University of Southern California Dissertations and Theses

PSSA-PVDF semi-IPN blends for direct methanol fuel cells

(USC Thesis Other)

PSSA-PVDF semi-IPN blends for direct methanol fuel cells

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

PSSA-PVDF Semi-IPN Blends for Direct Methanol Fuel Cells
The further development and investigation into the PSSA copolymer and its blends with PVDF

By

Adam Bruce Ung

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)

December 2022

Copyright 2022 Adam Bruce Ung

ii

Dedication

Dedicated to
My parents, siblings, relatives, and friends

iii

Acknowledgements
When I began my Ph.D. journey at USC, I was going through a rough time in my life. My
father had passed away and my studies had duly suffered. I spent an extra semester raising my
grades so that I could join a research group and was therefore behind the rest of my cohort. I was
depressed and unsure if I would continue to remain in the chemistry department. I had felt that
my passion and drive for theoretical chemistry was no longer present after me having performed
poorly in my initial graduate level courses, and due to stress with having just experienced a great
personal loss. I worked up the optimism, with help from the faculty, to find interest in doing
chemistry but perhaps in a different field. Fortunately, as I was searching for a group, a professor
from the Loker Hydrocarbon Research Institute had reached out about a research opportunity
involving polymers and electrochemical systems. I had just finished a graduate level course on
polymers and their applications and had been interested in energy systems all my life. My deepest
gratitude and appreciation both go to Dr. Thieo E. Hogen-Esch, for being my main research advisor
and for providing me the opportunity to work on an intellectually stimulating project that brought
together many facets of chemistry and engineering— from small molecule synthesis to device
evaluation, and everything in between. Dr. Hogen-Esch continues to be an invaluable source of
knowledge and tenacity for me to learn from, developed no doubt through his influential and
productive tenure at USC. I have learned a lot of meaningful career- and life-lessons from Dr.
Hogen-Esch that will be pivotal for tackling whatever great challenges await me in life. Secondly,
am very grateful to my committee chair, and current faculty research advisor, Professor Surya
Prakash (whose group collaborated closely with the Hogen-Esch group) for allowing me to join
his group after Dr. Hogen-Esch had earned professor emeritus status. This gesture allowed me to
continue my work in his Fuel Cell Lab. Both Prof. Prakash and Prof. Sri Narayan, and their

iv

collective past and present group members, were pivotal in providing me with very thoughtful
discussions that aided me when I was stuck. I would uniquely like to thank Dr. Prakash for
introducing me to: the legacy of the late Prof. George A. Olah, the methanol economy, his
welcoming and intelligent group members, Dr. Prakash’s own incredible research that spans many
notable industries, and significant research funding. The work done by the Prakash group, and
Prof. Prakash himself, have fundamentally shaped the way that I view the climate and energy
crisis. I would like to thank Prof. Steven Nutt, for serving as a faculty member on both my
Screening and Qualifying exam committees and for providing me the ability to acquire data
relating to thermal and mechanical properties of the materials I worked with. I would like to
specifically thank Prof. Nutt’s colleagues, Dr. Jon Lo, Dr. Yunpeng Zhang, and Dr. Zehan Yu for
doing the TGA, DSC, and DMA measurements for me.
For light scattering, aqueous SEC, and AFM measurements and instrument assistance, I
would like to thank Dr. Shuxing Li from The Center of Excellence in NanoBiophysics at USC.
From the USC Agilent Center of Excellence in Biomolecular Characterization, I would like to
thank Dr. Yasaman Dayani for the many SEC analyses of our polymer samples and for helpful
discussions regarding data interpretation. From the Core Center of Excellence in Nano Imaging,
I would like to express my appreciation for Dr. Carolyn Marks and Dr. Daniel Goodelman for their
assistance with microtoming, and TEM, SEM, and EDS measurements coupled with valuable
training experience.
Appreciation goes to my mentors, Dr. Sergey Muhkin, Dr. Marc Iuliucci, Dr. Dean Glass,
Dr. Eugene Kong, and Dr. Bo Yang taught me various practical methods for much of my laboratory
work and provided helpful discussions and suggestions when I was just beginning my work.
Previous group member, Dr. Ming Li, deserves mention for laying much of the groundwork

v

involving the PSSA-PVDF system. My mentees, Mr. Jung-Hoon Kim, Mr. Seo-Kyung Ahn, and
the late Mr. Jeremy Dawson, provided me the opportunity to teach what I had learned in my
research to another, and provided my work with useful investigations and repeatability data. My
gratitude goes to Dr. Alain Geoppert and Dr. Huong Dang for assistance with GC measurements,
and especially with general instrument maintenance. A very special thanks goes to Dr. Robert
Aniszfeld and Michele Dea who both worked very hard for all of us in the chemistry department
before, during, and after the 2020 worldwide pandemic, to ensure continued progress for everyone,
myself included.
My TA experience at USC was enjoyable and full of unforgettable life lessons. During my
time as a TA, my upmost gratitude and appreciation both go to Dr. Elizabeth Erickson for
mentoring me in a leadership role to help me grow into a person who is more: attentive to detail,
open-minded to work with others, and adaptable when working under conditions of high stress
with deadlines. To the many other past, and present, graduate students in the chemistry department
(for whom I had the pleasure of being head TA in some capacity), thank you for a superb TA
experience. Each group of TAs were excellent in unique ways, yet very productive to work with.
To my many peers and close friends at USC, whom I conversed with regularly about
various topics, some having little to do with chemistry, I would like to emphasize Dr. Eugene
Kong, Dr. Vicente Galvan, Dr. Amanda Baxter, Dr. Kevin Kossick, Mr. Juan Pablo de los Rios,
and Dr. Emma Singer. Thank you all for listening to what I have on my mind and never failing to
reciprocate with pleasant and sensible dialogue. A special thanks goes to Dr. Vicente Galvan, who
helped me obtain some valuable TEM data. My time spent in the Fuel Cell Lab at USC seemed
so short yet produced some of the most memorable moments of my life.

vi

Finally, I would like to thank my parents who both escaped the Khmer genocide to America
as refugees in order to guarantee a better life for my siblings and me. An infinitely compounding
sum of gratitude goes to my elder siblings and my elder relatives for their endless wisdom.

vii

Table of Contents
Dedication ................................................................................................................................... ii
Acknowledgements .................................................................................................................... iii
List of Tables ............................................................................................................................... x
List of Figures ............................................................................................................................ xi
Chapter 1 Miniature Review ........................................................................................................... 1
1.1 Anthropogenic Climate Change ............................................................................................ 1
1.2 Electricity from Fossil Fuels ................................................................................................. 3
1.3 Electrochemical Systems and Fuel Cells .............................................................................. 4
1.4 Chemistry of Direct Methanol Fuel Cells ............................................................................. 6
1.5 Thermodynamics of Fuel Cells ............................................................................................. 7
1.6 Materials for Fuel Cells ......................................................................................................... 8
1.7 Membrane Electrode Assembly (MEA) ................................................................................ 9
1.8 Performance of Fuel Cells ................................................................................................... 11
Chapter 2 Modified Synthesis of Tetrabutylammonium 4-Styrene Sulfonate and its
Homopolymerization .................................................................................................................... 13
2.1 Materials .............................................................................................................................. 14
2.2 Synthesis of TBASS ............................................................................................................ 15
2.3 NMR Characterization of TBASS ...................................................................................... 16
2.4 Chemical Shifts for TBASS ................................................................................................ 16
2.5 Melting point determination ................................................................................................ 17
2.6 X-ray crystallography .......................................................................................................... 17
2.7 Homopolymerization ........................................................................................................... 18
2.8 Molecular Weight Determination........................................................................................ 18
2.9 Thermogravimetric Analysis (TGA) ................................................................................... 19
2.10 Homopolymerization Kinetics .......................................................................................... 19
2.11 Crystal Structure of TBASS .............................................................................................. 19
2.12 TBASS monomer synthesis .............................................................................................. 23
2.13 Homopolymerization of TBASS ....................................................................................... 24
2.14 Evidence of Aggregation and Auto-acceleration .............................................................. 26
2.15 Supplemental Information ................................................................................................. 28
Chapter 3 Vinyl Copolymers of Tetrabutylammonium 4-Styrenesulfonate ................................. 31

viii

3.1 Materials .............................................................................................................................. 33
3.2 Copolymerization kinetics of TBASS and styrene ............................................................. 34
3.3 Copolymerization of styrene with TBASS.......................................................................... 34
3.4 Copolymerization of TBASS and MMA ............................................................................ 35
3.5 Thermal degradation ........................................................................................................... 35
3.6 Composition of Copolymers ............................................................................................... 36
3.7 NMR Spectral Measurements ............................................................................................. 37
3.8 Copolymerization kinetics TBASS and MMA ................................................................... 40
3.9 Thermal Properties of PS-Co-PTBASS .............................................................................. 42
3.10 Supplemental Information ................................................................................................. 48
Chapter 4 Blended Membrane Composites of TBASS Copolymers and Polyvinylidene Fluoride
....................................................................................................................................................... 53
4.1 Materials .............................................................................................................................. 55
4.2 Synthesis of PTBASS Copolymers ..................................................................................... 56
4.3 Composite Synthesis ........................................................................................................... 56
4.4 Ion Exchange Processing of Composites into Membranes ................................................. 57
4.5 Water Uptake Measurements .............................................................................................. 58
4.6 Ion-Exchange Capacity (IEC) ............................................................................................. 59
4.7 Conductance Measurements ................................................................................................ 59
4.8 Methanol Permeability ........................................................................................................ 60
4.9 Tensile Strength Measurements .......................................................................................... 61
4.10 Differential Scanning Calorimetry .................................................................................... 61
4.11 Scanning Electron Microscopy (SEM) ............................................................................. 61
4.12 Atomic Force Microscopy (AFM) .................................................................................... 61
4.13 Energy Dispersive X-ray Spectroscopy (EDS) ................................................................. 62
4.14 Degree of Crosslinking...................................................................................................... 62
4.15 Effect of Crosslinking on Modulus ................................................................................... 62
4.16 Synthesis of Copolymers ................................................................................................... 63
4.17 Synthesis of PSSA copolymer-PVDF Blends ................................................................... 64
4.18 Crosslinking in PSSA-PVDF Blends ................................................................................ 66
4.19 Crosslinking Mechanism ................................................................................................... 70
4.20 Modeling the Crosslinking Reaction ................................................................................. 71
4.21 Mechanical Measurements of PSSA-PVDF membranes .................................................. 73

ix

4.22 Proton Conductivities ........................................................................................................ 75
4.23 TGA of PSSA-PVDF ........................................................................................................ 75
4.24 Microscopy of PSSA-PVDF membranes .......................................................................... 79
4.25 Crystallinity of PVDF ....................................................................................................... 87
4.26 Differential Scanning Calorimetry (DSC)......................................................................... 88
4.27 Accelerated Oxidative Stress Test..................................................................................... 90
4.28 Proton Conductivity Dependence on Temperature ........................................................... 90
4.29 Supplemental Information ................................................................................................. 93
Chapter 5 Fuel Cell Performance .................................................................................................. 95
5.1 Perfluorinated Membranes .................................................................................................. 95
5.2 Non-perfluorinated hydrocarbon based membranes ........................................................... 96
5.3 Materials .............................................................................................................................. 98
5.4 Catalyst Ink Preparation Procedure ..................................................................................... 99
5.5 Painting Electrodes Procedure ............................................................................................ 99
5.6 MEA Hot-Pressing Procedure ........................................................................................... 101
5.7 Hydrating Procedure ......................................................................................................... 102
5.8 Fuel Cell Break-in Procedure ............................................................................................ 103
5.9 Performance Testing Protocol ........................................................................................... 104
5.10 Fabrication of MEA-5 ..................................................................................................... 107
5.11 MEA-5 Fuel Cell Results ................................................................................................ 108
5.12 Repeatability of MEA-5 .................................................................................................. 114
5.13 Fabrication of MEA-6 ..................................................................................................... 116
5.14 MEA-6 Fuel Cell Results ................................................................................................ 116
5.15 Membrane Electrode Assemblies Prepared with Carbon Supported Catalysts............... 123
5.16 Towards Thinner PSSA-PVDF Membranes ................................................................... 124
5.17 MEA-10 Fuel Cell Results .............................................................................................. 124
5.18 MEA-11 Fuel Cell Results .............................................................................................. 127
5.19 Fuel Cells with Nafion .................................................................................................... 128
5.20 Direct Comparison of Cell Performance ......................................................................... 129
5.21 Comparisons to other reports .......................................................................................... 140
5.22 Supplemental Section ...................................................................................................... 143
Bibliography ............................................................................................................................... 152

x

List of Tables
Table 1.1. Theoretical reversible cell potentials (E°rev) and
maximum intrinsic efficiencies for fuel cell reactions at STP adapted from Ref.
8
......................... 7
Table 1.2. Different Fuel Cells currently in use and development.
15
.......................................... 11
Table 2.1. TBASS homopolymers ............................................................................................... 25
Table 3.1. Composition of 70-90% conversion TBASS-styrene copolymers
in DCE at 65 °C as a function of initial comonomer feed composition. ....................................... 38
Table 3.2. Molecular weight and dispersity of copolymers after acidification ........................... 38
Table 4.1. Apparent molecular weight data for p(TBASS-S-CMS) terpolymers ........................ 64
Table 4.2. Composition and characterization of PSSA-PVDF Blended
Membranes using the 801010 copolymer of TBASS, styrene, and CMS. ................................... 68
Table 4.3. Composition and characterization of PSSA-PVDF blended membranes
using the 8020 copolymer of TBASS and CMS.
a.b
....................................................................... 68
Table 4.4. Reaction of Cumene and 4-isopropyl benzyl chloride (4IBC) .................................... 72
Table 4.5. Model Reaction of 4IBC and Anisole......................................................................... 72
Table 4.6. Thermodynamics of Proton Conductance for membranes
from Table 4.2 ............................................................................................................................... 90
Table 5.1. Thicknesses of Membranes used in MEAs ............................................................... 105

xi

List of Figures
Figure. 1.1. History and projection of global energy consumption
adapted from ref.
4
............................................................................................................................ 2
Figure 1.2. History and projection of global energy consumption
by sources of energy adapted from ref.
3
......................................................................................... 2
Figure 1.3. Schematic of the conversion of kinetic to electrical energy ........................................ 3
Figure 1.4. Balanced half-reactions of the DMFC system............................................................. 6
Figure 1.5. Typical polarization curve for a PEM fuel cell stack.
14
.............................................. 6
Figure 1.6. Schematic of the DMFC system .................................................................................. 8
Figure 2.1. Counter-ion distance in TBASS crystal structure. .................................................... 20
Figure 2.2. Crystal structure of cesium 4-methylbenzenesulfonate.
55
......................................... 21
Figure 2.3. Crystal structure of tetrapropylammonium hydrogen carbonate.
59
........................... 22
Figure 2.4. Conversion of Entry 2.6. ........................................................................................... 24
Figure 2.5. SEC Chromatogram plot of entry 2.7-16 in Table 2.2. ............................................. 27
Figure 3.1. TBASS styrene copolymer-as a function of
comonomer-composition at complete conversions. ...................................................................... 36
Figure 3.2. Kinetics of equimolar copolymerization of styrene and TBASS ............................... 37
Figure 3.3. Proton NMR of TBASS and MMA in CDCl3. .......................................................... 41
Figure 3.4. Conversion of TBASS and MMA as a function of time. .......................................... 41
Figure 3.5. Dependence of glass transition temperature of
partially sulfonated polystyrene on composition.
76
...................................................................... 43
Figure 3.6. Extrapolation plot of glass transition temperatures
of polystyrene copolymers containing small amounts of TBASS. ............................................... 43

xii

Figure 3.7. TGA curves of polystyrene, PTBASS, and
PS-co-TBASS (95:5 composition). ............................................................................................... 44
Figure 3.8. Isothermal TGA at 320 °C. ........................................................................................ 44
Figure 3.9. Isothermal TGA at 350 °C of
PS-co-TBASS (99:1 monomer feed ratio). ................................................................................... 46
Figure 4.1. Chemical structure of Nafion® ................................................................................. 54
Figure 4.2. Proposed mechanism for crosslinking reaction for a
terpolymer containing TBASS, styrene, and 4-chlormethystyrene (CMS). ................................. 71
Figure 4.3. Dynamic mechanical analysis (DMA) of fully
hydrated PVDF-PSSA blends (M8) .............................................................................................. 73
Figure 4.4. Effect of heating duration on the crosslinking .......................................................... 74
Figure 4.5. Linear relationship of PVDF-PSSA membranes (Table 4.2)
between proton conductivity and water content ........................................................................... 75
Figure 4.6. Conductance normalized to the acid concentration
of membranes from Table 4.2. ...................................................................................................... 76
Figure 4.7. UV-Vis spectra of a 33.6 wt% PSSA blended membrane......................................... 77
Figure 4.8. Sample was immobilized onto a circular metal puck ................................................ 80
Figure 4.9. TEM image of 30.4 wt% PSSA-PVDF
membrane with 20 nm scalebar. ................................................................................................... 81
Figure 4.10. TEM image of 30.4 wt% PSSA-PVDF
membrane with 5 nm scalebar. ..................................................................................................... 82
Figure 4.11. TEM image of 10 wt% PSSA-PVDF
membrane with 20 nm scalebar. ................................................................................................... 83
Figure 4.12. EDS of the freeze-fractured, cross-section
of a 30.4 wt% PSSA-PVDF membrane sample. ........................................................................... 84
Figure 4.13. EDS of the glass-facing side of a
30.4 wt% PSSA-PVDF membrane. .............................................................................................. 84

xiii

Figure 4.14. EDS of the atmosphere-facing side of a
30.4 wt% PSSA-PVDF membrane. .............................................................................................. 85
Figure 4.15. ATR-FTIR of M1-M8 from Table 4.2. ................................................................... 86
Figure 4.16. ATR-FTIR of 30 wt% PSSA-PVDF membrane ..................................................... 86
Figure 4.17. XRD patterns of a series of 5 membranes ranging
from low to high proton conductivity. .......................................................................................... 87
Figure 4.18. DSC of PSSA-PVDF membranes containing small
amounts of PSSA in wt% .............................................................................................................. 89
Figure 4.19. Arrhenius plot of Proton Conductivity of
membranes from Table 4.2. .......................................................................................................... 89
Figure 5.1. Effect of a) temperature, and b) methanol concentration.
9
........................................ 97
Figure 5.2. Simplified diagram of the electrode-electrolyte
interface in a fuel cell.
120
............................................................................................................... 98
Figure 5.3. Schematic of the painting process.
109
...................................................................... 100
Figure 5.4. Schematic of the 5 different layers of a
membrane electrode assembly (MEA). ....................................................................................... 101
Figure 5.5. Schematic of the hardware components
of a single fuel cell.
125
................................................................................................................. 102
Figure 5.6. Polarization curve of MEA-5 at 30 °C with
ambient pressures of oxygen and different methanol concentration. ......................................... 108
Figure 5.7. Power density curve of MEA-5 at 30 °C with
ambient pressures of pure oxygen and different methanol concentration. ................................. 109
Figure 5.8. Power density curve of MEA-5 at 30 °C with
ambient pressures of air and different methanol concentration. ................................................. 109
Figure 5.9. Polarization curve of MEA-5 at 60 °C and
ambient pressures of oxygen and different methanol concentrations. ........................................ 110

xiv

Figure 5.10. Power density curve of MEA-5 at 60 °C and
ambient pressures of pure oxygen and different methanol concentration. ................................. 110
Figure 5.11. Power density curve of MEA-5 at 60 °C and
ambient pressures of air and different methanol concentration. ................................................. 111
Figure 5.12. Polarization curve of MEA-5 at 90 °C and
ambient pressures of pure oxygen and different methanol concentration. ................................. 111
Figure 5.13. Power density curve of MEA-5 at 90 °C and
ambient pressures of pure oxygen and different methanol concentration. ................................. 112
Figure 5.14. Polarization Curves of MEA-5 at 30 °C with
humidified air at ambient pressure. ............................................................................................. 112
Figure 5.15. Power Density Curves of MEA-5 at 30 °C with
humidified air at ambient pressure. ............................................................................................. 113
Figure 5.16. Repeatability of MEA-5 tested using 5.0 M
methanol at 30 and 60 °C. ........................................................................................................... 115
Figure 5.17. Effect of temperature of 0.5 M methanol DMFC
on power density for MEA-6 ...................................................................................................... 116
Figure 5.18. Effect of 1.0 M methanol on power density for MEA-6. ...................................... 117
Figure 5.19. Effect of 2.0 M methanol on power density for MEA-6 ....................................... 117
Figure 5.20. Effect of 3.0 M methanol on power density for MEA-6. ...................................... 118
Figure 5.21. Performance of MEA-6 at 60 °C at different methanol
concentrations ............................................................................................................................. 119
Figure 5.22. Constant current plot for MEA-6 with 1.0 M methanol ........................................ 119
Figure 5.23. Constant current plot for MEA-6 with 5.0 M methanol. ....................................... 120
Figure 5.24. Hydrogen fuel cell performance of MEA-6 at several
temperatures using oxygen at ambient pressures ........................................................................ 121

xv

Figure 5.25. Constant current plot for MEA-6 using hydrogen at 60 °C
and air at ambient pressures ........................................................................................................ 122
Figure 5.26. Performance of MEA-10 at 60 °C with different
concentrations of methanol ......................................................................................................... 124
Figure 5.27. Performance of MEA-10 at 80 °C with different
concentrations of methanol ......................................................................................................... 124
Figure 5.28. Constant current experiment at 150 mA/cm
2
of
MEA-10 at 60 °C with pure methanol ........................................................................................ 125
Figure 5.29. Constant Current experiment at 400 mA/cm
2
for
MEA-11 at 90 °C and 1.0 M methanol ....................................................................................... 127
Figure 5.30. Comparison of MEA-5 and Nafion-117 with 2.0 M
MeOH, 30 °C, and oxygen .......................................................................................................... 130
Figure 5.31. Comparison of MEA-5 and Nafion-117 with 3.0 M
MeOH, 30 °C, and oxygen .......................................................................................................... 130
Figure 5.32. Comparison of MEA-5 and Nafion-117 with 4.0 M
MeOH, 30 °C, and oxygen .......................................................................................................... 131
Figure 5.33. Comparison of MEA-5 and Nafion-117 with 5.0 M
MeOH, 30 °C, and oxygen .......................................................................................................... 131
Figure 5.34. Comparison of MEA-5, MEA-10, Nafion-117, and
Nafion-212 with 1.0 M MeOH, 60 °C, and oxygen .................................................................... 132
Figure 5.35. Comparison of MEA-5, MEA-10, Nafion-117, and
Nafion-212 with 2.0 M MeOH, 60 °C, and oxygen .................................................................... 133
Figure 5.36. Comparison of MEA-5, MEA-10, Nafion-117, and
Nafion-212 with 3.0 M MeOH, 60 °C, and oxygen .................................................................... 133
Figure 5.37. Comparison of MEA-5, MEA-10, Nafion-117, and
Nafion-212 with 4.0 M MeOH, 60 °C, and oxygen .................................................................... 134
Figure 5.38. Comparison of MEA-5, MEA-10, Nafion-117, and
Nafion-212 with 5.0 M MeOH, 60 °C, and oxygen .................................................................... 134

xvi

Figure 5.39. Comparison of MEA-5 and Nafion-117 with different
cathode flow rates of oxygen ...................................................................................................... 135
Figure 5.40. Comparison of MEA-5, Nafion-117, and Nafion-212
with 1.0 M MeOH, 90 °C, and oxygen ....................................................................................... 136
Figure 5.41. Comparison of MEA-5, Nafion-117, and Nafion-212
with 2.0 M MeOH, 90 °C, and oxygen ....................................................................................... 136
Figure 5.42. Comparison of MEA-5, Nafion-117, and Nafion-212
with 3.0 M MeOH, 90 °C, and oxygen ....................................................................................... 137
Figure 5.43. Comparison of MEA-10 and Nafion-212 with 10.0 M
MeOH, 60 °C, and oxygen .......................................................................................................... 137
Figure 5.44. Comparison of MEA-10 and Nafion-212 with 15.0 M
MeOH, 60 °C, and oxygen .......................................................................................................... 137
Figure 5.45. Comparison of MEA-10 and Nafion-212 with 24.8 M
MeOH, 60 °C, and oxygen .......................................................................................................... 138

xvii

Abstract
The issue of climate change is briefly discussed with respect to carbon dioxide. The
Methanol Economy concept is quickly introduced. A short explanation of energy efficient
electrochemical systems is presented. The direct methanol fuel cell (DMFC) system is explored.
Materials used in fuel cell systems are presented and discussed. Focus on proton exchange
membranes is established. The introductory chapter then foreshadows the contents of the
remaining four chapters. Each chapter has its own Supplemental Information section containing
Figures, Tables, and sometimes explanations. The bibliography for this work is included after the
final chapter at the end.
The second chapter of this work begins with a description of the tetrabutylammonium 4-
styrene sulfonate (TBASS) and its novel crystal structure and polymer properties. Strong dipole-
dipole interactions in the polymeric form are corroborated by the large interionic distances
revealed in its monomeric structure. The electronic dipole effects are doubly demonstrated by
both the partial rejection of the polymer from size-exclusion chromatography analysis with strong
light scattering response and by the evidence of a Trommsdorff-Norrish type of auto-acceleration
when polymerized in 1,2-dichloroethane.
Once the properties of the monomer and its respective homopolymer were established, the
following chapter details TBASS copolymers with styrene and methylmethacrylate (MMA),
copolymerized in a common organic solvent. For both copolymerization systems, the reactivity
ratios were estimated, and it appears that TBASS readily copolymerizes with both styrene and
MMA to form moderately alternating copolymers with good control over composition. The
TBASS monomer imparts drastic deviations in properties of polystyrene, even at very low mole
percent. The pyrolysis of polystyrene and polystyrene containing a few mole percent of TBASS,

xviii

reveal some of these dramatic differences in polymer properties. Results from size-exclusion
chromatography also corroborate the large differences.
With better understanding of the copolymer kinetics and copolymer properties of the
TBASS and styrene system, 4-chloromethylstyrene (CMS) was introduced into the copolymer
system. Optimizations of the terpolymerization procedure are discussed with previous
investigations and findings referred to. Blends of poly(vinylidene fluoride) and TBASS
copolymers containing CMS were fabricated by direct blending in a common solvent. To our
surprise, crosslinked composite films, mixed at the molecular level, were readily fabricated in
under an hour. The film fabrication process does not require the use of any additional crosslinking
agents or compatibilizers. It was discovered that the crosslinking step proceeds unimpeded by the
absence of conventional benzylation catalysts. In the absence of conventional transition metal
catalysts, the crosslinking step is both quantified and further explored using small molecular
analogues to remove the effects of polymer chains. A possible mechanism of benzylation in the
absence of catalyst is proposed. To convert the composite PTBASS-PVDF semi-interpenetrating
network (semi-IPN) films, into PSSA-PVDF semi-IPN membranes, an optimized ion-exchange
and dialysis process is presented and discussed. Mechanical investigations on the membranes were
done and results presented. Relevant properties are tabulated for the PSSA-PVDF semi-IPN and
commercial standard Nafion-117 PEMs. Morphological investigation reveals extremely small
domain sizes in the PSSA-PVDF PEMs. Proton conductivity and methanol crossover values of
PSSA-PVDF semi-IPN PEMs were compared to commercial standard Nafion-117. Calorimetry
experiments reveal that proton conduction in the PSSA-PVDF membranes is achieved entirely
through proton hopping, since there are no free water molecules to sustain any level of vehicular
transport. In addition to calorimetry, x-ray diffraction (XRD), and attenuated total reflectance-

xix

Fourier transform infrared spectroscopy (ATR-FTIR) measurements deduce that the presence of
the ionic sulfonate copolymer, in any amount, enhances the polar and piezoelectric polymorphs of
semi-crystalline PVDF and suppresses the non-polar polymorphs.

1

Chapter 1 Miniature Review
Abstract
A component of climate change focused on carbon dioxide is discussed. A carbon-neutral
economy appears to be ideal for phasing out the use of fossil fuels. Hydrogen is the ideal clean
burning fuel, but its use is made more practical by using hydrogen carriers, such as methanol. The
use of methanol, as a hydrogen carrier, for fuel combines high volumetric energy density while
being carbon neutral when it is used for electrical energy. The concept and thermodynamics of
fuel cells are introduced with special interest in direct methanol fuel cells being established.
Common membrane materials are briefly reviewed, and the contents of the remaining chapters is
foreshadowed.

Introduction
1.1 Anthropogenic Climate Change
Climate change referred to here, is the long-term change of weather patterns across all
regions of planet Earth. Scientists agree that the earth is warming far more than natural climate
fluctuations would predict and that human activity is the cause.
1
The French mathematician Joseph
Fourier, in the 1800s, made groundbreaking discoveries in understanding heat and was the first the
propose that carbon dioxide content of the Earth’s atmosphere acted to raise the Earth’s
temperature. Hence, it had been known well into the 19
th
century that the Earth’s atmosphere,
absorbs light and infrared energy from the sun. When molecules absorb radiation, the energy is
dissipated as heat energy, which heats up the atmosphere.

2

The discussion of carbon dioxide is significant in the sense that it is one of the main
byproducts of combustion. Even today, combustion processes continue to produce the majority of
our electrical power
2
and this is illustrated in Figure 1.2.
3

Figure. 1.1. History and projection of global energy consumption adapted from ref.
4

Figure 1.2. History and projection of global energy consumption by sources of energy
adapted from ref.
3

3

1.2 Electricity from Fossil Fuels
Although the burning of fossil fuels had made the industrial revolution possible, mankind
is now burning at an unsustainable rate resulting in an unsustainable rate of warming.
5,6
Yet
currently, the combustion of fossil fuels remains the primary source of energy world-wide.
2
In
this process, the energy stored in the carbon-hydrogen bonds of fossil fuels is converted into
automotive and or electrical power, a process that is used to this day.

Figure 1.3. Schematic of the conversion of kinetic to electrical energy

The stator in the figure above, contains a conductive material rotating in a magnetic field.
If conductive materials, such as metals (usually copper) experience a changing magnetic field, the
loosely bound electrons in the outer most valence shells of the metal structure will respond to the
changing magnetic field and will flow in a manner to produce an opposing magnetic field.
7

Rotation about a circle produces a positive and negative current in the conductive material that are
equal and cancel out to zero, however, with the use of diodes, that permit the electrons in a circuit
to flow solely in one direction, a full rectifier bridge can be arranged to convert the sinusoidal
current flow generated in the conductive material into direct current flow of electrons. In addition,

4

this circuit requires capacitors, that store charge, to reduce voltage fluctuations, thus providing a
stable power output. Fundamentally, the 1
st
law of thermodynamics dictates that energy cannot be
destroyed, it can only change forms. The 2
nd
law of thermodynamics states that the entropy of any
isolated system always increases. It is observed in the amount of energy which is lost as heat
energy when converting one form of energy into another. For example, in an electric turbine,
rotational energy is lost to friction and is dissipated as heat energy that cannot be recovered. This
unavoidable loss of energy when converting from one form of energy into another is a universal
penalty of energy. Therefore, combustion systems that generate electricity from fuel will
inherently never be as energy efficient as systems that directly convert the energy from chemical
bonds into electrical energy or vice versa. Such systems that facilitate the direct conversion of
chemical into electrical energy are known as electrochemical systems. Thus, progress needs to be
directed towards more energy efficient systems such as fuel cells and batteries.

