2.6. Membrane 14-71 with 20 wt% PSSA has a much smaller water flux as
compared to Nafion
®
-117 under the same conditions despite similar ion exchange capacities. The
water flux of PSSA/PVDF polymer blend membranes increased with increasing of PSSA content
in membranes. For example, a significantly higher water flux of 2.14 x 10
-6
ml·s
-1
·arm
-1
was
observed for membrane 14-71 with 40 wt% PSSA, which is more than 28 times higher than that
of membrane 14-95 with 20 wt% PSSA. Further increase of PSSA content to 50 wt%,
PSSA/PVDF membranes eventually exhibit a slightly higher water flux than Nafion
®
-117 under
the same conditions.
PSSA/PVDF membranes generally exhibit smaller water diffusion coefficients even with
high PSSA loadings (40 wt%). As discussed previously, the less pronounced microphase
separation in PSSA/PVDF membranes, resulting in narrow, more branched with more dead-end
“pockets” water filled channels, may contribute to the distinctly smaller water flux. This unique
property is essential for DMFC applications, as high water flux and methanol crossover often
cause water flooding and mixed potential at the cathode, respectively.
84
Figure 2.26 Stirred Cell used for ultrafiltration experiments.
Table 2.6 Ultrafiltration property of PSSA/PVDF membranes and Nafion
®
-117.
Membrane PSSA Content Thickness (mm)
Hydraulic
Pressure (psi)
Water Flux
(10
-6
ml·s
-1
·atm
-1
)
Nafion
®
-- 0.17 60 3.57
14-71 20 wt% 0.17 60 0.075
14-95 40 wt% 0.15 60 2.14
14-96 50 wt% 0.15 60 4.38
2.5.12.2 Ultrafiltration experiments of PSSA/PVDF membranes
For PSSA/PVDF membranes, the hydrophilic PSSA domains evenly distributed in PVDF
matrix and form nano-scale and probably well-connected hydrophilic channels enabling water
transport. In principle, the number and the size of those hydrophilic channels can be fine tuned
by adjusting the PSSA content in the membrane, which promote us to explore the potential
application of PSSA/PVDF blends as ultrafiltration membranes.
90,183,184
85
Gregor et al.
90
studied the ultrafiltration properties of highly porous PSSA-PVDF
interpolymer ion-exchange membranes prepared by blending sulfonated polystyrene with PVDF.
The membrane was reported has excellent rejections of erythrosin (> 96%) and bovine serum
albumin (> 99%) and the rejection was relatively independent of the solute concentration. In
addition, those membranes also exhibited some rejection of KCl depending on its concentration
(0.001 M KCl: 100%; 0.01 M KCl: 55%, 1.0 M KCl: 0%). As those membranes were prepared
by direct blending sulfonated PS with PVDF, which may create voids and phase separation due
to the poor miscibility between PSSA and PVDF, the ultrafiltration property studied by Gregory
et al. might be different from our membranes.
In order to study the ultrafiltration property of PSSA/PVDF membranes, a dyed
polystyrene (PS) suspension with 29 nm diameter nano-spheres (Bangs Laboratories) and NaCl
with various concentrations were used as the solutes and membrane 14-95 with 40 wt% PSSA
was chosen as the filter membrane as it exhibits a moderate water flux as compared to Nafion
®
-
117. As shown in Figure 2.27, a feed solution of PS nano-spheres was ultrafiltered through
membrane 14-95 and become optically clear and colorless, indicating the red PS spheres in the
solution had been successfully filtered out. The removal of PS nano-spheres through
PSSA/PVDF membrane was also confirmed by the later optical charity measurement. As shown
in Figure 2.28, the striking contrast between the transmittance differences of original dyed PS
nano-sphere suspension (before ultrafiltration) and that of filtrate (after ultrafiltration) indicating
the PS nano-spheres were removed through PSSA/PVDF membranes.
86
Figure 2.27 Photographs of dyed polystyrene sphere solution (left) and the filtrate (right) after
ultrafiltration using PSSA/PVDF membrane with 40 wt% PSSA.
Figure 2.28 Comparison of transmittances of solutions before (blue) and after (red)
ultrafiltration.
87
In the case of NaCl rejection experiments, the effect of electrolyte concentration on solute
rejection was determined at a constant hydraulic pressure of 4 atm. Different concentrations of
NaCl were ultrafiltered and the salt rejection rates were qualitatively assessed by titrating the
filtrates with AgNO
3
. The results are demonstrated by comparing the photographs of the feed
solutions and the filtrates titrated with AgNO
3
in Figure 2.29. Compared to the feed solutions,
the filtrates exhibit less or no precipitate of AgCl depending on the concentration of NaCl. It is
clearly that the salt rejection increases as the concentration of the feed solution is decreased from
0.01 M to 0.001 M, which is consistent with the result observed by Gregor et al.
90
The
preliminary results of ultrafiltration experiment also suggest that PSSA/PVDF polymer blend
membranes emerge as possible applications in water treatment.
90,184
In principle, the
ultrafiltration performance of PSSA/PVDF polymer blend membranes can be further optimized
by adjusting the amount of PSSA or incorporating nano-particle fillers, such as graphene
oxide.
185
Figure 2.29 Photographs of feed solutions and ultrafiltrates of NaCl titrated with AgNO
3
.
88
2.5.13 Reproducibility measurements
Most sulfonated or similarly modified PEMs, including that of polystyrene, poly(arylene
ether)s, polyimides, and polyphosphazenes, and several others, are prepared based on post-
polymerization sulfonation or analogous chemical modification of suitable polymer precursors.
There are several potential drawbacks involved in post-polymerization sulfonation methods. For
instance, post polymerization methods typically require high temperatures or aggressive
reagents. These often result in a lack of control over the degree of reaction, the regiochemistry of
the introduced ionic groups and the often inevitable side reactions including polymer
degradation.
60
Chemical modification often implicitly affects both mechanical and proton
conductance properties and hence provides a lesser degree of control. Moreover, the opportunity
to control and/or increase molecular weight to enhance durability is not feasible if one choose to
conduct a post-reaction on an existing commercial product.
60
In conclusion, polymer electrolyte
membranes prepared by post-polymerization sulfonation often encounter poor reproducibilities.
The direct (co)polymerization of styrene sulfonate ensures the reproducibility of
membrane as the sulfonate group remains intact during polymerization. Hence the properties of
the membranes are expected to be subject to very small random fluctuations. To illustrate the
reproducibility of membranes, characterization of several membrane samples with the same
blend composition (20 wt% PSSA) but fabricated using different polymer precursors at different
days have been conducted and compared. As shown in Figure 2.30, proton conductivities of
three different membrane samples with the same composition exhibited nearly identical proton
conductivities (2.01 mS/cm ± 2.7%). This reproducibility was found as well for methanol
permeability as demonstrated by nearly overlapped methanol diffusion curves (Figure 2.30). The
properties of the membranes are subject to very small random fluctuations considering all
89
membranes are hand-made in our lab. These results indicated that membranes prepared using our
methodology were highly reproducible.
Figure 2.30 Reproducibility demonstration of proton conductivity and methanol permeability of
samples fabricated with different precursors on different days.
A high reproducibility is achievable when designing PEMs through polymer blends along
with direct copolymerization of sulfonated monomers, like BASS, is. Both the proton transport
component and the inert matrix polymer can be controlled independently with regard to the
molecular weight and polymer composition. The direct polymerization of ionic liquid monomer
with inherited sulfonic acid groups allows precisely control of the amount of sulfonic acid in
membranes, which is often difficult to achieve by post-polymerization sulfonation. Therefore,
the dedicated balance between proton conductivity and methanol permeability can be accurately
regulated by adjusting the composition of polymer blends.
90
2.6 References
(106) Manea, C.; Mulder, M. Journal of Membrane Science 2002, 206, 443.
(107) Mollá, S.; Compañ, V. International Journal of Hydrogen Energy 2014, 39, 5121.
(108) Wu, H.-L.; Ma, C.-C. M.; Kuan, H.-C.; Wang, C.-H.; Chen, C.-Y.; Chiang, C.-L.
Journal of Polymer Science Part B: Polymer Physics 2006, 44, 565.
(109) Wenwen, H.; Meng, Z.; Fan, Y.; Lorne, F.; Parisa, H.; Leonard James, M.; Tyler,
S.; Thomas, S.; Peggy, C. MRS Proceedings 2012, 1384.
(110) Ameduri, B. Chemical Reviews 2009, 109, 6632.
(111) Qiu, X. P.; Li, W. Q.; Zhang, S. C.; Liang, H. Y.; Zhu, W. T. Journal of the
Electrochemical Society 2003, 150, A917.
(112) Brandrup, J.; Immergut, E. H.; Grulke, E. A. Polymer Handbook; Wiley, 2003.
(113) Carretta, N.; Tricoli, V.; Picchioni, F. Journal of Membrane Science 2000, 166,
189.
(114) van Albada, M. P.; Cerfontain, H. Journal of the Chemical Society, Perkin
Transactions 2 1977, 1548.
(115) van Albada, M. P.; Cerfontain, H. Journal of the Chemical Society, Perkin
Transactions 2 1977, 1557.
(116) Dai, C.-A.; Liu, C.-P.; Lee, Y.-H.; Chang, C.-J.; Chao, C.-Y.; Cheng, Y.-Y.
Journal of Power Sources 2008, 177, 262.
(117) Arunbabu, D.; Sanga, Z.; Seenimeera, K. M.; Jana, T. Polymer International
2009, 58, 88.
(118) Laruelle, G.; Nicol, E.; Ameduri, B.; Tassin, J.-F.; Ajellal, N. Journal of Polymer
Science Part A: Polymer Chemistry 2011, 49, 3960.
(119) Yao, H.; Wilkie Charles, A. In Fire and Polymers; American Chemical Society:
2001; Vol. 797, p 125.
(120) Wang, Z.; Jiang, D. D.; McKinney, M. A.; Wilkie, C. A. Polymer Degradation
and Stability 1999, 64, 387.
(121) Rabek, J. F.; Lucki, J. Journal of Polymer Science Part A: Polymer Chemistry
1988, 26, 2537.
(122) Brauman, S. K. Journal of Polymer Science: Polymer Chemistry Edition 1979,
17, 1129.
91
(123) Li, J.; Wilkie, C. A. Polymer Degradation and Stability 1997, 57, 293.
(124) Grassie, N.; Gilks, J. Journal of Polymer Science: Polymer Chemistry Edition
1973, 11, 1985.
(125) Kogej, K. Advances in Colloid and Interface Science 2010, 158, 68.
(126) Kötz, J.; Kosmella, S.; Beitz, T. Progress in Polymer Science 2001, 26, 1199.
(127) Dobrynin, A. V.; Rubinstein, M. Progress in Polymer Science 2005, 30, 1049.
(128) Sen, A. K.; Roy, S.; Juvekar, V. A. ISRN Analytical Chemistry 2012, 2012, 5.
(129) Bottino, A.; Cameraroda, G.; Capannelli, G.; Munari, S. Journal of Membrane
Science 1991, 57, 1.
(130) Siemann, U. In Scattering Methods and the Properties of Polymer Materials;
Springer Berlin Heidelberg: 2005; Vol. 130, p 1.
(131) Park, Y. J.; Kang, Y. S.; Park, C. European Polymer Journal 2005, 41, 1002.
(132) Buonomenna, M. G.; Macchi, P.; Davoli, M.; Drioli, E. European Polymer
Journal 2007, 43, 1557.
(133) Moore, R. B.; Martin, C. R. Macromolecules 1988, 21, 1334.
(134) Sood, R.; Iojoiu, C.; Espuche, E.; Gouanvé, F.; Gebel, G.; Mendil-Jakani, H.;
Lyonnard, S.; Jestin, J. The Journal of Physical Chemistry C 2014, 118, 14157.
(135) Weber, J.; Janda, P.; Kavan, L.; Jegorov, A. Journal of Electroanalytical
Chemistry and Interfacial Electrochemistry 1986.
(136) Benz, M.; Euler, W. B. Journal of Applied Polymer Science 2003, 89, 1093.
(137) Gregorio, R. Journal of Applied Polymer Science 2006, 100, 3272.
(138) Li, M.; Wondergem, H. J.; Spijkman, M.-J.; Asadi, K.; Katsouras, I.; Blom, P. W.
M.; de Leeuw, D. M. Nat Mater 2013, 12, 433.
(139) Moore, R. B.; Martin, C. R. Analytical Chemistry 1986, 58, 2569.
(140) Gregor, H. P.; Wetstone, D. M. The Journal of Physical Chemistry 1957, 61, 147.
(141) Gryte, C. C.; Gregor, H. P. Journal of Polymer Science: Polymer Physics Edition
1976, 14, 1839.
(142) Wang, J. H.; Li, N. W.; Cui, Z. M.; Zhang, S. B.; Xing, W. Journal of Membrane
Science 2009, 341, 155.
92
(143) Kerres, J.; Zhang, W.; Ullrich, A.; Tang, C. M.; Hein, M.; Gogel, V.; Frey, T.;
Jörissen, L. Desalination 2002, 147, 173.
(144) Kopitzke, R. W.; Linkous, C. A.; Anderson, H. R.; Nelson, G. L. Journal of The
Electrochemical Society 2000, 147, 1677.
(145) Lucid, J.; Meloni, S.; MacKernan, D.; Spohr, E.; Ciccotti, G. The Journal of
Physical Chemistry C 2012, 117, 774.
(146) Kim, Y. W.; Lee, D. K.; Lee, K. J.; Kim, J. H. European Polymer Journal 2008,
44, 932.
(147) Cahan, B. D. Journal of The Electrochemical Society 1993, 140, L185.
(148) Zawodzinski, T. A.; Derouin, C.; Radzinski, S.; Sherman, R. J.; Smith, V. T.;
Springer, T. E.; Gottesfeld, S. Journal of The Electrochemical Society 1993, 140, 1041.
(149) Slade, S.; Campbell, S. A.; Ralph, T. R.; Walsh, F. C. Journal of The
Electrochemical Society 2002, 149, A1556.
(150) Sone, Y.; Ekdunge, P.; Simonsson, D. Journal of The Electrochemical Society
1996, 143, 1254.
(151) Moukheiber, E.; De Moor, G.; Flandin, L.; Bas, C. Journal of Membrane Science
2012, 389, 294.
(152) Peckham, T. J.; Schmeisser, J.; Rodgers, M.; Holdcroft, S. Journal of Materials
Chemistry 2007, 17, 3255.
(153) Tricoli, V.; Carretta, N.; Bartolozzi, M. Journal of The Electrochemical Society
2000, 147, 1286.
(154) Scott, K.; Taama, W. M.; Argyropoulos, P.; Sundmacher, K. Journal of Power
Sources 1999, 83, 204.
(155) Heinzel, A.; Barragán, V. M. Journal of Power Sources 1999, 84, 70.
(156) Ren, X.; Springer, T. E.; Zawodzinski, T. A.; Gottesfeld, S. Journal of The
Electrochemical Society 2000, 147, 466.
(157) Zhang, J.; Wang, Y. Fuel Cells 2004, 4, 90.
(158) Han, J.; Liu, H. Journal of Power Sources 2007, 164, 166.
(159) Tricoli, V. Journal of The Electrochemical Society 1998, 145, 3798.
(160) Jiang, R. Z.; Chu, D. R. Journal of the Electrochemical Society 2004, 151, A69.
(161) Tricoli, V. Journal of The Electrochemical Society 1998, 145, 3798.
93
(162) Schneider, N. S.; Rivin, D. Polymer 2006, 47, 3119.
(163) Hsu, W. Y.; Gierke, T. D. Journal of Membrane Science 1983, 13, 307.
(164) Gierke, T. D.; Munn, G. E.; Wilson, F. C. Journal of Polymer Science: Polymer
Physics Edition 1981, 19, 1687.
(165) Yeager, H. L.; Steck, A. Journal of The Electrochemical Society 1981, 128, 1880.
(166) Yeager, H. L.; Steck, A. Analytical Chemistry 1979, 51, 862.
(167) Xu, K.; Chanthad, C.; Gadinski, M. R.; Hickner, M. A.; Wang, Q. ACS Appl
Mater Interfaces 2009, 1, 2573.
(168) Lin, H. L.; Yu, T. L.; Han, F. H. Journal of Polymer Research 2006, 13, 379.
(169) Chu, D.; Gervasio, D.; Razaq, M.; Yeager, E. B. Journal of Applied
Electrochemistry 1990, 20, 157.
(170) Surowiec, J.; Bogoczek, R. Journal of Thermal Analysis 1988, 33, 1097.
(171) Samms, S. R.; Wasmus, S.; Savinell, R. F. Journal of The Electrochemical
Society 1996, 143, 1498.
(172) Nasef, M. M.; Saidi, H. Polymer Degradation and Stability 2000, 70, 497.
(173) Seulgi, K.; Hyunjung, L.; Dahee, A.; Hae Woong, P.; Taihyun, C.; Wonmok, L.
Journal of Membrane Science 2013, 427.
(174) Jung, H.-Y.; Park, J.-K. International Journal of Hydrogen Energy 2009, 34,
3915.
(175) Unveren, E. E.; Inan, T. Y.; Çelebi, S. S. Fuel Cells 2013, 13, n/a.
(176) Xue, S.; Yin, G. Polymer 2006, 47, 5044.
(177) Huang, X.; Solasi, R.; Zou, Y.; Feshler, M.; Reifsnider, K.; Condit, D.; Burlatsky,
S.; Madden, T. Journal of Polymer Science Part B: Polymer Physics 2006, 44, 2346.
(178) Hussein, I.; Zaidi, S. M. J. In Polymer Membranes for Fuel Cells; Zaidi, S. M. J.,
Matsuura, T., Eds.; Springer US: 2009, p 235.
(179) Ge, S.; Li, X.; Yi, B.; Hsing, I.-M. Journal of The Electrochemical Society 2005,
152, A1149.
(180) Mittelsteadt, C. K.; Staser, J. ECS Transactions 2011, 41, 101.
(181) Okada, T.; Xie, G.; Gorseth, O.; Kjelstrup, S.; Nakamura, N.; Arimura, T.
Electrochimica Acta 1998, 43, 3741.
94
(182) Jiao, K.; Li, X. Progress in Energy and Combustion Science 2011, 37, 221.
(183) Uragami, T.; Fujimoto, M.; Sugihara, M. Polymer 1981, 22, 240.
(184) Uragami, T.; Fujimoto, M.; Sugihara, M. Desalination 1980, 34, 311.
(185) Wang, Z.; Yu, H.; Xia, J.; Zhang, F.; Li, F.; Xia, Y.; Li, Y. Desalination 2012,
299, 50.
95
CHAPTER 3: Electrical Performance
Although DMFCs are more efficient with pressurized oxygen, but these conditions end
up causing so many losses in the complete system that the advantage of simplicity of DMFCs is
lost. Therefore, all electrical measurements are measured without back pressure (ambient) unless
otherwise noted.
3.1 Fuel cell performance of PSSA/PVDF membranes cast from acetone
Although the membranes cast from acetone suffered from wrinkled appearance and
uneven surface, several carefully selected membranes were fabricated into membrane electrolyte
assemblies (MEAs). The electrical performances of membranes cast from acetone were
measured in order to evaluate and compare DMFC performances of membranes prepared by
different methods.
3.1.1 Membrane electrolyte assembly fabrication
The membrane electrode assembly (MEA) is the heart of proton exchange membrane fuel
cells (PEMFCs), and determines both fuel cell performance and durability.
186
The structure of a
MEA for DMFC generally consists of an anode gas diffusion layer (GDK), an anode catalyst
layer (CL), a proton exchange membrane (PEM), a cathode catalyst layer, and a cathode gas
diffusion layer, as shown in Figure 3.1. An ideal MEA allows all active catalyst sites to be
accessible to fuels, protons and electrons, and minimizes the impediments for proton transport
and product removal. Over the past several decades, great efforts have been made to optimize the
catalyst layer and MEA. Significant improvements in MEA fabrication and the corresponding
96
electrical performance have since been made by the incorporation of Nafion
®
-H ionomer into the
catalyst layer.
The initial MEA fabrication procedure was based on the modified procedure described in
elsewhere.
92,187
This process involved two major steps, namely painting catalyst inks onto
Teflon-impregnated carbon electrodes, and assembling catalyst coated carbon electrodes and
membrane into a MEA using hot-press. The composition of catalyst inks was reported to play an
essential role in optimization of MEAs and their electrical performance. The catalyst ink
generally consists of Platinum-Ruthenium (Pt-Ru) (anode) or Platinum black (Pt) (cathode),
Nafion
®
-H ionomer (5 wt% solution dispersed in lower alcohols) and deionized water (DI H
2
O).
The detailed composition of catalyst inks for both anode and cathode is listed in Table 3.1. Other
important factors that affect the quality of MEAs are the temperature and the pressure used
during hot-press process. High temperatures (> 140 °C) and high pressures (> 250 psi) are
generally used to improve the interfacial bonding between electrode and electrolyte. However,
PSSA/PVDF membranes cast from acetone suffered from brittleness in the dry state, and such
conditions in hot-press process often end up with MEAs with cracks. Therefore, the initial
parameters used for hot-press of MEAs with membranes cast from acetone were adjusted to
milder conditions with a temperature of 100 °C and a pressure 125 psi.
97
Figure 3.1 Schematic illustration of the structure of a membrane electrolyte assembly (MEA)
used in DMFCs.
Table 3.1 Anode and cathode catalyst ink compositions for 25 cm
2
MEAs.
b
Electrode Catalyst/g Nafion
®
-H
a
/g DI H
2
O/g
Anode 0.2g Pt-Ru 0.2 g 0.8g
Cathode 0.2g Pt 0.2g 0.8g
a. 5% Nafion
®
-H ionomer solution with 1100 equivalent weight.
b. Catalyst ink was painted on carbon paper using hand-brushing method.
3.1.2 Electrical performance of MEA 12-20
The PSSA/PVDF membrane with 20 wt% of PSSA (membrane 12-20) was chosen for
preliminary testing of electrical performance as the membrane showed a similar proton
conductivity as Nafion
®
-117. The PSSA uptake and related properties of membrane 12-20 as
compared to Nafion
®
-117 is shown in Table 3.2. The MEA of membrane 12-20 (MEA 12-20)
was fabricated with the ink composition shown in Table 3.1 and used a pressure of 125 psi and a
temperature of 100 °C during the hot-press process. The MEA 12-20 had an open circuit voltage
98
(OCV) resistance of 69 mOhms at 30 °C, which was considerably higher than that of Nafion
®
-
117 (19 mOhms) at the same temperature despite the similar proton conductivities. The disparity
in resistance values suggests that the wrinkled surface and unfavorable conditions used in hot-
press could have contributed to the poor contact between the electrodes and the membrane.
Table 3.2 PSSA uptake and related properties of membrane 12-20 cast from acetone.
Membrane
PSSA
Uptake
Thickness
Proton
Conductivity
MEA
Resistance
Nafion
®
-117 N/A 7 mils 75 mS/cm 19 mΩ
12-20 20% 6.9 mils 78 mS/cm 69 mΩ
The initial results of electrical performance of MEA 12-20 were shown in Figure 3.2.
Despite the high resistance and unfavorable membrane morphology, MEA 12-20 achieved a cell
voltage of 0.367 V at 160 mA/cm² with a maximum power density of 40 mW/cm² and a
maximum current density of 337 mA/cm² at 90 °C utilizing ambient oxygen at the cathode. The
open circuit voltage (OCV) of MEA 12-20 was 0.90 V at 90 °C, which is higher than that of
Nafion
®
117 under the same conditions (< 0.8 V), indicating the significant reduced methanol
crossover rate of PSSA/PVDF membranes. The relatively linear slope of the polarization plot
maintained even at high current densities suggests that oxygen content is sufficient and the
absence of water flooding at the cathode. However, the high resistance of MEA 12-20 due to the
uneven membrane surface and unoptimized hot press conditions presumably caused the poor
electrode-electrolyte interfacial contact, and the maximum power density was limited to 40
mW/cm². Nevertheless, the initial electrical results using the PSSA/PVDF membrane cast from
acetone were very promising considering this was a non-optimized membrane.
99
Figure 3.2 Performance of MEA 12-20 in a direct methanol fuel cell at 90 °C with 2.0 M
methanol utilizing ambient oxygen at the cathode.
3.1.3 Temperature effect on the electrical performance of MEA 12-20
The operation temperature has a strong influence on the electrical performance of a
DMFC. On one hand, the catalyst activity improves significantly at higher temperatures; on the
other hand, the methanol and water crossover rates are also increased with increasing of
temperature. Figure 3.3 illustrates the temperature effect on the electrical performance of MEA
12-20 with 2.0 M methanol solution on the anode and 0.02 L/min oxygen at the cathode. The
electrical performance of MEA 12-20 at 90 °C is far better than that of 30 °C and 60 °C. This
indicates that the beneficial effect of increasing catalyst activity and utilization by temperature
outweighs any inhibitory affect of larger methanol crossover rates. The beneficial affect of
temperature reached a limit at 80 °C, above which there is not further improvement. For instance,
at low current densities (< 220 mA/cm²), the electrical performance at 80 °C is slightly higher
than that of at 90 °C, presumably due to the lower methanol crossover. However, at high current
100
densities (> 220 mA/cm²), MEA 12-20 worked slightly better at higher temperature (90 °C)
presumably as the anode consumes more methanol at high current densities thus alleviating
methanol crossover.
Figure 3.3 Temperature effect on the polarization behavior of MEA 12-20.
3.1.4 Methanol concentration effect on the electrical performance of MEA 12-20
The methanol effect on the electrical performance of MEA 12-20 cast from acetone was
investigated further. Zhao et al.
188
studied the effect of methanol concentration on the
performance of Nafion
®
membranes in DMFCs and they found out that at high methanol
concentrations (i.e. >2 M), the Nafion
®
MEAs suffer significant performance losses due to
increased methanol crossover rates. PSSA/PVDF membranes exhibit significantly lower
methanol crossover rates than Nafion
®
, therefore, better tolerance of high methanol
concentrations was expected. As shown in Figure 3.4, the electrical performance of MEA 12-20
increases with the increasing of methanol concentration up to 2.0 M. The tolerance of high
101
methanol concentrations is clearly due to the lower methanol crossover of membranes, which is
essential for the practical application of DMFCs giving them higher energy density. The
electrical performance of MEA 12-20 started to decrease when the methanol concentration was
further increased to 3.0 M due to the increased methanol crossover rates. It is important to note
that at the time of testing MEA 12-20, our experiment for measuring the methanol crossover
current density was not yet setup, so no methanol crossover measurement for MEA 12-20 was
conducted.
Figure 3.4 Methanol concentration effect on the performance of MEA 12-20.
3.1.5 Oxygen flow effect on the electrical performance of MEA 12-20
There are two plausible proton-conducting mechanism in PEMs: vehicle mechanism and
Grotthuss mechanism.