1.3 Electrochemical Systems and Fuel Cells
Nowadays electrochemistry is used for many processes. Electrons can be used to split
water to generate pure hydrogen and pure oxygen. Given the problematic role of carbon dioxide,
hydrogen is a desirable carbon-less fuel that gives only water upon oxidation in air. In theory, this
means that hydrogen could be a perfect clean-burning fuel and does have a high, gravimetric,
energy density. However, producing hydrogen through electrolysis, is an energy intensive process
that is not very efficient. The electrolysis of water essentially results in the electrical energy being
stored in the single bond of H2 and the double bond of O2. Then the hydrogen and oxygen, can, in
theory and practice, be recombined to produce energy and water. The process of producing energy
and water through the combination of hydrogen and oxygen can be done via combustion however

5

this is very dangerous as hydrogen is very explosive. Another issue is that hydrogen at ambient
conditions, is a gas, and needs to be pressurized and condensed into the liquid state. Hydrogen
has very low volumetric energy density under ambient conditions. Molecules that have many
hydrogens, known as hydrogen-carriers, can be used in place of hydrogen to obtain better
volumetric energy density. Generally, hydrogen carriers containing carbon are ideal for this
purpose because they can store more hydrogen atoms in a smaller volume than hydrogen at the
same pressure and temperature. To mitigate climate change brought about by increasing CO 2
emissions, the hydrogen-carriers that only contain one carbon atom per molecule would ideally be
used as a fuel. Hydrogen-carriers that meet this criterium are typically limited to methane,
methanol, formaldehyde, formic acid, and any of their conjugate bases. By using these one-carbon,
hydrogen-carriers, in the production of energy, there would be a net zero CO2 emission since only
1 molecule of CO2 is produced for every 1 molecule of hydrogen carrier. These single-carbon
based fuels are also very clean-burning because they do not emit any oxides of nitrogen (NOx)
which is toxic to humans. The concept of fuel cells, is part of the basis of "The Methanol
Economy" advocated by Nobel Laureate George Olah.
8
Another big part of the methanol economy
is that, starting from methanol, there are a variety of well justified synthetic pathways to produce
most of, if not all, the plastic products that are currently made from fossil fuels at scale.
8
Methanol
is the most practical of these one-carbon options since it is a liquid at ambient conditions, and it
has the least oxidized carbon (other than methane which is a gas), thereby affording it the
maximum amount of hydrogen atoms to produce energy and water from. Electrochemical cells
that produce electrical energy from methanol and oxygen are known as direct methanol fuel cells
(DMFCs).
8

6

1.4 Chemistry of Direct Methanol Fuel Cells
The methanol fuel cell concept was actualized by joint efforts of NASA-JPL and USC.
9,10,11

The thermodynamic theoretical energy conversion efficiency of the DMFC is predicted to be
97%.
12

Figure 1.4. Balanced half-reactions of the DMFC system

Relative to the standard hydrogen electrode (SHE), the electrochemical potentials for the anode
and cathode are 0.03 and 1.23 volts (V), respectively.
13
Therefore, the theoretical maximum cell
potential is 1.20 V. In practice, due to a combination of sluggish kinetics at the anode or cathode,
and the cathode corrosion reaction generating a small counter-potential, the typical maximum cell
potential is in the range of 0.9-1.0 V. These losses are generally known as activation losses or the
activation overpotential.

Figure 1.5. Typical polarization curve for a PEM fuel cell stack.
14

7

1.5 Thermodynamics of Fuel Cells
Assuming that the energy of the reagents are fully converted into electrical work, the
enthalphic cell voltage, UH, would be expressed by the following equation.
15

𝑈 𝐻 = −
𝛥𝐻
𝑅 𝑧𝐹
(1)
Where ΔHR is the overall reaction enthalpy at standard conditions, z is the number of electrons,
and F is the Faraday constant (96484.6 C/mol).
15
Due to the 2
nd
law of thermodynamics, there will
be a deviation in energy due to loss of heat relating to entropy. Thus, the maximum amount of
electrical work is better represented using Gibbs free energy, ΔGr.
𝑈 𝑟𝑒𝑣 =
𝛥 𝐺 𝑟 𝑧𝐹
=
𝛥 𝐻 𝑟 −𝑇𝛥 𝑆 𝑟 𝑧𝐹
(2)
Where T is the absolute temperature of the system, ΔSr is the entropy contribution of the system at
standard conditions.

Table 1.1. Theoretical reversible cell potentials (E°rev) and maximum intrinsic efficiencies
for fuel cell reactions at STP adapted from Ref.
8

Fuel Reaction n -ΔH°

(kJ/mol)
-ΔG°
(kJ/mol)
E°rev
(V)
E (%)
Hydrogen H2 + 0.5O2 → H2O 2 286.0 237.3 1.229 83.0
Methane CH4 + 2O2 → CO2 + H2O 8 890.8 818.4 1.060 91.9
Methanol CH3OH + 1.5O2 → CO2 + 2H2O 6 726.6 702.5 1.214 96.7
Formic Acid HCOOH + 0.5O2 → CO2 + H2O 2 270.3 285.5 1.480 105.6

In Table 1.1, adapted from the methanol economy book, we see that methanol is the least
problematic of these fuels and is coupled with decent thermodynamics and excellent maximum
efficiency.
8
Methane has the same storage and volumetric energy density issues as hydrogen

8

because it is a gas. In addition to that, the C-H bond activation for methane is difficult and therefore
methane fuel cells do not possess high power density.
16

1.6 Materials for Fuel Cells
This section will cover the materials in active research and development that would be used
in a fuel cell. General considerations for each cell component will be briefly discussed in the
following sections with focus on the heart of the fuel cell, the membrane electrode assembly
(MEA).

Figure 1.6. Schematic of the DMFC system

In a direct methanol fuel cell (DMFC) the anode and cathode are separated by a membrane
Fuel cell systems can be conceptualized as a battery, in the sense that a fuel cell is a chemical
battery that would only require that the reagents at both anode and cathode be replenished to
“recharge” the cell. DMFCs based around water, due to hydrogen and oxygen being gases, the
requirements for a gas diffusion layer (GDL) are that it needs to be porous to allow mass-transport,

9

strong, and electrically conductive. Typically for DMFCs, the anode GDL requires wettability so
that the aqueous methanol solution can reach the catalyst layer. The cathode GDL needs to have
some degree of hydrophobicity to reject water from the GDL pores as it is being formed as a
byproduct. The requirements for the catalyst are that it needs to be highly active and highly
selective. The requirements for the membrane include a lack of electrical conductivity, high
permeation of molecular ions (such as H3O
+
or OH
-
), but low methanol permeability, and finally
needs to be chemically and mechanically durable. The issue of methanol permeability is not
simply a loss of fuel but the cathode catalyst, generally, is also highly active in the oxidation of
methanol, thus generating an electrical potential in the wrong direction by producing a counter
current of that greatly reduces the energy efficiency of the cell. Currently, there exist no
membranes with both high ionic conductivities (>150 mS/cm @ 60 °C) and a perfect barrier (0.00
cm
2
/s) to methanol crossover. However, for proton exchange membrane fuel cells (PEMFCs), the
perfluorinated sulfonic acid polymeric materials such as Nafion, made by DuPont, have remained
the most used. Due to the similar conducting ion, the proton, membranes used for PEMFCs have
also been used for DMFCs. The topics discussed in later chapters of the dissertation will mostly
focus on the membrane component of the fuel cell.

1.7 Membrane Electrode Assembly (MEA)
The heart of the fuel cell is the MEA, which is a 5-layer sandwich comprising, in order:
the anode gas diffusion layer (GDL), the anode catalyst layer, the membrane, the cathode catalyst
layer, and the cathode GDL. Typically, the sandwich is fabricated by pressing the layers together
with heat and pressure in a hot-press process. Work done by Chen et al, indicate that the hot-
pressing conditions strongly affect the performance of MEAs.
17
In general, the pressing force

10

should be in the range of 2-5 MPa with the temperature being around 135 °C.
17,18
There has been
a lot of work done addressing the catalysts used at both the anode and cathode. For methanol fuel
cells, the methanol oxidation reaction (MOR) occurs at the anode and the oxygen reduction
reaction (ORR) occurs at the cathode.
19,20
In recent years, research has gone into increasing the
catalytic activity and stability at the anode by increasing the electrochemically active surface area
(ECSA) by: reducing the size of the catalyst particles
21
, increasing the tolerance of the platinum
catalyst to carbon monoxide
22
, supporting the platinum particles on a high surface area carbon
support
23
, or a combination of these strategies. The types of membranes used for DMFC
applications can be separated into cation-exchange and anion-exchange membranes. Generally,
the membranes are like ones used for PEMFCs except that the properties are tuned to suit DMFC
applications which require low methanol crossover. Due to concerns over PFAS
24
, membranes
that do not contain or use PFASs in synthesis, have been more attractive for ion-exchange
membranes. Perfluorinated membrane materials, however, have very good chemical stability
coupled with very good conductivity properties. One commonly studied membrane material, that
is not Nafion or Nafion-based, is sulfonated poly(ether ether ketone) membranes and their blends
with inorganic or organic materials.
25,26,27,28
The PSSA-based membranes also have been
investigated for DMFC applications
29
, although the degradation of the polymer backbone, and
therefore cell performance, is rapid under operating cell conditions.
30
Solutions to the problem of
PSSA degradation have been addressed by grafting the PSSA chains onto a different polymer
which is more stable towards degradation.
31,32
Another strategy is to make block copolymers of
PSSA with other polymers.
33
The styrene sulfonate portion of the copolymers will naturally form
ionic domains when mixed with a hydrophobic component.
34
Blends of sulfonic acid based
polymer is the topic of this dissertation. Previous work done by Madsen are the most similar to

11

our system through the use of the tetrabutylammonium cation, although fuel cell data was never
acquired for their system.
35

1.8 Performance of Fuel Cells
There are a variety of different fuel cells that can be made for different applications. Below
summarizes important features of each type of fuel cell (Table 1.1).
15
Fuel cells, in general, can
be used for various applications where continuous power is required and can operate in a wide
range of temperatures and are theoretically more efficient than any system governed by
combustion processes.

Table 1.2. Different Fuel Cells currently in use and development.
15

AFC
(Alkaline)
PEMFC
(Polymer
Electrolyte
Membrane)
DMFC
(Direct
Methanol)
PAFC
(Phosphoric
Acid)
MCFC
(Molten
Carbonate)
SOFC
(Solid
Oxide)
Temperature
(°C)
<100 60-120 60-120 160-220 600-800 500-
800
Applications Transportation, Space, Military,
Energy Storage Systems
Combined
heat and
power for
decentralized
stationary
power systems
Combined heat and
power for stationary
decentralized systems
and for transportation
(trains, boats, etc.)
Realized
Power
Small
plants
5-150 kW
modular
Small
plants
5-250 kW
modular
Small
plants
5 kW
Small-
medium plants
50kW-11MW
Small
power
plants
100kw-
2MW
Small
power
plants
100-
250 kW
Charge
Carrier in
Electrolyte
OH
-
H
+
H
+
H
+
CO3
2-
O2
2-

12

Conclusions
The main cause of climate change is the increased and uncontrolled emission of carbon
dioxide and is a global issue that demands immediate mitigation steps. The concept of "The
Methanol Economy" presents a valuable and sustainable solution towards that end.
Electrochemical cells, namely fuel cells, are an inevitable part of any sustainable energy economy
simply due to the inherent benefits in energy efficiency. In the following chapters, new DMFCs
membranes are fabricated and evaluated. These result in cell performances that either rival or
surpass the performance of commercial Nafion membranes, under certain conditions. Chapter 2
is limited to the description of the synthesis, characterization, and homopolymerization of
tetrabutylammonium 4-styrene sulfonate (TBASS) vinyl monomer that has received only limited
attention in the context of fuel cell membranes. Chapter 3 covers topics associated with the
copolymerization of TBASS with other common vinyl monomers used in polymer synthesis and
research with potential for DMFC uses. Chapter 4 covers the blended composites of these
copolymers and poly(vinylidene fluoride) for membrane properties that are critical towards
DMFCs. The properties of these blended composite membranes are compared to the properties of
commercial standard Nafion membranes. Finally, chapter 5 will detail the cell performance results
of the DMFCs made using our composite membranes. Cell performance data in chapter 5 is also
presented for DMFCs made using Nafion membranes as a direct, practical comparison to our
membranes.

13

Chapter 2 Modified Synthesis of Tetrabutylammonium 4-Styrene
Sulfonate and its Homopolymerization
Abstract
An improved synthesis of tetrabutylammonium styrene sulfonate (TBASS) is described
and characterized by X-Ray and NMR spectroscopy. The synthesis gives TBASS as a crystalline
solid with a melting point of 83-84 °C. The homopolymerization of TBASS in 1,2-dichloroethane
and N,N-dimethylformamide show conclusive evidence of aggregation and a Trommsdorff-
Norrish type auto-acceleration of the polymerization.

Introduction
Polystyrene sulfonic acid (PSSA) is used in many applications. These include heterogenous
catalysts in organic reactions or resins for cation exchange,
36
crosslinked PSSA or PSSA
copolymers as proton exchange membranes,
37
and as PSSA composites in fuel cells based on the
oxidation of hydrocarbon fuels using air or oxygen.
29
The sodium salt of PSSA has found
applications as a superplasticizer in cements and as an ionic plasticizer in batteries.
38,39
The direct
sulfonation of polystyrene by reaction of PS with concentrated sulfuric acid or SO 3 is the oldest
method to yield sulfonated polystyrene.
40,41
However, the degree of sulfonation, the position of
the sulfonic acid moiety on the phenyl groups and the occurrence of side reaction such as formation
of phenyl sulfones are not readily controlled.
42

PSSA vinyl copolymers, completely or partially neutralized with tetrabutylammonium
(TBA) bases allows control of its properties.
35
In addition, the presence of the TBA cations in
copolymers containing sulfonate groups increases hydrophobicity and hence helps to mediate the
formation of poly(vinylbenzyl sulfonic acid)-PVDF blend composites.
35
Tetrabutylammonium

14

styrene sulfonates have also been used in the synthesis of semiconductors.
43
Furthermore,
polythiophenes containing blocks of fluorinated PSSA synthesized through click chemistry
coupling followed by functionalization by tetrabutylammonium hydroxide have been reported to
form micellar structures for the in situ synthesis of well-defined nanostructures of titania.
44

Alternatively, sodium 4-styrenesulfonate (SSANa) can be polymerized followed by ion
exchange with protic acids to yield PSSA allowing better control of the structure and degree of
sulfonation.
7
However, the sodium salt of 4-styrene sulfonic acid (SSA) does not co-dissolve with
some common vinyl monomers such as styrene or MMA in several common solvents and requires
emulsion polymerization to generate a latex product.
45
The solubility of TBASS in polar aprotic
solvents such as 1,2 dichloroethane (DCE) and N,N-dimethylformamide (DMF) allows the
synthesis of PSSA copolymers through copolymerization of TBASS and vinyl comonomers
followed by acidification.
46
This allows PSSA copolymers with properties for instance enhanced
hydrophobicity and crosslinking. Here we report an improved synthesis and purification of
tetrabutylammonium 4-styrene sulfonate (TBASS) that is shown to be a white crystalline solid
(mp = 83-84 °C). In addition, we report on the homopolymerization of TBASS and its molecular
weight characterizations. This indicated that TBASS, in principle can be readily copolymerized
with a large number of vinyl monomers that do not dissolve in aqueous media.

Experimental
2.1 Materials
Unless otherwise stated, materials were used as received. Dichloromethane (DCM),
anhydrous magnesium sulfate (MgSO4), 1,2-dichloroethane (DCE), N,N-dimethylformamide
(DMF), tetrahydrofuran (THF), diethyl ether, (DEE), and toluene were purchased from EMD

15

Millipore. Sodium chloride (NaCl) was purchased from BDH VWR Analytical. Styrene was
purchased from EMD Millipore and purified by passing it through a column of basic activated
aluminum oxide (Alfa Aesar; Brockman Grade-I). Azobisisobutyronitrile (AIBN) was purchased
from Sigma Aldrich and was recrystallized from methanol. Tetrabutylammonium bromide (TBA-
Br) was purchased from Chem-Impex. Sodium 4-styrene sulfonate (SSNa) and Whatman filter
paper and lithium bromide (LiBr) were purchased from Sigma Aldrich. De-ionized (DI) water
which had 18.2 MΩ of electrical resistance was acquired from a Milli-Q system from EMD
Millipore.

2.2 Synthesis of TBASS
Sodium 4-styrenesulfonate (NaSS; 5.0 g; 24.2 mmol) and tetrabutylammonium bromide
(TBABr; 7.8 g; 24.2 mmol) were dissolved into 100 mL of DI water. The solution was transferred
to a separatory funnel and extracted with three portions of 25 mL of dichloromethane (DCM) and
the fractions combined. The DCM solution was washed with three aliquots of 50 mL of D.I. water
and then dried over anhydrous MgSO4 for at least 2 hrs with stirring. After routine filtration, the
DCM solution was partially concentrated (rotovap at 90-95°C) until the solution became viscous.
While still warm, the flask containing this solution was cooled quickly by immersion into liquid
nitrogen. The solution froze quickly and shattered into shards from shaking and swirling. These
were quickly transferred into a beaker that was placed into a vacuum chamber and any remaining
DCM was removed by evaporation under partial vacuum at room temperature during which the
frozen solid melted and began to crystallize. After further evacuation in vacuo (1 hr) the fully
crystallized material was crushed followed by re-evacuation. This was repeated until constant
mass. In addition, no traces of DCM or any other impurities were observed in the proton NMR.

16

The TBASS monomer, prepared in this manner, was an off-white, pale, crystalline solid, with a
yield of 86%.

2.3 NMR Characterization of TBASS
For proton and carbon NMR spectra, 50 mg of TBASS was dissolved in NMR tubes
containing 0.8-1.0 mL of D2O or CDCl3. Proton and carbon NMR spectra

were acquired using
Varian Mercury 400 or Varian VNMRS 600, operating at 400 MHz (9.4 Tesla) and 600 MHz (14.1
Tesla), respectively. Proton and carbon NMR spectra are presented in the supplemental section
(Fig. S1 and S2). No proton NMR signals of DCM or other impurities in TBASS could be
detected.

2.4 Chemical Shifts for TBASS
Proton 400 MHz NMR (CDCl3 Chemical shifts): δ: 0.91 ppm (triplet, 7.3 Hz, 12H, butyl, CH3), 1.32
ppm (sextet, 7.4 Hz, 8H, butyl, CH2), 1.52 ppm (pentet, 8.3 Hz, 8H, butyl, CH2), 3.16 ppm, (triplet, 8.5
Hz, 8H, butyl, CH2), 5.21 ppm (dd, 10.9 and 0.9 Hz, 1H, vinyl, =CH2), 5.70 ppm (dd, 17.6 and 0.9 Hz,
1H, vinyl, =CH2), 6.66 ppm (dd, 17.6 and 10.9 Hz, 1H, vinyl, -CH=), 7.31 ppm (doublet, 8.2 Hz, 2H,
aromatic), 7.81 ppm (doublet, 8.3 Hz, 2H, aromatic).
Proton 400 MHz NMR (D2O Chemical shifts): δ: 0.94 ppm (triplet, 7.4 Hz, 12H, butyl, CH3), 1.35
ppm (sextet, 7.4 Hz, 8H, butyl, CH2), 1.61 ppm (pentet, 8.0 Hz, 8H, butyl, CH2), 3.14 ppm, (triplet,
8.5 Hz, 8H, butyl, CH2), 5.44 ppm (doublet, 10.9 Hz, 1H, vinyl, =CH2), 5.97 ppm (doublet, 17.7
Hz, 1H, vinyl, =CH2), 6.85 ppm (doublet of doublets, 17.7 and 10.9 Hz, 1H, vinyl, -CH=), 7.63
ppm (doublet, 8.0 Hz, 2H, aromatic), 7.78 ppm (doublet, 7.4 Hz, 2H, aromatic).

17

Carbon-13 125 MHz NMR (CDCl3 Chemical shifts): δ 13.62, 19.61, 23.87, 58.48, 114.27, 125.56,
126.37, 136.36, 138.01, 146.68 ppm. Carbon-13 125 MHz NMR (D2O Chemical shifts): δ: 12.76,
19.65, 23.03, 58.00, 116.55, 125.78, 126.48, 135.49, 140.31, 141.55 ppm.

2.5 Melting point determination
A small amount of TBASS sample was loaded into a melting point capillary tube that was
evacuated in the presence of P2O5 for at least 12 hrs or until constant weight. It was introduced
into the tube under ultra-pure argon gas and sealed. Melting points (83
0
-84
0
) were determined
using a Dynalon Digital Melting Point Device (Model DMP100) with a heating rate of ~0.2
ºC/min.

2.6 X-ray crystallography
Single-crystal X-ray diffraction data were collected on a Bruker SMART APEX DUO 3-
circle platform diffractometer, equipped with an APEX II CCD, using Mo Kα radiation
(TRIUMPH curved-crystal monochromator) from a fine-focus tube Cu Kα radiation (multi-layer-
optics monochromator IuS microsource). The diffractometer was equipped with an Oxford
Cryosystems Cryostream 700 apparatus for low-temperature data collection. The frames were
integrated using the SAINT algorithm to give the hkl files corrected for Lp/decay.
47
The absorption
correction was performed using the SADABS program.
48
The structures were solved by intrinsic
phasing and refined on F2 using the Bruker SHELXTL Software Package and ShelXle.
49
All non-
hydrogen atoms were refined anisotropically. ORTEP drawings were prepared using the Mercury
CSD software.
50

18

2.7 Homopolymerization
TBASS, 0.1 g (0.235 mmol) and 2 mL of DMF or DCE, were added into a 5 mL reaction
vessel and stirred until dissolution. The vessel was sealed with a rubber septum and then purged
by bubbling ultra-pure argon (99.999%) through the solution for 30 min (flow rate: 50 mL/min;
18-gauge needle) and then placed in an oil bath heated to 65°C. After thermal equilibration, 50 μL
of an AIBN solution in DMF (7.8 mg AIBN/mL DMF) was injected into the reaction; the initiator
solution was purged (in the same way as done with the polymerization solution) prior to injection.
The monomer to initiator mole ratio used here was 100 unless indicated otherwise. The
polymerization proceeded for at least 10 hrs and was quenched by removal from heat and exposure
to air. A small aliquot of the reaction mixture was taken for MW analysis. The remaining reaction
solution was concentrated under reduced pressure to 1 mL and then dripped slowly into a DEE
bath having at least 5 times the volume of the concentrated reaction mixture; in this case, 5 mL of
diethyl ether. The polymer was separated from the non-solvent bath by decanting. The polymer
precipitate was then placed in a vacuum-oven chamber evacuated at 23°C to remove residual
solvents. The precipitate was then removed from the chamber and crushed and re-evacuated until
constant mass. The homopolymer of TBASS is a hygroscopic white and brittle solid (97-99%
precipitate yield) and is stored in an air-tight bottle, with the lid slightly unscrewed, and placed in
a sealed chamber in the presence of P2O5.

2.8 Molecular Weight Determination
Molecular weights and polydispersities (PDI's)were determined using size exclusion
chromatography (SEC) (1260 Infinity II Multi-Detector SEC/SEC from Agilent) with DMF (containing
0.05 M LiBr) as the mobile phase and column (Styragel HR 4, 104 Å, 5 µm, 7.8 mm X 300 mm from

19

Waters) calibrated using polystyrene standards. The refractive index detector was used for molecular
weight characterization. Injection volume of 100 μL and flow rate of 1 mL/min was used for analysis.

2.9 Thermogravimetric Analysis (TGA)
Data was acquired on a TGA Q5000 instrument from TA Instruments. Heating rate was 10
°C/min under a nitrogen atmosphere. To prepare samples, 3-5 mg of polymer and a platinum pan
was used.

2.10 Homopolymerization Kinetics
The polymerization of sulfonate monomer, TBASS, was monitored from zero to low, to
high conversion via proton NMR with pre-acquisition delay. The delay between the first and
second acquisition was 3 minutes followed by 1-minute incremental increases to the delay between
subsequent acquisitions. To be clear, the time between the 2
nd
and 3
rd
acquisition is 4 minutes, the
time between the 3
rd
and 4
th
acquisition is 5 minutes, and so on until the monomer is nearly fully
consumed. The conversion is estimated by the ratio of the vinyl signals of the monomer to the
tetrabutylammonium signals of the cation of the same monomer. The butyl groups of the
tetrabutylammonium cations do not change in chemical shift or intensity.

Results and Discussion
2.11 Crystal Structure of TBASS
As described in the Experimental section, TBASS was synthesized by ionic metathesis and
obtained as hygroscopic crystalline solid (m.p. 83-84°C) and characterized by NMR and X-ray
diffraction. The crystal structure and sharp melting point, together, indicated that TBASS is not a

20

liquid, as reported recently.
51,52
The crystallographic data indicate that the TBA cation is more
closely associated with two of the three sulfonate oxygens, with interionic distances of 4.53, 4.87,
and 6.25 Å (Figure 1).

Figure 2.1. Counter-ion distance in TBASS crystal structure. Distances was measured for
the closest anion-cation pair; units are in Ångstroms. Hydrogens were not shown. Depth
cue was added.

The unequal association of the cation with the 3 sulfonate oxygens is consistent with the typical
arrangement of sulfonate anion and cation structures.
53,54
The asymmetric interionic distances are
not that unusual and appear to be consistent with Coulombs law given the inverse square
dependence on interionic distance. For instance, the crystallographic interionic distances for the
cesium methylsulfonate, are reported as 3.22, 3.35 and 5.14 Å.
53,54
The crystal structure of cesium

21

4-methylbenzenesulfonate appears to have been reported to form a distorted octahedral
coordination structure with Cs-O distances ranging between 3.017 and 3.344 Å (Figure 2).
55
The
crystal structure of creatininium benzenesulfonate was reported previously.
56
The positively
charged nitrogen and the sulfonate group were close enough to hydrogen bond.
56
Hydrogen bond
lengths in systems such as creatininium benzenesulfonate, should be in the range of 2 to 3 Å and
so this means that the tetrabutylammonium group is the main contributor towards the interionic
distance in the TBASS crystal structure.

Figure 2.2. Crystal structure of cesium 4-methylbenzenesulfonate.
55

Other structures of benzene sulfonate containing crystal structures such as potassium 4-
methyl benzene sulfonate were documented in the literature.
57
However, these structures
containing a metal cation all had anion-cation distances that were well under 4 Å.
57,58
Crystal
structures containing the tetraakylammonium structures appear to offer a better description of this
large distance between styrene sulfonate oxygens and the quaternary ammonium in TBASS.

22

Figure 2.3. Crystal structure of tetrapropylammonium hydrogen carbonate.
59

As seen above based on the scale and depth, the carbonate anion is between 4 to 5 Å
away from the positively charged nitrogen of the tetrapropylammonium cation.
59
Therefore, the
reason for this increased equilibrium distance between the quaternary ammonium cation and
negatively charged oxygens in TBASS is likely due to both the steric and entropic contributions
of the tetraalkylammonium group. The increased interionic distance increases the electrical
potential of the sulfonate oxygens of TBASS, especially the furthest removed oxygen (Figure
2.1).

23

2.12 TBASS monomer synthesis
There are several other known ways to prepare TBASS. For example, in situ through ionic
metathesis using styrene sulfonic acid and TBA-OH followed by removal of water via
evaporation.
60
Another involves the treating an aqueous solution of NaSS with a TBA ion
exchange resin giving an aqueous TBASS solution.
21
As described in the Experimental section of
this chapter, the solid TBASS monomer was prepared by ionic metathesis in water followed by
extraction using DCM. The extraction process inevitably brings some tetrabutylammonium
bromide into the organic layer, as indicated by proton NMR which reveals that a small fraction of
SSNa remains in the aqueous layer. To increase the purity of TBASS, its solution in DCM was
washed with multiple (3-6) aliquots of DI water although this procedure will, of course, decrease
the overall yield of the TBASS. In this work, because TBASS was extracted from water using
DCM, this organic layer was washed repeatedly with small aliquots of DI water giving sufficiently
pure TBASS. Careful removal of solvents led to crystallization, producing a white solid with a
sharp melting point (83-84 °C). Increasing the monomer purity led to an observable increase
polymer yield. If the TBA-Br contamination of TBASS was purposely left unaddressed, the
resulting yields were lower due to the contamination comprising more of the monomer than
intended (sometimes up to 15% by mass). The contamination of TBASS with excess TBA-Br,
surprisingly, did not affect the crystallization of the monomer. Therefore, monomer purity could
only be confirmed through proton NMR. We observed that samples of TBASS of up to 5 wt% of
TBA-Br contamination, would still readily crystallize but gave a lower m.p. (52-67 °C) as a range
of temperatures. However, when obtaining the proton NMR spectra and/or the melting point, it
was instantly obvious that there was TBA-Br contamination. In the proton NMR, the relative
signal integrations suggested an excess of butyl groups. In the melting point measurements, if the

24

contamination was present, the melting point dropped by more than 15 °C and was no longer a
sharp melting point. Typical melting point ranges for crystalline TBASS samples, contaminated
with TBA-Br, ranged from 52-67 °C. Careful preparation of the melting point capillary tube (to
avoid all moisture) did not result in any narrowing of the melting point range for the samples
contaminated with TBA-Br.
However, after washing the organic layer (that contained the TBA-Br contaminant as
excess reagent) with DI water, followed by full removal of moisture (via drying agent) and organic
solvent (via rotovap and then gradual evacuation in a sealed chamber) a sharp melting point of 83-
84 ºC was observed. This indicated a higher degree of purity compared with monomeric TBASS,
or SSTBA, described in previous reports, which does not mention TBASS as a solid.
46,51,61

Figure 2.4. Conversion of Entry 2.6.

2.13 Homopolymerization of TBASS
As indicated in the Experimental section, AIBN initiated radical polymerizations were
carried out in DCE or DMF at 65

°C with TBASS:AIBN molar ratios of about 100:1 (Table 2.2,
0
0.5
1
1.5
2
2.5
3
0 30 60 90 120 150 180 210 240 270 300
ln(M
0
/M)
Time (min)
Homopolymerization of 0.3 M TBASS

25

entries 2.1-2 through 2.8-2). The homopolymers were precipitated in diethyl ether (DEE). The
polymer yields were near quantitative (97-99%) homopolymer yields. When polymerizing
TBASS at concentrations of 0.15 M in DMF there appeared to be unexpected changes in the
PTBASS SEC analysis (Table 2.2, entries 2.7-16 through 2.9-16). When polymerizing TBASS at
concentrations of 0.30 M in DMF at about 80 % conversion the rates of polymerization increase
(Figure 2.5) consistent with a Trommsdorff-Norrish effect due to lower rates of radical
recombination or disproportionation.
62
At monomer concentrations above 0.75 M, an unusually
strong increase in viscosity was observed with reaction time consistent with the strong aggregation
indicated above. This combination is unusual for radical polymerizations and is likely caused by
the strong association due to the polyelectrolyte nature of the PTBASS.
63
As a result, the apparent
molecular weights are excluded from the column resulting in bimodal distributions and hence
partial exclusion from the column (Table 2.2). Due to exclusion from the column the PDI's were
found to be much lower than for typical radical polymerizations.

Table 2.1. TBASS homopolymers
Entry Polymerization
Solvent
TBASS
Concentration
(M)
TBASS:AIBN
Mole Ratio
Yield
(%)
a
Mn
(kDa)
b
Mw
(kDa)
b
PDI
b
2.1-2 1,2-DCE 3.0 100 99 3,000 3,200 1.1
2.2-2 1,2-DCE 1.5 100 98 800 870 1.1
2.3 1,2-DCE 0.75 100 97 n/a n/a n/a
2.4-2 1,2-DCE 0.15 100 95 340 400 1.2
2.5-16 DMF 0.75 100 96 210 360 1.7
2.6 DMF
b,c
0.30 100 95 n/a n/a n/a
2.7-16 DMF 0.15 100 n/a 18 34 1.9
2.8-2 DMF
d
0.15 100 n/a 1.4 2.8 2.0
2.9-16 DMF 0.15 300 n/a 140 180 1.3
2.10-16 DMF 0.75 500 n/a 580 800 1.4
2.11-16 DMF 0.15 1000 n/a 200 270 1.4
(a)Yield was determined by the mass of homopolymer precipitate. (b) Molecular weight data was determined using
SEC. Some samples were only made for SEC analysis therefore some samples were not precipitated and yield
information was not collected. (c) Polymerization was done in an NMR tube for kinetics. (d) Polymerization had a
CHCl 3 as a chain transfer agent.

26

It was clear that dipole-dipole interactions mediated the formation of polymer aggregates during
polymerization and were not fully reversible upon dilution, hence complicating SEC analysis. All
molecular weight data for this homopolymer of TBASS are highly suspect due to strong
aggregation, discussed later.

2.14 Evidence of Aggregation and Auto-acceleration
Due to high aggregation in all samples of TBASS homopolymer (Figure 2.6 is an example
of the least aggregated sample we obtained SEC data for), all MW data are highly suspicious and
should be taken with a grain of salt. The homopolymer made at 0.30 M TBASS to high monomer
conversion, using the same monomer to initiator ratio of 100, showed an increase in rate of
polymerization at high (above 80%) conversions (Figure 2.5). This was not entirely unexpected
given the increase in viscosity, possibly due to the Trommsdorff effect. This increase, given the
polymerization time scale, is not due to polymerization driven heating that would also increases
polymer rates. The current effect appears to be largely due to due to viscosity mediated decrease
in radical termination rate hence increasing the concentration of polymer radicals.
64

27

Figure 2.5. SEC Chromatogram plot of entry 2.7-16 in Table 2.2.

The homopolymers formed at monomer concentrations of 0.15 M TBASS appear to be less
aggregated and have a larger fraction of non-aggregated chains for which we could obtain MW
information on. The aggregation effect must be a function of both the degree of polymerization
and the extent of inter-polymer interactions that are likely due to multiple electrostatic dipole-
dipole interactions. The TBASS concentration during homopolymerization, however, directly
affects the degree of polymerization through a minor auto-acceleration effect (Figure 2.5) when
the polymerization progresses towards very high conversion (>90%).

Conclusions
We describe the optimized synthesis and the crystal structure of tetrabutylammonium 4-
styrene sulfonate (TBASS) and its radical polymerization in 1,2-DCE and DMF. The TBA cation
is positioned nearly equally to two oxygens (4.53 Å-4.87 Å) but much further (6.25 Å) from a
third. The apparent molecular weights and MW distributions of all TBASS homopolymers
contained evidence of strong polymer aggregation. When using 0.30 M TBASS or higher

28

concentrations, the polymerization rates were affected by an auto-acceleration effect. The
increased viscosity throughout the reaction increased the steady state polymer radical
concentrations and hence increased polymerization rates even more.

2.15 Supplemental Information

Figure S2.1.
1
H NMR of TBASS monomer in CDCl3 (400 MHz). Peak at 7.27 ppm is
CHCl3.

29

Figure S2.2.
1
H NMR of TBASS monomer in D2O (400 MHz). Peak at 4.79 ppm is H2O.

Figure S2.3.
13
C NMR of TBASS monomer in CDCl3 (400MHz).

30

Figure S2.4.
13
C NMR of TBASS monomer in D2O (400MHz).

31

Chapter 3 Vinyl Copolymers of Tetrabutylammonium 4-
Styrenesulfonate
Abstract
The radical copolymerization kinetics of tetrabutylammonium 4-styrene sulfonate
(TBASS) with, styrene (S) and methylmethacrylate (MMA) is reported. Despite significant vinyl
signal overlap, the copolymerization with styrene appears to produce a moderately alternating
random copolymer with reactivity ratios of 0.36 and 0.30 for TBASS and styrene, respectively,
using Fineman-Ross linearization methods. The reactivity ratios of TBASS and MMA were
estimated to be 0.67 and 0.30, respectively, by the same methods, giving nearly perfect random
copolymers. These copolymers can, in principle, be converted to form partially sulfonated PMMA
that, unlike PS-PSSA, cannot be synthesized by sulfonation of PMMA. Investigative pyrolysis
(isothermal TGA degradation under nitrogen) experiments on polystyrene copolymers containing
0.0, 1.0, and 6.4 mole % of TBASS, indicate that at 1.0 mole percent TBASS, the polystyrene
appears to be more resistant to thermal degradation, while at 5 mole percent, the thermal
degradation of the copolymer appears to be accelerated. This behavior was consistently observed
at multiple pyrolysis temperatures.

Introduction
The structure and properties of copolymers consisting of a hydrophilic and hydrophobic
monomers are of great interest, as these materials can access a wide range of ionic contents and
morphologies.
65,66
The properties of polystyrene can be modified through copolymerization and
or blending with other materials, polymeric or otherwise, for instance to increase its hydrophilic
character in a systematic way.
67
Typical copolymers of styrene include that with acrylonitrile,

32

alpha-substituted styrene, acrylamides, and acrylates or methacrylates such as MMA or
hexafluoroisopropylmethacrylate (HFPMA).
65-68
Some copolymers of polystyrene show desirable
changes in its properties such as increased or decreased glass transition temperatures and have
been studied for decades.
69-70
The copolymers of styrene, and MMA and other vinyl monomers
with TBASS are all soluble in very common aprotic solvents such as dichloromethane (DCM),
1,2-dichloroethane (DCE), DMF, DMSO, acetone and many other solvents. Copolymerization of
styrene and MMA with hydrophilic and/or ionic comonomers have been less studied, and when
they are they are usually done with emulsion techniques.
71
Possible comonomers with styrene
such as acrylamide
71
, vinylpyrrolidone
72
, and acrylic acid
73
have been investigated but many of
these require bi-phasic emulsion conditions, or copolymerization at increased pressures and
temperatures in reactor cells. The copolymerization kinetics of tetrabutylammonium 4-styrene
sulfonate (TBASS) with styrene and MMA in 1,2-DCE at relatively low temperatures (65 °C) and
at ambient pressures (1 atm), are described in this chapter. TBASS as a monomer is of special
interest as it is fully soluble in common organics, it is hydrophilic, and it is highly ionic due to the
frustrated ionic structure. For instance, TBASS can be used to produce polymeric sulfonic acid or
impart a specific amount of hydrophilicity to a more hydrophobic vinyl polymer.
The formation of polystyrene sulfonate copolymers through sulfonation of PS is well
known but results in unwanted reactions such as meta-sulfonation and/or formation of sulfone
linkages.
74
The glass-transition temperatures of polystyrenesulfonic acid (PSSA) has been
estimated to be 440 °C and cannot readily be determined directly due to thermal degradation.
75,76

The glass transition temperature (Tg) of sodium polystyrene sulfonate has been reported as
220°C.
77
However, according to more recent work by Balding et al., the endotherm found at 220
°C is likely due to degradation only found in commercial sodium polystyrenesulfonates and not a

33

thermal transition associated with backbone reptation.
75
Balding synthesized the homopolymer
starting from purified sodium 4-styrene sulfonate by reversible-deactivation radical
polymerization (RDRP) to compare the DSC behavior with commercial homopolymer samples.
75

The synthesis of copolymers of styrene containing small mole fractions of TBASS, could allow
the determination of Tg for polyTBASS and/or its PSSA conjugate acid. Most commercial sources
of polystyrene sulfonate are produced via the post-sulfonation method and contain sulfonation
levels between 85-99%.
75
The copolymerization kinetics of TBASS and styrene, as well as
TBASS and MMA, are presented here along with exploratory thermal degradation properties of
TBASS copolymers. The results indicate that extension of TBASS, or PSSA, to novel copolymers
is possible.

Experimental
3.1 Materials
Unless otherwise stated, materials were used as received. Styrene (S) and
methylmethacrylate (MMA) was purchased from EMD Millipore and passed through a column of
basic activated aluminum oxide purchased from Alfa Aesar (Brockman Grade I, 58 Ångstroms;
150 m
2
/g surface area) before use. Anhydrous dimethylformamide (DMF), and sulfuric acid
(H2SO4) were purchased from EMD Millipore. De-ionized (D.I.) water with 18.2 MΩ electrical
resistance was obtained from a Milli-Q system from EMD Millipore. Sodium chloride (NaCl) was
purchased from BDH VWR Analytical. Sodium hydroxide (NaOH) pellets were purchased from
Macron Fine Chemicals. Tetrabutylammonium 4-styrene sulfonate (TBASS) was synthesized in
our laboratory (see Experimental section in Chapter 2).

34

3.2 Copolymerization kinetics of TBASS and styrene
The kinetics of copolymerization of TBASS and styrene were studied in 1,2-
dichloroethane-d4 using VT-NMR at 65ºC with toluene as internal standard on a Varian VNMRS-
600 3-Channel NMR Spectrometer. Copolymer composition was confirmed using data at low
conversions (0-20%) and high conversions (65-80%). Single-acquisition spectra were acquired
for up to 1 hour of reaction time, except for the TBASS and MMA system for which spectral data
were acquired for only the first 18 minutes of reaction time.

3.3 Copolymerization of styrene with TBASS
For a typical PS-TBASS copolymerization 1.12 g styrene (95 mol%), 240 mg of TBASS
(5 mol%), 5 mL of DCE, and a PTFE magnetic stir bar were all added into a 50 mL round bottom
flask. The contents were stirred until dissolution. The flask was sealed with a rubber septum and
then subjected to at least 3 freeze-pump-thaw cycles and filled with ultra-pure argon and placed in
an oil bath heated to 65°C. After thermal equilibration, a purged 0.5 mL solution of DCE
containing 19.4 mg (1 mol% with respect to monomers) of dissolved AIBN was injected into the
flask. The polymerization proceeded for at least 12 hrs and was quenched by removing from heat
and exposure to air. The polymer solution was concentrated by rotovap and then precipitated by
slow addition of the polymerization solution into methanol at around 0°C. The precipitate was
dried in a convection oven at 90°C overnight and was partially crushed into a powder and stored
in an air-tight bottle, with the lid slightly unscrewed, and placed in a sealed chamber in the presence
of P2O5. Copolymers containing 1, 2, and 5 mol% TBASS were precipitated in cold methanol and
worked up as stated above. Precipitation of 10 mol% TBASS occurred in diethyl ether at 0°C
while precipitation of 20, 35, 50, 65, and 80 mol% TBASS occurred in THF at 0°C. All

35

copolymers were dried in vacuo in the presence of P2O5 at 90ºC. Solid precipitates were repeatedly
crushed and re-evacuated until constant mass. The color the copolymers ranged from the pure
white to a very faint yellow at higher TBASS contents. The copolymers containing ≥20 mol%
TBASS were brittle. Copolymer compositions were determined by proton NMR in the presence
of an internal standard. The moles of tetrabutylammonium groups directly correlated to the moles
of sulfonate groups due to the purity of the TBASS monomer (see chapter 2).

3.4 Copolymerization of TBASS and MMA
The copolymerization kinetics were carried out in DCE-d4 in the same manner as that of
TBASS and styrene involving equimolar monomers and was monitored in an NMR tube.
Molecular weight characterizations were not carried out in either case but, given the kinetics, it is
expected that the MW's should roughly be comparable to that of polystyrene itself. The copolymer
of TBASS and MMA (equimolar comonomers) used for SEC analysis was carried out in DCE at
identical conditions to the styrene and TBASS system.

3.5 Thermal degradation
Both thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were
used to study the TBASS-styrene copolymers. The glass transition temperatures of PS copolymers
containing TBASS, were determined by DSC between room temperature and a variety of
temperatures such as 220, 260, 300, 320, and 350 °C with a ramp rate of 10 °C/min under a nitrogen
atmosphere. Isothermal TGA was used to track how fast copolymers of TBASS-styrene
copolymers lost mass based on composition and temperature.

36

Results and Discussion
3.6 Composition of Copolymers
The TBASS-S copolymers were synthesized via conventional radical polymerization in
DCE at 65°C with initiator concentration of 1 mol% (with respect to total moles of monomer)

as
described in the Experimental section.

Figure 3.1. TBASS styrene copolymer-as a function of comonomer-composition at
complete conversions.

The copolymer precipitates collected from the non-solvent baths were characterized by
SEC in DMF with 0.05 M LiBr, using polystyrene standards for calibration (Figure S3.8).
Monomer compositions of TBASS and styrene copolymers were determined by proton NMR (in
the presence of an internal standard) by measuring the moles of butyl groups as the moles of
TBASS in the polymer.
y = 1.0261x
R² = 0.9974
0
10
20
30
40
50
60
70
80
90
100
0 10 20 30 40 50 60 70 80 90 100
mole % TBASS in copolymer
mole % TBASS in comonomer feed
Composition of Copolymer and Comonomer feed

37

Based on the composition of a series of styrene and TBASS copolymers and its comonomer
feed ratio, it appears that the reactivity ratios of both may be close to one another. To assess the
kinetics of the system at low conversions, an experiment where the disappearance of comonomers
was tracked in real-time was employed to estimate the reactivity ratios near the start of copolymer
formation. Copolymerization rates were monitored via proton NMR by comparing vinyl proton
signals of both monomers against an internal standard. The signals were deconvoluted due to
overlap. The deconvolution selects the peaks belonging to both monomers, then calculates two
new NMR spectra, each containing the peaks of only one monomer. From the two calculated
spectra, an estimated integration of each monomer can be measured.

Figure 3.2. Kinetics of equimolar copolymerization of styrene and TBASS in 1,2-
dichloroethane-d4 at 65 °C at a total monomer concentration of 0.75 M

3.7 NMR Spectral Measurements
The ratio of vinyl resonances of the comonomers, visually appeared to remain relatively
consistent as the copolymerization proceeded from low to moderate (~30 %) conversions (Figure
3.2). As seen in Figure S3.1, in the SI section, the NMR signals were difficult to integrate
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 10 20 30 40 50 60
ln(M
0
/M)
Time (min)
Comonomer Conversion
Styrene
TBASS

38

accurately due to virtually identical chemical shifts and hence severe signal overlap. Thus,
computational deconvolution was used to estimate integration values. However, copolymer
contents, based on the low conversion NMR spectra, closely correlate with comonomer content,
indicating that both monomers appear to react at approximately the same rate. Using Fineman-
Ross treatments indicate that in DCE, the TBASS (1) and styrene (2) reactivity ratios r1 and r2 are
0.36 and 0.30, respectively (Figure 3.2 and Figure S3.1).
78
The general procedure is described in
the SI section.

Table 3.1. Composition of 70-90% conversion TBASS-styrene copolymers in DCE at 65 °C
as a function of initial comonomer feed composition.
Entry
No.
TBASS
mole %
a

TBASS mol% in
copolymer
b

Copolymer Yield (%)
c
Mn
(kDa)
d
Mw
(kDa)
d

PDI
d

1 80 78.6 77.3 38.3 56.4 1.5
2 65 68.9 85.2 35.8 54.2 1.5
3 50 57.4 80.9 36.5 64.3 1.8
4 35 37.6 86.2 20.4 42.5 2.1
5 20 21.0 72.4 8.05 17.3 2.0
6 10 12.1 82.0 13.6 20.3 1.5
7 5 6.40 79.8 10.1 13.1 1.3
8 2 2.74 77.1 11.1 14.6 1.3
9 1 1.04 80.7 7.17 9.77 1.4
10
e
69.5 50 88.6 93.6 136 1.5
(a) Based on comonomer molar feed ratios. (b) TBASS copolymer molar contents. Determined using proton NMR of
the methyl resonances of the butyl groups with internal standard. (c) This was based on the precipitated fraction of the
copolymers (d) Apparent MW's based on SEC using polystyrene standards. SEC samples were 1 mg polymer/mL
DMF with 0.05 M LiBr. (e) This copolymer was made using TBASS and MMA, not styrene, with identical conditions.

Table 3.2. Molecular weight and dispersity of copolymers after acidification
Entry
no.
a
Copolymer
Composition
b
Mn
(kDa)
d
Mw
(kDa)
d
PDI
d
10 SSA:MMA(50:50)
c
38.5 118 3.0
6 SSA:S (10:90) 8.17 15.9 1.9
5 SSA:S (20:90) 9.14 17.8 2.0
4 SSA:S (35:65) 18.7 37.3 2.0
3 SSA:S (50:50) 27.5 56.6 2.1
2 SSA:S (65:35) 25.5 52.8 2.1
1 SSA:S (80:20) 31.0 55.4 1.8

39

(a) From chapter 3 Table 3.1. Entries 7-9 were insoluble in water and were not acidified. (b) The composition
written in the parentheses were the comonomer feed ratios used, not the measured composition of the copolymer
after synthesis. (c) This copolymer was only made once, its measured composition is 69.5 mol% SSA:S, determined
by NMR in the presence of an internal standard of certain amount. (d) Determined using SEC. Samples were
roughly 10 mg/mL.

Given the SEC results of Chapter 2, it should be stressed that the molecular weights and
PDI values are merely apparent values, due to intense aggregation arising from the strong dipole
from the TBASS units. Indeed, as seen in Table 3.1, the apparent MW's tend to decrease with
increasing styrene content. The increase in apparent MW estimations was due to an increase in
aggregation due to an increase of TBASS. The relatively low PDI values obtained at very low
TBASS fractions appear to be due to incomplete precipitation. At high TBASS fractions the
apparent MW's and PDI's are qualitatively like that reported in Chapter 2. As shown in Table 3.1,
a variety of copolymers of styrene and TBASS can be readily synthesized with good control over
composition. Results from chapter 2 indicate that degree of polymerization can, in principle, be
affected by the monomer concentration and amount of AIBN initiator. The relatively good control
of composition is rather significant since most syntheses of sulfonated polystyrene do not show
good control over composition or have unwanted side reactions. For example, for the emulsion
polymerization of sodium styrene sulfonate and styrene, the estimated reactivity ratios for NaSS(1)
and styrene(2), reactivity ratios r1 and r2 of 10 and 0.5 were reported respectively.
77
Thus, the
above reactivity ratios and composition data, indicate that the copolymerization of TBASS and
styrene, favor formation of true copolymers rather than separate homopolymers.
As indicated above, the SEC analysis of these styrene copolymers containing TBASS
contained evidence of aggregation as documented more fully in Chapter-2. At early retention
times, for samples containing 10 mol % TBASS or higher, we observed intense light scattering
with virtually no response from other detectors (Figure S3.3 and S3.4). This is highly indicative

40

of large, low-concentration, light-scattering aggregates of polymer chains. These aggregates do
not trigger more concentration dependent detectors, such as UV or RI. The aggregation of these
polystyrenes containing a few mole % of TBASS are likely due to dipole-dipole interactions
arising from the TBASS units along the polymer backbone. Despite the SEC mobile phase
containing added LiBr as an electrolyte, evidence of aggregation was still present in all samples
that contained TBASS in nearly any amount. In Figure S3.3, for the 1 mole % TBASS copolymer
of polystyrene, the light scattering signal at low retention times was small, however the signal was
present, nonetheless. This indicates to us, that all the molecular weight data and the
polydispersities in Table 3.1, and in the previous chapter, are only apparent values estimated from
the fraction of the polymer sample, which was not excluded from the column during analysis, due
to aggregation.
When the styrene and TBASS copolymers were acidified into corresponding PSSA and
styrene copolymers, the aggregation in the SEC was far less pronounced and more reasonable
MWs and PDIs were estimated (Table 3.2). These results indicate that the styrene copolymers
containing roughly 20 mol% TBASS (or higher) were able to be converted into the corresponding
PSSA copolymers. This was predicated based on solubility in DI water. The styrene copolymers
containing less than about 20 mol % were insoluble in DI water and therefore could not be
effectively converted into the corresponding PSSA copolymer of styrene.

3.8 Copolymerization kinetics TBASS and MMA
The copolymerization of methylmethacrylate (MMA) and TBASS, had virtually no
overlap of vinyl resonances, as seen in Figure 3.3, the non-integrated signals at 5.4 and 6.0 ppm

41

are for MMA. The conversions of the two comonomers could be easily plotted yielding a mostly
straight line due to low monomer conversion.

Figure 3.3. Proton NMR of TBASS and MMA in CDCl3. Unlabeled signals are for MMA.
Small signal at 7.27 is for CHCl3.

Figure 3.4. Conversion of TBASS and MMA as a function of time.

For TBASS (M1) and MMA (M2), reactivity ratios of 0.67 and 0.30 was also estimated
using the Finemann-Ross linearization method.
78
The conversions appear linear due to the low
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0 2 4 6 8 10 12 14 16 18
ln(M
0
/M)
Time (min)
Copolymerization of TBASS and MMA
TBASS
MMA

42

conversion (~15%). In contrast to the TBASS and styrene system, the conversions of TBASS and
MMA from the raw data were linear enough that we did not artificially represent the conversions
as a perfectly straight line (explained in the SI section). This was only done for the TBASS and
styrene system due to the severe overlap of their vinyl resonances. The TBASS and MMA system,
as seen in Figure 3.3, did not have any signal overlap and did not require any deconvolution of
signals. Hence, the copolymerization of TBASS and MMA proceeds giving with a copolymer that
will have a greater fraction of TBASS than in the comonomer feed as time progresses. Both
copolymerization systems appear to slightly favor formation of alternating copolymers.
3.9 Thermal Properties of PS-Co-PTBASS
In addition to degradation studies, measurements of copolymers of styrene containing 0-
12 mole % of TBASS, were carried out to measure the onset temperature for the endotherms which
were plotted as a function of mol % TBASS in the polystyrene. This generated an approximately
linear relationship which we used to extrapolate to 100 mol% TBASS as seen in Figure 3.6, below.
At 100 mole %, PTBASS is predicted to have a Tg of about 500 °C. This prediction is slightly
higher than the sodium form of 4-styrenesulfonate polymer which was predicted to be around 440
°C.
75
The extrapolated Tg of PTBASS is also slightly higher than 440 °C which was estimated for
polystyrene sulfonic acid using partially sulfonated polystyrenes.
76
The work done by Wallace in
1971, appear to reveal a linear relationship (Figure 3.5).
76
For sulfonation contents above 20 mole
%, it is only theorized that the behavior remains linear.

43

Figure 3.5. Dependence of glass transition temperature of partially sulfonated polystyrene
on composition.
76

Figure 3.6. Extrapolation plot of glass transition temperatures of polystyrene copolymers
containing small amounts of TBASS.

y = 4.0366x + 98.813
R² = 0.9816
0
25
50
75
100
125
150
175
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Glass Transition
Temperature (°C)
mol % TBASS

44

Figure 3.7. TGA curves of polystyrene, PTBASS, and PS-co-TBASS (95:5 composition).

Figure 3.8. Isothermal TGA at 320 °C.

The TGA of the styrene-TBASS copolymers containing small amounts (5 mol%) of
TBASS were interesting when compared to the TGAs of polystyrene (Figure 3.7).
79
Both the
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500
Weight %
Temperature ( °C)
PS 95:5 poly(S-co-TBASS) polyTBASS
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500
Mass % remaining
Time (min)
Isothermal TGA at 320°C
PS PS with 1 mol% TBASS PS with 5 mol% TBASS

45

TBASS homopolymer and the copolymer containing 5 mole percent TBASS had mass loss at
temperatures far below that of polystyrene synthesized under identical conditions starting below
40 °C for PTBASS and around 200 °C for the styrene copolymer that is 95:5 styrene to TBASS
molar ratio (Figure 3.7). The mass of the homopolymer TBASS in Figure 3.7 at low temperatures
was likely caused by moisture that the polymer had absorbed from the atmosphere that did not
leave the system until around 100 °C. The PS synthesized by us for these thermal studies, had an
apparent Mw of 71 kDa and an Mn of 44 kDa. It appeared that for the 5 mole percent TBASS
copolymer, even temperatures as low as 220 °C (Figure S3.5) were sufficient to bring the mass
loss towards the linear range, starting at around 90 mins, where most of the mass is lost likely due
to degradation of the polystyrene backbone. It is plausible that the decomposition of TBASS
occurs in two stages: desulfonation and depolymerization. That could account for the patterns we
see. The isothermal data at 320 °C indicated that in a little over 2 hrs, much of the polystyrene in
the 95:5 copolymer had degraded into monomeric styrene, and other small-molecule byproducts,
and volatilized (Figure 3.8). The ceiling temperature is, defined here, as the temperature where
the polymer backbone begins to decompose to form monomers, occurring in the bulk polymer
material, not in solution. The ceiling temperature of commercial polystyrene (polymerized using
bulk polymerization) is around 395 °C when heated under a nitrogen atmosphere in a flow system
where the gaseous products are removed by the flow of nitrogen.
79
In some pyrolysis studies,
polystyrene does not appreciably lose mass until the temperature is within 50 °C of the ceiling
temperature.
79,80
To be clear, the ceiling temperature noted here is different than in conventional
polymer science where the concentrations of monomers is around 1 molar. The ceiling
temperature here is specifically used for pyrolysis where there virtually is no monomer

46

concentration to speak of since it is volatilized and flowed away from the system during thermal
degradation.

Figure 3.9. Isothermal TGA at 350 °C of PS-co-TBASS (99:1 monomer feed ratio). PS
refers to polystyrene synthesized under identical conditions to the copolymers.

The sample named “PS” was synthesized in our lab with identical conditions to the TBASS
homopolymerization, as well as the styrene copolymers. Industrially, most polystyrene is
produced through bulk polymerization.
81
Despite this, it is still useful to compare the degradation
behavior of the styrene homopolymer, made in this research with the same end-groups, with the
poly(S-coTBASS) polymers of 1 and 5 mole % TBASS. At 220 °C, polystyrene hardly degrades
at all over 6 hrs, as seen in Figure S3.6. While at 220 °C, the 95:5 PS-co-TBASS sample loses
more than 15 wt% in 3 hrs, possibly losing another 15 wt% if an additional 3 hrs of heating was
done. It was seen from Figure 3.8, that increasing the molar content of TBASS in the copolymer
from 1 to 5 mole percent, the 5 mol% TBASS copolymer lost nearly all of its mass within 2 hrs at
320 °C. On the other hand, at 320 °C, the 1 mole % of TBASS copolymer appears to reduce the
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500
Mass % remaining
Time (seconds)
Isothermal TGA at 350°C
PS PS with 1 mol% TBASS

47

rate of thermal degradation significantly in that it appears to be more thermally stable than PS,
synthesized without any TBASS. Theories of precisely what causes this thermal degradation
behavior is outside the scope of this report, however, it is likely that dipole-dipole interactions play
a major role.

Conclusions
The ionic monomer TBASS in copolymerization systems with both styrene and MMA was
investigated. Reactivity ratios were estimated using linearization methods. A variety of
polystyrenes with varying levels of sulfonation were synthesized and characterized by SEC
analysis. All copolymers of styrene, containing any amount of TBASS, had evidence of
aggregation in the SEC chromatograms; MW estimations are highly suspect. Glass transition
temperatures was acquired for some of these polystyrene copolymers using DSC, in attempts to
extrapolate towards 100 mole percent TBASS. Reactivity ratios indicate that TBASS can be
readily copolymerized with at least two significant comonomers, styrene and MMA, with good
control over composition. This could allow one to impart a controlled amount of ionic character
to copolymers made using TBASS with a slightly alternating comonomer sequence along the
polymer backbone. Exploratory pyrolysis data was acquired for polystyrene and polystyrene
containing a few mole % of TBASS using isothermal TGA. Initial data indicate that the presence
of TBASS can drastically affect the thermal stability of the polymer.

48

3.10 Supplemental Information

Figure S3.1. NMR spectrum of the TBASS-S copolymerization at low conversion. Only the
vinyl signals are shown. Relative signal intensities between comonomers at higher
conversions were visually identical.

Figure S3.2. Conversion of 1.0 M styrene as a function of time in DCE at 65 °C.

0
0.05
0.1
0.15
0.2
0.25
0 15 30 45 60 75 90
ln(M
0
/M)
Time (min)
Styrene Homopolymerization

49

Figure S3.3. SEC chromatogram plot of polystyrene copolymer containing 1 mol %
TBASS
The green trace is the light scattering at 90° and the red trace is the refractive index

Figure S3.4. SEC chromatogram plot of polystyrene copolymer containing 10 mol %
TBASS

50

Figure S3.5. Isothermal TGA at 220 °C of PS-co-TBASS (99:1 monomer feed ratio). PS
synthetic refers to polystyrene synthesized under identical conditions to the copolymers.

Figure S3.6. Isothermal TGA at 300 °C of PS-co-TBASS (99:1 monomer feed ratio). PS
refers to polystyrene synthesized under identical conditions to the copolymers.

0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500
Mass % remaining
Time (min)
Isothermal TGA at 220°C
PS PS with 1 mol% TBASS PS with 5 mol% TBASS
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600
Mass % remaining
Time (min)
Isothermal TGA at 300°C
PS PS with 1 mol% TBASS

51

Figure S3.7. Isothermal TGA at 260 °C of PS-co-TBASS (95:5 composition)

Figure S3.8. SEC Calibration plot of PS standards in DMF with 0.05 M LiBr

Table S3.1. Polystyrene standards
Mn (kDa) Mw (kDa) PDI
101 103 1.02
14.9 15.2 1.02
2.03 2.04 1.00
0.31 0.32 1.03

60
65
70
75
80
85
90
95
100
0
50
100
150
200
250
300
0 30 60 90 120 150 180
Weight (%)
Temperature(°C )
Time (mins)
Temperature
Weight %

52

Figure S3.9. Fineman-Ross plot of G vs F.