189
In the vehicle mechanism, protons migrate though the membrane along
with a “vehicle” or proton carriers such as H
3
O
+
, H
5
O
2
+
, and H
9
O
4
+
. On the other hand, in the
Grotthuss mechanism, protons are transferred from one site to another via the formation and
102
breaking of hydrogen bonds (proton hopping), so a vehicle or proton carrier is not needed. The
proton transport mechanism of sulfonic acid based polymer exchange membranes belongs to the
former. Therefore, the proton conductivity of sulfonic acid based PEMs relies on the amount of
water in the membranes. Figure 3.5 illustrates the oxygen flow rate effect on the fuel cell
performance of MEA 12-20. The performance of MEA 12-20 improves with the initial
increasing of oxygen flow rates (from 0.01 to 0.02 L/min), implying the cathode activity is
enhanced by the oxygen concentration in the cathode. This is presumably due to the effective
removal of water from the cathode making more available sites for oxygen diffusion. However,
when the oxygen flow rates was above 0.1 L/min, the electrical performance of MEA 12-20 was
sharply depressed as the cell voltage at 0.2 L/min was smaller than that at 0.1 L/min during the
entire current density range. The resistance of MEA 12-20 at high oxygen flow rate (0.2 L/min)
was measured to be 109 mOhms, which is considerably higher the resistance (69 mOhms) at low
oxygen flow rate (0.02 L/min). Since the proton conductivity of sulfonic acid membranes
strongly relies on the amount of water in membranes, the increased resistance of MEA 12-20
suggests that the cathode exhibits severe drying due to evaporative water losses caused by
increased flow rates.
103
Figure 3.5 Oxygen flow effect on the electrical performance of MEA 12-20.
Although several other membranes were cast from acetone, the electrical performance
measurement was not continued due to high MEA resistances, poor interfacial contacts between
electrolyte and electrode or the membrane bulk or surface structure imperfections. Also no
methanol crossover measurement was investigated as membrane preparation techniques became
our top priority. Nevertheless, the preliminary electrochemical results were promising as the
membranes were based on a completely different membrane system and membrane preparation
and MEA fabrication conditions were not optimized yet. The impressive initial results motivated
the development of improved membrane fabricating techniques.
3.2 Fuel cell performance of membranes cast from DMF with 20 wt% PSSA loading
Membranes prepared using acetone as the casting solvent suffer from wrinkled and
irregular surfaces, resulting in poor interfacial catalyst-membrane contact. Also, as pointed out
above, those membranes tend to be brittle in dry state. Using DMF as the casting solvent turned
104
out to give significantly improved membrane flexibilities and mechanical strengths. These
membranes developed were transparent, flexible and tough even in dry state, allowing the
optimization of the MEA fabrication using higher temperatures and pressures.
3.2.1 Membrane electrode assembly optimization
There are several factors involved in the final quality of MEA: the catalyst ink
composition, the hot-press temperature and pressure. Passalacqua et al.
190
studied the effect of
Nafion
®
ionomer content in the catalyst layer on the fuel cell performance of Nafion
®
-117
membranes and found that the amount of Nafion
®
had a strong influence on the electrical
performance of fuel cells. According to his percolation model, either too much or insufficient
Nafion
®
ionomer in the catalyst layer had negative effects on the fuel cell performance: at low
Nafion
®
loading, ionic conduction become a problem resulting in low catalyst utilization; on the
other hand, the excess of Nafion
®
ionomer in the catalyst layer cuts off the percolation path of
catalyst particles and hinders the gas diffusion in the reaction sites. Only with the optimum
amount of Nafion
®
ionomer, both electronic and ionic conductivity can be optimized and the
mass transport limitation of fuels can be minimized. As MEAs prepared with membranes cast
from acetone have considerable higher resistance as compared to that of Nafion
®
, we reasoned
that the increasing of Nafion
®
ionomer content in the catalyst may create a better contact
between the catalyst and the membrane. Therefore, a new formula of catalyst inks with a higher
content of Nafion
®
ionomer was used to prepare the catalyst ink.
105
Table 3.3 Improved composition of catalyst inks.
New formula Previous formula
Anode Cathode Anode Cathode
Nafion
®
ionomer
solution (5 wt%)
0.3 g 0.25 g 0.2 g 0.2 g
DI water/g 1.0 g 1.0 g 0.8 g 0.8 g
Pt-Ru 0.2 g - 0.2 g -
Pt - 0.2 g - 0.2 g
As shown in Table 3.3, Nafion
®
ionomer contents in the anode and cathode were
increased by 50% and 25%, respectively, as compared to the previous formula used for MEA 12-
20. The amount of Nafion
®
ionomer at the cathode is less than that at the anode, thus facilitating
the gas diffusion into the reaction sites and minimizing the mass transport limitation of oxygen.
In order to prepare well-dispersed catalyst ink suspensions, we also paid great attention to every
single detail. For example, a typical preparation of catalyst ink is as follow: to a vial charged
with 0.2 g catalyst (Pt-Ru for anode and Pt for cathode), 1 ml deionized water was added. The
mixture was shaken on a vortex mixer (Fisher Scientific G-560) for 30 min until a well
dispersion was obtained. After that the appropriate amount of ice-cold Nafion
®
ionomer solution
was added the dispersion slowly while shaking the vial on the vortex mixer. After all Nafion
®
ionomer solution was added, the mixture was continued to mix for additional 30 min to obtain a
well-dispersed suspension.
The most important parameters involved in the MEA fabrication are the processing
temperature and the pressure, both of which have strong influences on the integrity and electrical
performance of MEAs. The parameters used for MEA fabrication were primary based on
empirical methods. It was generally agreed that increasing MEA fabrication hot-press
temperature and pressure would have an advantageous effect on interfacial bonding at the
106
catalyst layer-membrane interfaces. Our newly developed membranes are flexible and tough
even in dry state, which allow us to fabrication using high temperatures and pressures. Therefore,
in order to create a better contact between catalyst layers and membranes, the hot-press pressure
was increased to 375 psi compared to 125 psi for MEA 12-20, and the temperature was raised to
140 °C from 100 °C. Three membranes cast from DMF with the same PSSA loading (20 wt%)
were fabricated into MEAs using our optimized MEA fabrication procedure. The properties of
those MEAs were summarized in Table 3.4.
Table 3.4 Properties of MEAs based on membranes cast from DMF with a 20 wt% PSSA
loading.
Membranes PSSA Loading Thickness
Proton
Conductivity
MEA Resistance
at 30 °C
Nafion
®
-117 N/A 7 mils 75 mS/cm 19 mΩ
MEA 13-42 20 wt% 6 mils 79 mS/cm 35 mΩ
MEA 13-28 20 wt% 5.9 mils 80 mS/cm 30 mΩ
MEA 14-122 20 wt% 6.2 mils 76 mS/cm 39 mΩ
3.2.2 Fuel cell performance of MEA 13-42
The improved membrane morphology and the optimization of MEA fabrication had an
immediate impact on the quality of MEAs. As shown in Figure 3.6, all three membranes had an
OCV resistance around 35 mΩ at 30 °C, which accounts for nearly half of that of MEA 12-20
(69 mΩ). However, these values were still considerably higher than that of Nafion
®
-117 (19
mΩ). These results suggested that the bonding between catalyst layers and membranes can be
further improved through modifications of hot-press or catalyst compositions. Nevertheless, the
107
decreased resistance of MEAs demonstrated for membranes cast with DMF is a clear
improvement.
Figure 3.6 Performance of MEA 14-42 in a DMFC at 90 °C with 2.0 M methanol utilizing 0.02
L/min ambient oxygen at the cathode.
As PSSA/PVDF membranes exhibits lower methanol permeability than Nafion
®
-117, the
electrical behaviors of the three MEAs were evaluated with a particular interest paid to high
methanol concentrations. The fuel cell performance of MEA 14-42 was shown in Figure 3.6,
achieving a cell voltage of 0.43 V at 160 mA/cm² at 90 °C with 2.0 M methanol utilizing 0.02
L/min ambient oxygen at the cathode. This represents a significant improvement compared to
our initial cell measurements of MEA 12-20, as a maximum power density of 79.8 mW/cm²
(MEA 12-20: 39.6 mW/cm²) and a maximum current density of 536 mA/cm² (MEA 12-20: 371
mA/cm²) were achieved. Notice that both MEAs have the same PSSA content (20 wt%) in the
membranes, therefore, the striking enhancement of electrical performance of MEA 13-42 was
108
attributed to the improved properties of membranes, such as a more even distribution of PSSA in
PVDF matrix, a smoother membrane surface, and a better interfacial contact with catalyst layers.
Under the given conditions, the cell voltage of MEA 13-42 was much higher than that of
Nafion
®
-117 over the entire current density range presumably due to the lower methanol
crossover of MEA 12-42 despite it has a higher cell resistance. The nearly linear polarization plot
of MEA 12-42 also suggests that oxygen concentration is enough even such a low oxygen flow
rate (0.02 L/min). This would seem advantageous for practical applications of DMFCs as
portable power sources. In contrast to MEA 13-42, the polarization plot of Nafion
®
-117 dropped
sharply at current density above 200 mA/cm², indicating that Nafion
®
-117 suffers severe water
flooding at the cathode aggravated by high methanol crossover under low oxygen flow rates.
3.2.3 Temperature effect on electrical performance of MEA 13-42
Similar to MEA 12-20, the electrical performance of MEA 13-42 exhibits an ascendant
trend as increasing of temperature. Jung et al.
191
found an increase of the performance of
DMFCs when the operating temperature increased, using Nafion
®
-117 and 2.5 M methanol.
They attributed this higher performance to the combined effects of a reduction of ohmic
resistance and enhancement of catalyst activity. In deed, the ionic conductivity of polymer
electrolyte membranes increases with increasing temperature as the ohmic resistance of MEA
13-42 decreases from 35 mΩ at 30 °C to 21 mΩ at 90 °C, in agreement with other reports.
25
Therefore, despite the fact that the increase of the temperature increases methanol crossover
rates, as previously mentioned, the electrical performance of MEA 13-42 improved with
increasing of temperature as shown in Figure 3.7.
109
Figure 3.7 Temperature effect on the electrical performance of MEA 13-42 with 2.0 M methanol
and 0.02 L/min ambient oxygen at the cathode.
It is worth noting that the ability of MEA 13-42 to operate at high temperatures with such a
small oxygen flow rate (0.02 L/min) is unique as Nafion
®
-117 does not seem to operate well
under these conditions. In the case of Nafion
®
-117, this involves high water permeability and
hence severe water flooding at the cathode aggravated by high methanol crossover especially at
high current densities as shown in Figure 3.8.
110
Figure 3.8 Effect of temperature on the performance of Nafion
®
-117 in a DMFC with 2.0 M
methanol utilizing 0.02 L/min ambient oxygen at the cathode.
3.2.4 Methanol concentration effect on the electrical performance of MEA 13-42
If methanol crossover is a cause of a reduction in cell voltage it would be expected that a
higher concentration of methanol in the feed to the anode would decrease the cell performance.
This agrees with the fact that Nafion
®
-117 membrane has optimal fuel cell performance when
operating with low methanol concentrations (< 2.0 M). For instance, as demonstrated in Figure
3.9, the electrical performance of Nafion
®
-117 increased slightly when the concentration of
methanol was increased from 0.5 M to 1.0 M, however, it decreased sharply when the methanol
concentration is above 2.0 M. These facts suggest that methanol crossover and cathode
polarization losses become problematic for Nafion
®
-117 with high methanol concentrations.
111
Figure 3.9 Methanol concentration effect on electrical performance of Nafion
®
-117 at 90 °C
with 0.02 L/min ambient oxygen at the cathode.
Thus, it is necessary to find the optimal concentration of methanol for the specific type of
polymer electrolyte membrane used in DMFCs. As shown in Figure 3.10, the fuel cell
performance of MEA 13-42 increases when the methanol concentration is increased from 0.5 M
to 2.0 M, implying a high concentration of methanol is beneficial to the reaction kinetics at the
anode as the mass transfer limitation of methanol is alleviated with a high concentration of
methanol. Above 2.0 M, the fuel cell performance started to decline but in a much slow
magnitude as compared to Nafion
®
-117. This suggested that the polarization loss due to
methanol crossover and mixed potential outweigh any anode potential gain when the methanol
concentration exceeds a certain value. Therefore, for the case of MEA 13-42, the optimal
concentration of methanol for producing higher power densities was around 2.0 M. The ability to
operate at high methanol concentrations is a great advantage for DMFCs as portable power
sources as it increases the energy density.
112
Figure 3.10 Methanol concentration effect on the electrical performance of MEA 13-42 at 90 °C
with 0.02 L/min ambient oxygen at the cathode.
3.2.5 Oxygen flow effect on the electrical performance of MEA 13-42
The oxygen supply mode is essential to the water management at the cathode and thus the
fuel cell performance of a DMFC. Nakagawa et al.
192
investigated the effect of oxygen flow rates
on the electrical performance of a DMFC using Nafion
®
membranes and found out that a
moderate oxygen flow rate gave the optimum fuel cell performance, and either too low or too
high oxygen flows resulted in reduction of fuel cell performances due to water flooding or severe
dryness at the cathode, respectively. As shown by Chen et al.,
193
the production of water at the
cathode, especially at high current densities, strongly inhibited the performance of an air-
breathing DMFC as the water blocked the air feeding into the cathode. The problem can be
alleviated by the use of an air flow toward the cathode and operating the cell at an elevated
temperature.
The results shown in Figure 3.11 show that the electrical performance of Nafion
®
-117
was significantly increased when the oxygen flow rate was increased from 0.02 L/min to 0.1
113
L/min. This indicates that Nafion
®
-117 suffers significant water flooding at the cathode when
operated at low oxygen flow rates and high current densities. The high O
2
flow rates help to
remove excess water resulting from electro-osmosis drag and the concentration gradient
mediated diffusion. However, it was also pointed out that increasing oxygen flow to 1.0 L/min
had a negative effect on cell performance due to a convective cooling/drying effect.
192
Figure 3.11 Oxygen flow effect on the electrical performance of Nafion
®
-117 at 90 °C with 2.0
M methanol utilizing ambient oxygen at the cathode.
As discussed in the ultrafiltration property session, PSSA/PVDF membranes have much
small water flux under the same hydraulic pressure compared to Nafion
®
-117. Therefore, water
flooding is not an issue for PSSA/PVDF membranes even when operating at high temperatures
and low oxygen flow rates. As shown in in Figure 3.12, the increase of oxygen flow rates at the
cathode had a negative effect on MEA 13-42 fuel cell performance and increased resistivity due
to a drying effect caused by water removal at high oxygen flow rates. It was reported that PSSA-
114
PVDF semi-interpenetrating composite membranes was able to operate at a static pressured
oxygen with only occasionally purging necessary to remove product water.