Estimation of Reactivity Ratios for TBASS and Styrene.
For estimating the reactivity ratios for the TBASS and styrene system, the first 20 mins of
conversion from Figure 3.2, is approximated as a straight line due to low comonomer conversion.
Then, the monomer concentrations of the first 10 data points were then extracted out of the
linearized conversions of each comonomer by computing the inverse natural log of the value. The
monomer concentrations are then computed into variables X and Y.
78
The variable X is the ratio
of M1 and M2 for the first 10 data points.
78
The variable Y is the ratio of the change in monomer
concentration for both M1 and M2 between the current point of reaction time and the initial
monomer concentration at the beginning of the reaction.
78
Then, variables G and F are calculated.
The variable G is equal to X(Y-1)/Y and the variable F is equal to X
2
/Y.
78
Finally, F is plotted vs
G to yield a straight line where the slope is equal to r1 and the intercept is equal to -r2.
78

y = 0.355x - 0.3012
0.038
0.0382
0.0384
0.0386
0.0388
0.039
0.0392
0.0394
0.0396
0.0398
0.955 0.956 0.957 0.958 0.959 0.96 0.961
G vs F

53

Chapter 4 Blended Membrane Composites of TBASS Copolymers
and Polyvinylidene Fluoride
Abstract
Crosslinked poly 4-styrenesulfonic acid (PSSA) blended membranes were made with
poly(vinylidene fluoride) (PVDF) using a poly(4-styrenesulfonate) precursor polymer. The
precursor to the crosslinked PSSA was a PTBASS copolymer containing CMS, described in the
previous chapter. Crosslinked composite films were readily prepared by co-dissolution of PVDF
and uncrosslinked PTBASS copolymer in a common solvent, N,N-dimethylformamide (DMF),
followed by solvent casting at 165 °C. Membranes were processed from the crosslinked composite
films via an optimized ion-exchange bath to convert the crosslinked PTBASS into an equally
crosslinked PSSA. The ion-exchange capacity, water content, and proton conductivities were
measured for a series of PSSA-PVDF membranes and commercial standard Nafion-117. The
PSSA-PVDF membranes were also characterized by TGA, DSC, DMA, AFM, UV-Vis, ATR-
FTIR, XRD, SEM, EDS, and TEM. Results of DSC indicate that the PSSA-PVDF membranes,
when fully hydrated, contains water which does not freeze in the temperature range of -15 to 0 °C.
Microscopy results indicate that the domain sizes are extremely small. Attempts were made to
quantify the crosslink density in PSSA-PVDF membranes and to elucidate a mechanism of
crosslinking in the absence of transition metal catalysts. Activation energies of proton
conductivity of Nafion-117 and some PSSA-PVDF membranes are comparable.

54

Introduction
The most versatile and commonly used membrane for cation-exchange fuel cells is
"Nafion" invented by W. Grot in the 1960s as a separator for chlor-alkali cells and was later
commercialized by DuPont.
82
Since then it has been used in many applications ranging from
electrochemical devices to catalysis.
83,84
For fuel cells, namely (DMFC), Nafion membranes have
been and continue to be used in hydrogen- and the direct methanol fuel cells. Disadvantages
include high ruthenium catalyst content as well as methanol crossover from anode to cathode
(MCO).
85
In addition its manufacture requires the use of perfluorinated alkyl substances (PFAS).
24

Figure 4.1. Chemical structure of Nafion®

Considerable research has been done to fabricate new proton conducting membranes that
do not contain or use PFAS molecules.
26,27,29,82,86
Although some of these have been claimed to
out-perform Nafion in one or more properties, Nafion membranes and their production have not
yet been phased out.
Herein, we describe the synthesis and characterization of proton conducting, low methanol
crossover, low water content, PSSA-PVDF blended membranes with suitable mechanical

55

properties required for Nafion based DMFCs. The membranes were prepared by the physical
blending of PVDF with PTBASS copolymer containing CMS in a common solvent, casting,
followed by protonolysis and processing in a single step giving an unprecedented dispersion of
PSSA into PVDF. These PSSA-PVDF semi-IPNs have been shown to have membrane properties
that are of potential interest in DMFC and other electrochemical applications.

Experimental
4.1 Materials
Unless otherwise stated, materials were used as received. Poly(vinylidene fluoride)
powder (Mw ~530 kDa) was obtained from commercial sources Alfa Aesar and Sigma Aldrich.
Styrene (S) and 4-chloromethylstyrene (CMS) were purchased from EMD Millipore and Sigma
Aldrich, respectively, and were each passed through a column of basic activated aluminum oxide
purchased from Alfa Aesar (Brockman Grade I, 58 Å, 150 m
2
/g surface area) to remove inhibitors
prior to use. Anhydrous dimethylformamide (DMF), sulfuric acid (H2SO4), lead nitrate
(Pb(NO3)2), hydrogen peroxide (H2O2) and iron sulfate (FeSO4) were purchased from EMD
Millipore. De-ionized (D.I.) water with 18.2 MΩ electrical resistance was obtained from a Milli-
Q system from EMD Millipore. Sodium chloride (NaCl) was purchased from BDH VWR
Analytical. Sodium hydroxide (NaOH) pellets were purchased from Macron Fine Chemicals.
Tetrabutylammonium styrene sulfonate was synthesized in our lab according to experimental
procedures described elsewhere (see Chapter 2). Fenton’s reagent (aqueous 3% H2O2 and 2ppm
FeSO4 solution) was prepared in our lab immediately before use. Detergents Alconox and sodium
dodecylbenzenesulfonate (NaDDBS) was acquired from VWR vendor on USC campus and Sigma
Aldrich, respectively.

56

4.2 Synthesis of PTBASS Copolymers
The copolymerization, of 15 g of TBASS (35.3 mmol), 0.672 g of CMS (4.40 mmol, 0.459
g of S (4.41 mmol), 45 mL of DCE, and teflon magnetic stir bar were added into a 250 mL round
bottom flask. The contents were stirred until dissolution. The flask was sealed with a rubber
septum and then subjected to at least 3 freeze-pump-thaw cycles and filled with ultra-pure argon
and placed in an oil bath heated at 65°C. After thermal equilibration, an argon-purged 1 mL
solution of DCE containing 7.2 mg (1 mol% of total moles) of AIBN was injected into the flask.
The polymerization proceeded for at least 12 hrs and was quenched by removing from heat,
opening to air and dilution with 40 mL of DCM. The contents of the flask were mixed until it was
homogenous and then transferred into an addition funnel. The terpolymer solution was added to
a DEE bath of at least 5 times its volume to ensure complete precipitation, redissolved in 10 mL
of DCM and reprecipitated once more in excess DEE. The terpolymer was then evacuated crushed
and re-evaporated until constant mass. The terpolymer was a pale, cream-colored, off-white,
hygroscopic, brittle solid (97-99% yield). It was stored under ultra-pure argon in the presence of
P2O5. The synthesis of a TBASS-CMS copolymer with the same content (80 mol%) of TBASS as
the copolymer, was carried out using the same procedure and polymer product was precipitated
and dried in an identical manner, visually appeared the same, giving similar yields.

4.3 Composite Synthesis
A PVDF-PTBASS composite calculated to contain 24 wt percent PSSA after hydrolysis,
requires 0.5 g of TBASS-S-CMS (or TBASS-CMS) and 0.5 g of PVDF powder to be dissolved in
5 mL of DMF. Generally, the contents and a stir bar were all added to a sufficiently large vessel,

57

usually a 20 mL sample vial. The mixture was stirred at, or left at, room temperature until the
copolymer dissolved, then at 40-50 °C until the PVDF was fully dissolved forming an optically
clear solution. When using the PTBASS copolymer that did not contain styrene, the solution
would never become fully optically clear at room temperature after all solids had dissolved; it is
unknown what causes this. The DMF solutions were then poured into Petri dishes and placed into
a Yamato Convection Oven, 3.2 ft
3
(Model # DKN-402C) pre-heated to 165 ºC. The convection
was turned off as this was shown to give rise to non-homogeneous film surface. After 1 to 5 hrs,
the samples were removed, and 20 mL of room temperature DI water was immediately poured into
the dish to submerge the film. After several hours, the composite samples would swell and separate
from the glass substrate and could be peeled out. The films were optically transparent and
colorless, unless heated for 5 hrs or longer during fabrication where they would gain a yellow tint.
For uniform thickness films, a 10-20 wt % blend solution (meaning total mass of PVDF and
PTBASS copolymer collectively would comprise 10-20 wt%, and the remaining 80-90 wt% would
be DMF, or other casting solvent) on a flat glass plate gave best results. For instance, PVDF and
TBASS copolymer weight compositions of 50% PVDF and 50% TBASS copolymer by mass,
would require the total mass of both polymers to comprise no less than 10 wt% when it is dissolved
in DMF. If the total mass of both polymers in the casting solvent was less than 10 wt%, the wet
blend solution would spill over the sides of the flat glass plate when transferring to oven for
heating.

4.4 Ion Exchange Processing of Composites into Membranes
In order to convert the crosslinked PTBASS into crosslinked PSSA, a one liter bath of 1.0
M H2SO4, containing a 1 wt% "Alconox" detergent or equivalent surfactant was apparently

58

required, since incomplete TBA
+
-H
+
exchange had been observed in the absence of detergent.
61,46

The composite films were submerged in an ion-exchange bath of 30 g of sodium dodecyl
benzenesulfonate (NaDDBS) dissolved in one liter of 1.0 M H2SO4 (or equivalent Alconox bath
with 15 g of detergent) followed by heating to 70-90 ºC at least 12 hours. No magnetic stirring
was used as this was shown to damage the typically thin composite membranes. These were then
removed from the ion-exchange bath and dialyzed in 1.0 liter DI water baths at 60-80 ºC, replacing
the DI water until the pH of the aqueous solution was neutral and detergent-free i.e. below the
critical micelle concentration. Finally, the PSSA-PVDF membranes were placed in a fresh DI
water bath inside a Ziploc bag and kept in a hydrated state. Later, the ion-exchange process and
dialysis against DI water was optimized to use sealed sample vials instead of a large bath in a
beaker to decrease the total amount of water wasted from each bath.

4.5 Water Uptake Measurements
After processing sample into membrane form, the membrane was dried in an oven set to
100 °C for at least 10 hrs, usually overnight, and then was immediately weighed (Wdry) upon
removal from oven. The immediacy was because these membranes are hygroscopic in the
anhydrous form. The membrane was then submerged in water overnight at room temperature,
excess water droplets were wiped off the surface of the sample and was weighed (Wwet)
immediately after. Water uptake %, or water content %, of a membrane sample is calculated using:
𝑊𝑎𝑡𝑒𝑟 𝑈𝑝𝑡𝑎𝑘𝑒 =
𝑊 𝑤𝑒𝑡 −
𝑊 𝑑𝑟𝑦 𝑊 𝑑𝑟𝑦 ∗ 100%. (1)
where Wwet is the mass of the membrane when fully hydrated and Wdry is the mass of the membrane
when dried at 100 ºC overnight. Drying the membrane further, by using reduced pressures, heating

59

at 110 ºC, and in the presence of P2O5 as a drying agent, were shown to only lose an additional
0.1% mass in the form of water.

4.6 Ion-Exchange Capacity (IEC)
The membranes were fully hydrated by submersion in DI water for several hours and then
rolled up and placed into vials. A saturated NaCl aqueous solution (using 18.2 MΩ de-ionized
water and excess NaCl the supernatant was used) was poured into the vial to completely submerge
the membrane, the vial was sealed, and contents were left to equilibrate at room temperature for at
least 24 hours. The salty, acidic supernatant was then titrated against a standardized solution of
NaOH until neutral using phenolphthalein as the pH indicator. The IEC is then calculated using:

𝐼𝐸𝐶 =
𝐶 𝑁𝑎𝑂𝐻 𝑉 𝑁𝑎𝑂𝐻 𝑊 𝑑𝑟𝑦 (2)
where CNaOH and VNaOH are the concentrations and volumes respectively and Wdry were weights
when dried at 100ºC. The hydration number (λ = [H2O/[H
+
]) can be calculated by dividing water
uptake (as a w/w fraction) by the IEC (in mol/g) then converting the mass of water into moles.

4.7 Conductance Measurements
In-plane membrane conductivities were measured using a Solatron® 1260 Impedance/Gain
Analyzer (Schribner Associates), a BT-110 conductivity clamp (Schribner Associates), and the
ZView and ZPlot software. A small strip (0.8 cm width x 3-4 cm length) was cut out using a metal
rectangular tool bit blank (0.8 cm x 0.8 cm x 4 cm) and a razor blade so that all cut strips would
have the same width of 0.8 cm. The strips were then boiled in deionized water for 10 minutes to
remove surface contamination and then stored in sample vials filled with de-ionized water. For
impedance measurement, the membrane strip was removed from its sample vial and clamped into

60

the conductivity clamp. The conductivity clamp (4-probe electrochemical impedance
spectroscopy) with membrane strip equipped were placed in a beaker filled with enough de-ionized
water to submerge the membrane strip with Pt electrode wires protruding out of the beaker and
connected via alligator clips to the potentiostat instrument. The frequency was scanned between
1,000,000 Hz to 10 Hz with 5 mVAC applied to the counter and working electrodes (outer wires)
using "ZPlot" and "ZView" software. The high frequency intercept at the real axis was taken as
the in-plane resistance of the membrane sample (a quarter-circle fit from a comparable circuit
diagram was also performed with the 2-probed method, but it gave the same value of resistance
each time as the 4-probed method),
𝜎 =
𝐿 𝑅 ∙𝐴 (3)
where L, R, and A represents the distance between the two inner Pt electrode wires, the measured
impedance, and cross-sectional area of the membrane (0.8 cm width times thickness of membrane
area between the electrode wires), respectively.

4.8 Methanol Permeability
The diffusion of methanol across membranes designed for DMFC applications is an
important property and is determined by placing the membrane sample between two stirred
reservoirs, one containing an aqueous methanol solution of known concentration, and the other
containing DI water. Several small aliquots of the DI water reservoir as a function of time were
taken to analyze its methanol concentration by gas chromatography.

61

4.9 Tensile Strength Measurements
Membrane samples were stored in a vial with Millipore water and cut into simple
rectangles and analyzed on a DMA Q800 from TA Instruments. Once cut, the excess water
droplets were wiped off the surface of the rectangular membrane samples using tissue paper. The
samples were then clamped into the DMA instrumentation and the stress strain curve was
generated.

4.10 Differential Scanning Calorimetry
This was performed to characterize the fraction of freezable water in these proton
conducting membranes. Instrument used was DSC Q2000 from TA Instruments. Samples were
boiled in DI water for 10 minutes, then stored in room temperature DI water, and then excess water
droplets were wiped off the surfaces of the samples before analysis. Samples were equilibrated at
-15 ºC for 45 minutes before heating a rate of 5 ºC/min to 15 ºC. None of the fabricated membrane
samples exhibited any endotherm for the melting of ice at 0 ºC.

4.11 Scanning Electron Microscopy (SEM)
Images were obtained for the freeze-fractured surfaces as well as both sides of a membrane
sample on a FEI Helios G4 PFIB Xe instrument. Samples were coated with conducting carbon
prior to imaging.

4.12 Atomic Force Microscopy (AFM)
AFM imaging was performed on an Innova® atomic force microscope instrument in
tapping mode. The sample was immobilized on to a stainless-steel puck using double sided tape.

62

Both sides of the membrane were imaged but appeared to be visually identical. Membranes in the
TBA
+
and H
+
form were imaged also but appeared to be visually identical.

4.13 Energy Dispersive X-ray Spectroscopy (EDS)
EDS images for the elemental mapping of fluorine and sulfur were obtained on a FEI
Helios G4 PFIB Xe instrument using an Oxford Instruments Symmetry Detector. Elemental
mapping was used to estimate the proportions of PSSA and PVDF found on either side of the
material as well as the cross-section.

4.14 Degree of Crosslinking
In this case, the PTBAS-S-CMS composites were contained in a round bottom flask.
Attached was a water-cooled glass condenser connected to a liquid nitrogen trap. The contents of
the flask were heated to 165 °C for 5 hrs and the vapors captured in the trap. The DMF-HCl
solution in the trap was then diluted with a minimal amount of DI water and then titrated against
a calibrated solution of silver chromate to determine the molar amounts of HCl, and hence the
degree of crosslinking. This can be done by dividing the moles of HCl by the total volume of the
membrane.

4.15 Effect of Crosslinking on Modulus
Casting the blend at the same temperature for increasing duration increases the crosslink
density, defined here as the number of crosslinks per unit volume. Five identical blended samples
were cast in an oven set at 165 °C and every hour, one sample was removed from the oven and

63

rapidly cooled. These samples were subjected to tensile strength measurements to determine the
modulus of a series of membranes of various crosslinking densities, based on heating duration.

Results and Discussion
4.16 Synthesis of Copolymers
Initially, a series of terpolymerizations of 4-chloromethylstyrene, (CMS) styrene (S), and
TBASS were carried out at 65 °C in DCE with total monomer concentrations being kept at 3.0 M.
These terpolymerizations resulted in high viscosity solutions and insoluble gels when heated for
longer than 5 hrs at 65ºC. In the early stages of optimizing the terpolymerization conditions, the
initial concentration TBASS comparable to roughly 50 mass% where the other 50% was comprised
of styrene, CMS, solvent, and AIBN. At this intensely high concentration of TBASS, often,
insoluble gels would be produced from the crosslinking of the terpolyer during formation. With
excess initial solvent the gels could be rendered soluble by increasing DCE solvent by a factor of
2 or 3. However, even then evidence of some chain coupling via crosslinking was observed.
Compared to TBASS-S copolymers, the CMS containing copolymers had higher apparent MW's
by factors of 2-4, indicating CMS mediated coupling (Tables 4.1). The copolymer system of
styrene and TBASS never produced any polymers larger than about 50 kDa (previous chapter
Table 3.1). Therefore, systems containing CMS that produced larger apparent MW’s clearly were
larger due to crosslinking at 65 °C during copolymerization. Furthermore, in preliminary
copolymerizations, the precipitated (PTBASS-S-CMS) terpolymers, while drying in a vacuum-
oven set at 60 ºC, also produced insoluble copolymers with or without the presence of DCE or
other solvents. Hence, when removing residual non-solvent (DEE) from the precipitated

64

PTBASS-S-CMS (or PTBASS-CMS) polymer, heating was avoided, as this was shown to result
in spontaneous crosslinking. The TBASS copolymer containing CMS would crosslink easily in
the anhydrous state at relatively mild temperatures of 60 °C and ambient pressures in air.

4.17 Synthesis of PSSA copolymer-PVDF Blends
In the synthesis of PSSA copolymer-PVDF blends, it is worth noting that the TBA cation
comprises roughly 54% of the copolymer mass (for the 801010 terpolymer for instance). Hence
after ion-exchange, the total dry mass of the blend with PVDF is decreased accordingly. The
number of sulfonic acid groups in the neat PSSA terpolymer is estimated assuming the PTBASS
terpolymer to be in the protonated form. The sulfonate mole % of the terpolymer is known from
proton NMR in the presence of an internal standard by measuring the amount of butyl signal.
Thus, the approximate maximum acid composition for the TBASS copolymer, by itself, is known.
The calculated IEC and subsequently the corresponding PSSA content of any blend composition
with PVDF can be determined solely from the weighed-out proportions of each polymer. The
PVDF and the TBASS copolymer containing CMS were co-dissolved in DMF and then solvent
cast at 165 °C to give a homogenous, crosslinked, composite of PTBASS and PVDF. The
homogenous blends (not yet membranes) were then subjected to an optimized ion-exchange
process to fully convert the TBA cations into protons, followed by repeated dialysis against DI
water to give the PSSA-PVDF membrane. A series of membranes based on different blend
compositions, or different TBASS copolymer precursor, can readily be made with PVDF. The
membranes can undergo the ion-exchange process in the same bath at the same time.

Table 4.1. Apparent molecular weight data for p(TBASS-S-CMS) terpolymers
Code(TBASS:S:CMS) Mn (kDa)
a
Mw (kDa)
a
PDI
a
80:10:10 108 157 1.46

65

80:15:05 196 268 1.37
80:17:03 155 216 1.40
80:00:20 166 234 1.41
90:00:10 174 229 1.31
95:00:05 111 153 1.37
a. Apparent MW's and PDI's using 1.0 M NaCl aqueous SEC. Narrow MW distribution PSSA standards were used
for calibration (previous chapter Figure S3.8).

Initially for early composite samples, the IEC yields after processing IEC yields were
≤75% of the calculated values which were based on the total moles of sulfonate groups inside the
composite. This was first noticed by an unusually low mass loss. We had believed the low IEC
was due to incomplete crosslinking and partial leaching out of the polyelectrolyte from the PVDF
matrix. To test this, we prepared membranes using various Lewis acid catalysts commonly used
for Friedel-Crafts chemistry as well as a membrane with no catalyst. This was done to assess if
different Friedel-Crafts catalysts were going to increase the IEC yield. Some of these catalysts
possessed higher catalytic activity towards Friedel-Crafts alkylation when compared to the ZnCl2
that was used initially. We found that there were negligible improvements to the corresponding
IEC of these membranes, including to the membrane prepared without catalyst; none of them had
IECs that were sufficient (≤60-75% IEC yield was obtained). These results, however, do not imply
that the number of crosslinks were the same for membranes prepared with different catalysts,
catalyst concentrations, or no catalyst at all. In the absence of conventional catalysts entirely, it
appeared that the crosslinking was sufficient in trapping the polyelectrolyte in the PVDF matrix.
Elemental analysis of C, H, N, and S, of these membrane samples indicated that there was excess
tetrabutylammonium remaining in the membranes accounting for the low IECs.
Later, it was discovered that the use of surfactants was essential for the ion exchange
process to be completed as intended. For example, Madsen et al. used a TBA-OH compatibilizer
in a similar crosslinked blend of a poly(4-vinylbenzylsulfonic acid) (PVBSA) and PVDF, but had

66

only achieved an IEC yield of approximately 82% after ion-exchange with a purely inorganic acid
bath.
35
Multiple combinations of H2SO4 or HCl, in the presence of NaCl, LiCl, or several other
inorganic ions at concentrations ranging from 1-3M at 80-90 °C for periods lasting from 3-7 days,
gave IEC yields of 75% (removal of TBA
+
with H
+
) or lower. In those cases, dried membranes
were slightly heavier than predicted, suggesting that the TBA was not fully removed. However,
in the presence of 5-10 g of “Alconox”, the resulting dried masses were much closer to predicted
values, the resulting IEC yields exceeded 90%, and elemental analysis did not detect nitrogen
(Tables 4.2 and 4.3).

4.18 Crosslinking in PSSA-PVDF Blends
As shown in Figure 4.2, later in this chapter, the crosslinking reaction was proposed to
occur through a combination of an SN-2 displacement of Cl anion of the copolymer benzyl
chloride by sulfonate anion resulting in a benzyl sulfonate. A subsequent electrophilic aromatic
substitution (EAS), i.e. benzylation of phenyl from another chain, gives a crosslink. This results
in the formation of HCl as a volatile byproduct. The volatility of HCl was demonstrated by heating
a blend solution to roughly 165 °C and distilling the DMF and HCl condensate. This condensate
solution formed AgCl upon adding AgNO3 to the DMF solution (explained below). For a blend
using a PTBASS-styrene (4:1 ratio) copolymer as a control trial, this AgCl, obviously, was not
formed. The above indicated that the quantification of the number of crosslinks per mass, or
volume, was feasible and provides a better quantitative understanding of the crosslinked polyacid
structure.
For setting up the quantification experiment, a 1.0 g (0.5 g PVDF; 0.5 g terpolymer; 10 mL
DMF) composite blend solution, once all solids dissolved, was placed in a round bottom flask

67

equipped with glass adapter and condenser and heated to 170 ºC for 2 hrs, to ensure that all volatiles
including DMF, and especially HCl, were collected quantitatively using liquid nitrogen cooling to
ensure the capture of all HCl. Aluminum foil was also wrapped around the top of the flask, the
glass adapter and condenser to encourage even heating. The determination of the IEC of the neat
copolymer using proton NMR, the composition of the copolymer is expressly known. This
allowed the approximation of the molar amount of CMS per gram of copolymer so that a calculated
Cl
-
maximum yield of benzylation events could reasonably be predicted. The experimental amount
of Cl
-
was determined, in the collected DMF vapor, using Mohr’s method of chloride
determination after dilution of the DMF vapor condensate with DI water.
87
This quantification
method relies on CrO4
2-
reacting with an excess of Ag
+
ions once all Cl
-
have been reacted to form
the insoluble AgCl(s), thus the formation of the crimson Ag2CrO4 is the indication of the end-
point.
87
The amount of HCl produced in this investigation corresponded to a crosslink, or
benzylation, yield of 6.8%. Meaning that 6.8% of the CMS groups put in the blended system were
converted into crosslinks. The volume of this membrane prepared in a round bottom flask, to
capture all vapors and HCl, was approximately 1.8 cm
3
. Assuming a degree of polymerization
(DP) of approximately 100 for the TBASS-S-CMS system, we calculate a crosslink density of
roughly 6.4 x 10
18
crosslinks/cm
3
, 6.4 million crosslinks/μm
3
, or 0.01 crosslinks/nm
3
. Meaning
that within 100 cubic nanometers (approximately the volume of a 5 nm cube), we expect to find
approximately 1 crosslink, on average. The approximation of DP as 100 is only theoretical,
however, since all polymers with TBASS showed evidence of aggregation, complicating the
estimation of its true DP (discussed in chapter 2 and later in this chapter). If the DP was larger or
smaller, that would only mean that there were less or more chains, respectively, in the blended
system. In principle, a different DP should not affect the total number of CMS groups in the

68

composite (and blend system), only the number and length of the copolymer chains. We cannot
comment on whether more than one benzylation reactions occur per ring, especially for the 8020
copolymer of TBASS and CMS.

Table 4.2. Composition and characterization of PSSA-PVDF Blended Membranes using
the 801010 copolymer of TBASS, styrene, and CMS.
Membrane
a
PSSA wt%
b
IEC (mmol/g)
c
IEC % yield
d
λ
e
σ (mS/cm)
f
P MeOH
(x 10
-7
cm/s)
g

M1 7.9 0.43 92.6 6.0 4.03
n/a
M2 11.5 0.62 93.3 8.8 9.64
n/a
M3 15.5 0.84 95.8 9.0 22.7
n/a
M4 19.4 1.06 94.5 11 38.3
n/a
M5 23.2 1.26 94.0 13 56.6
n/a
M6 26.6 1.45 93.7 15 77.0
6.29
M7 30.4 1.65 92.5 16 94.2
9.72
M8 33.6 1.82 91.9 17 111
11.2
Nafion-117
h
n/a 0.91 n/a 22 78.0
14.7
(a) Poly(TBASS-S-CMS) terpolymer with 80:10:10 (TBASS:S:CMS) monomer molar ratio. (b) based on the
measured IEC. (c) Measured via titration. (d) Proton NMR is used to measure IEC of neat terpolymer after synthesis
and was used to determine the calculated IEC of resulting blends. (e) Moles of water molecules per mole of SO 3H. (f)
Calculated from resistance determined from the real Z’ axis intercept via 4-probe electrochemical impedance
spectroscopy (EIS) with 5 mV alternating current scanning frequency between 1 MHz to 10 Hz at room temperature.
(g) Determined via GC analysis with 1-butanol as internal standard only for samples with conductivities comparable
or higher than Nafion-117. (h) Nafion-117 properties were experimentally determined using the same methods as the
synthesized membranes.

Table 4.3. Composition and characterization of PSSA-PVDF blended membranes using the
8020 copolymer of TBASS and CMS.
a.b
Membrane
a
PSSA wt%
b
IEC (mmol/g)
c
IEC % yield
d
λ
e
σ (mS/cm)
f
M9 6.48 0.35 93.4 3.27 3.68
M10 18.7 1.02 96.8 10.07 35.5
M11 32.2 1.75 93.9 14.6 80.7
M12 6.48 0.35 93.4 2.98 3.72
M13 18.8 1.02 96.8 9.89 35.6
M14 32.3 1.75 93.9 14.1 82.1
Nafion-117
h
n/a 0.91 n/a 22.0 78.0
(a) Poly(TBASS-CMS) terpolymer with 80:20 (TBASS:CMS) monomer molar ratio. Samples M9-11 were allowed
to cool gradually in the oven from 165 °C to 23 °C. Samples M12-14 were rapidly cooled from 165 °C by thermal
quenching in a water bath. (b) Experimentally determined PSSA wt% based on the measured IEC. (c) Measured via
titration. (d) Proton NMR is used to measure IEC of neat copolymer after synthesis and was used to determine the
calculated IEC of resulting blends. (e) Moles of water molecules per mole of SO 3H. (f) Calculated from resistance

69

determined from the real Z’ axis intercept via 4-probe electrochemical impedance spectroscopy (EIS) with 5 mV
alternating current scanning frequency between 1 MHz to 10 Hz at room temperature. (g) Determined via GC analysis
with 1-butanol as internal standard. (h) Nafion-117 properties were experimentally determined using the same
methods as the synthesized membranes.

This was confirmed via elemental analysis in the membranes processed using detergents.
Analysis of elements C, H, N, and S did not detect any nitrogen and the measured IEC values were
found to be close to the values calculated in Table 4.2 and 4.3. Proton NMR in the presence of an
internal standard was used to calculate the IEC for a membrane that was 100% copolymer and 0%
PVDF. This calculation assumes that 6.8% of the CMS groups are converted into crosslinks and
the system loses mass in the form of HCl accordingly. Once the hypothetical membrane used for
calculation is ion-exchanged, all the tetrabutylammonium groups would be changed to protons and
all the remaining benzyl chlorides are converted into benzyl alcohol and the mass of the membrane
would also change accordingly. This calculation of IEC also considers the mass (or lack thereof)
of styrene in the copolymer backbone. Finally, the weight percent of PSSA, for any blend, only
considers the mass of monomeric PSSA units, and does not consider the mass of anything else
(styrene, crosslinks, benzyl alcohol, tetrabutylammonium, water, etc). Therefore, it is possible to
determine the IEC for a membrane that is any composition of PVDF and TBASS copolymer
containing CMS, if the composition of the synthetic TBASS copolymer is known. When using an
inorganic ion-exchange bath, the persistence of some of the TBA ion in the blend is presumably
due to their interactions with hydrophobic PVDF and PSSA copolymer backbone as well as the
sulfonate anions. The favorable role of the surfactant may be due to several factors that mitigate
all these interactions in the blend, hence allowing the TBA cation to disassociate into the bulk
solution. The TBASS copolymer feasibly could yield surfactant-like effects due to intense dipole-
dipole interactions, giving rise to micellar structures. These structures could be comprised of
small, but clustered, hydrophobic domains of the PVDF that could bind the TBA cation very

70

effectively while the hydrophilic domains remain anhydrous during solvent casting in the oven at
165 °C. The difficulty in TBA extraction is enhanced as the PVDF and PSSA copolymer domains
appear to be extremely small (Figure 4.9-4.11). The membranes and their properties are stable
when stored for over 5 years, submerged in DI water. The composite blends (before ion-exchange
into membranes) were also stable in the dry state for over 5 years and could be readily ion-
exchanged and dialyzed into the membrane form with no differences in properties. The IEC,
proton conductivity, and dry mass of a set of membrane samples ranging from 7.9 to 33.6% PSSA
from Table 4.2 and 4.3 were re-measured and these values showed negligible changes, indicating
that the membranes and ion conductive properties were unchanged.

4.19 Crosslinking Mechanism
This could be a rare example of a sulfonate group, a conjugate base of a strong acid, acting
as a nucleophile, but only when paired with specific types of cations. The sulfonate ester
intermediate shown in the mechanism was never isolated or detected by GC-MS. We believe this
is because the arylsulfonate is a very good leaving group and therefore the group could be unstable
under the reaction conditions as well as unstable during mass spectrometry conditions. It could
also be that the steps in the proposed mechanism are coordinated with one another in a way that
the sulfonate ester never quite forms. The distance between the nitrogen and sulfur in TBASS is
large compared to analogous sulfonates paired with alkali or organic cations (see chapter 2).
53-57

From the X-ray data, the TBA cation is positioned much closer to two oxygens with the remaining
oxygen being much further away.
2,3,58
Plausibly, this remaining oxygen is nucleophilic enough to
displace the chloride anion with enough heat.

71

Figure 4.2. Proposed mechanism for crosslinking reaction for a terpolymer containing
TBASS, styrene, and 4-chlormethystyrene (CMS).