194,195
These
phenomena suggest that the water flux through MEA 13-42 is sufficient to facilitate proton
transport while avoiding cathode flooding at the cathode. As water flux through MEAs always
comes along with methanol crossover, the absence of water flooding in the cathode also supports
the experimental low methanol crossover data.
Figure 3.12 Oxygen flow effect on the electrical performance of MEA 13-42 at 90 °C with 2.0
M methanol utilizing ambient oxygen at the cathode.
3.2.6 Reproducibility of electrical performance
For sulfonated hydrocarbon PEMs, including sulfonated polystyrene and its derivatives,
sulfonated polyarylene polymers, sulfonated polyimides, sulfonated polyphenylenes, and
sulfonated polyphosphazenes, can be prepared using any or combinations of the following
115
methods: post-polymerization sulfonation, direct copolymerization of sulfonated monomers, and
physicochemical modification of the sulfonated polymers.
95
Post-polymerization sulfonation
method have been extensively used for fuel cell membrane applications due to large scope of
available polymer precursors that can be sulfonated to introduce sulfonic acid groups.
196
However, post-polymerization modification technique offers limited control over the location
and degree of sulfonation, and the size and connectivity of these ionic structures are poorly
controlled, and may create undesirable byproducts or degrade the polymer structure. Therefore,
polymer electrolyte membranes prepared by post-polymerization sulfonation always end up with
poor reproducibilities.
48,60,197
In CHAPTER 2, the characterizations of PSSA/PVDF membranes, including proton
conductivity, methanol permeability as well as water uptake, were highly reproducible.
Therefore, the electrical performance of PSSA/PVDF polymer blend membranes is expected to
have a fairly high reproducibility. The reproducibility of polarization curves was determined by
comparing three different MEAs with membranes have the same PSSA content but prepared
using different polymer precursors on different days. The polarization curves shown in Figure
3.13 represents the typical variation in the cell performance carried out under the same operating
conditions. As expected, all three membranes showed very reproducible and nearly overlapped
polarization curves. These three different MEAs exhibited maximum current densities of 536,
537, and 520 mA/cm², respectively, corresponding a maximum variation less than 3.0%. In
addition, these MEAs showed very similar power densities of 78.4, 73.1, and 80.2 mW/cm²,
respectively, corresponding to a largest variation of 8%. Considering that all catalyst inks and
MEAs were prepared by hand, such a small variation in fuel cell performance indicated our
membranes are highly reproducible, which is often difficult to achieve for polymer electrolyte
116
membranes using post-polymerization sulfonation. The highly reproducible fuel cell
performance also suggests PSSA/PVDF membranes studied in the present thesis are very
promising alternative polymer electrolyte membranes in both prospects of manufacturing and
commercialization.
Figure 3.13 Reproducibility measurement of MEAs at 90 °C with the same PSSA uptake but
fabricated on different days with different polymer precursors.
3.2.7 Fuel cell performance of MEA 13-42 in ambient air
As DMFC has a potential in application as portable power sources it would be of interest
to investigate the performance of PSSA/PVDF polymer blend membranes in air. As illustrated in
Figure 3.14, MEA 13-42 in 0.10 L/min air with 0.5 M methanol gave a cell voltage of 0.37 V at
the current density of 100 mA/cm² at 60 °C. Similar to utilizing oxygen at the cathode, the cell
performance was enhanced by increasing of operation temperature and reached a cell voltage of
117
0.43 V at the current density of 100 mA/cm² at 90 °C. It is worth noting that the air flow rate has
a strong influence on the fuel cell performance in ambient air, especially at high current
densities. For instance, the polarization curve of MEA 13-42 at 90 °C drops quickly above 170
mA/cm² due to the limited mass transfer of oxygen to the cathode.
Figure 3.14 Performance of MEA 13-42 at various temperatures with 0.5 M methanol utilizing
0.10 L/min ambient air at the cathode.
As shown in Figure 3.15, the effect of methanol concentration on the electrical
performance of MEA 13-42 is optimal with 1.0 M methanol and is only slightly decreased when
the methanol concentration was changed to 4.0 M. On the other hand, the impact of methanol
concentration on the performance of Nafion
®
-117 in air is demonstrated in Figure 3.16. The
steep drop of FC performance of Nafion
®
-117 in ambient air above 2.0 M methanol indicates
severe limitations. In contrast, MEA 13-42 exhibited a larger cell voltage over the entire current
density range at 30 °C with 2.0 M methanol and ambient air as shown in Figure 3.17.
118
Figure 3.15 Methanol concentration effect on electrical performance of MEA 13-42 at 30 °C
utilizing 0.10 L/min ambient air at the cathode.
Figure 3.16 Methanol concentration effect on electrical performance of Nafion
®
-117 at 30 °C
utilizing 0.10 L/min ambient air at the cathode.
119
Figure 3.17 Fuel cell performances of Nafion
®
-117 and MEA 13-42 at 30 °C with 2.0 M
methanol utilizing 0.10 L/min ambient air at the cathode.
3.3 Fuel cell performance of PSSA/PVDF membranes with 25 wt% PSSA loading
For PSSA/PVDF polymer blend membranes, the amount of PSSA in membranes has a
profound influence on the electrical performance of membranes, as well as the water flux, proton
transport, methanol crossover and even mechanical strength. The relatively high resistance of
MEA 13-42 (39 mΩ at 30 °C) indicated that higher fuel cell performance is obtainable by
increasing the PSSA uptake in membranes.
Therefore membrane 14-127 was prepared with 25 wt% PSSA loading using our
optimized membrane casting procedure. And the corresponding MEA 14-127 was fabricated
using the same condition as MEA 13-42. The beneficial effect of a higher PSSA loading was
immediately noticed as the OCV resistance of MEA 14-127 was reduced to 18 mΩ at 30 °C,
which is very closed to that of Nafion
®
-117 (19 mΩ at 30 °C). It was important to note that even
though MEA 14-127 has a similar resistance as compared to MEAs prepared with Nafion
®
-117,
120
PSSA/PVDF membranes with 25 wt% PSSA give much lower methanol permeability as
discussed earlier. The distinct low methanol permeability of PSSA/PVDF membranes despite the
similar MEA resistances suggests that the electrical performance can be benefited from low
methanol crossover during a fuel cell operation.
3.3.1 Effect of PSSA content on electrical performance of PSSA/PVDF membranes
The electrical performance of MEA 14-127 with 25 wt% PSSA was studied and
compared with MEA 13-42. As shown in Figure 3.18, MEA 14-127 achieved a current density
of 245 mA/cm² at a cell voltage of 0.4 V with 1.0 M methanol, which is a significant
enhancement of fuel cell performance as compared to MEA 13-42 (0.33 V at 245 mA/cm²). In
addition, the maximum power density of MEA 14-127 was 106 mW/cm², representing a 30%
improvement over that of MEA 13-42 (81 mW/cm²). Although membrane 14-127 has a higher
methanol permeability as reported in methanol permeability session, the large improvement in
cell performance suggests that the benefit gained from the higher IEC outweighs any mixed
potential caused by the slightly higher methanol crossover.
121
Figure 3.18 Performance of MEAs 14-127 and 13-42 in DMFCs at 90 °C with 1.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode.
3.3.2 Electrical performance comparison with Nafion
®
-117
Up to now, Nafion
®
membrane is considered to be the benchmark of polymer electrolyte
membranes for DMFCs even though it has several major drawbacks including high costs and
methanol crossover issues. Over the past decade, advances in fuel cell technology have come
primarily from improved electro-catalysts, membrane electrode assembly fabrication strategies,
and cell/stack/system engineering. Apart from Nafion
®
, new ion conducting polymeric materials
have played only a minor role in significantly increasing cell performance.
60
In an effort to create
a high performance and low cost alternative polymer electrolyte membrane based on polymer
blends, the cell performance of MEA 14-127 was evaluated and compared with Nafion
®
-117
under the same operation conditions.
122
As Nafion
®
-117 membrane has an optimum performance in an ambient oxygen flow rate
of 0.10 L/min, we decided to measure the cell performance of MEA 14-127 and Nafion
®
-117
under the same oxygen flow rate (0.10 L/min). As shown in Figure 3.19 MEA 14-127 has a
better fuel cell performance compared to Nafion
®
-117 attaining a cell voltage of 0.4 V at a
current density of 183 mA/cm². The distinction of cell performance was even more pronounced
at higher current densities (> 200 mA/cm²), as the polarization curve of Nafion
®
-117 started to
drop sharply presumably due to the water flooding that prevents oxygen diffusing into the
cathode catalyst and mixed potential caused by the methanol crossover.
Figure 3.19 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 0.5 M methanol using
0.10 L/min ambient oxygen flow at the cathode.
The performance of MEA 14-127 and Nafion
®
-117 in DMFCs with 1.0 M methanol is
shown in Figure 3.20. MEA 14-127 exhibits a much higher fuel cell performance as compared
to Nafion
®
-117 at 60 °C with 1.0 M methanol utilizing 0.10 L/min ambient oxygen at the
cathode, achieving a current density of 200 mA/cm² at 0.4 V, which is 68% higher that that of
123
Nafion
®
-117 at 0.4 V (119 mA/cm²). The higher performance of PSSA/PVDF membranes might
be benefited from several factors including lower methanol permeability, and higher proton
conductivity.
Figure 3.20 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 1.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode.
Motivated by superior cell performances at lower concentration of methanol solutions,
we further measured the cell performances of MEA 14-127 at 2.0, 3.0 and 4.0 methanol and
compared with those of Nafion
®
-117. The cell performance of MEA 14-127 and Nafion
®
-117 at
60 °C with 2.0 M methanol utilizing 0.10 L/min ambient oxygen was illustrated in Figure 3.21,
the cell voltage of MEA 14-127 is higher than that of Nafion
®
-117 over the entire range of
current densities. For example, at the cell voltage of 0.4 V, MEA 14-127 achieved a current
density of 165 mA/cm², which is 43% higher that of Nafion
®
-117 (115 mA/cm²) under the same
operation conditions.
124
Figure 3.21 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 1.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode.
Similar but more striking results are seen with 3.0 and 4.0 M methanol as shown in
Figure 3.22 and Figure 3.23, respectively. Again MEA 14-127 exhibits much higher cell
performance over entire range of current densities. It is important to note that the cell
performance of MEA 14-127 with 3.0 M methanol was quite similar to that with 2.0 M methanol
despite the potentially higher methanol crossover rate with 3.0 M, indicating that the potential
gain at the anode due to higher methanol concentration compensates the mixed potential due to
methanol crossover. The ability to be operated at high methanol concentrations further supports
the significantly lower methanol crossover of PSSA/PVDF membranes.
125
Figure 3.22 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 3.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode.
Figure 3.23 Performance of MEAs 14-127 and Nafion
®
-117 at 60 °C with 4.0 M methanol using
0.10 L/min ambient oxygen flow at the cathode.
126
3.3.3 Effect of temperature on the performance of MEA 14-127
Voltage-current characteristics of MEA 14-127 in a DMFC were measured over a range
of temperatures with 1.0 M methanol utilizing 0.10 L/min ambient oxygen at the cathode. Figure
3.24 reveals a pronounced increase in performance with increasing of temperature over the range
of 30 to 90 °C. For example, at a potential of 0.5 V, the current density outs are 25, 109, and 140
mA/cm² at temperatures of 30, 60 and 90 °C, respectively. The trend of increased output with
increase in temperature is in accord with that exhibits by Nafion
®
-117 membrane. The increased
output at higher temperatures is attributed to a combination of factors consisting of a reduction of
cell ohmic resistance, activation polarization, and concentration polarization.
198
Figure 3.24 Effect of temperature on the performance of MEA 14-127 with 1.0 M methanol
utilizing ambient oxygen at the cathode.
However, temperature effects at various methanol concentrations, rarely mentioned in
previous studies seem to show some apparent anomalies. As shown in Figure 3.25, the cell
127
performance of MEA 14-127 with 4.0 M methanol and 0.10 L/min ambient oxygen actually
decreased from 60 to 90 °C. especially at higher current densities (> 150 mA/cm²). Therefore,
the impact of one parameter like temperature on the cell performance of a DMFC is closely
depends on other operational parameters including methanol concentration, oxygen flow rate at
the cathode.
Figure 3.25 Effect of temperature on the cell performance of MEA 14-127 with 4.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode.
3.3.4 Impact of methanol flow rate on electrical performance of MEA 14-127
In an active DMFC, the methanol fuel is usually supplied and circulated with a
combination of a mechanical pump and a methanol solution reservoir. The circulation of
methanol solution not only provides a nearly constant concentration of methanol, but also
removed the carbon dioxide gas released at the anode. Therefore, it is important to study the
0.0
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
Cell Voltage (V)
Current Density (mA/cm²)
4.0 M Methanol
Oxygen Flow Rate: 0.10 L/min
Electrode Area: 2'' x 2''
MEA 14-127: 90 °C
MEA 14-127: 60 °C
MEA 14-127: 30 °C
128
impact of methanol flow rate at the anode on the cell performance of PSSA/PVDF membranes in
DMFCs.
As more carbon dioxide gas is released when a DMFC is operated at temperatures and
higher current densities, the impact of methanol flow rate on the cell performance was measured
at 90 °C with 1.0 M methanol and 0.10 L/min oxygen. As shown in Figure 3.26, the voltage of
MEA 14-127 increased with increasing of methanol flow rates, especially at higher current
densities. It was assumed that mass transport limitation was involved in this phenomenon.
Higher methanol flow rates help to remove carbon dioxide gas from gas diffusion layer more
effectively, thus creating more available sites for methanol to diffuse into the catalyst layer. The
beneficial impact of methanol flow rate on the electrical performance is less pronounced when
the cell is operated with high methanol concentration (> 2M) as mass transport problem is
minimized with high concentration methanol solutions and the methanol crossover might be
exacerbated by high methanol flow rates.
Figure 3.26 Effect of methanol flow rate on the performance of MEA 14-127 at 90 °C with 1.0
M methanol utilizing 0.10 L/min ambient oxygen at the cathode.
129
3.3.5 Electrical performance of MEA 14-127 in air
As mentioned before, for portable DMFC power sources, it is important to function
effectively using air at the cathode. The cell performance of MEA 14-127 and Nafion
®
-117 at 60
°C with 0.5 M methanol was illustrated in Figure 3.27. MEA 14-127 exhibits a cell voltage of
0.4 V at the current density of 100 mA/cm², indicating a similar performance as Nafion
®
-117.
This is one of the few instances we could find comparable performance of these two membranes.
Figure 3.27 Performance of MEA 14-127 and Nafion
®
-117 at 60 °C with 0.5 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode.
The polarization curves of MEA 14-127 and Nafion
®
-117 in a DMFC at 60 °C with 1.0
M methanol utilizing 0.10 L/min ambient air at the cathode were shown in Figure 3.28. The
benefit of low methanol permeability was immediately noticed as the MEA 14-127 performed
far better than Nafion
®
-117, accomplishing a cell voltage of 0.45 V at the current density of 100
130
mA/cm² (Nafion
®
-117: 0.37 V at 100 mA/cm²). The cell performance of Nafion
®
-117 drops
sharply above 150 mA/cm², giving a maximum current density of 195 mA/cm². On the contrast,
MEA 14-127 achieves a maximum current density output of 345 mA/cm², which is 75% higher
than that of Nafion
®
-117.
Figure 3.28 Performance of MEA 14-127 and Nafion
®
-117 at 60 °C with 1.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode.
Figure 3.29 demonstrates the cell performance of MEA 14-127 and Nafion
®
-117 with
2.0 M methanol at 60 °C. The difference of cell performance between MEA 14-127 and
Nafion
®
-117 is even more manifest. In fact, Nafion
®
-117 can not function properly with 2.0 M
methanol utilizing 0.10 L/min ambient air at the cathode. The low open circuit voltage of 0.51 V
indicates Nafion
®
-117 suffers from severe methanol crossover with high methanol
concentrations. The cell performance of Nafion
®
-117 become highly sensitive to the methanol
131
concentration when operated in air. On the contrast, MEA 14-127 performed far much better
than Nafion
®
-117, obtaining a cell potential of 0.42 V at the current density of 100 mA/cm². In
addition, MEA 14-127 reaches a maximum current density output of 280 mA/cm² at 2.0 M
methanol, about double that of Nafion
®
-117 (136 mA/cm²).
Figure 3.29 Performance of MEA 14-127 and Nafion
®
-117 at 60 °C with 2.0 M methanol
utilizing 0.10 L/min ambient oxygen at the cathode.
DMFCs have been considered as promising power sources for small electrical devices,
such as cell phones, laptops, and cameras. The ability of polymer electrolyte membranes to
function effectively at room temperature has special relevance. Figure 3.30 shows voltage-
current characteristics of MEA 14-127 and Nafion
®
-117 at near room temperature (30 °C) with
1.0 M methanol utilizing 0.10 L/min air at the cathode. MEA 14-127 shows significantly higher
voltages over the entire range of current densities and reaches a maximum power density of 38
132
mW/cm², 27 percent higher than that of Nafion
®
-117 (30 mW/cm²). All these data suggest that
PSSA/PVDF polymer blend membranes outperform Nafion
®
-117 with air. It is important to note
that although the cell performance of PEMs based on sulfonated polymers such as sulfonated
polystyrene and polystyrene derivatives, sulfonated poly(arylene ether)s, polyimides, and
polyphosphazenes, is extensively reported, most of these reports only use oxygen at the cathode.
Figure 3.30 Performance of MEA 14-127 and Nafion
®
-117 at 30 °C with 1.0 M methanol
utilizing 0.10 L/min ambient air at the cathode.
3.4 Methanol crossover analysis of PSSA/PVDF membranes
Methanol crossover of different MEAs was extensively measured by analyzing the CO
2
concentration in the cathode exhaust using a CO
2
analyzer. The CO
2
concentration data detected
by the analyzer was further converted to the crossover current density according to the following
relationships:
195
133
the mole percentage of carbon dioxide:
the crossover current density:
where P is an atmosphere pressure (1 atm), V is the volume fraction of CO
2
in the cathode
exhaust, F is Faraday constant (96485), z is the number of electron generated by oxidation of one
methanol molecule, and A is the area of electrode, respectively. CO
2
measurements on the
cathode exhaust provide an evaluation of methanol crossover at various current densities
assuming the complete oxidation of all methanol that is transported through MEAs. The
crossover current density obtained by above method was plotted with fuel cell current density in
order to demonstrate methanol crossover at given current densities.
3.4.1 Effect of temperature on methanol crossover
The methanol crossover of MEA 13-42 and Nafion
®
-117 were measured with 2.0 M
methanol under various operating temperatures and the results were presented in Figure 3.31.
The results indicate that MEA 13-42 has significantly lower methanol crossover current densities
as compared to Nafion
®
-117 in the temperature range of 30-90 °C. For instance, the average
crossover current density for Nafion
®
-117 at 60 °C was 158 mA/cm², while that of MEA 13-42
was 40 mA/cm², which only accounts for 25% of that of Nafion
®
-117. Clearly, the methanol
crossover current density is highly temperature dependent as evidenced by the notable increasing
of crossover current densities with increasing of temperatures for both MEA 13-42 and Nafion
®
-
117. Similar results were also found by Narayanan et al.
25,198
However, the methanol crossover
current density of Nafion
®
-117 increased far more than that of MEA 13-42. For example, the
134
crossover current density of Nafion
®
-117 increased from 50 mA/cm² at 30 °C to 158 mA/cm² at
90 °C (3 folds), while that of MEA 13-42 only increased by less than 20 mA/cm² (< 2 folds).
Narayanan et al.
25
studied the dependence of the crossover rate of Nafion
®
-117 on the
current density with 1.0 M methanol, and found out that the crossover current density decreases
with increasing current density due to an increased consumption of methanol at high current
densities. However, what we found here was the opposite trend, in which the crossover rate
slightly increased with increasing current density. The difference of methanol concentration used
for measuring methanol crossover rates may contribute to the difference, as we used 2.0 M
methanol rather than 1.0 M. At higher methanol concentration, the methanol diffusion caused by
concentration gradients and the electro-osmosis drag is also higher, which offsets any methanol
depletion at the anode due to greater methanol consumption at high current densities.
However, it is important to note that the performance of a DMFC generally increases
with temperature, due to the increase of reaction kinetics and the reduction of ohmic resistance.
Therefore, despite the fact that methanol crossover increases with the increase of the
temperature, the results indicate that the increase of temperature generally improves the cell
performance. For this reason, vapor-feed DMFCs are receiving more attention.
199-203
135
Figure 3.31 Methanol crossover current density of MEA 13-42 and Nafion
®
-117 measured with
2.0 M methanol in the temperature range of 30-90 °C.
3.4.2 Effect of methanol concentration on methanol crossover
Methanol crossover through MEAs occurs in at least two ways: the spontaneous
methanol diffusion caused by concentration gradient and methanol transport mediated by electro-
osmotic drag.
156
Experimental studies on the cell performance and methanol crossover under
different methanol concentrations can help to identify the optimum operating conditions for a
DMFC without excessive methanol crossover. Wang et al.
204
found that most of the methanol
crossover reacts to form carbon dioxide at the cathode. They also found that cathode open-circuit
potential is inversely proportional to the amount of CO
2
formed and the poison effects on cathode
catalyst. Higher methanol crossover were also by observed Jung et al.
191
in that the open circuit
voltages decreased with increasing methanol concentration.
The impact of methanol concentration on methanol crossover of Nafion
®
-117 was
illustrated in Figure 3.32. Clearly, the crossover current density of Nafion
®
-117 is highly
136
sensitive to the methanol concentration and increases proportionally with increasing of methanol
concentration. For example, the average crossover current density of Nafion
®
-117 with 1.0 M
methanol was 79.6 mA/cm², and the value increased more than 1 fold to 167.0 mA/cm² with 2.0
M methanol.
Figure 3.32 Effect of methanol concentration on the methanol crossover current density of
Nafion
®
-117.
The impact of methanol concentration on methanol crossover of PSSA/PVDF polymer
blend membranes were also measured and shown in Figure 3.33 and Figure 3.34. Both MEAs
13-42 and 14-127 exhibit an increase of methanol crossover current density with increasing of
methanol concentration, however, the magnitudes of increase were substantially smaller than that
of Nafion
®
-117. The correlation between the average crossover current density and methanol
concentration are summarized in Figure 3.35. The plot indicates that the crossover current
137
densities of PSSA/PVDF membranes are less sensitive to methanol concentration due to the low
permeability property, which explains the high performance of PSSA/PVDF membranes at
higher methanol concentrations (> 2.0 M).
Figure 3.33 Effect of methanol concentration on crossover current density of MEA 13-42.
Figure 3.34 Effect of methanol concentration on crossover current density of MEA 14-127.
138
Figure 3.35 Crossover current density vs. methanol concentration for MEAs 13-42, 14-127, and
Nafion
®
-117.
3.4.3 Effect of PSSA uptake on the methanol crossover
For PSSA/PVDF polymer blend membranes, two components have distinct
functionalities, in which hydrophilic PSSA provides proton transport while hydrophobic PVDF
endows membrane with stability and toughness. In the membrane characterization session, we
demonstrated that methanol diffusion coefficient increases with increasing of PSSA uptake. A
too high PSSA content results in excessive water swelling and high methanol crossover, while an
insufficient PSSA uptake leads to poor proton conductivity. Therefore, the amount of PSSA in
PSSA/PVDF polymer blend membranes need to be optimized in order to achieve high cell
performances.
The impact of PSSA uptake on methanol crossover for PSSA/PVDF membranes and
Nafion
®
-117 was shown in Figure 3.36. MEA 14-127 contains 25 wt% PSSA and exhibits a
larger methanol crossover as compared to MEA 13-42, which has a 20 wt% PSSA uptake. This
139
result indicates that methanol crossover of PSSA/PVDF membranes increases with the increases
of PSSA content in membranes, which is agreed with the methanol diffusion experiment
discussed in the membrane characterization session. It is important to note that although MEA
14-127 exhibits a larger methanol crossover than MEA 13-42, the magnitude of its methanol
crossover is still much smaller than that of Nafion
®
-117 under the same operating conditions.
Figure 3.36 Effect of PSSA uptake on methanol crossover of PSSA/PVDF polymer blend
membranes.
3.5 Fuel utilization of PSSA/PVDF polymer blend membranes
The most immediate consequence of methanol crossover in direct methanol fuel cells is
the reduction of fuel utilization. The reduction in methanol crossover when using PSSA/PVDF
membranes in DMFCs indicates that higher fuel utilization values are obtainable. The fuel
utilization (also called Faraday efficiency) of an operating DMFC can be calculated according to
the following equation:
140
where I
Cell
is the current density output of a DMFC and I
Crossover
is the equivalent current density
due to methanol crossover.
The fuel utilization of MEAs 13-42 and 14-127 at 60 °C with 1.0 M methanol was
illustrated in Figure 3.37. The data indicates that PSSA/PVDF polymer blend membranes have
much higher fuel utilization compared to Nafion
®
-117. For instance, at a current density of 100
mA/cm², MEAs 13-42 and 14-127 exhibited fuel efficiencies of 84% and 68%, representing 52%
and 24% increases compared to that of Nafion
®
-117.
The impact of PSSA uptake on fuel utilization of PSSA/PVDF follows the same trend on
methanol crossover rates. A higher PSSA content in membranes usually reduce fuel utilization in
DMFCs due to higher methanol crossover rates. However, it is important to note that MEA 14-
127 still exhibits much higher fuel utilization than Nafion
®
-117 despite the relatively high PSSA
uptake (25 wt% PSSA).
141
Figure 3.37 Effect of PSSA content on the fuel utilization of PSSA/PVDF polymer blend
membranes.
3.5.1 Temperature effect on fuel utilization
The impact of temperature on fuel utilization is demonstrated in Figure 3.38 and Figure
3.39. The fuel utilization was usually higher at low temperatures and high current densities.
However, the variation in fuel utilization at different temperatures becomes smaller at high
current densities. For example, at the current density of 100 mA/cm², the fuel utilization of MEA
13-42 at 30 and 60 °C was 88% and 74%, representing a utilization difference of 14%, and at
260 mA/cm², the discrepancy of fuel utilization became much smaller (96% for 30 °C and 94%
for 60 °C). These data indicate that the increase of operating temperature in a certain range is
helpful to increase fuel utilization while avoiding excessive methanol crossover.
142
Figure 3.38 Effect of temperature on the fuel utilization of MEA 13-42 and Nafion
®
-117 in a
DMFC with 1.0 M methanol.
Figure 3.39 Fuel utilization of MEA 14-127 in a DMFC at various temperatures with 1.0 M
methanol.
143
3.5.2 Effect of methanol concentration on fuel utilization
The fuel utilizations of Nafion
®
-117, MEAs 13-42 and 14-127 were determined using a
variety of methanol concentration at 60 °C, as shown in Figure 3.40 to Figure 3.42. The data
indicate that the fuel utilization values increases with decreasing methanol concentration as a
result of the reduction in methanol crossover rates, trends previously reported in methanol
crossover section. Inspection of these data reveals that methanol concentration has less influence
on the fuel utilization of PSSA/PVDF membranes than Nafion
®
-117. For instance, at the current
density of 100 mA/cm², MEA 13-42 exhibited fuel efficiencies of 89% and 83% for 0.5 M and
1.0 M methanol, respectively, representing only a 6 percent decrease. However, Nafion
®
-117
exhibited a utilization difference of 30% under the given current density (68% for 0.5 M
methanol and 38% for 1.0 M methanol). The tolerance of higher methanol concentrations
indicated PSSA/PVDF membranes serve a good barrier against methanol crossover while
maintaining the needed proton conductivity.