4.20 Modeling the Crosslinking Reaction
To verify the nature of the crosslinking reactions in the polymer blends, low MW model
reactions of 4-isopropylbenzyl chloride (4IBC) with cumene or anisole catalyzed by
tetrabutylammonium tosylate (TBAOTs). These were carried out with a large excess of cumene
serving as the solvent. Thus, the benzylation of cumene or anisole with benzyl chlorides in the
presence (and absence) of TBAOTs were carried out in the absence of solvent (Tables 4.4 and
4.5). Due to the TBAOTs having a high molecular weight of roughly 413.6 g/mol, its absence, or
presence, would drastically change the volume of the system, therefore, the amounts used in these

72

model reactions were expressed in mole ratios. The sulfonate ester intermediate under the reaction
conditions may have been unstable under the reaction conditions of 165 °C in the case of cumene
and benzyl chloride, and 140 °C in the case of anisole and benzyl chloride, both systems were
solvent-less. The benzyl tosylate group should be an excellent leaving group and was never
isolated or detected by us, under these specific reaction conditions. Our experiments showed that
benzylation of cumene or anisole progressed faster and in higher yields in the presence of
TBAOTs, when compared to controls. In the case of cumene and benzyl chloride, benzylation
only occurred at 165 °C when in the presence of TBA-OTs (Table 4.4). The sodium form of p-
toluenesulfonate was also attempted but did not result in any benzylation products for the cumene
and benzyl chloride system at 165 °C.
Table 4.4. Reaction of Cumene and 4-isopropyl benzyl chloride (4IBC)
Model Reaction
a
TBAOTs
(mole ratio)
4IBC
(mole ratio)
Cumene
(mole ratio)
Yield (%)
1 4 1 8 21 (24 hr)
30 (40 hr)
2 0 1 8 0 (24 hr)
0 (40 hr)
3 4
b
1 8 0 (24 hr)
a. Both reactions were heated in the same bath at 165°C for 24 hr with an additional 16 hrs
of heating. b. Sodium tosylate was used here instead of TBAOTs.

Table 4.5. Model Reaction of 4IBC and Anisole
Model Reaction
a
TBAOTs
(mole ratio)
4IBC
(mole ratio)
Anisole
(mole ratio)
Yield (%)
3 0.25 1 1 42
4 0.025 1 1 28
5
b
0.25 1 1 41
6 0 1 1 19
a. Reactions at 140°C for 24 hrs without solvent. b. This reaction had DMF at mole
ratio of 1:1:1:0.25 (anisole:4IBC:DMF:TBAOTs).

73

4.21 Mechanical Measurements of PSSA-PVDF membranes
The fabricated membranes had some deviations in physical properties based on
fabrication procedures. Dynamic Mechanical Analysis (DMA) was used to investigate some of
the results of different fabrication techniques for the blends of PVDF with PTBASS copolymer
containing CMS.

Figure 4.3. Dynamic mechanical analysis (DMA) of fully hydrated PVDF-PSSA blends
(M8) prepared either from a 20 wt% solution using a doctoral blade or a 15 wt% solution
solvent-cast in a Petri dish.

0
10
20
30
40
50
60
0 50 100 150 200 250 300 350 400 450
Stress (MPa)
Strain (%)
Hydrated 33.6% PSSA-PVDF Membrane
Doctors Bladed Membrane Petri Dish Membrane

74

Figure 4.4. Effect of heating duration on the crosslinking as indicated by elastic modulus
for the 8020 copolymer

The membrane made using the Petri dishes cast from more dilute solutions, has a slightly
higher Young’s Modulus (~1.5 MPa difference) while having a lower ultimate strength (Figure
4.3). The degree of crosslinking should influence the mechanical properties of the membranes and
indeed with longer heating an increase was seen in membrane modulus (Table 4.6). The slightly
decreased moduli of the Doctor’s blade membranes relative to the Petri dish protocols was
consistent with its decreased crosslinking. Additional studies had shown that blend solutions of
20 wt% of both polymers dissolved in DMF, were highly viscous, and appeared to not fully
crosslink because after ion-exchange it was discovered that up to 20% of the sulfonate groups were
missing in some samples cast from DMF or N-methyl-2-pyrrolidone (NMP). The casting solvent
NMP has a slightly higher boiling point than DMF, it was speculated that the increased boiling
point would allow for more crosslinking; that speculation was wrong. In contrast, membranes cast
from a blend solution that was below 15 wt % of both polymers dissolved in DMF, all resulted in
IEC yields above 90% indicating better crosslinking (Tables 4.2 and 4.3) of all polyelectrolyte
chains.
0
10
20
30
40
50
60
0 1 2 3 4 5 6
Modulus (MPa)
Heating duration at 165 °C (hours)
Fully Hydrated 30.4 wt% PSSA-PVDF Membranes

75

4.22 Proton Conductivities
As indicated earlier, the fully processed PSSA-PVDF blended membranes have IECs that
were very close to calculated values (Tables 4.2 and 4.3) indicating adequate ion-exchange. The
sulfonate mediated crosslinking was effective at trapping the terpolymer within the PVDF matrix
to form a semi-IPN PEM, shown by nearly complete (90-96% IEC yields). The 5-8% difference
between calculated and measured IEC is probably due to tightly bound water that artificially
increases Wdry (Equation 1). It is likely that for the PSSA-PVDF semi-IPN system, the
anhydrous state is inaccessible without degrading the material (Figure S4.4).

Figure 4.5. Linear relationship of PVDF-PSSA membranes (Table 4.2) between proton
conductivity and water content

4.23 TGA of PSSA-PVDF
Some of the first mass lost (~5-8% mass loss at 300 ºC) in the TGA from the blends is
likely tightly bound water (Figure S4.4). When dried under more rigorous conditions (110 ºC in
presence of P2O5 and under vacuum for several days), it was found that the samples only lost an
y = 2.1926x - 8.3358
R² = 0.9977
0
20
40
60
80
100
120
0 10 20 30 40 50 60
Proton Conductance (mS/cm)
Water Uptake % or Water wt%

76

additional 0.3% of its mass before the mass became constant. Thus, while it may be possible to
remove all moisture content of the membrane, reaching a truly anhydrous state would risk
degradation (Figure S4.4). Generally, these temperatures of irreversible degradation would not be
reached in the DMFC setting. The next significant mass loss in the TGA was considered to be
desulfonation of the crosslinked polyacid, followed by depolymerization of the polymer backbone
into small volatile compounds.
88,89
It is worth noting that the hydration number appeared to increase with increasing PSSA
content. The reason for this may involve a polyelectrolyte aggregation effect as well as an effect
of simply having less hydrophobic PVDF matrix in the composite to restrict water content.
Heating duration appeared to slightly reduce the water content of the composites, probably through
further crosslinking. However, the difference in water content between heating at 165 ºC for 30m
and 120m in resulting processed membranes is only about 5 wt%. It is seen from Figure 4.4, that
much of the mechanical benefit from crosslinking is achieved within 3 hours.

Figure 4.6. Conductance normalized to the acid concentration of membranes from Table
4.2.
0
10
20
30
40
50
60
70
80
90
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0
mS·g/cm·mmol H
+
PSSA wt%
Conductivity/mol H+ Nafion-117

77

The differences between the proton conductivity of the PSSA-PVDF composites and
Nafion seen in Figure 4.5 are consistent with the greater acidities of perfluoroalkylsulfonic acid
compared to arylsulfonic acids. The pKa of Nafion is estimated to be -6 while the pKa of sulfonated
polystyrene is -2.1.
90,91
The 26.6% PSSA sample from Table 4.2 has similar proton conductivity
but clearly requires a higher concentration of sulfonic acid. The trend appears to have a sigmoidal
shape. As one can see, the amount of conductivity afforded per density of protons (as measured
by mmol H
+
/gram) levels off as higher PSSA wt% is achieved in these membranes. This apparent
plateau of the conductivity, is an illustrative representation of the difference in pKa between
perfluorosulfonic acid found in commercial Nafion membranes and aryl sulfonic acid found in our
PSSA-PVDF membranes. There are other factors influencing the number and mobility of protons
between the two systems, such as the size of the hydrophilic domains and contact resistance with
electrodes.

Figure 4.7. UV-Vis spectra of a 33.6 wt% PSSA blended membrane.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
200 250 300 350 400 450 500 550 600 650 700 750 800 850
Absorbance (a.u.)
Wavelength (nm)
33.6% PSSA

78

As supported by the UV-Vis data in Figure 4.7, the PSSA-PVDF semi-IPNs are fully
transparent in the visible range (> 400 nm). Samples heated for 5 hrs had a slight yellow tint when
compared to samples heated for only 1 hr. When sample solutions were cast at temperatures lower
than 60 °C, the samples developed a slight opacity. The opacity was more intense if the blend
solution was cast at room temperature. The solvents that can be used to cast this blend solution
were investigated, with DMF being the most used for this work. Other polar aprotic solvents used
were: acetone, methylethylketone (MEK), dimethylsulfoxide (DMSO), N,N-dimethylacetamide
(DMAc), N-methyl-2-pyrrolidone (NMP), but some solvents (acetone and MEK) gave opaque
films. When these samples were then heated at 165 °C for any amount of time, up to 10 hrs, this
opacity remained, indicating that additional phase mixing does not occur under these conditions.
When samples were solvent-cast at 165 °C, the blended composite appeared transparent and
remained so even if allowed to cool slowly from 165 °C to room temperature over 5 hours. The
lack of changes to opacity when cooled slowly, indicated that additional phase de-mixing does not
occur under these conditions. The persistence of transparency, or lack thereof, once casting solvent
was sufficiently evaporated (usually within 15-30 minutes, depending on casting temperature)
indicated that crosslinking occurs rapidly and prevents any further mixing or de-mixing of the
phases. Therefore, it was imperative that the PVDF and PTBASS copolymer containing CMS
were both fully dissolved in sufficient amount of casting solvent, to encourage adequate mixing of
the phases before casting. The casting of the blend solution from a more concentrated solution in
DMF was explored in Figure 4.4. The doctor bladed membrane, from a more concentrated and
viscous solution, had an inferior elastic modulus, indicative of less crosslinking. For the doctor
bladed membrane, the titration and conductivity measurements revealed missing sulfonate groups,

79

indicative of incomplete crosslinking. The cause of this is likely due to incomplete mixing of the
PVDF and PTBASS copolymer containing CMS, due to not enough casting solvent. Additionally,
the cyclic crosslinking mechanism, proposed in the previous chapter, requires the system to have
some molecular motion to have any level of turnover. Since the doctors bladed composite blend
solution started out more viscous than the Petri blend solution, when the solvent was evaporating,
there was less Brownian motion generated from the evaporation of the solvent and therefore less
crosslinking. We did not collect much data on these membranes made from a composite blend
solution of 20 wt% in DMF, since they had inferior membrane properties compared to membranes
made from blend solutions of 10-15 wt% in DMF.

4.24 Microscopy of PSSA-PVDF membranes
Due to the direct blending of an ionic polymer with PVDF, we wanted to investigate the
morphology of the blends using techniques such as AFM, SEM, EDS, and TEM. Only the SEM
produced images where there was nothing worthy of note. This was due to the extremely small
domain sizes.

80

Figure 4.8. Sample was immobilized onto a circular metal puck using double-sided tape.
Sample is M5 from Table 4.2, in the TBA form.

AFM results indicate that the surface is quite smooth with roughly a 20 nm variation in
height. The vertical axis was scaled up to better visualize the density of peaks and valleys. The
acid form of the material looked virtually identical to the TBA-form, at least on the surfaces. Both
samples were in the dry state during measurement. The smoothness of the surface is consistent
with the homogeneity of the blended system. The identical appearance of the acid form to the
TBA-form indicates that the hydrophilic domains must be extremely small.

81

Figure 4.9. TEM image of 30.4 wt% PSSA-PVDF membrane with 20 nm scalebar.

The TEM images of membranes reported here were selected to be comparable in proton
conductivity to the commercial standard Nafion membranes. The ionic sulfonate domains were
proton exchanged with Pb ions using lead nitrate solutions with the lead ions appearing dark in the
TEM images (Figures 4.8-4.10).

82

Figure 4.10. TEM image of 30.4 wt% PSSA-PVDF membrane with 5 nm scalebar.

In Figure 4.9, which is the zoomed top of Figure 4.8 near the tear in the sample, that was
the thinnest region we observe stained hydrophilic domains that were on the order of 1 nanometer.

83

Figure 4.11. TEM image of 10 wt% PSSA-PVDF membrane with 20 nm scalebar.

In Figure 4.10 (10 wt% PSSA-PVDF membrane) we see dark hydrophilic PSSA domains
that appear to still be continuous at the nanometer scale. We also observed in the ATR-FTIR and
the XRD, that composites of PVDF and practically any amount of sulfonate copolymer,
predominantly gave rise to the beta-phase crystallinity in PVDF. It has been determined that
nanofibers of -phase PVDF, also produce dark TEM regions being slightly less than 1 nm. Hence

84

it is plausible that the polyelectrolyte and PVDF are practically molecularly dispersed.
92,93
This
incorporation of a PSSA polyelectrolyte within PVDF is similar to work by Madsen et al who used
both a compatibilizer and a crosslinking agent,
35
which were not needed in this work.

Figure 4.12. EDS of the freeze-fractured, cross-section of a 30.4 wt% PSSA-PVDF
membrane sample. Wt% was normalized by removing all elements except fluorine and
sulfur.

Figure 4.13. EDS of the glass-facing side of a 30.4 wt% PSSA-PVDF membrane. Wt% was
normalized by removing all elements except fluorine and sulfur.

85

Figure 4.14. EDS of the atmosphere-facing side of a 30.4 wt% PSSA-PVDF membrane.
Wt% was normalized by removing all elements except fluorine and sulfur.

Elemental mapping was obtained for the 3 different surfaces for the 30.4 wt% PSSA
membrane system as this was the membrane that had proton conductivity values comparable to
that of the Nafion membranes. Due to the nature of the film being cast from solution, the direction
of solvent evaporation all but guarantees some level of anisotropy. There is also the chemical
composition of the surfaces. We predicted that the surface of the blend interfacing with the
atmosphere would have a higher fluorine content, due to the nature of the atmosphere being more
hydrophobic than the glass substrate. We also predicted that relative to the atmosphere-interface,
the substrate interface would have a lower fluorine content, and would be relatively more
hydrophilic. This is more or less, what we observed in the elemental mapping of elements fluorine
and sulfur. The fluorine content could only arise from presence of PVDF and the sulfur content
could only arise from the crosslinked polyelectrolyte.

86

Figure 4.15. ATR-FTIR of M1-M8 from Table 4.2.

Figure 4.16. ATR-FTIR of 30 wt% PSSA-PVDF membrane

87

Figure 4.17. XRD patterns of a series of 5 membranes ranging from low to high proton
conductivity and neat PVDF film prepared identically to the membranes. Intensities
between samples are arbitrary as the thicknesses were not controlled for.

4.25 Crystallinity of PVDF
It is well known in the literature that PVDF is a ferroelectric polymer.
94,95
The
piezoelectricity arises from the crystalline phases of PVDF with the polar β-phase crystallinity
contributing most towards piezoelectricity.
96
Both the ATR-FTIR and the XRD indicate that
blending PVDF with our TBASS copolymer causes the crystalline phases of PVDF to be converted
mostly to the polar β-phase. This is illustrated in Figure 4.15, where the red arrows indicate the
signals most indicative of the α-phase.
97
This is likely due to strong dipole-dipole interactions
between the two polymers thereby nucleating the growth of β-phase crystallites in the PVDF
material. We observe in the TEM images that there appears to be rows of gray to white dots with
0
2000
4000
6000
8000
10000
12000
14000
16000
10 20 30 40 50 60
Intensity cps
2 Θ
XRD
Recast PVDF powder
15% PSSA
20% PSSA
25% PSSA
30% PSSA
35% PSSA

88

domain sizes slightly under 1 nm in diameter. These could be molecular chains of PVDF in the β-
phase that matches well to images obtained by Lolla et al. using Transmission Electron
Abberation-corrected Microscopy (TEAM).
92
The β-phase PVDF was also simulated using
molecular dynamics and the computed TEM image had similar appearance with gray and white
dots appearing along the PVDF chains.
93
An image is included in the supplemental section.

4.26 Differential Scanning Calorimetry (DSC)
The instrument used for analyzing the membrane samples (Exp. section) could not be used
to obtain data near or below 10 ºC. We did not attempt to detect the Tg of amorphous PVDF since
it is known in the literature to be well below 0 ºC.
94
The endotherm we were attempting to observe
should be close to the weighted average Tg of the polyacid and the amorphous PVDF. PSSA is
extremely brittle and has a predicted Tg value far above its decomposition temperature as discussed
in the previous chapter.
76
Instead, new endotherm was observed that appears to increase with
PSSA content and partially overlaps with the endotherms for the β, and γ crystalline phases for
PVDF.
98
This is not unheard of, as there are many examples in the literature of polar additives
affecting the endotherms for β-PVDF.
94,98,95,99
This suggests that the polar sulfonate groups of the
crosslinked PSSA is interacting with the polar crystalline phases of PVDF as it melts and that
interaction absorbs more energy with increasing amounts of the polyelectrolyte. Perhaps, with
increasing PSSA content, the -phase crystallinity in PVDF is more stable to melting. It is
noteworthy that the PSSA-PVDF membranes contained no fraction of water, when fully hydrated,
that would freeze in the range of -15 to 0 °C and there was no endotherms found at or near 0 °C
(Figure S4.2 and S4.3).

89

Figure 4.18. DSC of PSSA-PVDF membranes containing small amounts of PSSA in wt%

Figure 4.19. Arrhenius plot of Proton Conductivity of membranes from Table 4.2.
Percentages in the figure indicate PSSA wt%.

2
2.2
2.4
2.6
2.8
3
3.2
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
5
5.2
5.4
5.6
5.8
2.7 2.8 2.9 3 3.1 3.2 3.3 3.4
ln(Conductivity))
1000/T (1000/K)
11.5 wt%
15.5 wt%
19.4 wt%
23.2 wt%
30.4 wt%
33.6 wt%
Nafion-117

90

Table 4.6. Thermodynamics of Proton Conductance for membranes from Table 4.2
PSSA wt% Slope(1/K) Intercept ΔH
‡
kcal/mol ΔS
‡
(cal/mol·ºC)
11.5 -1585 7.81 3.15 15.52
15.5 -1530 8.43 3.04 16.76
19.4 -1376 8.54 2.73 16.97
23.2 -1252 8.46 2.49 16.81
26.6 -1263 8.54 2.51 16.98
30.4 -1190 8.74 2.37 17.37
33.6 -1208 8.94 2.40 17.76
Nafion-117 -1268 8.61 2.52 17.12
.

4.27 Accelerated Oxidative Stress Test
The membrane samples were not stable to Fenton’s reagent or 3% hydrogen peroxide at
100 °C. After boiling in either solution for 30 minutes, the dimensions of the membrane samples
would irreversibly change. These samples were losing mass rapidly and therefore losing volume,
causing the material to buckle. Plausibly, the hydrogen peroxide with heating forms hydroxyl
radicals that abstract a methine hydrogens from the corresponding carbons along the PSSA
backbone.
30
This behavior to hydroxyl radicals is in stark contrast with Nafion membranes, that
have a virtual Teflon backbone which is far more robust towards chemical decomposition.

4.28 Proton Conductivity Dependence on Temperature
From temperature studies of Nafion-117, not only are the conducting properties strongly
dependent on water content and hydration level but temperature as well.
100,101
This temperature
dependence becomes significant for fuel cell applications when not at room temperature. Although
the proton conductance relationship with temperature was not perfectly linear, it was linear enough
to approximate some thermodynamic information for proton conduction (Figure 4.18 and Table

91

4.7). However, the data collected for Nafion-117 were also not perfectly linear, indicating a
systemic lack of precise temperature control in the thermocouple used. The membrane that is
closest to Nafion-117 (Table 4.2) in proton conductance appears to have rather similar
thermodynamic data and temperature dependence. This is surprising since these PSSA-PVDF
semi-IPNs are morphologically and compositionally different than Nafion-117. One significant
observation from DSC measurements of our membranes was that all our membranes contain zero
percent freezable water at 0 °C (Figures S4.2 and S4.3). This is a significant finding, since most
PEMs (when fully hydrated), including Nafion, contain water that is both freezable and non-
freezable.
7,102,103,104
Another important observation was that even at very low PSSA content (7.9%
PSSA), the PSSA-PVDF membranes still were conductive despite the polyacid content being
lower than the proposed threshold in the percolation model for ion conductance in conductive
particles.
105
The crosslinked and molecularly dispersed nature of the polyacid in PVDF must be
allow proton conduction along the path of the polyelectrolyte chains. Due to the lack of free water
in the PSSA-PVDF membranes, this conduction of protons is likely executed via the Grotthuss
proton hopping mechanism since the vehicular mechanism requires free water to exist in the
membrane.
27,106,107,108,109

Conclusions
Blended membranes of PSSA and PVDF were fabricated starting from a copolymer of a
TBASS, with or without styrene, containing a CMS crosslinker (described in the previous chapter)
and PVDF. Microscopy imaging of the membranes, indicate that the two polymers form stable
homogenous blends at what appears to be the molecular level. Crosslinking and ion-exchange
results in the conversion of TBASS copolymer blends with PVDF into PSSA-PVDF proton

92

conducting membranes. Some compositions of PSSA-PVDF membranes had proton
conductivities equal to or higher than commercial Nafion membranes. These membranes can be
used for fuel cell applications provided that the presence of hydroxyl radicals formed at the cathode
is remedied.

93

4.29 Supplemental Information

Figure S4.1. Simulated TEM image comparison with experimental TEM image of β-phase
PVDF nanofibers.
92,93

Figure S4.2. DSC of fully hydrated 30.4% PSSA-PVDF heated in oven for 1 hour.

94

Figure S4.3. DSC of fully hydrated 30.4% PSSA-PVDF heated in oven for 2 hours.

Figure S4.4. TGA of PVDF and 20% PSSA-PVDF blended membranes.

0
10
20
30
40
50
60
70
80
90
100
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700
% Mass Remaining
Temperature (°C)
TGA
Blend A Blend B PVDF

95

Chapter 5 Fuel Cell Performance
Abstract
Poly(4-styrenesulfonic acid)-polyvinylidene fluoride (PSSA-PVDF) blends utilized in
membrane electrode assemblies (MEAs) for direct methanol fuel cells are described. The
performance in similar MEAs using commercial membranes Nafion 117 and 212 are reported for
comparison. Fabrication techniques and cell "break-in" procedures are described. Evaluation of
MEA performance was mainly done by obtaining polarization and power density curves at various
cell conditions such as fuel concentration, temperature, and flow rates. Fuel cell performances of
the MEAs made using these PSSA-PVDF blended membranes with hydrogen as fuel were also
evaluated. For comparison of the PSSA-PVDF vs the above Nafion membranes the catalyst
loading for both anode and cathode electrodes were kept high (~8 mg/cm
2
). Carbon supported
catalysts were attempted with the PSSA-PVDF membranes, but these always resulted in MEAs
with prohibitively high internal resistance. Data indicate that the PSSA-PVDF blended
membranes are promising for DMFCs, especially at very high methanol concentrations and low
cathode flow rates.

Introduction
5.1 Perfluorinated Membranes
One of the most commonly used membrane for fuel cells is Nafion® (DuPont).
101
It has
properties that are most suitable ("ideal" overstates the case) in hydrogen fuel cells.
110
Its use in
direct methanol fuel cells (DMFCs), has been studied extensively and was found to involve
methanol crossover (MCO) that accounts for roughly 50% of performance losses.
86
The addition
of ruthenium to the anode catalyst prevents the gradual poisoning of the anode platinum catalyst

96

allows stable fuel cell operation.
111
Without ruthenium, the platinum catalyst loses catalytic
activity via formation of carbon monoxide, and other methanol oxidation intermediates that bind
irreversibly to the platinum catalyst active sites.
111,19
Another major issue with Nafion membranes
in DMFCs, is the high ruthenium crossover at the Pt-Ru anode, although the precise mechanism
of Pt/Ru degradation has not been fully described.
85
There have been numerous attempts to
address methanol crossover (MCO) issues in Nafion.
86
In hydrogen PEMFCs, Nafion membranes
have shown unsurpassed cell longevity exceeding 60,000 hrs but its durability in DMFCs without
additives has only seen operation times of 500-4,000 hrs.
86,112
One method for reducing the MCO
of Nafion, without proton conductivity losses, has been the use of inorganic fillers such as
zirconium, silica, titanium, or cerium at the Pt-Ru anode.
113
These modified Nafion membranes
were shown to be more thermally stable at temperatures above 80 °C where pristine Nafion shows
dehydration resulting in severe decreases in proton conduction.
86
Another strategy involves
composite membranes, comprising, for instance, PVDF doped with proton conducting acids or
polymers such as sulfuric acid, phosphoric acid, Nafion ionomer, or silica gel.
114
Due to the high
cost of Nafion ($600-1200 m
-2
, depending on thickness) and the environmental hazards of the
synthesis and/or disposal of Nafion membranes, efforts have been made to create non-
perfluorinated membrane materials.
86,115,29

5.2 Non-perfluorinated hydrocarbon based membranes
Methanol fuel cells without the use of Nafion-based membranes were developed in the
early-mid 1990s by joint efforts of NASA-JPL and USC.
9,116
Since then, these fuel cells with
kWh capacity have been commercialized. Earlier research into FC's involving the electro-
oxidation of methanol had begun in the late 1980s with some of these showing promising
performance.
117,118

97

Figure 5.1. Effect of a) temperature, and b) methanol concentration.
9

The cell performance is a function of several different parameters. Temperature is a key
parameter affecting all FC processes with all electrode reactions occurring faster at higher
temperatures. Fuel concentration at the electrodes is another key parameter. Increased fuel
transport to the electrodes increases cell performance while increased fuel transport to the opposing
electrode does the opposite. Similar effects are seen for gaseous reagents with pressures playing
an important role by increasing or decreasing the fuel concentrations. In addition, a triple-phase
boundary exists between the hydrophilic ion-conducting channels of the exchange membrane, the
active sites on the surface of catalyst particles, and finally the hydrophobic gas phase of both
reagent(s) and product(s). In conventional DMFCs, the anode is contacted by an aqueous methanol
but there will also be a carbon dioxide gaseous phase generated by the oxidation of methanol. At
the cathode oxygen gas is the oxidant but must reach the catalyst active sites to allow protons and
electrons to react. For these reasons the triple-phase boundary (TPB) is one of the most critical
factors (Figure 5.2).
83,119
The TPB, also facilitates the conduction of charged molecular species

98

from the hydrophilic domains of the membrane, the electrodes of the MEA, and electrons from the
external circuit.

Figure 5.2. Simplified diagram of the electrode-electrolyte interface in a fuel cell.
120

Experimental
5.3 Materials
Unless stated otherwise, all materials were used as received. Platinum black (Pt Blk) and
Platinum Ruthenium Black (PtRu Blk) were obtained from Alfa Aesar and Sigma Aldrich,
respectively. Nafion membranes (117, 212), Toray carbon paper (E-TEK, non-teflonized and 10
% wet-proofing), and the fuel cell hardware components (current collectors, bipolar plates, etc.)
were obtained from Fuel Cell Store. Liquion binder was purchased from Ion Power. Paint brushes
were purchased from commercial sources. A square punch was custom cut by a machine shop to
punch out the carbon paper consistently for each membrane electrode assembly (MEA). Cerium
Oxide nanopowder (<25 nm particle size, confirmed by BET analysis) was purchased from Sigma-
Aldrich.

99

5.4 Catalyst Ink Preparation Procedure
Catalyst dispersions ("inks") were painted onto a 5 cm
2
gas diffusion layers (GDLs) made
of porous carbon paper. Non-teflonized carbon paper was used for the anode GDL while carbon
paper with 10% “wet-proofing” was used for the cathode GDL. For the anode catalyst ink solution,
42 mg of PtRu black, 200 mg of Millipore water, and 200 mg of Liquion was added into a sample
vial of appropriate size such that the solution was at least 5 mm in height. The contents of the vial
were subjected to probe sonication at 20 kHz with sonication tip at 10% strength with 2 second
pulses followed by 1 second pauses, iteratively, for 5 minutes. For the cathode catalyst ink
solution, an identical formulation and sonication procedure was used except that Pt black was used
in place of PtRu black. For carbon supported catalysts to fabricate electrodes with lower metal
loadings, the metal to Liquion ratio was changed to 1:1 and the amount of water was increased by
a factor of two.

5.5 Painting Electrodes Procedure
The catalyst coated substrate (CCS) method was used instead of the catalyst coated
membrane (CCM) method.
121
We found that the CCS gave us the most reproducible and
repeatable results. All substrates were punched out to be precisely 5 cm
2
. Therefore, the active
surface area for all MEAs in this work is 5 cm
2
. Using a small brush with soft bristles, typically
used for fine detailing in watercolor painting techniques, the anode and cathode gas diffusion
layers (GDLs) were painted immediately after sonication of the catalyst solution. Thin layers of
the catalyst dispersion/ink were iteratively applied to the respective GDL with the brush and then
dried using a stream of dry nitrogen; this was repeated until all the ink was used. If the contents
of the vial became too dry to effectively paint, then a small amount of an aqueous solution of

100

isopropanol was added. The amount used was 10-100 μL of a solution of isopropanol to water
(ratio of roughly 1:1 v/v), depending on the remaining mass of catalyst in the vial. The
combination of a small brush with soft bristles, addition of isopropanol and water for redispersion,
allowed nearly all the catalyst particles of the dispersion/ink to be transferred out of the vial and
directly onto the GDL. Similar painting processes have already been described elsewhere.
46,122,123

A figure was included, adapted from the literature, for clarity of the process of loading the GDL
with catalyst and binder (Figure 5.3).
109
In our method, the brush is simply a delivery system for
the catalyst and particles. To be clear, this means, the instant that the brush, loaded up with catalyst
and binder, touches the surface of the GDL (which is very dry) capillary and adhesion forces of
the GDL and dispersion solvent, respectively, will transfer catalyst particles from the brush and
onto the GDL. Once this occurs, we load the brush again and then barely touch the brush to a
different spot on the GDL surface. This was done until a uniform layer was achieved before
drying. Also, only one side of the GDL should be painted.

Figure 5.3. Schematic of the painting process.
109

101

5.6 MEA Hot-Pressing Procedure
The painted electrodes are dried in an oven preset to 100 ºC. The dried electrodes were
weighed once thermally equilibrated to room temperature after removal from the oven. The dried
electrodes were loosely assembled on a bottom steel plate into a sandwich that goes in order from
bottom to top: cathode electrode (GDL) with catalyst layer facing up, membrane, and anode
electrode (GDL) with catalyst layer facing down. The arrangement is vital since the two CCS
electrodes do not have the same catalyst or level of teflonization. Teflon gaskets were also
included in the sandwich, but this was done to prevent adhesion of MEA to the steel plates.

Figure 5.4. Schematic of the 5 different layers of a membrane electrode assembly (MEA).
The size is not to scale; this diagram is to demonstrate the correct ordering of the layers.

102

A top steel plate was placed on the bottom steel plate thus pressing together the components
shown in Figure 5.4 above. The two steel plates have cut geometries with sufficient tolerances to
prevent the top plate and bottom plate shifting. The two steel plates were then placed in a hot-
press and pressed with 500 lbs (~625 psi) of force while heating from room temperature to 140 ºC
over 40 mins and then cooling to room temperature over 10 mins.
124
When calculating the amount
of weight that is required to be used, it is important to use the surface area of the GDLs as these
are always the thickest part of the sandwich and would experience all the force from the press.