Figure 3.40 Effect of methanol concentration on the fuel utilization of Nafion
®
-117 at 60 °C.
144
Figure 3.41 Effect of methanol concentration on the fuel utilization of MEA 13-42 with 20 wt%
PSSA loading at 60 °C.
Figure 3.42 Effect of methanol concentration on the fuel utilization of MEA 14-127 with 25
wt% PSSA loading at 60 °C.
145
3.6 Fuel cell efficiency of PSSA/PVDF polymer blend membranes
The fuel utilization only represents one part of the overall picture when evaluating the
efficiency of DMFCs utilizing PSSA/PVDF membranes. Due to the polarization losses, the
DMFC experimental voltage is always smaller than the reversible DMFC voltage. In order to
study the fuel cell polarization loss behavior, the voltage efficiency is also need to be considered
when calculating the overall efficiency of DMFCs. Since not all the fuel chemical energy in a
DMFC is converted into electric work, the thermodynamic efficiency limited by the fuel intrinsic
properties is also needed to take into account when evaluating the overall efficiency of a DMFC.
The voltage efficiency of the direct methanol fuel cell is determined as the ratio of the
real operating voltage (V
cell
) of a fuel cell to the thermodynamic reversible voltage (V
reversible
):
as mentioned before, the thermodynamic efficiency of a DMFC can be calculated by the
following equation:
Therefore, the overall efficiency is defined as the product of the thermodynamic
efficiency, fuel efficiency and potential efficiency, which is given by the following equation:
as the thermodynamic efficiency of the DMFC is constant and independent of materials, it can be
neglected in the calculation of the overall efficiency of DMFCs. Then the overall efficiency of
DMFCs can be expressed:
146
3.6.1 Comparison of fuel cell efficiency between Nafion
®
-117 and PSSA/PVDF membranes
The overall fuel cell efficiencies of MEAs 13-42 and 14-127 were determined based on
the fuel utilization and voltage efficiency with various operating conditions utilizing 0.10 L/min
of ambient oxygen at the cathode. Overall, both MEA13-42 and 14-127 exhibit higher overall
efficiency as compared to Nafion
®
-117 in various operating conditions due to the lower
methanol crossover and larger fuel utilization discussed in previous sessions. A typical overall
efficiency of comparison is shown in Figure 3.43. The data indicates both MEA 13-42 and 14-
127 have much higher fuel cell efficiency than Nafion
®
-117. For example, the fuel cell
efficiencies of 33.5% and 35.5%, are observed for MEAs 13-42 and 14-127, respectively, at the
current density of 120 mA/cm² with 0.5 M methanol, which represent 35% and 43%
improvement as compared to Nafion
®
-117 (fuel cell efficiency 24.8% at 120 mA/cm²).
Inspection of the data reveals that MEA 13-42 has higher overall DMFC efficiency as compared
to Nafion
®
-117 up to 220 mA/cm² due to higher fuel utilization. Above 220 mA/cm², Nafion
®
-
117 take over the lead as a result of higher voltage efficiency at low methanol concentration (0.5
M). On the other hand, MEA 14-127 with 25 wt% PSSA content exhibits higher efficiencies
over the entire range of current density due to the combined benefits of higher fuel utilization
and higher voltage efficiency as compared to Nafion
®
-117.
147
Figure 3.43 Fuel cell efficiencies of MEA 13-42, 14-127 and Nafion
®
-117 at 90 °C with 0.5 M
methanol utilizing 0.10 L/min ambient oxygen at the cathode.
3.6.2 Effect of methanol concentration on fuel cell efficiency
The methanol concentration has both positive and negative effect on the fuel cell
performance. For one hand, the increase of methanol concentration alleviate the mass transfer
limitation of methanol thus increase the reaction kinetics at the anode; on the other hand, a too
high methanol concentration also increases the methanol crossover, which decreases fuel
utilization and voltage efficiency due to fuel loss and mixed potential.
The effect of methanol morality on the DMFC efficiency using Nafion
®
-117 was shown
in Figure 3.44. The data indicates that the fuel cell efficiency of Nafion
®
-117 achieves highest
values with 0.5 M methanol up to 200 mA/cm² due to alleviated methanol crossover with diluted
methanol concentration. However, the fuel cell efficiency of Nafion
®
-117 with 0.5 M drops
sharply thereafter as a result of reducing of voltage efficiency due to mass transfer limitation of
148
methanol at the anode. Beyond 200 mA/cm², the fuel efficiency values with 1.0 M methanol take
over the lead, indicating the increase in voltage efficiency outweighs losses in fuel utilization. As
discussed before, Nafion
®
-117 is highly sensitive to methanol concentration in terms of the
methanol crossover rate, therefore, further increases in methanol concentration have a adverse
effect on the fuel cell efficiency of Nafion
®
-117, implying the decrease in fuel utilization
eventually outweigh any gain brought by increasing of concentration.
The impact of methanol concentration on MEA 13-42 is shown in Figure 3.45. The data
indicates that MEA 13-42 exhibits a higher overall efficiency with 0.5 M methanol when the
current density is less than 60 mA/cm² due to the lower methanol crossover thus higher fuel
utilization. However, when the current density is > 60 mA/cm², the fuel cell efficiency with 1.0
M methanol is superior which is attributed to the increase in voltage efficiency. Beyond 1.0 M,
the fuel cell efficiency of MEA 13-42 exhibits a decreasing trend with increasing methanol
concentration. This is due to the increase of methanol crossover with increasing concentration,
which offset any benefit concentration may have on the voltage efficiency.
Figure 3.46 illustrates the effect of methanol concentration on MEA 14-127 that consists
of 25 wt% PSSA. The data indicates that MEA 14-127 exhibits high fuel cell efficiencies with
0.5 M methanol up to 260 mA/cm², achieving a fuel cell efficiency of 35.1% at the current
density of 100 mA/cm². In fact, the impact of methanol concentration on the fuel cell efficiency
of MEA 14-127 exhibits a similar tendency as compared to Nafion
®
-117 except higher fuel cell
efficiencies were obtainable for MEA 14-127. For instance, at the current density beyond 260
mA/cm², the overall efficiency with 1.0 and 2.0 M methanol is superior to that of 0.5 M
methanol, attributing to the increase in voltage efficiency as a result of alleviated mass transfer
149
limitations of methanol. Further increase in methanol concentration to 3.0 M did not bring
beneficial effects on the overall efficiency of MEA 14-127 due to the decrease in fuel utilization.
Figure 3.44 Effect of methanol concentration on the fuel cell efficiency of Nafion
®
-117 at 60 °C
utilizing 0.10 L/min ambient oxygen at the cathode.
150
Figure 3.45 Effect of methanol concentration on the fuel cell efficiency of MEA 13-42 at 60 °C
utilizing 0.10 L/min ambient oxygen at the cathode.
Figure 3.46 Effect of methanol concentration on the fuel cell efficiency of MEA 14-127 at 60 °C
utilizing 0.10 L/min ambient oxygen at the cathode.
151
3.6.3 Effect of temperature on fuel cell efficiency of PSSA/PVDF membranes
The impact of one operating parameter on the fuel cell efficiency of a DMFC is often
inherently linked to other operating parameters. Figure 3.47 to Figure 3.49 demonstrated the
impact of methanol concentration on the fuel cell efficiency of Nafion
®
-117, MEA 14-127 and
MEA 13-42 with 0.5 M methanol utilizing 0.1 L/min ambient oxygen at the cathode. The
increase in temperature generally has a favorable effect on the overall efficiency of a DMFC,
especially at high current densities. This is due to the combined effect of reduction of lower
ohmic resistance, and higher reaction kinetics on both anode and cathode, which outweighs the
decrease in fuel utilization as a result of higher methanol crossover. For instance, Nafion
®
-117
exhibits higher overall efficiencies at 90 °C when the current density is larger than 120 mA/cm²
and similarly MEA 14-127 shows superior fuel cell efficiency at 90 °C at current densities
beyond 80 mA/cm². The beneficial effect of temperature to the fuel cell efficiency of MEA 13-
42 is even more pronounced as the overall efficiency increased monotonically with increasing of
temperature in the range of 30 to 90 °C over the range of current density. This is due to a
significant improvement of fuel cell performance as a result of temperature increases and a
relatively minor decrease in fuel utilization due to methanol crossover.
152
Figure 3.47 Effect of temperature on the fuel cell efficiency of Nafion
®
-117 in a DMFC with 0.5
M methanol utilizing 0.10 L/min ambient oxygen at the cathode.
Figure 3.48 Temperature effect on the fuel cell efficiency of MEAs with 20 wt% PSSA.
153
Figure 3.49 Effect of temperature on the fuel cell efficiency of MEA 14-127 with 0.5 M
methanol utilizing 0.1 L/min ambient oxygen at the cathode.
However, the impact of temperature on the fuel cell efficiency of a DMFC presents a
different tendency with high methanol concentrations. For example, with 3.0 M methanol, the
fuel cell efficiency of MEA 13-42 exhibits higher fuel cell efficiency at 60 °C than 90 °C over
the entire range of current density as shown in Figure 3.50. This suggests that at higher methanol
concentrations, the decrease in fuel utilization brought by increasing of temperature eventually
outweighs the voltage efficiency gain as a result of higher temperatures.
154
Figure 3.50 Effect of temperature on the fuel cell efficiency of MEA 13-42 with 3.0 M methanol
utilizing 0.1 L/min ambient oxygen at the cathode.
Figure 3.51 reveals the effect of temperature on the overall efficiencies of MEA 13-42
and Nafion
®
-117 with 2.0 M methanol and 0.10 L/min ambient oxygen at the cathode.
Comparing the fuel cell efficiency of PSSA/PVDF membranes and that of Nafion
®
-117, it is not
difficult to find out that PSSA/PVDF membranes exhibit superior overall DMFC efficiency, due
to lower methanol crossovers with comparable or higher fuel cell performances. These results
indicate that PSSA/PVDF polymer blend membranes are very promising polymer electrolyte
membranes to achieve both high fuel cell performances and efficiencies.
155
Figure 3.51 Fuel cell efficiency of MEAs with 20 wt% PSSA and Nafion
®
-117 with 2.0 M
methanol utilizing ambient oxygen at the cathode.
3.7 Conclusions
The development and commercialization of DMFC call for new materials that are
inexpensive, high performance, and durable. The multiple criteria for polymer electrolyte
membranes in DMFCs make the designing of materials with all desirable properties relevant to
fuel cell operations challenging. Polymer blending offers an attractive mean to overcome the
shortcomings of individual polymer and obtain inexpensive materials with comprehensive
desirable properties by combining the advantages of each polymer component together.
However, the immiscibility between polymers needs to be resolved in order to developed
feasible DMFC membranes based on polymer blends.
New pathway to fabricated homogenous polymer blend membranes based on
PSSA/PVDF was developed involving the direct copolymerization of a ionic liquid monomer
156
TBASS. The homogeneity of PSSA/PVDF membrane is achieved by a “camouflage” method by
blending the hydrophobic terpolymer P(BASS-S-CMS) with PVDF. Upon crosslinking and ion
exchange, which converted PTBSS moiety back to analog PSSA, membranes with a
homogeneous morphology with superior mechanical properties were obtained. The direct
copolymerization of the ionic liquid monomer BASS not only allows for precise control of the
sulfuric acid content in membranes but also avoids any side reactions associated with post-
sulfonation method, resulting in highly reproducible membranes.
The morphology of polymer blend membrane prepared by solution-cast was found to be
strongly dependent on the preparation conditions, particularly the choice of solvent, the casting
temperature, as well as the evaporation speed. Rapid solvent speed with low boiling point solvent
often causes pronounced phase-separation and inferior mechanical properties. The optimization
of the membrane fabrication has been achieved by utilizing high b.p. polar aprotic solvents with
carefully calibration of the annealing temperature and the solvent evaporation speed, yielding
highly homogeneous, flexible and tough membranes.
The membrane composition-property relationship was extensively investigated. The
precise control of the sulfonic acid content in membranes allows for the systematical study of the
role of PSSA content in the fuel cell relevant properties such as water uptake, methanol
permeability, IEC, proton conductivity, mechanical strength, ion cluster microstructure,
distribution of sulfonic acid groups, as well as ultrafiltration property. Of particular interest,
membranes based on PSSA/PVDF exhibited excellent mechanical strength and flexibility, and a
distinct microstructure from Nafion
®
-117 with narrower and more branched hydrophilic
pathways, which is believed to be responsible for the lower water uptake and the smaller
methanol crossover.
157
The control of the chemical nature of the polymer afforded by direct copolymerization of
ionic liquid monomer BASS and the repeatability of the reactions allow us to gain a more
systematic understanding of property-performance correlation of PSSA/PVDF membranes. As a
basis for an alternative membrane design, fundamental research was performed to address the
effect of PSSA uptake in membranes on the fuel cell performance and identify the optimum
membrane composition for DMFCs. In addition, the effects of various fuel cell operation
parameters, such as temperature, methanol concentration, and methanol/oxygen flow rate, on
electrical performance, methanol crossover, fuel utilization and overall efficiency have been
explored. It is found that PSSA/PVDF membranes with optimized composition exhibit uniform
and superior electrochemical performance compared to Nafion
®
-117 under the same operation
conditions due to the lower water and methanol crossover and high proton conductivity nature.
As a result, PSSA/PVDF modular membranes demonstrated both higher methanol utilizations
and higher fuel cell efficiencies, which are the strongly desirable properties as PEMs in DMFC
applications.
In conclusion, PSSA/PVDF proton exchange membranes involved direct
copolymerization of ionic liquid monomer TBASS appear competitive to Nafion
®
-117 in terms
of preparation, performance, and cost-effectiveness. The attractiveness of this method is that it
allows accurate control the amount of sulfonic acid groups in membranes by fine-tuning the
membrane composition, which is often of difficulty for perfluorinated sulfonic acid membranes
like Nafion
®
-117. In addition, the method allows the use of a wide range of ionic liquid
monomers and base polymers. Furthermore, with regards to commercialization, polymer
blending is a promising process for industrial applications and easy to integrate into current
membrane production processes.
158
Recommendations for further investigations are given below. Preliminary studies
demonstrate that PSSA/PVDF membranes can effectively filter out nano-particles and low
concentration salts. This suggest that the potential application in ultrafiltration and reverse
osmosis. Further optimization of membrane, such as incorporation of nano-particles, is needed in
order to increase the water flux. Additional long-term (up to 5,000 hours) durability test under
DMFC operation conditions is needed to evaluate the endurance of PSSA/PVDF and their
commercial viability. In addition, the exploration of the application of PSSA/PVDF membranes
in hydrogen fuel cells as an alternative material to Nafion
®
is of great interest. Finally, the study
of other ionic liquid monomers, including the ammonium salts of 4-vinyl sulfonic acid (VSA), 2-
acrylamido-2-methylpropane sulfonic acid) (AMPS), and 3-sulfopropyl methacrylate (SPM) as
polymers electrolyte components in polymer blends using similar methodologies for fuel cell
applications is also worth to investigate.
159
3.8 References
(186) Zhang, H.; Wang, X.; Zhang, J.; Zhang, J. PEM Fuel Cell Electrocatalysts and …
2008, 889.
(187) Zhang, H.; Wang, X.; Zhang, J.; Zhang, J. In PEM Fuel Cell Electrocatalysts and
Catalyst Layers; Zhang, J., Ed.; Springer London: 2008, p 889.
(188) Liu, J. G.; Zhao, T. S.; Chen, R.; Wong, C. W. Electrochemistry Communications
2005, 7, 288.
(189) Zuo, Z.; Fu, Y.; Manthiram, A. Polymers 2012, 4, 1627.
(190) Passalacqua, E.; Lufrano, F.; Squadrito, G.; Patti, A.; Giorgi, L. Electrochimica
Acta 2001, 46, 799.
(191) Hwan Jung, D.; Hyeong Lee, C.; Soo Kim, C.; Ryul Shin, D. Journal of Power
Sources 1998, 71, 169.
(192) Abdelkareem, M. A.; Nakagawa, N. Journal of Power Sources 2007, 165, 685.
(193) Chen, C. Y.; Yang, P. Journal of Power Sources 2003, 123, 37.
(194) Atti, R. A. Ph.D. Thesis, University of Southern California, 2000.
(195) Yang, B. Ph.D. Thesis, University of Southern California, 2009.
(196) Yee, R. S.; Zhang, K.; Ladewig, B. P. Membranes (Basel) 2013, 3, 182.
(197) Oikonomou, E. K.; Bethani, A.; Bokias, G.; Kallitsis, J. K. European Polymer
Journal 2011, 47, 752.
(198) Surampudi, S.; Narayanan, S. R.; Vamos, E.; Frank, H.; Halpert, G.; Laconti, A.;
Kosek, J.; Prakash, G. K. S.; Olah, G. A. Journal of Power Sources 1994, 47, 377.
(199) Rice, J.; Faghri, A. International Journal of Heat and Mass Transfer 2008, 51,
948.
(200) Xu, C.; Faghri, A.; Li, X. Journal of The Electrochemical Society 2010, 157,
B1109.
(201) Yuan, W.; Tang, Y.; Zhang, Z.; Zhou, B.; Deng, J. Chin. Sci. Bull. 2014, 59,
3216.
(202) Halim, F. A.; Hasran, U. A.; Masdar, M. S.; Kamarudin, S. K.; Daud, W. R. W.
APCBEE Procedia 2012, 3, 40.
(203) Kim, H. Journal of Power Sources 2006, 162, 1232.
(204) Wang, J. T. Journal of The Electrochemical Society 1996, 143, 1233.
160
Bibliography
(1) Zhang, H.; Shen, P. K. Chem Soc Rev 2012, 41, 2382.
(2) Steele, B. C.; Heinzel, A. Nature 2001, 414, 345.
(3) Olah, G. A. Angew Chem Int Ed Engl 2005, 44, 2636.
(4) Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; McGrath, J. E. Chem
Rev 2004, 104, 4587.
(5) Berber, M.; Fujigaya, T.; Sasaki, K.; Nakashima, N. Scientific reports 2013, 3,
1764.
(6) Prakash, G. K. S.; Smart, M. C.; Wang, Q.-J.; Atti, A.; Pleynet, V.; Yang, B.;
McGrath, K.; Olah, G. A.; Narayanan, S. R.; Chun, W.; Valdez, T.; Surampudi, S. Journal of
Fluorine Chemistry 2004, 125, 1217.
(7) Narayanan, S. R.; Kindler, A.; Jeffries-Nakamura, B.; Chun, W.; Frank, H.;
Smart, M.; Valdez, T. I.; Surampudi, S.; Halpert, G.; Kosek, J.; Cropley, C. In Battery
Conference on Applications and Advances, 1996., Eleventh Annual 1996, p 113.
(8) McGrath, K. M.; Prakash, G. K. S.; Olah, G. A. J. Ind. Eng. Chem. 2004, 10,
1063.
(9) Zhou, Z.; Dominey, R. N.; Rolland, J. P.; Maynor, B. W.; Pandya, A. A.;
DeSimone, J. M. J Am Chem Soc 2006, 128, 12963.
(10) Zhou, Z. Ph.D. Thesis, Georgia Institute of Technology, 2007.
(11) Zhang, N.; Zhang, G.; Xu, D.; Zhao, C.; Ma, W.; Li, H.; Zhang, Y.; Xu, S.; Jiang,
H.; Sun, H.; Na, H. International Journal of Hydrogen Energy 2011, 36, 11025.
(12) Zhang, J.; Liu, H. Electrocatalysis of Direct Methanol Fuel Cells: From
Fundamentals to Applications; Wiley, 2009.
(13) Zhang, H.; Shen, P. Chemical Society reviews 2012, 41, 2382.
(14) Zaidi, J.; Matsuura, T. Polymer Membranes for Fuel Cells; Springer, 2009.
(15) Yuan, W.; Zhou, B.; Deng, J.; Tang, Y.; Zhang, Z. C.; Li, Z. T. International
Journal of Hydrogen Energy 2014, 39, 6689.
(16) Yin, Y.; Yamada, O.; Tanaka, K.; Okamoto, K.-I. Polym. J 2006, 38, 197.
(17) Ye, Y.-S.; Rick, J.; Hwang, B.-J. Journal of Materials Chemistry A 2013, 1, 2719.
(18) Yang, T. International Journal of Hydrogen Energy 2008, 33, 6772.
161
(19) Yakovlev, S.; Wang, X.; Ercius, P.; Balsara, N. P.; Downing, K. H. Journal of the
American Chemical Society 2011, 133, 20700.
(20) Xing, P.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.; Wang, K.;
Kaliaguine, S. Journal of Membrane Science 2004, 229, 95.
(21) Witham, C. K.; Chun, W.; Valdez, T. I.; Narayanan, S. R. Electrochemical and
Solid-State Letters 2000, 3, 497.
(22) Wasmus, S.; Küver, A. Journal of Electroanalytical Chemistry 1999, 461, 14.
(23) Wang, Y.-J.; Qiao, J.; Baker, R.; Zhang, J. Chemical Society reviews 2013, 42,
5768.
(24) Wang, J.; Li, N.; Cui, Z.; Zhang, S.; Xing, W. Journal of Membrane Science
2009, 341, 155.
(25) Wang, F.; Hickner, M.; Kim, Y. S.; Zawodzinski, T. A.; McGrath, J. E. Journal of
Membrane Science 2002, 197, 231.
(26) Wainright, J. S.; Wang, J. T.; Weng, D.; Savinell, R. F.; Litt, M. Journal of The
Electrochemical Society 1995, 142, L121.
(27) Vielstich, W.; Lamm, A.; Gasteiger, H. A. Handbook of Fuel Cells:
Fundamentals, Technology, and Applications; Wiley: New York, 2003.
(28) Vasquez, L. O. Fuel Cell Research Trends; Nova Science Publishers, 2007.
(29) Unveren, E. E.; Erdogan, T.; Çelebi, S. S.; Inan, T. Y. International Journal of
Hydrogen Energy 2010, 35, 3736.
(30) Tsou, Y.-M.; Zhu, Z.; Cao, L.; De Castro, E. ECS Transactions 2007, 2, 19.
(31) Tripathi, B. P.; Shahi, V. K. Progress in Polymer Science 2011, 36, 945.
(32) Thomas, S. C.; Ren, X.; Gottesfeld, S.; Zelenay, P. Electrochimica Acta 2002, 47,
3741.
(33) Taylor, E. P.; Landis, F. A.; Page, K. A.; Moore, R. B. Polymer 2006, 47, 7425.
(34) Surampudi, S.; Narayanan, S. R.; Vamos, E.; Frank, H.; Halpert, G.; LaConti, A.;
Kosek, J.; Prakash, G. K. S.; Olah, G. A. Journal of Power Sources 1994, 47, 377.
(35) Subianto, S. Polymer International 2014, 63, 1134.
(36) Springer, T. E.; Zawodzinski, T. A.; Gottesfeld, S. Journal of The
Electrochemical Society 1991, 138, 2334.
(37) Souzy, R.; Boutevin, B.; Ameduri, B. Macromolecules 2012, 45, 3145.
162
(38) Smitha, B.; Sridhar, S.; Khan, A. A. Journal of Membrane Science 2005, 259, 10.
(39) Smitha, B.; Sridhar, S.; Khan, A. A. Macromolecules 2004, 37, 2233.
(40) Singha, S.; Jana, T. Polymer 2014, 55, 594.
(41) Silva, V. S.; Mendes, A. M.; Madeira, L. M.; Nunes, S. P. In Advances in fuel
cells; Zhang, X., Ed.; Research Signpost: Kerala, India, 2005.
(42) SHRIVASTAVA, N. K.; THOMBRE, S. B. International Journal of Engineering
Science and Technology 2011.
(43) Shin, J. P.; Chang, B. J.; Kim, J. H.; Lee, S. B.; Suh, D. H. Journal of Membrane
Science 2005, 251, 247.
(44) Sharma, S.; Pollet, B. G. Journal of Power Sources 2012, 208, 96.
(45) Sharaf, O. Z.; Orhan, M. F. Renewable and Sustainable Energy Reviews 2014, 32,
810.
(46) Sgreccia, E.; Di Vona, M. L.; Knauth, P. International Journal of Hydrogen
Energy 2011, 36, 8063.
(47) Seung Hyun, H.; Jin-Nam, P. Asia-Pacific Journal of Chemical Engineering
2013, 8.
(48) Service, R. F. Science 1999, 285, 682.
(49) Rozière, J.; Jones, D. J. Annual Review of Materials Research 2003, 33, 503.
(50) Robeson, L. M.; Hwu, H. H.; McGrath, J. E. Journal of Membrane Science 2007,
302, 70.
(51) Rikukawa, M.; Sanui, K. Progress in Polymer Science 2000, 25, 1463.
(52) Richard, P.; Zhigang, Q.; Qunhui, G.; Leng, M.; John, F. E.; Bin, D. In Materials
for the Hydrogen Economy; CRC Press: 2007, p 251.
(53) Ren, X.; Springer, T. E.; Gottesfeld, S. Journal of The Electrochemical Society
2000, 147, 92.
(54) Ren, S.; Sun, G.; Li, C.; Wu, Z.; Jin, W.; Chen, W.; Xin, Q.; Yang, X. Materials
Letters 2006, 60, 44.
(55) Rashidi, R.; Dincer, I.; Naterer, G. F.; Berg, P. Journal of Power Sources 2009,
187, 509.
(56) Rana, D.; Matsuura, T.; Zaidi, S. M. J. In Polymer Membranes for Fuel Cells;
Zaidi, S. M. J., Matsuura, T., Eds.; Springer US: 2009, p 401.
163
(57) Prakash, G. K. S.; Olah, G. A.; Smart, M. C.; Narayanan, S. R.; Wang, Q. J.;
Surumpudi, S.; Halpert, G.; University Of Southern California: US, 2002.
(58) Prakash, G. K. S.; Krause, F. C.; Viva, F. A.; Narayanan, S. R.; Olah, G. A.
Journal of Power Sources 2011, 196, 7967.
(59) Peighambardoust, S. J.; Rowshanzamir, S.; Amjadi, M. International Journal of
Hydrogen Energy 2010, 35, 9349.
(60) Park, J.-S.; Choi, Y.-W. Chemistry Letters 2013, 42, 998.
(61) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. The Journal of Organic Chemistry
2008, 74, 487.
(62) Oikonomou, E. K.; Bethani, A.; Bokias, G.; Kallitsis, J. K. European Polymer
Journal 2011, 47, 752.
(63) O'Hayre, R. P.; Cha, S. W.; Colella, W.; Prinz, F. B. Fuel cell fundamentals;
Wiley: New York, 2006.
(64) Ng, C. W. A.; MacKnight, W. J. Macromolecules 1996, 29, 2412.
(65) Neburchilov, V.; Martin, J.; Wang, H. J.; Zhang, J. J. Journal of Power Sources
2007, 169, 221.
(66) McNicol, B. D.; Rand, D. A. J.; Williams, K. R. Journal of Power Sources 1999,
83, 15.
(67) Mauritz, K. A.; Moore, R. B. Chemical Reviews 2004, 104, 4535.
(68) Manea, C.; Mulder, M. Journal of Membrane Science 2002, 206, 443.
(69) Maier, G.; Meier-Haack, J. In Fuel Cells II; Scherer, G., Ed.; Springer Berlin
Heidelberg: 2008; Vol. 216, p 1.