5.7 Hydrating Procedure
The MEA from the hot-press is nearly anhydrous and would not have adequate proton
conductivity. There are different ways to hydrate the MEA, primarily the membrane.

Figure 5.5. Schematic of the hardware components of a single fuel cell.
125

The MEA from the hot-press was immediately assembled into fuel cell testing hardware (Figure
5.5). Millipore water was then flowed through both anode and cathode compartments at 5-10
mL/min and the cell heated to 60 ºC. Alternatively, another way to hydrate the MEA is to leave

103

it submerged in a sealed bag of DI water overnight. The internal resistance of the cell was
measured either doing 2-probed electrochemial impedance or using an instrument that
automatically applies a small alternating current at high frequencies to the cell to induce a small
but detectable oscillation in cell potential. The latter procedure is most practical as it allows the
monitoring of cell resistance in real-time. The results of both methods were repeatedly observed
to be in good agreement. The hydration procedure involved monitoring the cell resistance in real-
time with DI water flowing through the cell at 60 ºC until the cell resistance was stable for at least
1 hour.

5.8 Fuel Cell Break-in Procedure
Once stable cell resistance was achieved, the cell was assumed to be hydrated enough to
sustain the "break-in" procedure followed by performance testing. The MEA was assembled into
the cell hardware and evaluated using a Fuel Cell Test System 890B (Schribner Associates Inc.).
The cell was connected to a pump that fed a 1.0 M MeOH solution through the cell on the anode
side and a line was connected to the cathode side of the cell for the controlled flow of humidified
ultra-pure oxygen. The “break-in” procedure of the MEA involved operating the cell at 90 ºC,
using 1.0 M aqueous methanol solution at 5 mL/min to the anode and 50 mL/min of humidified
oxygen to the cathode. A constant voltage experiment was conducted at 0.6 V until current density
was stable for more than 30 mins, followed by a second constant voltage experiment at 0.3 V until
current density was stable for more than 30 mins.

104

5.9 Performance Testing Protocol
Performance was tested under various conditions: the flow of the methanol solution
through the anode was kept at a fixed rate of 5 mL/min at 80 ºC using humidified oxygen or - air.
At the cathode the air or oxygen flow rates ranged from 20-1500 mL/min. Testing sequence was
generally: cathode flow rate, cathode gas (air vs oxygen), followed by testing the effects of
temperature, and finally the effects of methanol concentrations. For instance, testing involved
measuring a polarization- and power density-curve by scanning current. This began with low (1.0
M) methanol concentrations and low temperatures (30 ºC), with oxygen tested first and then air,
and low cathode flow rates (10 mL/min) increasing up to 500 mL/min. Performance was assessed
first for low priority parameters followed by assessment of higher priority parameters. For
example, the performance of the cell using a range of flow rates (from low to high flow) for the
cathode were tested for oxygen and then air. Next, the temperature was increased from 30 ºC to
60 ºC and then cell performance was evaluated using the same range of flow rates with oxygen
first and air thereafter. Once this iteration of testing was done, once the conditions reached the
testing temperature of 90 ºC, the methanol concentration was increased to 2.0 M and the cell
temperature lowered to 30 ºC. The cathode stream was then switched back to oxygen, the cathode
flow rate was reduced to 10 mL/min, and then the same order of evaluation of the cell performance
was carried out at various parameters for the next iteration. Between acquisitions of polarization
and power density curves (roughly 30-60 seconds of data acquisition at each set of conditions) the
conditions of the DMFC were changed and then new data was collected after allowing the cell the
equilibrate for 60 seconds at the new conditions.

105

Results and Discussion
A total of 11 different MEAs were produced with many having internal resistances that
were far too high to justify further study. Only, MEAs with resistances below 100 mΩ, after
hydration were evaluated further. The PSSA-PVDF membranes were fabricated to be far thinner
than Nafion-117 to have adequate internal cell resistances with electrode sizes of 5 cm
2
.

Table 5.1. Thicknesses of Membranes used in MEAs
Membrane Thickness (μm)
c
Internal Resistance of MEA (mΩ)
d
26.6 wt% PSSA-PVDF (MEA-1)
a
200 3200
26.6 wt% PSSA-PVDF (MEA-2)
a
175 2500
26.6 wt% PSSA-PVDF (MEA-3)
a
120 1300
30.4 wt% PSSA-PVDF (MEA-4)
a
120 300
30.4 wt% PSSA-PVDF (MEA-5)
a
80 95
30.4 wt% PSSA-PVDF (MEA-5 2nd)
a
75 87
30.4 wt% PSSA-PVDF (MEA-6)
a
55 70
30.4 wt% PSSA-PVDF (MEA-7)
a
60 230
30.4 wt% PSSA-PVDF (MEA-8)
b
45 180
30.4 wt% PSSA-PVDF (MEA-9)
b
30 330
30.4 wt% PSSA-PVDF (MEA-10)
b
33 65
30.4 wt% PSSA-PVDF (MEA-11)
b
50 77
Nafion-117 183 33
Nafion-212 54 10
a. Membranes were made using the 801010 copolymer of TBASS, styrene, and CMS. b. Membranes were made
using the 8020 copolymer of TBASS and CMS. c. Thickness was measured using a digital micrometer after the
membrane was fully hydrated in a bag of DI water overnight. d. Measured via impedance spectroscopy using the 2-
probed method.

The earlier MEAs made in this work had high internal cell resistances. Attempts to make
the membrane the same conductivity and the same thickness Nafion-117 which is roughly 183 μm,
turned out to be a mistake. When working with 5 cm
2
the early PSSA-PVDF semi-IPN
membranes, relative, to the active surface area of 5 cm
2
, was much too thick, indicated by high
internal cell resistance. The high resistance of these early MEAs also demonstrated the
incompatibility of the PSSA-PVDF membranes and the Nafion-ionomer binder used in the catalyst
layer. The membrane was reformulated to be more conductive and made much thinner despite

106

different conductivities and thickness when compared to commercial Nafion membranes. The
membranes used for MEAs 7-9 were highly experimental. The membrane used for MEA-7
contained 0.1 wt% graphene oxide (GO) which appeared to darken the membrane into a brown
color. This MEA containing GO, did not conduct very well and this was seen in the resistance
value measured in Table 1. This was likely caused by the GO particles impeding the proton
hopping mechanism. The membrane for MEA-8 was made using a TBASS copolymer containing
CMS and hexafluoroisopropylmethacrylate (HFPMA). It was hypothesized that inclusion of a
fluorinated component to the sulfonate copolymer would increase the blending of the copolymer
with PVDF and would make the hydrophilic channels more discrete. The addition of the HFPMA
appeared to have decreased the conductivity to a point where the MEA had too much resistance.
This is likely due to the hydrophobic comonomer interfering with proton conduction along the
polystyresulfonic acid backbone of the PSSA-PVDF semi-IPN. If proton conduction is tightly
correlated with water content and the PSSA-PVDF membranes normally do not contain any free
water, then it stands to reason that incorporating HFPMA reduced proton conductivity by
enhancing the hydrophobicity of the membrane therefore deceasing water content. The membrane
used for MEA-9 contained 1 wt% CeO2 nanopowder dispersed into the PSSA-PVDF semi-IPN as
a radical scavenger during solvent casting. The ceria, however, appeared to complicate the
membrane fabrication and these membranes were less conductive. The carbon-supported catalysts
were attempted for MEA-12 and MEA-13 (discussed later), however, these MEAs appeared to
have inadequate TPB formation. The membranes used for MEA-12 and MEA-13, were using
membranes with proton conductivities much greater than Nafion, but these MEAs had very high
internal cell resistance. This concludes the descriptions of the MEAs that were made with
membranes with poor properties or unfortunate material dimensions (too thick), that formed MEAs

107

simply not worth evaluating due to prohibitively high internal cell resistance. The fabrication of
membranes was described in the previous chapter, but a more specific procedure is included here
for at least 1 membrane of a working MEA.

5.10 Fabrication of MEA-5
The first MEA that met the evaluation criteria of <100 mΩ of internal cell resistance was
MEA-5, which was an MEA fabricated with a 30.4 wt% PSSA-PVDF membrane that was 80 μm
thick. The sulfonate copolymer used in the blended membrane for MEA-5 was a poly(TBASS-
co-styrene-co-CMS) 80/10/10 terpolymer reported in the previous chapter. The membrane was
made by mixing 0.59 g of the 80/10/10 terpolymer of TBASS with 0.41 g of PVDF powder into a
sample vial. Then 9.0 g of DMF was added into the vial. The vial was allowed to rest for 15 mins
to allow the terpolymer to dissolve. Then once the terpolymer was dissolved, the vial was heated
to 40-50 °C using a heat gun and shaken to dissolve the remaining PVDF. Once all solids were
dissolved, and the vial cooled back down to room temperature, the viscous polymer composite
solution was poured onto a flat glass plate. Then the diameter of the circular puddle of the viscous
solution was increased 3-fold by spreading the puddle by tilting the glass plate to use gravity. This
was done to decrease the membrane thickness to below 100 μm. The glass plate with the sample
on top as a wet solution was then heated to 165 °C for 2 hour in an oven. Then the membrane was
retrieved from the glass plate by hydration-induced swelling, immediately after removal from the
oven. This was done just by spraying water over the top of the membrane and allowing it to
hydrate and swell slowly over 30 mins. The membrane peeled from the substrate easily and was
placed into a vial containing 1.0 M H2SO4 and 1.0 g of Alconox of a volume that practically fills
the entire sample vial. The vial was then sealed with the cap and then heated to 90 °C for 12 hrs

108

(6 hrs if using Alconox). The vial was then decanted, and the membrane was subjected to dialysis
with DI water to remove all excess acid and detergent.

5.11 MEA-5 Fuel Cell Results
This section includes performance data for MEA-5 operated as a DMFC at cell
temperatures of 30, 60, and 90 °C. Oxidant streams used were humidified pure oxygen or
humidified air, both at ambient pressures. Methanol concentrations used were between 1.0 and
17.1 M. The 17.1 M represents an equimolar mixture of methanol and water. It was noted that at
cell temperatures above 60 °C, the methanol solutions that were 4.0 M or above in concentration,
appeared to be boiling out of the anode outlet. For this reason, for MEA-5, data for the 4.0, 5.0,
and 17.1 M at 90 °C, was not acquired.

Figure 5.6. Polarization curve of MEA-5 at 30 °C with ambient pressures of oxygen and
different methanol concentration.
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
Cell Potential (V)
Current Density (mA/cm
2
)
30 °C, O
2
2M MeOH
3M MeOH
4M MeOH
5M MeOH
17.1M MeOH

109

Figure 5.7. Power density curve of MEA-5 at 30 °C with ambient pressures of pure oxygen
and different methanol concentration.

Figure 5.8. Power density curve of MEA-5 at 30 °C with ambient pressures of air and
different methanol concentration.

0
10
20
30
40
50
60
70
0 100 200 300 400 500
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
30 °C, O
2
2M MeOH
3M MeOH
4M MeOH
5M MeOH
17.1M MeOH
0
5
10
15
20
25
30
35
40
45
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 50 100 150 200 250 300
Powe Density (mW/cm
2
)
Cell Potential (V)
Current Density (mA/cm
2
)
30 °C, air
2M MeOH potential
3M MeOH potential
2M MeOH power
3M MeOH power

110

Figure 5.9. Polarization curve of MEA-5 at 60 °C and ambient pressures of oxygen and
different methanol concentrations.

Figure 5.10. Power density curve of MEA-5 at 60 °C and ambient pressures of pure oxygen
and different methanol concentration.

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400 500 600 700 800 900 1000
Cell Potential (V)
Current Density (mA/cm
2
)
60 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH
0
20
40
60
80
100
120
140
0 100 200 300 400 500 600 700 800 900 1000
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
60 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH

111

Figure 5.11. Power density curve of MEA-5 at 60 °C and ambient pressures of air and
different methanol concentration.

Figure 5.12. Polarization curve of MEA-5 at 90 °C and ambient pressures of pure oxygen
and different methanol concentration.

0
10
20
30
40
50
60
70
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 50 100 150 200 250 300 350 Power Density (mW/cm
2
)
Cell Potential (V)
Current Density (mA/cm
2
)
60 °C, air
1M MeOH potential
2M MeOH potential
3M MeOH potential
1M MeOH power
2M MeOH power
3M MeOH power
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 200 400 600 800 1000
Cell Potential (V)
Current Density (mA/cm
2
)
90 °C, O
2
1M MeOH
2M MeOH
3M MeOH

112

Figure 5.13. Power density curve of MEA-5 at 90 °C and ambient pressures of pure oxygen
and different methanol concentration.

Figure 5.14. Polarization Curves of MEA-5 at 30 °C with humidified air at ambient
pressure.

0
20
40
60
80
100
120
140
160
180
0 200 400 600 800 1000
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
90 °C, O
2
1M MeOH
2M MeOH
3M MeOH
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 50 100 150 200 250
Cell Potential (V)
Current Density (mA/cm
2
)
MEA-5, 30 °C, air
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH
10M MeOH
15M MeOH
17.1M MeOH

113

Figure 5.15. Power Density Curves of MEA-5 at 30 °C with humidified air at ambient
pressure.

This fuel cell performance was superior to performances acquired by previous
investigations using the PSSA-PVDF membranes via the direct blending method.
46,61
Overall,
previous work used the same catalysts (PtRu black for anode and Pt black for cathode) and catalyst
loading (~8 mg/cm
2
), however, sometimes an active surface area of 25 cm
2
was used instead of 5
cm
2
which was used for the present work. Previous work by Muhkin in 2016 and Li in 2014
achieved max power densities of roughly 100 mW/cm
2
and did not investigate degradation of cell
performance during operation. The enhanced performance of this work was likely due to a variety
of factors one being that we used a purer source of TBASS monomer due to optimizations.
Another reason was careful investigation into the crosslinking mechanism. The reaction
mechanism (discussed in the previous chapter) allowed the conceptualization of better reaction
conditions for synthesizing TBASS copolymers containing CMS. Previous conditions to
synthesize the TBASS copolymer containing CMS operated with a concentration that would all
but guarantee crosslinking. Often, at those conditions, more than 70 % of the copolymer would
become crosslinked to the point of insolubility in all solvents, during copolymerization. After
copolymerization, the crosslinked mass of polymer and solvent can be dried up and crushed into a
0
5
10
15
20
25
30
35
0 50 100 150 200 250
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
MEA-5; 30 °C, air
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH
10M MeOH
15M MeOH
17.1M MeOH

114

fine glassy powder, however, these would still never dissolve in any solvents including DMF
which was used to fabricate blends with PVDF. Membranes made from these highly crosslinked
polymers were inferior due to a total lack of control of copolymer structure and therefore
membrane properties. It was later discovered upon SEC analysis that even with the dilution of
the polymerization system down to 0.75 M, evidence of crosslinking was still present, indicated
by increased apparent molecular weight by factors of 2-4 when compared to an analogous
copolymer without any CMS.

5.12 Repeatability of MEA-5
The repeat of MEA-5 was done by making a new batch of the 801010 terpolymer, followed by
preparing the membrane identically to the membrane used in MEA-5 (see previous section for
membrane fabrication for MEA-5). The thickness of MEA-5 during the second attempt was
roughly 5 μm thinner than the first attempt and had a small excess of 0.05 mg/cm
2
of PtRu Black
on the anode, meaning the anode catalyst layer had an excess of 0.25 mg of PtRu Black.

115

Figure 5.16. Repeatability of MEA-5 tested using 5.0 M methanol at 30 and 60 °C.

The difference in thickness and catalyst amounts between the 2 attempts were both
unintentional, however, and likely due to the painting process. Everything else about the MEA
was identical to the first attempt of MEA-5 and the MEA was evaluated for cell performance. We
see in Figure 5.16, that the cell performance of MEAs made using our PSSA-PVDF membranes,
was relatively repeatable, despite the slight difference of MEA thickness and catalyst loading.
Generally, several factors of MEA fabrication can have deviations based on how they are prepared.
Since the paint brush method was used it could sometimes be difficult to precisely paint the catalyst
slurry onto the substrate in the same manner for each new MEA. For example, because some
catalyst particles will always stick to the bristles of the brush, attempts were made to factor for this
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 100 200 300 400 500 600 700 800
Cell Potential (V)
Current Density (mA/cm
2
)
Repeatability of MEA-5; using 5.0 M MeOH, O
2
MEA-5, 30C
MEA-5 (2nd), 30C
MEA-5, 60C
MEA-5 (2nd), 60C

116

by including a small excess of catalyst when we are formulating the ink. However, when painting,
there is no good control of how much catalyst ink will stick to the bristles of the brush.

5.13 Fabrication of MEA-6
The second membrane electrode assembly investigated, MEA-6, was made similarly to
MEA-5 (identical PSSA wt% of 30.4%), except that 1 wt% graphite oxide was added to the
membrane during casting (see section 5.8 for the membrane fabrication for MEA-5). Another
difference (Table 5.1) was that the membrane was 55 μm thick instead of 80 μm. Since MEA-6
had a membrane that was of similar composition to MEA-5, we wanted to evaluate the
performance of it under conditions of low methanol concentrations (0.5-3.0 M MeOH), with
hydrogen (instead of methanol) as anode fuel, and under constant current conditions with both
hydrogen and methanol solutions.

5.14 MEA-6 Fuel Cell Results

Figure 5.17. Effect of temperature of 0.5 M methanol DMFC on power density for MEA-6

The membrane used for MEA-6 was investigated with 0.5 M methanol and oxygen at
ambient pressures. The data of Figure 5.17, indicated a strong temperature dependence. With 0.5
0
20
40
60
80
100
120
0 100 200 300 400 500
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
0.5M MeOH, O
2
30C
60C
90C

117

M methanol, there should have been less overall crossover of methanol compared to other
concentrations of methanol seen later. However, at low methanol concentrations, the cell could
also have sluggish mass transport of methanol to the active sites in the anode catalyst layer,
especially at high current densities where there is higher demand for methanol at the anode.

Figure 5.18. Effect of 1.0 M methanol on power density for MEA-6.

Figure 5.19. Effect of 2.0 M methanol on power density for MEA-6

0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600 700
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
1.0M MeOH, O
2
30C
60C
90C
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600 700
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
2M MeOH, O
2
30C
60C
90C

118

Figure 5.20. Effect of 3.0 M methanol on power density for MEA-6.

Compared to 0.5 M methanol, 1-3 M methanol revealed that methanol crossover was more
severe with these concentrations, as indicated by Figures 5.15-5.20. The power density’s
dependence on temperature was apparently not as strong between 60 and 90 °C. Interestingly, the
difference in power density between 30 and 60 °C indicates that, between the various methanol
concentrations used, MEA-6 performs best with 2 M methanol and at 60 °C. Due to loss of power
density when increasing methanol concentration, this membrane containing graphite oxide
appeared to have higher MCO. Intriguingly, at 1-3 M methanol, the increased reaction kinetics of
MEA-6, appeared to have been canceled out by increased MCO between the cell temperatures of
60 and 90 °C. Compared to the previous MEA, MEA-6 has this behavior due to the decreased
membrane thickness (Table 5.1). Crossover is not only a function of membrane thickness but also
of methanol concentration at the anode. It is also a function of the rate of methanol consumption
at the anode. In principle, the methanol contacts the catalyst layer before the membrane.
Therefore, if there is enough catalytic activity at the anode, balanced by sufficient catalytic activity
of the ORR at the cathode, the cell could theoretically consume the methanol before it can
crossover.
0
10
20
30
40
50
60
70
80
90
0 100 200 300 400 500 600 700
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
3.0M MeOH, O
2
30C
60C
90C

119

Figure 5.21. Performance of MEA-6 at 60 °C at different methanol concentrations

Figure 5.22. Constant current plot for MEA-6 with 1.0 M methanol

0
20
40
60
80
100
0 100 200 300 400 500 600 700
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
MEA-6 Summary plot at 60 °C, O
2
0.5M MeOH
1M MeOH
2M MeOH
3M MeOH
y = -1.20E-03x + 4.62E+01
0
10
20
30
40
50
0 1000 2000 3000 4000 5000 6000
Power Density (mW/cm
2
)
Time (seconds)
(1.0 M MeOH) DMFC Constant Current
90C (1M MeOH/air) 120 mA/cm2

120

Figure 5.23. Constant current plot for MEA-6 with 5.0 M methanol.

Figures 5.22 and 5.23, indicate that the slope of power density decay over 100 minutes was
smaller in the 5.0 M methanol case versus the 1.0 M methanol case for MEA-6. The constant
current demanded from the cell was different between the 5.0 M and 1.0 M MeOH experiments,
and the decay in power density appears to reflect this. The smaller current requires fewer chemical
reactions per unit time on both electrodes. This meant that the poisoning of the cathode catalyst
from methanol oxidation (consequence of MCO) was likely reduced. This also meant that less
hydrogen peroxide was forming from the oxygen reduction reaction (ORR) at the cathode.

y = -1.07E-04x + 1.51E+01
0
2
4
6
8
10
12
14
16
0 1000 2000 3000 4000 5000 6000
Power Density (mW/cm
2
)
Time (seconds)
(5.0M MeOH) DMFC Constant Current
Power 90C (5M MeOH/air) 80 mA/cm2

121

Figure 5.24. Hydrogen fuel cell performance of MEA-6 at several temperatures using
oxygen at ambient pressures

This is the first hydrogen fuel cell data collected for the PSSA-PVDF membranes
fabricated via direct blending of PVDF with TBASS copolymer containing CMS. The data
indicates that at temperatures exceeding 60 °C, the PSSA-PVDF membrane of MEA-6 appeared
to be drying out and losing proton conductivity based on the loss of power density when heating
the cell from 60 to 80 °C, the power density and the max current density drop considerably.
Interestingly, there was not much difference in performance between 30 and 60 °C.

0
100
200
300
400
500
600
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
PEMFC; O
2
200sccm
30C 60C 80C 60C (after constC exp)

122

Figure 5.25. Constant current plot for MEA-6 using hydrogen at 60 °C and air at ambient
pressures

The constant current experiment for MEA-6 using hydrogen and oxygen at ambient
pressures for the anode and cathode reagents, respectively, we observe relatively stable
performance, as indicated in Figure 5.25 above. We attribute the variations of power density to
the fact that the anode gas diffusion layer (GDL) was not teflonized since the MEA was evaluated
with methanol solutions, mentioned previously. This meant that hydrogen molecules had high
mass transfer resistance due to the pores of the GDL being filled with water. The anode gas stream
was humidified and the pores (GDL porosity of ~78%)
126
of the anode GDL had, macroscopically,
been shown to become partially clogged with water. This meant that condensation of water
droplets could feasibly block out some regions of the GDL surface, reducing mass transport, and
therefore power production was not perfectly stable due to high transport resistance.

0
20
40
60
80
100
120
140
160
0 2000 4000 6000 8000 10000
Power Density (mW/cm
2
)
Time (seconds)
PEMFC Constant Current Experiment
60C (H2/air) 200mA/cm2

123

5.15 Membrane Electrode Assemblies Prepared with Carbon Supported Catalysts
When PSSA-PVDF membranes were prepared with state-of-the-art carbon-supported
catalysts (Pt/C; commonly used in hydrogen fuel cells) it was discovered that the MEAs had
internal resistances greater than 1.0 Ω rather than < 0.1 Ω. This indicated that a triple phase
boundary (TPB) was not formed well enough to allow adequate proton conduction between
electrodes. Even when using 40% PSSA-PVDF membrane that has a proton conductivity of over
120 mS/cm at room temperature, the internal cell resistance of that MEA, using Pt/C, would be
>0.5 Ω. Values above 0.1 Ω were far too high of a cell resistance to produce any meaningful
power from these MEAs. Causes of this high cell resistance are likely due to the binder (Nafion
ionomer) being incompatible with the surface of our PSSA-PVDF membranes, thereby resulting
in a poorly formed triple phase boundary (TPB). This is not surprising since the two materials
have different morphologies. Attempts to improve upon the TPB was not the focus of this work.
However, attempts were made to improve the TPB anyway; the MEAs made with carbon
supported catalysts was further investigated by sulfonating the carbon supports using concentrated
sulfuric acid. The MEAs fabricated with unsulfonated Pt/C appeared to have enhanced
hydrophobic character, likely from the excess carbon. The carbon support was sulfonated with
concentrated sulfuric acid so that it could lower the internal cell resistance of the MEAs. However,
this only decreased the internal cell resistance of the MEAs to about 0.5 Ω, still too high to warrant
a complete evaluation. The maximum power density for the MEAs made using carbon-supported
catalysts was about 20-25 mW/cm
2
(depending on flow rates) and this only got worse with
increasing temperature. Hence efforts with carbon supported catalysts was discontinued in favor
of PtRu black and Pt black electrodes.

124

5.16 Towards Thinner PSSA-PVDF Membranes
The third MEA that met the evaluation criteria of <100 mΩ of internal cell resistance was
MEA-10, which was an MEA fabricated with a 30.4 wt% PVDF-PSSA membrane that was 33 μm
thick. The sulfonate copolymer used in the blended membrane was the 4:1 mole ratio copolymer
of PTBASS and CMS.

5.17 MEA-10 Fuel Cell Results

Figure 5.26. Performance of MEA-10 at 60 °C with different concentrations of methanol

Figure 5.27. Performance of MEA-10 at 80 °C with different concentrations of methanol

0
10
20
30
40
50
60
70
80
90
100
110
0 100 200 300 400 500 600 700 800
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
60 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH
10M MeOH
15M MeOH
24.8M MeOH
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
0 200 400 600 800 1000 1200
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
80 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH

125

Figure 5.28. Constant current experiment at 150 mA/cm
2
of MEA-10 at 60 °C with pure
methanol

The membrane thickness used in MEA-10 was previously attempted, unsuccessfully, using
the 801010 copolymer of 80 mole % TBASS, 10 mole % styrene, and 10 mole % CMS. The
objective for the membrane, to be used for MEA-10, was a thickness less than 50 μm. The 801010
terpolymer of TBASS was able to form a 40 μm thick, uniform membrane. However, it was later
observed from conductivity measurements (and titration), that the 801010 did not fully crosslink
in the membrane during casting and was missing approximately 20% of its conductivity and 15%
of its sulfonate groups. This was likely due to the dramatically decreased thickness of the sample.
To remedy this, a new copolymer, where the styrene fraction of the 801010 terpolymer was
replaced with more CMS (8020 composition), was synthesized to allow for more extensive
crosslinking in our membranes, when made thin. The synthetic procedure for the polymer
composed of 80 mole % TBASS and 20 mole % CMS is described in the previous chapter,
although it is a straightforward replacement of styrene with CMS. The new membrane made using
the 8020 copolymer, was found to retain all of the sulfonate groups into the blend solution, checked
by titration, indicating that the PSSA was fully incorporated into the blend. Results from this
membrane (MEA-10) were noteworthy. Pure methanol was tried for the first time and
surprisingly, with MEA-10 being more than twice as thin as MEA-5 (first and second attempts),
y = -1.01E-03x + 2.12E+01
0
5
10
15
20
25
0 500 1000 1500 2000 2500 3000 3500
Power Density (mW/cm
2
)
Time (s)
24.8M (pure) MeOH 60 °C
O2 1600 sccm
Linear (O2 1600 sccm)

126

the DMFC performed relatively well unlike the corresponding Nafion membranes due to
catastrophically higher methanol crossover (see below).
The very thin membrane used for MEA-10 did well in the fuel cell but only while acquiring
the polarization curves. However, when operating at 1.0, 5.0, or 24.8 M methanol, power was not
stable, even within 1 hour of constant current. This would be consistent with two conditions the
first of which involves MCO that should gradually poison the cathode side catalyst.
127
Second,
there is a possible 2-electron pathway for the oxygen reduction reaction (ORR) that produces
hydrogen peroxide at the cathode instead of water
128
that can diffuse into the membrane. Upon
decomposition into hydroxy radicals, these can attack the methyne hydrogen of the PSSA portion
of the blended membranes.
30
Both effects would be most clear under constant current conditions
and over extended periods. Most polarization and power density curves only required 60 seconds
and almost no measurable degradation of performance could be seen within that timeframe.
Therefore, constant current experiments where a consistent amount of current is demanded from
the cell for longer durations, was required to assess the degradation behaviors of MEAs made using
the PSSA-PVDF membranes. Degradation data was not collected for Nafion membranes. After
the constant current experiments with pure methanol, MEA-10 was no longer adequate for further
evaluation, as the cathode catalyst was poisoned by (MCO).
After careful consideration, CeO2 nanoparticles were introduced into the next MEA. For
example, commercial membrane Nafion XL, includes CeO 2 nanoparticles dispersed inside the
membrane acting as radical scavengers. For some membranes, CeO2 nanoparticles were added
into the blend solution before casting, however, the addition of CeO2 had apparently adversely
affected the internal cell resistance of the MEAs as the resistance was too high (>200 mΩ) to
warrant further evaluation. In addition, the proton conductivity of these ceria containing

127

membranes was around 30 percent lower than the matching unmodified membranes. Since the
peroxide generation occurs at the cathode, we surmised that if the radical scavenger were to be
placed in the catalyst layer, that would be the best option to protect the backbone of the crosslinked
PSSA in our membranes. The mechanism of PSSA degradation is described elsewhere with some
mitigation strategies.
30

To evaluate the effects of including cerium oxide into the MEA, the membrane used for
MEA-10 was prepared with a thickness of 50 μm instead of 34 μm (still using the 8020) but in this
case, both anode and cathode catalyst layers included 0.1 mg/cm
2
of <25 nm particles of CeO2.
This new version of MEA-10, thicker and containing the radical scavenger CeO2, was named
MEA-11.

5.18 MEA-11 Fuel Cell Results

Figure 5.29. Constant Current experiment at 400 mA/cm
2
for MEA-11 at 90 °C and 1.0 M
methanol

The addition of radical scavengers to the MEA system appeared to have allowed the cell
to operate with relatively stable power. We observe that over time, the power increases slightly,
we attribute this to the methanol concentration gradually reducing as the cell consumes the
0
10
20
30
40
50
60
70
0 5000 10000 15000 20000 25000 30000 35000
Power Density (mW/cm
2
)
Time (seconds)
400 mA/cm
2
constant current, 1 M MeOH, MEA-11, 90 °C, O
2

128

methanol from the reservoir. As the methanol concentration in the reservoir gradually decreases
over a long period, the cell can gradually perform better due to less MCO. Eventually, the reservoir
run out of fuel all together, and cell power will begin to decrease due to the lack of fuel. This
began to occur around 19000 seconds, near halfway of the experiment when the reservoir of 1.0
M methanol was replaced with a new solution of 1.0 M methanol. With MEA-11, pure methanol
constant current data was not collected since the power density was already far too low to be stable.
This was due to a higher internal cell resistance of MEA-11, compared to MEA-10. The 1.0 M
methanol constant current data (Figure 5.29) for MEA-11, with its start contrast to MEA-6 (See
Figure 5.22), illustrates the difference of using CeO2 as a radical scavenger over constant cell
operation.