(70) Lucid, J.; Meloni, S.; MacKernan, D.; Spohr, E.; Ciccotti, G. The Journal of
Physical Chemistry C 2012, 117, 774.
(71) Liu, H.; Song, C.; Zhang, L.; Zhang, J.; Wang, H.; Wilkinson, D. P. Journal of
Power Sources 2006, 155, 95.
(72) Liu, F.; Hashim, N. A.; Liu, Y.; Abed, M. R. M.; Li, K. Journal of Membrane
Science 2011, 375, 1.
(73) Li, W.; Manthiram, A.; Guiver, M. D. Journal of Membrane Science 2010, 362,
289.
(74) Li, M.; Wondergem, H. J.; Spijkman, M.-J.; Asadi, K.; Katsouras, I.; Blom, P. W.
M.; de Leeuw, D. M. Nat Mater 2013, 12, 433.
164
(75) Li, L.; Zhang, J.; Wang, Y. Journal of Membrane Science 2003, 226, 159.
(76) Larminie, J.; Dicks, A. Fuel Cell Systems Explained; 2nd ed.; Wiley: New York,
2003.
(77) Landis, F. A.; Moore, R. B. Macromolecules 2000, 33, 6031.
(78) Kreuer, K. D. Journal of Membrane Science 2001, 185, 29.
(79) Kordesch, K. V.; Simader, G. R. In Fuel Cells; Wiley-VCH Verlag GmbH & Co.
KGaA: 2006, p 9.
(80) Knights, S. D.; Colbow, K. M.; St-Pierre, J.; Wilkinson, D. P. Journal of Power
Sources 2004, 127, 127.
(81) Kirt, A. P.; Brandon, W. R. In Polymers for Energy Storage and Delivery:
Polyelectrolytes for Batteries and Fuel Cells; American Chemical Society: 2012; Vol. 1096, p
147.
(82) Kim, Y. S.; Dong, L.; Hickner, M. A.; Glass, T. E.; Webb, V.; McGrath, J. E.
Macromolecules 2003, 36, 6281.
(83) Kim, D. H.; Choi, J.; Hong, Y. T.; Kim, S. C. Journal of Membrane Science 2007,
299, 19.
(84) Kerres, J. A.; Chromik, A.; Krajinovic, K.; Katzfuss, A.; Seyb, C.; Aniol, K.;
Atanasov, V.; Kaz, T.; Haering, T.; Kruger, A.; Schoeman, H.; Bessarabov, D.; Krieg, H. ECS
Transactions 2011, 41, 1621.
(85) Kerres, J. A. Fuel Cells 2005, 5, 230.
(86) Kerres, J. A. Journal of Membrane Science 2001, 185, 3.
(87) Kerres, J.; Zhang, W.; Jörissen, L.; Gogel, V. J. New Mat. Electrochem. Systems
2014.
(88) Kerres, J. In Polymer Membranes for Fuel Cells; Javaid, Z. S. M., Matsuura, T.,
Eds.; Springer US: 2009, p 1.
(89) Kerres, J. Solid State Ionics 1999, 125, 243.
(90) Karim, M. R.; Hatakeyama, K.; Matsui, T.; Takehira, H.; Taniguchi, T.;
Koinuma, M.; Matsumoto, Y.; Akutagawa, T.; Nakamura, T.; Noro, S.; Yamada, T.; Kitagawa,
H.; Hayami, S. J Am Chem Soc 2013, 135, 8097.
(91) Kamarudin, S. K.; Achmad, F.; Daud, W. R. W. International Journal of
Hydrogen Energy 2009, 34, 6902.
(92) Jung, B.; Kim, B.; Yang, J. M. Journal of Membrane Science 2004, 245, 61.
165
(93) Jones, D. J. In Polymer Electrolyte Membrane and Direct Methanol Fuel Cell
Technology; Hartnig, C., Roth, C., Eds.; Woodhead Publishing: 2012; Vol. 1, p 27.
(94) Jannasch, P. Current Opinion in Colloid & Interface Science 2003, 8, 96.
(95) Jagur-Grodzinski, J. Polymers for Advanced Technologies 2007, 18, 785.
(96) Ibrahim, H.; Ilinca, A.; Perron, J. Renewable and Sustainable Energy Reviews
2008, 12, 1221.
(97) Houchins, C.; Kleen, G.; Spendelow, J.; Kopasz, J.; Peterson, D.; Garland, N.;
Ho, D.; Marcinkoski, J.; Martin, K.; Tyler, R.; Papageorgopoulos, D. Membranes 2012, 2, 855.
(98) Higashihara, T.; Matsumoto, K.; Ueda, M. Polymer 2009, 50, 5341.
(99) Hamnett, A. Catalysis Today 1997, 38, 445.
(100) Fu, Y.-Z.; Manthiram, A.; Guiver, M. D. Electrochemical and Solid-State Letters
2007, 10, B70.
(101) Dutta, K.; Kumar, P.; Das, S.; Kundu, P. P. Polymer Reviews 2014, 54, 1.
(102) Dupuis, A. C. Progress in Materials Science 2011, 56, 289.
(103) Dresselhaus, M. S.; Thomas, I. L. Nature 2001, 414, 332.
(104) Drake, J. A.; Wilson, W.; Killeen, K. Journal of The Electrochemical Society
2004, 151, A413.
(105) Dillon, R.; Srinivasan, S.; Arico, A. S.; Antonucci, V. Journal of Power Sources
2004, 127, 112.
(106) Di Vona, M. L.; Knauth, P. Zeitschrift für Physikalische Chemie 2013, 227, 595.
(107) DeLuca, N. W.; Elabd, Y. A. Journal of Polymer Science Part B: Polymer
Physics 2006, 44, 2201.
(108) Deimede, V.; Voyiatzis, G. A.; Kallitsis, J. K.; Qingfeng, L.; Bjerrum, N. J.
Macromolecules 2000, 33, 7609.
(109) Chu, S.; Majumdar, A. Nature 2012, 488, 294.
(110) Choi, J.; Kim, D. H.; Kim, H. K.; Shin, C.; Kim, S. C. Journal of Membrane
Science 2008, 310, 384.
(111) Chikh, L.; Delhorbe, V.; Fichet, O. Journal of Membrane Science 2011, 368, 1.
(112) Chang, H.; Kim, H.; Choi, Y.; Lee, W. In Polymer Membranes for Fuel Cells;
Zaidi, S. M. J., Matsuura, T., Eds.; Springer US: 2009, p 307.
166
(113) Chandan, A.; Hattenberger, M.; El-Kharouf, A.; Du, S. F.; Dhir, A.; Self, V.;
Pollet, B. G.; Ingram, A.; Bujalski, W. Journal of Power Sources 2013, 231, 264.
(114) Carrette, L.; Friedrich, K. A.; Stimming, U. Fuel Cells 2001, 1, 5.
(115) Carrette, L.; Friedrich, K. A.; Stimming, U. Chemphyschem 2000, 1, 162.
(116) Buchmüller, Y.; Wokaun, A.; Gubler, L. Fuel Cells 2013, 13, 1177.
(117) Brandrup, J.; Immergut, E. H.; Grulke, E. A. Polymer Handbook; Wiley, 2003.
(118) Bose, S.; Kuila, T.; Thi, X. L. N.; Kim, N. H.; Lau, K. T.; Lee, J. H. Progress in
Polymer Science 2011, 36, 813.
(119) Blomen, L. J. M. J.; Mugerwa, M. N. Fuel Cell Systems; Springer, 1993.
(120) Bernardi, D. M.; Verbrugge, M. W. Journal of The Electrochemical Society 1992,
139, 2477.
(121) Behling, N. H. Fuel Cells: Current Technology Challenges and Future Research
Needs; Elsevier, 2013.
(122) Bashyam, R.; Zelenay, P. Nature 2006, 443, 63.
(123) Asensio, J. A.; Sanchez, E. M.; Gomez-Romero, P. Chemical Society Reviews
2010, 39, 3210.
(124) Arunbabu, D.; Sannigrahi, A.; Jana, T. The Journal of Physical Chemistry B
2008, 112, 5305.
(125) Aricò, A. S.; Srinivasan, S.; Antonucci, V. Fuel Cells 2001, 1, 133.
(126) Aricò, A. S.; Baglio, V.; Antonucci, V. In Electrocatalysis of Direct Methanol
Fuel Cells; Wiley-VCH Verlag GmbH & Co. KGaA: 2009, p 1.
(127) Appleby, A. J.; Foulkes, F. R. Fuel cell handbook; Van Nostrand Reinhold, 1989.
(128) Ajji, A.; Utracki, L. A. Polymer Engineering & Science 1996, 36, 1574.
Abstract (if available)
Abstract
As direct methanol fuel cells (DMFCs) hold great promise for wide applications in transportation, stationary and portable power sources, the study of fuel cell materials and their contribution to the overall cost and performance are of considerable practical importance. One of the key components that receives much research attention is the proton electrolyte membrane (PEM). Perfluorosulfonic acid (PFSA) membranes such as Nafion® are considered to be the current benchmark membranes for DMFCs. Despite of the merit of high proton conductivity and stability, PFSA membranes suffer from high cost and excessive methanol crossover. Therefore, alternative cost-effective and high performance materials are strongly desired. In this work, high performance and cost-effective polymer electrolyte membranes (PEMs) based on tetra-butylammonium poly(styrene sulfonate) (PTBASS)/PVDF polymer blends are developed. The PTBASS/PVDF holds promise as a substitute membrane in DMFCs, and the methodology to fabricate membranes in the present study also provides a tool to elucidate the complexities of the structure-performance relationship in PEMs. ❧ In Chapter I, the background of fuel cells, the working principle of DMFCs and the polymer electrolyte membranes for DMFCs are reviewed and discussed. Various strategies including sulfonation of aromatic polymer, impregnation of polyelectrolytes in inert polymers, organic/inorganic composite, and polymer blending have been employed to develop alternative membranes for DMFCs. In spite of some progress have been made in terms of reduction in methanol crossover, development of a sturdy inexpensive substitute to perfluorosulfonic acid membranes is yet to materialize. ❧ In chapter II, a seemingly immiscible polystyrene sulfonic acid/polyvinylidene fluoride (PVDF-PSSA) nano-composite homogeneous composite blend was synthesized by a “polymer camouflage” approach. Terpolymerization of an ionic liquid monomer, tetra-butylammonium styrene sulfonate (BASS), with styrene and 4-chloromethyl styrene (CMS) is carried out with free radical polymerization to give a terpolymer poly(BASS-S-CMS). Polymer blend membranes consisting of poly(BASS-S-CMS) and PVDF was fabricated through solution-cast technique followed by annealing, crosslinking and ion exchange, giving highly flexible, transparent and tough PSSA/PVDF Nafion®-like membranes. Fuel cell relevant properties of PSSA/PVDF membranes, such as the methanol permeability, water uptake, ion exchange capacity (IEC), proton conductivity, membrane morphology and microstructure, and ultrafiltration property were investigated. The relationship between membrane composition and property is systematically studied. The experimental techniques including Gas Chromatography (GC), Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), Energy Dispersive X-ray spectroscopy (EDS), Thermogravimetric Analysis (TGA), UV-Visible Spectroscopy, and Electrochemical Impedance Spectroscopy (EIS) are included. ❧ In Chapter III, the electrochemical performance, methanol crossover, fuel utilization, and fuel cell efficiency of PSSA/PVDF modular membranes in DMFCs were extensively measured and analyzed. The effect of PSSA content on the fuel cell performance, methanol crossover and fuel cell efficiency are addressed. The impact of various operation parameters including temperature, methanol concentration, methanol and oxygen flow rate on the performance of PSSA/PVDF in DMFCs were tested in both ambient oxygen and air environment. An understanding of membrane property-performance relationship is examined and discussed. The present PSSA/PVDF membranes exhibit comparable or superior electrochemical performance compared to Nafion®-117 due to lower methanol crossover and higher proton conductivity. As a result, high fuel utilizations and overall fuel cell efficiency were achieved using PSSA/PVDF membranes in DMFCs. ❧ The polymer electrolyte membranes based on the polymer blend of PSSA/PVDF shows a promising potential in DMFC applications with impressive economic advantages. The methodology employed in membrane fabrication provides a general method for exploring many other monomer, including ammonium salts of 4-vinyl sulfonic acid (VSA), 2-acrylamido-2-methylpropane sulfonic acid) (AMPS), and 3-sulfopropyl methacrylate (SPM) as alternative polymers electrolyte materials for DMFCs applications. The conclusions and outlook are discussed in the last chapter.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Blends of polystyrene sulfonic acid copolymers and polyvinylidene fluoride as polyelectrolyte membranes
PDF
Development of polymer electrolyte membranes for fuel cell applications
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
Direct C−H arylation for the synthesis of conjugated polymers
PDF
The modification of catalysts and their supports for use in various fuel cells
PDF
Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction
Asset Metadata
Creator
Li, Ming
(author)
Core Title
Development of polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry
Publication Date
11/11/2014
Defense Date
09/15/2014
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
ammonium substituted styrene sulfonate,direct copolymerization of styrene sulfonate,direct methanol fuel cell,DMFC,ionic liquid monomer,methanol crossover,OAI-PMH Harvest,PEM,polymer blend,polymer electrolyte membrane,PSSA,PVDF
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Hogen-Esch, Thieo E. (
committee chair
), Gupta, Malancha (
committee member
), Prakash, G. K. Surya (
committee member
)
Creator Email
mingli1@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-516256
Unique identifier
UC11298627
Identifier
etd-LiMing-3075.pdf (filename),usctheses-c3-516256 (legacy record id)
Legacy Identifier
etd-LiMing-3075.pdf
Dmrecord
516256
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Li, Ming
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
ammonium substituted styrene sulfonate
direct copolymerization of styrene sulfonate
direct methanol fuel cell
DMFC
ionic liquid monomer
methanol crossover
PEM
polymer blend
polymer electrolyte membrane
PSSA
PVDF