5.19 Fuel Cells with Nafion
To evaluate our membranes in the proton exchange membrane fuel cell (PEMFC) systems,
we thought it necessary to make direct comparisons to the commercial standard Nafion. For initial
DMFCs, Nafion-117 was used due to its increased thickness. The idea of using a thick membrane
was the greater the distance between anode and cathode meant the greater the distance the fuel
must traverse to fully crossover. It was later discovered that the increased thickness had other
drawbacks that offset whatever benefit was gained from the reduced crossover. For example, a
thicker membrane simply uses more material and is inherently less atom economical and adds to
the internal cell resistance (distance between anode and cathode). Also, since there is more
material in the cross-sectional area of the membrane, during swelling, there is enhanced pressure
applied along the normal plane of the membrane and its surroundings. These enhanced pressures
can be the cause of catastrophic failure in the cell, especially at any sharp corners or junctions that

129

come from the flow-field cut into the bipolar plates and/or gaskets of the cell.
121
Nevertheless,
below is the DMFC data that was collected for Nafion-117 and Nafion-212 directly compared with
the MEAs made using PSSA-PVDF membranes that had best performance at the indicated
conditions. The fabrication procedure for the MEAs made using commercial Nafion membranes
were identical to that of the MEAs made using the PSSA-PVDF semi-IPN PEMs. Polarization
and power density curves were collected for MEA-5 when using humidified air at ambient
pressures (Figures 5.14 and 5.15). However, with MEAs made using Nafion membranes, the
performance data with air at the cathode (instead of oxygen) was not collected. Therefore, all
following comparisons of the MEAs made using PSSA-PVDF direct blended membranes, as
opposed to commercial Nafion membranes, were done using humidified oxygen at the cathode
during cell evaluation.

5.20 Direct Comparison of Cell Performance
Performance data for the Nafion-117 MEA and Nafion-212 membranes were collected at
various conditions. The data for the MEAs made using the PSSA-PVDF membranes were
evaluated, often (but not always) at comparable testing conditions. The data below illustrate the
performances of MEAs (comparing commercial membranes to the PSSA-PVDF membranes)
where the only variable is the membrane used. It is stressed again that, all other parameters
dealing with MEA fabrication were kept as consistent as possible between the different
membranes investigated.

130

Figure 5.30. Comparison of MEA-5 and Nafion-117 with 2.0 M MeOH, 30 °C, and oxygen

Figure 5.31. Comparison of MEA-5 and Nafion-117 with 3.0 M MeOH, 30 °C, and oxygen

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 100 200 300 400 500 600 700
Cell Potential (V)
Current Density (mA/cm
2
)
2.0M MeOH, 30 °C, O
2
2M MeOH MEA-5
2M MeOH Nafion-117
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 100 200 300 400 500 600 700
Cell Potential (V)
Current Density (mA/cm
2
)
3.0M MeOH, 30 °C, O
2
3M MeOH MEA-5
3M MeOH Nafion-117

131

Figure 5.32. Comparison of MEA-5 and Nafion-117 with 4.0 M MeOH, 30 °C, and oxygen

Figure 5.33. Comparison of MEA-5 and Nafion-117 with 5.0 M MeOH, 30 °C, and oxygen

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 Potential (V)
Current Density (mA/cm
2
)
4.0M MeOH, 30 °C, O
2
4M MeOH MEA-5
4M MeOH Nafion-117
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 Potential (V)
Current Density (mA/cm
2
)
5.0M MeOH, 30 °C, O
2
5M MeOH MEA-5
5M MeOH Nafion-117

132

At methanol concentrations of 4.0 and 5.0 M and cell temperature of 30 °C, we already
observe that MEA-5 was performing comparably to an analogous MEA made using Nafion-117.
However, only in Figure 5.33, do we see that the open circuit potential (OCV) of MEA-5, surpass
that of Nafion-117. This is probably due to the difference of membrane thickness among other
differences in parameters such as proton conductivity and water content of the membranes. Next,
are the direct comparisons at 60 °C.

Figure 5.34. Comparison of MEA-5, MEA-10, Nafion-117, and Nafion-212 with 1.0 M
MeOH, 60 °C, and oxygen

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 200 400 600 800 1000
Cell Potential (V)
Current Density (mA/cm
2
)
1.0M MeOH, 60 °C, O
2
1M MeOH MEA-5
1M MeOH MEA-10
1M MeOH Nafion-117
1M MeOH Nafion-212

133

Figure 5.35. Comparison of MEA-5, MEA-10, Nafion-117, and Nafion-212 with 2.0 M
MeOH, 60 °C, and oxygen

Figure 5.36. Comparison of MEA-5, MEA-10, Nafion-117, and Nafion-212 with 3.0 M
MeOH, 60 °C, and oxygen

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-100 100 300 500 700 900 1100
Cell Potential (V)
Current Density (mA/cm
2
)
2.0M MeOH, 60 °C, O
2
2M MeOH MEA-5
2M MeOH MEA-10
2M MeOH Nafion-117
2M MeOH Nafion-212
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 200 400 600 800 1000
Cell Potential (V)
Current Density (mA/cm
2
)
3.0M MeOH, 60 °C, O
2
3M MeOH MEA-5
3M MeOH MEA-10
3M MeOH Nafion-117
3M MeOH Nafion-212

134

Figure 5.37. Comparison of MEA-5, MEA-10, Nafion-117, and Nafion-212 with 4.0 M
MeOH, 60 °C, and oxygen

Figure 5.38. Comparison of MEA-5, MEA-10, Nafion-117, and Nafion-212 with 5.0 M
MeOH, 60 °C, and oxygen

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 200 400 600 800
Cell Potential (V)
Current Density (mA/cm
2
)
4.0M MeOH, 60 °C, O
2
4M MeOH MEA-5
4M MeOH MEA-10
4M MeOH Nafion-117
4M MeOH Nafion-212
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 100 200 300 400 500 600 700 800
Cell Potential (V)
Current Density (mA/cm
2
)
5.0M MeOH, 60 °C, O
2
5M MeOH MEA-5
5M MeOH MEA-10
5M MeOH Nafion-117
5M MeOH Nafion-212

135

From the cell performance at 60 °C (Figures 5.34-5.38), MEA-5 consistently had the
highest OCV compared to the other MEAs evaluated, including the ones made with Nafion
membranes. The MEA made using Nafion-117 and MEA-5 were the thickest membranes
evaluated (see Table 5.1) and they both had relatively stable performance as the methanol
concentration was increased.

Figure 5.39. Comparison of MEA-5 and Nafion-117 with different cathode flow rates of
oxygen

0
20
40
60
80
100
120
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 200 400 600 800
Power Density (mW/cm
2
)
Cell Potential (V)
Current Density (mA/cm
2
)
5.0 M MeOH, 60 °C, O
2
MEA-5, 100 mL/min O2 Nafion-117, 100 mL/min O2
MEA-5, 1500 mL/min O2 Nafion-117, 1500 mL/min O2

136

Figure 5.40. Comparison of MEA-5, Nafion-117, and Nafion-212 with 1.0 M MeOH, 90 °C,
and oxygen

Figure 5.41. Comparison of MEA-5, Nafion-117, and Nafion-212 with 2.0 M MeOH, 90 °C,
and oxygen
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 200 400 600 800 1000 1200 1400
Cell Potential (V)
Current Density (mA/cm
2
)
1.0M MeOH, 90 °C, O
2
1M MeOH MEA-5
1M MeOH Nafion-117
1M MeOH Nafion-212
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 200 400 600 800 1000 1200 1400
Cell Potential (V)
Current Density (mA/cm
2
)
2.0M MeOH, 90 °C, O
2
2M MeOH MEA-5
2M MeOH Nafion-117
2M MeOH Nafion-212

137

Figure 5.42. Comparison of MEA-5, Nafion-117, and Nafion-212 with 3.0 M MeOH, 90 °C,
and oxygen

Figure 5.43. Comparison of MEA-10 and Nafion-212 with 10.0 M MeOH, 60 °C, and
oxygen

Figure 5.44. Comparison of MEA-10 and Nafion-212 with 15.0 M MeOH, 60 °C, and
oxygen
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 200 400 600 800 1000 1200 1400
Cell Potential (V)
Current Density (mA/cm
2
)
3.0M MeOH, 90 °C, O
2
3M MeOH MEA-5
3M MeOH Nafion-117
3M MeOH Nafion-212
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 100 200 300 400 500 600 700
Cell Potential (V)
Current Density (mA/cm
2
)
10.0M MeOH, 60 °C, O
2
10M MeOH MEA-10
10M MeOH Nafion-212
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 100 200 300 400 500 600
Cell Potential (V)
Current Density (mA/cm
2
)
15.0M MeOH, 60 °C, O
2
15M MeOH MEA-10
15M MeOH Nafion-212

138

Figure 5.45. Comparison of MEA-10 and Nafion-212 with 24.8 M MeOH, 60 °C, and
oxygen

The plots pertaining only to Nafion-117 and Nafion-212 and their performance with respect
to methanol concentration and cathode flow rates are included in the supplemental section. In
addition to this, MEAs made using Nafion membranes were not evaluated for its performance
under constant current conditions. Typically, the MEAs made using Nafion membranes were too
severely damaged after acquiring polarization and power density curves. For instance, the MEA
made using Nafion-212, had swelled so drastically and became inoperable. It was known that
Nafion membranes had higher methanol permeability compared to PSSA-PVDF membranes of
comparable proton conductivity. Also, the MEA made using the thinner Nafion-212, has more
MCO than Nafion-117 which is clear from the lower OCVs when comparing these two commercial
membranes at identical conditions. The most interesting comparisons of the PSSA-PVDF semi-
IPN PEMs and commercial Nafion membranes, were the data acquired 60 °C as the PSSA-PVDF
membranes had severe water management issues at temperatures at both 80 and 90 °C. The issue
of water management is illustrated by the fuel cell performance (Figure 5.24). From the data at 60
°C (Figures 5.34-5.42), MEA-5 using a 30.4 wt% PSSA-PVDF membrane that was 80 μm thick,
had the best open circuit potential (OCV). The OCV is a qualitative measure of passive methanol
diffusion through the membrane under cell conditions when there is no current or power demanded
0
0.1
0.2
0.3
0.4
0.5
0.6
0 50 100 150 200 250 300 350 400
Cell Potential (V)
Current Density (mA/cm
2
)
24.8M MeOH, 60 °C, O
2
24.8M MeOH MEA-10
24.8M MeOH Nafion-212

139

from the cell. Despite MEA-5 being roughly 100 μm thinner than the MEA made using Nafion-
117, it still had less MCO. It was observed that during the experiment where 15.0 M methanol
and 1500 sccm of O2 was used (Figure S5.15 in supplemental), the MEA made with Nafion-212
began to fail from the excess methanol. During the 24.8 M methanol measurements (Figure
S5.16), we can clearly see the opposite trend (seen in Figures S5.4 through S5.14) of power and
cathode flow rate, suggesting that the MEA was failing in real-time during the measurements in
pure methanol. Finally, immediately after acquiring the last polarization curve for the MEA made
using Nafion-212, the membrane developed a large hole. This was not seen with MEAs made
using the PSSA-PVDF semi-IPN membranes. The PSSA-PVDF semi-IPNs swelled but the MEAs
remained stable enough with high methanol concentration (15.0-17.1 M), and even pure methanol
(Figures 5.6, 5.7, 5.14-5.16, 5.26 and 5.28), for several hours of cell operation and evaluation.
Thus, the PSSA-PVDF semi-IPN membranes tolerate pure methanol without catastrophic
swelling, presumably because the PSSA-PVDF based membranes are both crosslinked (PSSA
copolymer portion) and semi-crystalline (PVDF beta-phase crystallinity portion), both of which
add strength to the material and resist swelling effects. It was noticed that our membranes did
worse with higher cathode flow rates while Nafion membranes generally did better with higher
cathode flow rates (Figure 5.39). This may be due to drying effect at the teflonized cathode GDL.
The non-teflonized anode side, generally, always has an aqueous solution of methanol in contact
in a DMFC. Therefore, drying could only occur at the cathode side. From the previous chapter
we understood that our PSSA-PVDF membranes in relation to Nafion membranes, do not contain
any fraction of conventional water which is “free” and freezable at 0 °C. This meant that all the
water in our membranes participated in proton conduction via a proton hopping mechanism and
there was virtually no vehicular mechanism occurring.
27,51,107,108
The PSSA-PVDF membranes,

140

as they lost water from either increased cell temperatures or higher cathode flow rates (or both),
did not have a molecular reservoir of free, non-solvation, water to draw from, to use to conduct
protons. There is no hydration buffer for these PSSA-PVDF membranes; we predict that proton
conductivity drops rapidly when relative humidity of the membrane falls below 90% RH. This
can further explain why our membranes did not perform well for hydrogen fuel cell, where both
electrodes could experience a drying effect.
5.21 Comparisons to other reports
Many others have reported on DMFC performance. For example, Hosseinpour et al.
obtained a maximum power density of around 75 mW/cm
2
when using multilayers Nafion-212
and Nafion-211 membranes and around 1.0 mg/cm
2
of catalyst.
129
Their DMFC operated at 70
°C, 1.0 M methanol, and air as the oxidant.
129
Goor and Peled et al. described a DMFC that
produced over 180 mW/cm
2
at 80 °C with very low air pressures of 0.05 atm.
18
They used a
nanoporous membrane based on a mixture of poly(tetrafluoroethylene) and nanoparticles of
several inorganic clays and ceramics.
18
They also attributed their enhanced cell performance due
to modification of the flow channels of the bipolar plates.
18
They use a ceramic-polymer
nanocomposite blend where they fill the pores with sulfuric acid to increase the conductivity to
above 200 mS/cm.
130
To achieve an excellent power density, Peled and workers ran their DMFC
with relatively high catalyst loading of unsupported Pt catalyst (5 mg/cm
2
) in each catalyst layer,
a solution of 1.0 M methanol and 1.9 M triflic acid. This allowed their DMFC to run at
favorable fuel concentrations and with lower pH in the membrane to further increase the
maximum current density that can be produced by the MEA.
18
Clearly, others in the literature
also suffer from triple-phase boundary issues, so they too increase the catalyst loading. An
interesting example of a membrane like our PSSA-PVDF semi-IPN was reported by Kumar et al.

141

in 2014.
131
In general, it was observed in the literature that lower catalyst loadings were used for
Nafion membranes, or nafion-containing membranes. This is due to the binder being Nafion-
ionomer and therefore very capable of forming an adequate interface with the surface of the
membrane because they are identical polymers. The outlook for methanol fuel cells looks
promising as methanol tolerant cathode catalysts have also been reported.
127,132
The Pt-alloy
with gold was shown to be more tolerant towards methanol oxidation and performed better than
Pt as the cathode catalyst.
132
This was demonstrated in achieving 120 mW/cm
2
of power with
only 2.0 mg/cm
2
of catalyst loading on each catalyst layer, when using the Pt-alloy of gold vs 80
mW/cm
2
when using Pt under the identical cell conditions of 70 °C.
132
Their active surface area
was 25 cm
2
and their MEA was using Nafion-117.
132
These strategies of improving upon the
cathode catalyst will improve the efficiency of the cell drastically. If the cathode catalysts were
tolerant to methanol, there would be less losses from counter-potentials generated from MCO.
Others have also attempted pure methanol, and their max power densities are also in the 20-30
mW/cm
2
range when using a variety of catalysts, catalyst loadings, membranes, and testing
conditions.
133-134
Many of the DMFCs using pure methanol, however, are using a more complex
vapor fed system or pervaporation system requiring an additional membrane.
133,135
Conclusions
Membranes based on PSSA-PVDF were accessed through the direct blending of a TBASS
copolymer containing CMS and PVDF and tested as methanol fuel cells. Performance data in this
work compared to similar previous membranes in part due to better TBASS puirity.
46,61
With
these PSSA-PVDF membranes, fuel cell performance data was collected with hydrogen, methanol
at very high concentrations (>5.0 M), and constant current data were collected. Select membranes
had ionic properties comparable to Nafion were fabricated into MEAs with identical procedures

142

and then their performance was evaluated for application in DMFCs. The PSSA-PVDF
membranes were severely affected by water management issues but generally these issues can be
optimized through engineering. Nafion membranes catastrophically swelled in pure methanol and
therefore durability data for Nafion could not be acquired when using pure methanol. The PSSA-
PVDF membranes, in contrast, have power densities are more stable than Nafion membranes, in
all methanol concentrations. Under identical methanol concentrations and oxidant, the PSSA-
PVDF membranes had a higher open circuit voltage (OCV), especially at elevated temperatures.
The OCV is indicative of the lower MCO of the PSSA-PVDF membranes compared to commercial
Nafion membranes at comparable proton conductivities and methanol concentrations. The PSSA-
PVDF membranes, however, usually had a larger immediate drop in cell potential. This indicated
more resistance associated with the triple phase boundaries and a larger contribution towards
activation losses. The robustness against degradation of both PSSA-PVDF semi-IPNs and
commercial Nafion membranes can be improved upon by using ceria as a radical scavenger to
protect the membranes from side products of the oxygen reduction reaction (ORR) occurring at
the cathode. Future work investigation should address issues dealing with formation of the TPB
and water management of the MEA.

143

5.22 Supplemental Section
Nafion-117 Fuel Cell Results

Figure S5.1. Polarization curve of Nafion-117 at 60 °C with different methanol
concentrations

Figure S5.2. Power density curve of Nafion-117 at 60 °C with different methanol
concentrations

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 100 200 300 400 500 600 700 800 900 1000
Cell Potential (V)
Current Density (mA/cm
2
)
60 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH
0
20
40
60
80
100
120
140
160
0 100 200 300 400 500 600 700 800 900 1000 Power Density (mW/cm2)
Current Density (mA/cm
2
)
60 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH

144

Figure S5.3. Polarization and power density curves of Nafion-117 at 90 °C with different
methanol concentrations

Nafion-212 Fuel Cell results

Figure S5.4. Performance of Nafion-212 at 60 °C and 1.0 M methanol and different
cathode flow rates
0
25
50
75
100
125
150
175
200
225
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 200 400 600 800 1000 1200
Power Density (mW/cm
2
)
Cell Potential (V)
Current Density (mA/cm
2
)
Nafion-117, 8mg/cm
2
; 90 °C
1M MeOH cell potential
2M MeOH cell potential
3M MeOH cell potential
1M MeOH power density
2M MeOH power density
3M MeOH power density
0
20
40
60
80
100
120
140
0 100 200 300 400 500 600 700 800
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
1M MeOH 60 °C, O
2
100 sccm
200 sccm
300 sccm
400 sccm
500 sccm

145

Figure S5.5. Performance of Nafion-212 at 90 °C and 1.0 M methanol and different
cathode flow rates

Figure S5.6. Performance of Nafion-212 at 60 °C and 2.0 M methanol and different
cathode flow rates
0
50
100
150
200
250
300
0 200 400 600 800 1000 1200 1400
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
1M MeOH 90 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm
1100 sccm
0
20
40
60
80
100
120
140
160
0 200 400 600 800 1000 1200
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
2M MeOH 60 °C, O
2
300 sccm
500 sccm
700 sccm

146

Figure S5.7. Performance of Nafion-212 at 90 °C and 2.0 M methanol and different
cathode flow rates

Figure S5.8. Performance of Nafion-212 at 60 °C and 3.0 M methanol and different
cathode flow rates

0
50
100
150
200
250
0 200 400 600 800 1000 1200
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
2M MeOH 90 °C, O
2
500 sccm
700 sccm
900 sccm
1100 sccm
1300 sccm
0
20
40
60
80
100
120
140
0 100 200 300 400 500 600 700 800
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
3M MeOH 60 °C, O
2
300 sccm
500 sccm
700 sccm

147

Figure S5.9. Performance of Nafion-212 at 90 °C and 3.0 M methanol and different
cathode flow rates

Figure S5.10. Performance of Nafion-212 at 60 °C and 4.0 M methanol and different
cathode flow rates

0
20
40
60
80
100
120
140
160
180
0 100 200 300 400 500 600 700 800 900 1000
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
3M MeOH 90 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm
1100 sccm
0
10
20
30
40
50
60
70
0 100 200 300 400 500 600
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
4M MeOH 60 °C, O
2
300 sccm
500 sccm
700 sccm

148

Figure S5.11. Performance of Nafion-212 at 90 °C and 4.0 M methanol and different
cathode flow rates

Figure S5.12. Performance of Nafion-212 at 60 °C and 5.0 M methanol and different
cathode flow rates

0
20
40
60
80
100
120
0 100 200 300 400 500 600 700 800 900 1000
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
4M MeOH 90 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm
1100 sccm
1300 sccm
1500 sccm
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600 700 800 900
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
5M MeOH 60 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm

149

Figure S5.13. Performance of Nafion-212 at 90 °C and 5.0 M methanol and different
cathode flow rates

Figure S5.14. Performance of Nafion-212 at 60 °C and 10.0 M methanol and different
cathode flow rates
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600 700 800 900
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
5M MeOH 90 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm
1100 sccm
1300 sccm
0
10
20
30
40
50
60
70
0 100 200 300 400 500 600 700
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
10M MeOH 60 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm
1100 sccm
1300 sccm
1500 sccm

150

Figure S5.15. Performance of Nafion-212 at 60 °C and 15.0 M methanol and different
cathode flow rates

Figure S5.16. Performance of Nafion-212 at 60 °C and pure methanol and different
cathode flow rates

0
5
10
15
20
25
30
35
40
0 50 100 150 200 250 300 350 400 450
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
15M MeOH 60 °C, O
2
500 sccm
700 sccm
900 sccm
1100 sccm
1500 sccm
0
2
4
6
8
10
12
14
0 20 40 60 80 100 120 140 160 180 200
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
24.8M (pure) MeOH 60 °C, O
2
300 sccm
500 sccm
700 sccm
900 sccm
1100 sccm

151

Figure S5.17. Performance of Nafion-212 at 60 °C and with different methanol
concentrations

Figure S5.18. Performance of Nafion-212 at 90 °C and with different methanol
concentrations

0
20
40
60
80
100
120
140
160
0 200 400 600 800 1000 1200
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
Summary Plot at 60 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH
10M MeOH
15M MeOH
24.8M (pure) MeOH
0
50
100
150
200
250
0 200 400 600 800 1000 1200 1400
Power Density (mW/cm
2
)
Current Density (mA/cm
2
)
Summary plot at 90 °C, O
2
1M MeOH
2M MeOH
3M MeOH
4M MeOH
5M MeOH

152

Bibliography
(1) Vijaya, V. S.; Raman, S.; Iniyan, R. G. A Review of Climate Change, Mitigation and
Adaptation. Renew. Sustain. Energy Rev. 2012, 16 (1), 878–897.
(2) Ritchie, H.; Roser, M.; Rosado, P. Fossil Fuels. Our World Data 2020.
(3) Sourmehi, C. EIA Projects Nearly 50% Increase in World Energy Use by 2050, Led by
Growth in Renewables.
(4) Kahan, A. EIA Projects Nearly 50% Increase in World Energy Usage by 2050, Led by
Growth in Asia.
(5) Climate change: a threat to human wellbeing and health of the planet. Taking action now
can secure our future — IPCC https://www.ipcc.ch/2022/02/28/pr-wgii-ar6/ (accessed Sep
20, 2022).
(6) Monastersky, B. Y. R. Near Worrisome Milestone. Nature 2013, 497, 13–14.
(7) Sadiku, M. N. O. Elements of Electromagnetics, 4th ed.; Oxford University Press: New
York & Oxford, 2007.
(8) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol
Economy: Second Edition. Beyond Oil Gas Methanol Econ. Second Ed. 2009, 1–334.
https://doi.org/10.1002/9783527627806.
(9) S.Surampudi, S. R. Narayanan, E. Vamos, H. Frank and G. Halpert, A. Laconti, J. Kosek,
G. K. Surya Prakash, and G. A. O. Advances in Direct Methanol Fuel Cells. J. Power
Sources 1994, 47, 377–385.
(10) Surampudi, S.; Narayanan, S.; Vamos, E.; Frank, H. A.; Halpert, G.; Olah, G. A.; Prakash,
G. K. S. Aqueous Liquid Feed Organic Fuel Cell Using Solid Polymer Electrolyte
Membrane, 1993.
(11) Surampudi, S.; Narayanan, S.; Vamos, E.; Frank, H. A.; Halpert, G.; Olah, G. A.; Prakash,
G. K. S. Direct Methanol Feed Fuel Cell and System, 1998.
(12) Demirci, U. B. Direct Liquid-Feed Fuel Cells: Thermodynamic and Environmental
Concerns. J. Power Sources 2007, 169 (2), 239–246.
https://doi.org/10.1016/J.JPOWSOUR.2007.03.050.
(13) Scott, K.; Xing, L. Direct Methanol Fuel Cells. Adv. Chem. Eng. 2012, 41, 145–196.
https://doi.org/10.1016/B978-0-12-386874-9.00005-1.
(14) Considerations for Fuel Cell Design https://www.fuelcellstore.com/blog-
section/considerations-for-fuel-cell-design (accessed Sep 20, 2022).
(15) Carrette, L.; Friedrich, K. A.; Stimming, U. Fuel Cells - Fundamentals and Applications.

153

Fuel Cells 2001, 1 (1), 5–39. https://doi.org/10.1002/1615-6854(200105)1:1<5::aid-
fuce5>3.0.co;2-g.
(16) Joglekar, M.; Nguyen, V.; Pylypenko, S. Organometallic Complexes Anchored to
Conductive Carbon for Electrocatalytic Oxidation of Methane at Low Temperature. J. Am.
Chem. Soc. 2016, 138 (1), 116–125.
(17) Chen, C. Y.; Tsao, C. S. Characterization of Electrode Structures and the Related
Performance of Direct Methanol Fuel Cells. Int. J. Hydrogen Energy 2006, 31 (3), 391–
398. https://doi.org/10.1016/J.IJHYDENE.2005.05.012.
(18) Goor, M.; Menkin, S.; Peled, E. High Power Direct Methanol Fuel Cell for Mobility and
Portable Applications. Int. J. Hydrogen Energy 2019, 44 (5), 3138–3143.
https://doi.org/10.1016/j.ijhydene.2018.12.019.
(19) Yajima, T.; Uchida, H.; Watanabe, M. In-Situ ATR-FTIR Spectroscopic Study of Electro-
Oxidation of Methanol and Adsorbed CO at Pt-Ru Alloy. J. Phys. Chem. B 2004, 108 (8),
2654–2659.
https://doi.org/10.1021/JP037215Q/ASSET/IMAGES/LARGE/JP037215QF00008.JPEG.
(20) Li, X.; Chao, T.; Duan, Y. E.; Qu, Y.; Liu, Y.; Tan, Q. High-Performance, Stable, and
Flexible Direct Methanol Fuel Cell Based on a Pre-Swelling Kalium Polyacrylate Gel
Electrolyte and Single-Atom Cathode Catalyst. ACS Sustain. Chem. Eng. 2021, 9 (45),
15138–15146.
https://doi.org/10.1021/ACSSUSCHEMENG.1C03047/SUPPL_FILE/SC1C03047_SI_00
1.PDF.
(21) Zhang, Z.; Liu, J.; Wang, J.; Wang, Q.; Wang, Y.; Wang, K.; Wang, Z.; Gu, M.; Tang, Z.;
Lim, J.; Zhao, T.; Ciucci, F. Single-Atom Catalyst for High-Performance Methanol
Oxidation. Nat. Commun. 2021 121 2021, 12 (1), 1–9. https://doi.org/10.1038/s41467-
021-25562-y.
(22) Kwon, S.; Jin Ham, D.; Kim, T.; Kwon, Y.; Geol Lee, S.; Cho, M. Active Methanol
Oxidation Reaction by Enhanced CO Tolerance on Bimetallic Pt/Ir Electrocatalysts Using
Electronic and Bifunctional Effects. 2018. https://doi.org/10.1021/acsami.8b09053.
(23) Wang, Z. B.; Yin, G. P.; Shao, Y. Y.; Yang, B. Q.; Shi, P. F.; Feng, P. X. Electrochemical
Impedance Studies on Carbon Supported PtRuNi and PtRu Anode Catalysts in Acid
Medium for Direct Methanol Fuel Cell. J. Power Sources 2007, 165 (1), 9–15.
https://doi.org/10.1016/J.JPOWSOUR.2006.12.027.
(24) Cheryl Hogue. How to Say Goodbye to PFAS. C&EN Glob. Enterp. 2019, 97 (46), 22–25.
https://doi.org/10.1021/CEN-09746-FEATURE2.
(25) Othman, M. H. D.; Ismail, A. F.; Mustafa, A. Physico-Chemical Study of Sulfonated
Poly(Ether Ether Ketone) Membranes for Direct Methanol Fuel Cell Application.
Malaysian Polym. J. 2007, 2 (1), 10–28.

154

(26) Awang, N.; Jaafar, J.; Ismail, A. F. Thermal Stability and Water Content Study of Void-
Free Electrospun SPEEK/Cloisite Membrane for Direct Methanol Fuel Cell Application.
Polymers (Basel). 2018, 10 (2). https://doi.org/10.3390/polym10020194.
(27) Rambabu, G.; Bhat, S. D. Sulfonated Fullerene in SPEEK Matrix and Its Impact on the
Membrane Electrolyte Properties in Direct Methanol Fuel Cells. Electrochim. Acta 2015,
176, 657–669. https://doi.org/10.1016/J.ELECTACTA.2015.07.045.
(28) Ilbeygi, H.; Ismail, A. F.; Mayahi, A.; Nasef, M. M.; Jaafar, J.; Jalalvandi, E. Transport
Properties and Direct Methanol Fuel Cell Performance of Sulfonated Poly (Ether Ether
Ketone)/Cloisite/Triaminopyrimidine Nanocomposite Polymer Electrolyte Membrane at
Moderate Temperature. Sep. Purif. Technol. 2013, 118, 567–575.
https://doi.org/10.1016/J.SEPPUR.2013.07.044.
(29) 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. High Efficiency
Direct Methanol Fuel Cell Based on Poly(Styrenesulfonic) Acid (PSSA)-Poly(Vinylidene
Fluoride) (PVDF) Composite Membranes. J. Fluor. Chem. 2004, 125 (8), 1217–1230.
https://doi.org/10.1016/j.jfluchem.2004.05.019.
(30) Yu, J.; Yi, B.; Xing, D.; Liu, F.; Shao, Z.; Fu, Y.; Zhang, H. Degradation Mechanism of
Polystyrene Sulfonic Acid Membrane and Application of Its Composite Membranes in
Fuel Cells. Phys. Chem. Chem. Phys. 2003, 5 (3), 611–615.
https://doi.org/10.1039/b209020a.
(31) Kang, S.; Jung, D.; Shin, J.; Lim, S.; Kim, S. K.; Shul, Y.; Peck, D. H. Long-Term
Durability of Radiation-Grafted PFA-g-PSSA Membranes for Direct Methanol Fuel Cells.
J. Memb. Sci. 2013, 447, 36–42. https://doi.org/10.1016/J.MEMSCI.2013.07.005.
(32) Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; Mcgrath, J. E. Alternative
Polymer Systems for Proton Exchange Membranes (PEMs). Chem. Rev. 2004, 104, 4587–
4612. https://doi.org/10.1021/cr020711a.
(33) Shi, Z.; Holdcroft, S. Synthesis and Proton Conductivity of Partially Sulfonated
Poly([Vinylidene Difluoride-Co-Hexafluoropropylene]-b-Styrene) Block Copolymers.
Macromolecules 2005, 38, 4193–4201. https://doi.org/10.1021/ma0477549.
(34) Ding, J.; Chuy, C.; Holdcroft, S. A Self-Organized Network of Nanochannels Enhances
Ion Conductivity through Polymer Films. Chem. Mater. 2001.
https://doi.org/10.1021/cm010144s.
(35) Hou, J.; Li, J.; Mountz, D.; Hull, M.; Madsen, L. A. Correlating Morphology, Proton
Conductivity, and Water Transport in Polyelectrolyte-Fluoropolymer Blend Membranes.
J. Memb. Sci. 2013, 448, 292–299. https://doi.org/10.1016/j.memsci.2013.08.019.
(36) Sahu, S. K.; Meshram, P.; Pandey, B. D.; Kumar, V.; Mankhand, T. R. Removal of
Chromium(III) by Cation Exchange Resin, Indion 790 for Tannery Waste Treatment.
Hydrometallurgy 2009, 99 (3–4), 170–174.

155

https://doi.org/10.1016/j.hydromet.2009.08.002.
(37) Aranda, P.; Chen, W. J.; Martin, C. R. Water Transport Across Polystyrenesulfonate
Alumina Composite Membranes. J. Memb. Sci. 1995, 99 (2), 185–195.
https://doi.org/10.1016/0376-7388(94)00214-J.
(38) El-Hosiny, F. I.; Gad, E. A. M. Effect of Some Superplasticizers of the Mechanical and
Physicochemical Properties of Blended Cement Pastes. J. Appl. Polym. Sci. 1995, 56,
153–159. https://doi.org/https://doi.org/10.1002/app.1995.070560205.
(39) Lee, W. J.; Jung, H. R.; Lee, M. S.; Kim, J. H.; Yang, K. S. Preparation and Ionic
Conductivity of Sulfonated-SEBS/SiO2/Plasticizer Composite Polymer Electrolyte for
Polymer Battery. Solid State Ionics 2003, 164 (1–2), 65–72.
https://doi.org/10.1016/S0167-2738(03)00298-4.
(40) Harvey, A. W.; Stegeman, G. The Sulfonation of Benzene. Ind. Eng. Chem. 1924, 16,
842–845. https://doi.org/10.1021/ie50176a030.
(41) Proceedings of the Chemical Society, Vol. 6, No. 85. Proc. Chem. Soc. London 1890, 6
(85), 95–106. https://doi.org/10.1039/PL8900600095.
(42) Coughlin, J. E.; Reisch, A.; Markarian, M. Z.; Schlenoff, J. B. Sulfonation of Polystyrene:
Toward the “Ideal” Polyelectrolyte. J. Polym. Sci. Part A Polym. Chem. 2013, 51 (11),
2416–2424. https://doi.org/10.1002/POLA.26627.
(43) Ramadan, K. S.; Sameoto, D.; Evoy, S. A Review of Piezoelectric Polymers as Functional
Materials for Electromechanical Transducers. Smart Mater. Struct. 2014, 23 (3), 033001.
https://doi.org/10.1088/0964-1726/23/3/033001.
(44) Reichstein, P. M.; Brendel, J. C.; Drechsler, M.; Thelakkat, M. Poly(3-Hexylthiophene)-
Block-Poly(Tetrabutylammonium-4-Styrenesulfonate) Block Copolymer Micelles for the
Synthesis of Polymer Semiconductor Nanocomposites. 2019.
https://doi.org/10.1021/acsanm.9b00108.
(45) Sunkara, H. B.; Jethmalani, J. M.; Ford, W. T. Synthesis of Cross-Linked Poly(Styrene-
Co-Sodium Styrene-Sulfonate) Latexes. J. Polym. Sci. Part A-Polymer Chem. 1994, 32
(8), 1431–1435. https://doi.org/10.1002/pola.1994.080320804.
(46) Li, M. Development of Polystyrene Sulfonic Acid-Polyvinylidene Fluoride (PSSA-PVDF)
Blends for Direct Methanol Fuel Cells (DMFCs); 2014.
(47) SAINT+ Vol. 8.27B; Bruker AXS: Madison, WI, 2011.
(48) SADABS Vol. 2012-1; Bruker AXS: Madison, WI, 2012.
(49) AG. M. Sheldrick, Acta Crystallogr A 2008, 64, 112-122; BG. M. Sheldrick, Acta
Crystallogr C 2015, 71, 3-8; CC. B. Hubschle, G. M. Sheldrick, B. Dittrich, J Appl
Crystallogr 2011, 44, 1281-1284; DSHELXT, G. M. Sheldrick, 2012; ESHELXTL 2014/7;

156

Bruker AXS: Madison, WI, 2014.
(50) Mercury CSD; CCDC, 2017.
(51) Dai, C. A.; Liu, C. P.; Lee, Y. H.; Chang, C. J.; Chao, C. Y.; Cheng, Y. Y. Fabrication of
Novel Proton Exchange Membranes for DMFC via UV Curing. J. Power Sources 2008,
177 (2), 262–272. https://doi.org/10.1016/J.JPOWSOUR.2007.11.097.
(52) Pan, J.; Liu, L.; Tao, Y.; Zhao, L.; Yu, X.; Wu, B.; Zhao, X.; Liu, L. Green Fabrication of
Tertrabutylammonium Styrene Sulfonate Cation-Exchange Membranes via a Solvent-Free
Photopolymerization Strategy. Ind. Eng. Chem. Res. 2021, 60 (47), 17055–17064.
https://doi.org/10.1021/acs.iecr.1c03274.
(53) Brandon, J. K.; Brown, I. D. Crystal Structure of Cesium Methylsulfonate, CsCH3SO3.
Can. J. Chem. 1966, 45 (12), 1385–1390. https://doi.org/10.1139/V67-229.
(54) Bolte, M.; Griesinger, C.; Sakhaii, P. Pyridinium Methylsulfonate: A Pseudosymmetric
Structure. Acta Crystallogr. Sect. E Struct. Reports Online 2001, 57 (5), o458–o460.
https://doi.org/10.1107/S1600536801006894/CI6022ISUP2.HKL.
(55) Sun, B.; Zhao, Y.; Wu, J. G.; Yang, Q. C.; Xu, G. X. Crystal Structure and FT-IR Study of
Cesium 4-Methylbenzenesulfonate. J. Mol. Struct. 1998, 471 (1–3), 63–66.
https://doi.org/10.1016/S0022-2860(98)00385-8.
(56) Sindhusha, S.; Padma, C. M.; Gunasekaran, B. Crystal Structure, Spectroscopic
Characterization, Mechanical, Thermal and Theoretical Investigations on Creatininium
Benzenesulfonate - A New Organic NLO Single Crystal. J. Mol. Struct. 2020, 1221,
128863. https://doi.org/10.1016/J.MOLSTRUC.2020.128863.
(57) Amirthakumar, C.; Valarmathi, B.; Chakkaravarthi, G.; Kumar, R. M. Synthesis,
Structural, Crystal Growth, Electrical and Mechanical Properties of Potassium 4-Methyl
Benzene Sulphonate. Bull. Mater. Sci. 2020, 43 (1), 1–10. https://doi.org/10.1007/S12034-
020-02241-0/TABLES/4.
(58) Broomhead, J. M.; Nicol, A. D. I. Crystal Structures of Zinc and Magnesium Benzene
Sulphonates. Nature 1947, 160 (4075), 795. https://doi.org/10.1038/160795a0.
(59) Wang, P.; Wang, P.; Bai, M. Crystal Structure of Tetrapropylammonium Hydrogen
Carbonate, C13H29NO3. Zeitschrift fur Krist. - New Cryst. Struct. 2017, 232 (5), 713–
714. https://doi.org/10.1515/NCRS-2016-0361/DOWNLOADASSET/SUPPL/NCRS-
2016-0361_SUPPL.CIF.
(60) Wang, Y.; Chen, Y.; Gao, J.; Yoon, H. G.; Jin, L.; Forsyth, M.; Dingemans, T. J.; Madsen,
L. A. Highly Conductive and Thermally Stable Ion Gels with Tunable Anisotropy and
Modulus. Adv. Mater. 2016, 2571–2578. https://doi.org/10.1002/adma.201505183.
(61) Mukhin, S. Blends of Polystyrene Sulfonic Acid Copolymers and Polyvinylidene Fluoride
as Polyelectrolyte Membranes; 2016.

157

(62) Trommsdorff, V. E.; Köhle, H.; Lagally, P. Zur Polymerisation Des
Methacrylsäuremethylesters. Die Makromol. Chemie 1948, 1 (3), 169–198.
https://doi.org/10.1002/MACP.1948.020010301.
(63) Young, A. M.; Garcia, R.; Higgins, J. S.; Timbo, A. M.; Peiffer, D. G. Dynamic Light
Scattering and Rheology of Associating Sulfonated Polystyrene Ionomers in Non-Polar
Solvents. Polymer (Guildf). 1998, 39 (8–9), 1525–1532. https://doi.org/10.1016/S0032-
3861(97)00452-7.
(64) Trommsdorff, V. E.; Köhle, H.; Lagally, P. Zur Polymerisation Des
Methacrylsäuremethylesters1. Die Makromol. Chemie 1948, 1 (3), 169–198.
https://doi.org/10.1002/MACP.1948.020010301.
(65) Zhang, L.; Eisenberg, A. Multiple Morphologies of “Crew-Cut” Aggregates of
Polystyrene-b-Poly(Acrylic Acid) Block Copolymers. Science (80-. ). 1995, 268 (5218),
1728–1731. https://doi.org/10.1126/SCIENCE.268.5218.1728.
(66) Hadjichristidis, N.; Pitsikalis, M.; Pispas, S.; Iatrou, H. Polymers with Complex
Architecture by Living Anionic Polymerization. Chem. Rev. 2001, 101 (12), 3747–3792.
https://doi.org/10.1021/CR9901337.
(67) Weiss, R. A.; Turner, S. R.; Lundberg, R. D. Sulfonated Polystyrene Ionomers Prepared
by Emulsion Copolymerization of Styrene and Sodium Styrene Sulfonate. J. Polym. Sci.
Polym. Chem. Ed. 1985, 23 (2), 525–533. https://doi.org/10.1002/POL.1985.170230226.
(68) Mansky, P.; Liu, Y.; Huang, E.; Russell, T. P.; Hawker, C. Controlling Polymer-Surface
Interactions with Random Copolymer Brushes. Science (80-. ). 1997, 275 (5305), 1458–
1460. https://doi.org/10.1126/SCIENCE.275.5305.1458.
(69) Abu, H.; Shaari, H.; Johar, N.; Mukhsin, N. Synthesis and Characterization of PS-b-
PDMS Block Copolymer. 2018, 2031, 20006. https://doi.org/10.1063/1.5066962.
(70) Weiss, R. A.; Lundberg, R. D.; Turner, S. R. Comparisons of Styrene Ionomers Prepared
by Sulfonating Polystyrene and Copolymerizing Styrene with Styrene Sulfonate. J. Polym.
Sci. Polym. Chem. Ed. 1985, 23 (2), 549–568.
https://doi.org/10.1002/POL.1985.170230228.
(71) Fares, H. M. M.; Azzam, E. M. S.; Abd El-Salam, H. M. Synthesis and Characterization
of Poly (Styrene-Co-Acrylamide)-Graft-Polyanilines as New Sorbents for Mercuric
Present in Aqueous Hydrocarbon Liquids. Beni-Suef Univ. J. Basic Appl. Sci. 2022, 11
(1), 1–17. https://doi.org/10.1186/S43088-022-00239-7/TABLES/6.
(72) Azza, M.; Abdelfatah, B.; Nahla, M.; Mohamed, F.; Maher, E. Preparation of Emulsion
Polymerization from Styrene Vinylpyrrolidone and Studying Their Thermal Stability and
Electrical Conductivity. J. Nano- Electron. Phys. 2013, 5 (4), 04063–04069.
(73) Wang, P. H.; Pan, C. Y. Preparation of Styrene/Acrylic Acid Copolymer Microspheres:
Polymerization Mechanism and Carboxyl Group Distribution. Colloid Polym. Sci. 2002

158

2802 2002, 280 (2), 152–159. https://doi.org/10.1007/S003960100588.
(74) Balding, P.; Cueto, R.; Russo, P. S.; Gutekunst, W. R. Synthesis of Perfectly Sulfonated
Sodium Polystyrene Sulfonate over a Wide Molar Mass Range via Reversible-
Deactivation Radical Polymerization. J. Polym. Sci. Part A Polym. Chem. 2019, 57 (14),
1527–1537. https://doi.org/10.1002/POLA.29415.
(75) Balding, P.; Borrelli, R.; Volkovinsky, R.; Russo, P. S. Physical Properties of Sodium
Poly(Styrene Sulfonate): Comparison to Incompletely Sulfonated Polystyrene.
Macromolecules 2022, 55 (5), 1747–1762.
https://doi.org/10.1021/ACS.MACROMOL.1C01065/ASSET/IMAGES/MEDIUM/MA1
C01065_M006.GIF.
(76) Wallace, R. A. Glass Transition in Partially Sulfonated Polystyrene. J. Polym. Sci. Part A-
2 Polym. Phys. 1971, 9 (7), 1325–1332. https://doi.org/10.1002/POL.1971.160090711.
(77) Arunbabu, D.; Sanga, Z.; Seenimeera, K. M.; Jana, T. Emulsion Copolymerization of
Styrene and Sodium Styrene Sulfonate: Kinetics, Monomer Reactivity Ratios and
Copolymer Properties. Polym. Int. 2009, 58 (1), 88–96. https://doi.org/10.1002/PI.2497.
(78) Fineman, M.; Ross, S. D.; Fineman, M.; Ross, S. D. Linear Method for Determining
Monomer Reactivity Ratios in Copolymerization. JPoSc 1950, 5 (2), 259–262.
https://doi.org/10.1002/POL.1950.120050210.
(79) Park, K. B.; Jeong, Y. S.; Guzelciftci, B.; Kim, J. S. Two-Stage Pyrolysis of Polystyrene:
Pyrolysis Oil as a Source of Fuels or Benzene, Toluene, Ethylbenzene, and Xylenes. Appl.
Energy 2020, 259, 114240. https://doi.org/10.1016/j.apenergy.2019.114240.
(80) Ordaz-Quintero, A.; Monroy-Alonso, A.; Saldívar-Guerra, E. Thermal Pyrolysis of
Polystyrene Aided by a Nitroxide End-Functionality. Experiments and Modeling. Process.
2020, Vol. 8, Page 432 2020, 8 (4), 432. https://doi.org/10.3390/PR8040432.
(81) Urgen Maul, J.; Frushour, B. G.; Kontoff, J. R.; Eichenauer, H. “Polystyrene and Styrene
Copolymers,” in: Ullmann’s Encyclopedia of Industrial Chemistry.
https://doi.org/10.1002/14356007.a21_615.pub2.
(82) Dhanapal, D.; Xiao, M.; Wang, S.; Meng, Y. A Review on Sulfonated Polymer
Composite/Organic-Inorganic Hybrid Membranes to Address Methanol Barrier Issue for
Methanol Fuel Cells. Nanomaterials 2019, 9 (5). https://doi.org/10.3390/NANO9050668.
(83) Glass, D. E.; Prakash, G. K. S. Effect of the Cathode Catalyst Layer Thickness on the
Performance in Direct Methanol Fuel Cells. Electroanalysis 2019, 31 (4), 718–725.
https://doi.org/10.1002/ELAN.201800628.
(84) Harmer, M. A.; Farneth, W. E.; Sun, Q. High Surface Area Nafion Resin/Silica
Nanocomposites: A New Class of Solid Acid Catalyst. J. Am. Chem. Soc. 1996, 118 (33),
7708–7715.
https://doi.org/10.1021/JA9541950/ASSET/IMAGES/LARGE/JA9541950F00009.JPEG.

159

(85) Piela, P.; Eickes, C.; Brosha, E.; Garzon, F.; Zelenay, P. Ruthenium Crossover in Direct
Methanol Fuel Cell with Pt-Ru Black Anode. J. Electrochem. Soc. 2004, 151 (12), A2053.
https://doi.org/10.1149/1.1814472/XML.
(86) Neburchilov, V.; Martin, J.; Wang, H.; Zhang, J. A Review of Polymer Electrolyte
Membranes for Direct Methanol Fuel Cells. J. Power Sources 2007, 169, 221–238.
https://doi.org/10.1016/j.jpowsour.2007.03.044.
(87) Sheen, R. T.; Kahler, H. L. Effect of Ions of Mohr Method for Chloride Determination.
Ind. Eng. Chem. 1938, 10 (11), 628–629.
(88) Baek, K. Y. Synthesis and Characterization of Sulfonated Block Copolymers by Atom
Transfer Radical Polymerization. J. Polym. Sci. Part A Polym. Chem. 2008, 46 (18),
5991–5998. https://doi.org/10.1002/POLA.22568.
(89) Zhang, J.; Xie, Z.; Zhang, J.; Tang, Y.; Song, C.; Navessin, T.; Shi, Z.; Song, D.; Wang,
H.; Wilkinson, D. P.; Liu, Z. S.; Holdcroft, S. High Temperature PEM Fuel Cells. J.
Power Sources 2006, 160, 872–891. https://doi.org/10.1016/J.JPOWSOUR.2006.05.034.
(90) Kreuer, K. D.; Ise, M.; Fuchs, A.; Maier, J. Proton and Water Transport in Nano-
Separated Polymer Membranes. J. Phys. IV JP 2000, 10 (7).
https://doi.org/10.1051/jp4:2000756.
(91) Suhail, M.; Fang, C. W.; Minhas, M. U.; Wu, P. C. Preparation, Characterization,
Swelling Potential and in-Vitro Evaluation of Sodium Poly(Styrene Sulfonate)-Based
Hydrogels for Controlled Delivery of Ketorolac Tromethamine. Pharmaceuticals 2021, 14
(4). https://doi.org/10.3390/ph14040350.
(92) Lolla, D.; Gorse, J.; Kisielowski, C.; Miao, J.; Taylor, P. L.; Chase, G. G.; Reneker, D. H.
Polyvinylidene Fluoride Molecules in Nanofibers, Imaged at Atomic Scale by Aberration
Corrected Electron Microscopy. Nanoscale 2016, 8 (1), 120–128.
https://doi.org/10.1039/C5NR01619C.
(93) Miao, J.; Reneker, D. H.; Tsige, M.; Taylor, P. L. Molecular Dynamics Simulations and
Morphology Analysis of TEM Imaged PVDF Nanofibers. Polymer (Guildf). 2017, 125,
190–199. https://doi.org/https://doi.org/10.1016/j.polymer.2017.07.086.
(94) Harrison, J. S. Piezoelectric Polymers. 2001.
(95) Dodds, J. S.; Meyers, F. N.; Loh, K. J. Piezoelectric Characterization of PVDF-TrFE Thin
Films Enhanced with ZnO Nanoparticles. IEEE Sens. J. 2012, 12 (6), 1889–1890.
https://doi.org/10.1109/JSEN.2011.2182043.
(96) Zhu, G. D.; Zeng, Z. G.; Zhang, L.; Yan, X. J. Piezoelectricity in β-Phase PVDF Crystals:
A Molecular Simulation Study. Comput. Mater. Sci. 2008, 44 (2), 224–229.
https://doi.org/10.1016/J.COMMATSCI.2008.03.016.
(97) Cai, X.; Lei, T.; Sun, D.; Lin, L. A Critical Analysis of the α, β and γ Phases in

160

Poly(Vinylidene Fluoride) Using FTIR. RSC Adv. 2017, 7 (25), 15382–15389.
https://doi.org/10.1039/C7RA01267E.
(98) Dutta, B.; Kar, E.; Bose, N.; Mukherjee, S. Significant Enhancement of the Electroactive
β-Phase of PVDF by Incorporating Hydrothermally Synthesized Copper Oxide
Nanoparticles. RSC Adv. 2015, 5 (127), 105422–105434.
https://doi.org/10.1039/C5RA21903E.
(99) Danz, R.; Elling, B.; Kunstler, W.; Pinnow, M.; Schmolke, R.; Wedel, A.; Geiss, D. Piezo
and Pyroelectricity and Structure of Doped Polymers. IEEE Trans. Electr. Insul. 1990, 25
(2), 325–330. https://doi.org/10.1109/14.52379.
(100) Satterfield, M. B.; Benziger, J. Mechanical and Water Sorption Properties of Nafion and
Composite Nafion/Titanium Dioxide Membranes for Polymer Electrolyte Membrane Fuel
Cells; 2007.
(101) H. L. Yeager and A. Steck. Cation and Water Diffusion in Nafion Ion Exchange
Membranes: Influence of Polymer Structure. J. Electrochem. Soc. 1981, 128, 1880–1884.
(102) Gloukhovski, R.; Freger, V.; Tsur, Y. Understanding Methods of Preparation and
Characterization of Pore-Filling Polymer Composites for Proton Exchange Membranes: A
Beginner’s Guide. Rev. Chem. Eng. 2018, 34 (4), 455–479. https://doi.org/10.1515/revce-
2016-0065.
(103) Hara, N.; Ohashi, H.; Ito, T.; Yamaguchi, T. Rapid Proton Conduction through
Unfreezable and Bound Water in a Wholly Aromatic Pore-Filling Electrolyte Membrane.
J. Phys. Chem. B 2009, 113, 4656–4663. https://doi.org/10.1021/jp810575u.
(104) Tasaka, M.; Suzuki, S.; Ogawa, Y.; Kamaya, M. Freezing and Nonfreezing Water in
Charged Membranes. J. Memb. Sci. 1988, 38 (2), 175–183. https://doi.org/10.1016/S0376-
7388(00)80878-9.
(105) Ukshe, A.; Glukhov, A.; Dobrovolsky, Y. Percolation Model for Conductivity of
Composites with Segregation of Small Conductive Particles on the Grain Boundaries. J.
Mater. Sci. 2020, 55 (15), 6581–6587. https://doi.org/10.1007/S10853-020-04408-
W/FIGURES/4.
(106) Mahato, N.; Jang, H.; Dhyani, A.; Cho, S. Recent Progress in Conducting Polymers for
Hydrogen Storage and Fuel Cell Applications. Polymers. MDPI AG November 1, 2020,
pp 1–40. https://doi.org/10.3390/polym12112480.
(107) Agmon, N. The Grotthuss Mechanism. Chem. Phys. Lett. 1995, 244 (5–6), 456–462.
https://doi.org/10.1016/0009-2614(95)00905-J.
(108) Chen, C.; Tse, Y. L. S.; Lindberg, G. E.; Knight, C.; Voth, G. A. Hydroxide Solvation and
Transport in Anion Exchange Membranes. J. Am. Chem. Soc. 2016, 138 (3), 991–1000.
https://doi.org/10.1021/jacs.5b11951.

161

(109) Bhosale, A. C.; Ghosh, P. C.; Assaud, L. Preparation Methods of Membrane Electrode
Assemblies for Proton Exchange Membrane Fuel Cells and Unitized Regenerative Fuel
Cells: A Review. Renew. Sustain. Energy Rev. 2020, 133, 110286.
https://doi.org/10.1016/J.RSER.2020.110286.
(110) Chen, S. L.; Krishnan, L.; Srinivasan, S.; Benziger, J.; Bocarsly, A. B. Ion Exchange
Resin/Polystyrene Sulfonate Composite Membranes for PEM Fuel Cells. J. Memb. Sci.
2004, 243 (1–2), 327–333. https://doi.org/10.1016/J.MEMSCI.2004.06.037.
(111) Sahin, O.; Kivrak, H. A Comparative Study of Electrochemical Methods on Pt–Ru DMFC
Anode Catalysts: The Effect of Ru Addition. Int. J. Hydrogen Energy 2013, 38 (2), 901–
909. https://doi.org/10.1016/J.IJHYDENE.2012.10.066.
(112) Roziére, J.; Jones, D. J. Non-Fluorinated Polymer Materials for Proton Exchange
Membrane Fuel Cells. http://dx.doi.org/10.1146/annurev.matsci.33.022702.154657 2003,
33, 503–555. https://doi.org/10.1146/ANNUREV.MATSCI.33.022702.154657.
(113) Sigwadi, R.; Mokrani, T.; Msomi, P.; Nemavhola, F. The Effect of Sulfated Zirconia and
Zirconium Phosphate Nanocomposite Membranes on Fuel-Cell Efficiency. Polymers
(Basel). 2022, 14 (2), 1–18. https://doi.org/10.3390/polym14020263.
(114) Peled, E.; Duvdevani, T.; Aharon, A.; Melman, A. Direct Methanol Fuel Cell Based on a
Novel Low-Cost Nanoporous Proton-Conducting Membrane. Electrochem. Solid-State
Lett. 2000, 3 (12), 525–528. https://doi.org/10.1149/1.1391198.
(115) Laruelle, G.; Nicol, E.; Ameduri, B.; Tassin, J. F.; Ajellal, N. Synthesis of
Poly(Vinylidene Fluoride)-b-Poly(Styrene Sulfonate) Block Copolymers by Controlled
Radical Polymerizations. J. Polym. Sci. Part A Polym. Chem. 2011, 49 (18), 3960–3969.
https://doi.org/10.1002/pola.24836.
(116) US Patents 5,599,638 and 5,773,162.
(117) Iwasita, T.; Vielstich, W. New In-Situ IR Results on Adsorption and Oxidation of
Methanol on Platinum in Acidic Solution. J. Electroanal. Chem. Interfacial Electrochem.
1988, 250 (2), 451–456. https://doi.org/10.1016/0022-0728(88)85184-2.
(118) Lamy, C.; Léger, J.-M.; Srinivasan, S. Direct Methanol Fuel Cells: From a Twentieth
Century Electrochemist’s Dream to a Twenty-First Century Emerging Technology. Mod.
Asp. Electrochem. 2002, 53–118. https://doi.org/10.1007/0-306-46923-5_3.
(119) Glass, D. E.; Olah, G. A.; Prakash, G. K. S. Effect of the Thickness of the Anode
Electrode Catalyst Layers on the Performance in Direct Methanol Fuel Cells. J. Power
Sources 2017, 352, 165–173. https://doi.org/10.1016/J.JPOWSOUR.2017.03.106.
(120) Bieberle, A.; Meier, L. P.; Gauckler ; R. O’hayre, L. J.; Prinz, F. B. The Pt/Nafion/Air
Triple Phase Boundary: Model, Experiment, and Implications for PEM Fuel Cells. J.
Fleig. Annu. Rev. Mater. Res. 1991, 138 (7).

162

(121) Xing, Y.; Li, H.; Avgouropoulos, G. Research Progress of Proton Exchange Membrane
Failure and Mitigation Strategies. Materials (Basel). 2021, 14 (10), 2591.
https://doi.org/10.3390/ma14102591.
(122) Atti, A. R. Development of High Performance Polymer Electrolyte Membranes for Direct
Methanol Fuel Cells. Dissertation 2000.
(123) Kocha, S. S. Handbook of Fuel Cells: Fundamentals, Technology and Applications; 2003.
(124) Liang, Z. X.; Zhao, T. S.; Xu, C.; Xu, J. B. Microscopic Characterizations of Membrane
Electrode Assemblies Prepared under Different Hot-Pressing Conditions. Electrochim.
Acta 2007, 53 (2), 894–902. https://doi.org/10.1016/J.ELECTACTA.2007.07.071.
(125) FuelCellStore. Fuel Cell Schematic.
(126) Sadeghi, E.; Djilali, N.; Bahrami, M. Effect of Compression on the Effective Thermal
Conductivity and Thermal Contact Resistance in PEM Fuel Cell Gas Diffusion Layers.
ASME Int. Mech. Eng. Congr. Expo. Proc. 2012, 7 (PARTS A AND B), 385–393.
https://doi.org/10.1115/IMECE2010-39284.
(127) Lee, D.; Gok, S.; Kim, Y.; Sung, Y.-E.; Lee, E.; Jang, J.-H.; Youn Hwang, J.; Joong
Kwon, O.; Lim, T. Methanol Tolerant Pt−C Core−Shell Cathode Catalyst for Direct
Methanol Fuel Cells. Cite This ACS Appl. Mater. Interfaces 2020, 12, 44596.
https://doi.org/10.1021/acsami.0c07812.
(128) Choi, C. H.; Kwon, H. C.; Yook, S.; Shin, H.; Kim, H.; Choi, M. Hydrogen Peroxide
Synthesis via Enhanced Two-Electron Oxygen Reduction Pathway on Carbon-Coated Pt
Surface. J. Phys. Chem. C 2014, 118 (51), 30063–30070.
https://doi.org/10.1021/JP5113894/SUPPL_FILE/JP5113894_SI_001.PDF.
(129) Hosseinpour, M.; Sahoo, M.; Perez-Page, M. Improving the Performance of Direct
Methanol Fuel Cells by Implementing Multilayer Membranes Blended with Cellulose
Nanocrystals. Int. J. Hydrogen Energy 2019, 44 (57), 30409–30419.
(130) Reichman, S.; Duvdevani, T.; Aharon, A.; Philosoph, M.; Golodnitsky, D.; Peled, E. A
Novel PTFE-Based Proton-Conductive Membrane. J. Power Sources 2006, 153 (2), 228–
233. https://doi.org/10.1016/J.JPOWSOUR.2005.05.038.
(131) Kumar, P.; Dutta, K.; Das, S.; Kundu, P. P. Membrane Prepared by Incorporation of
Crosslinked Sulfonated Polystyrene in the Blend of PVdF-Co-HFP/Nafion: A Preliminary
Evaluation for Application in DMFC. Appl. Energy 2014, 123, 66–74.
https://doi.org/10.1016/j.apenergy.2014.02.060.
(132) Selvarani, G.; Selvaganesh, S. V.; Krishnamurthy, S.; Kiruthika, G. V. M.; Dhar, S.;
Pitchumani, S.; Shukla, A. K. A Methanol-Tolerant Carbon-Supported Pt-Au Alloy
Cathode Catalyst for Direct Methanol Fuel Cells and Its Evaluation by DFT. J. Phys.
Chem. C 2009, 113 (17), 7461–7468. https://doi.org/10.1021/jp810970d.

163

(133) Ali Abdelkareem, M.; Nakagawa, N. DMFC Employing a Porous Plate for an Efficient
Operation at High Methanol Concentrations. J. Power Sources 2006, 162 (1), 114–123.
https://doi.org/10.1016/j.jpowsour.2006.07.012.
(134) Gottesfeld, S. Passive Water Management Techniques in Direct Methanol Fuel Cells, US
Patent 7,282,293., 2007.
(135) Eccarius, S.; Krause, F.; Beard, K.; Agert, C. Passively Operated Vapor-Fed Direct
Methanol Fuel Cells for Portable Applications. J. Power Sources 2008, 182 (2), 565–579.
https://doi.org/10.1016/j.jpowsour.2008.03.091.

Abstract (if available)

Abstract The issue of climate change is briefly discussed with respect to carbon dioxide. The Methanol Economy concept is quickly introduced. A short explanation of energy efficient electrochemical systems is presented. The direct methanol fuel cell (DMFC) system is explored. Materials used in fuel cell systems are presented and discussed. Focus on proton exchange membranes is established. The introductory chapter then foreshadows the contents of the remaining four chapters. Each chapter has its own Supplemental Information section containing Figures, Tables, and sometimes explanations. The bibliography for this work is included after the final chapter at the end.
The second chapter of this work begins with a description of the tetrabutylammonium 4-styrene sulfonate (TBASS) and its novel crystal structure and polymer properties. Strong dipole-dipole interactions in the polymeric form are corroborated by the large interionic distances revealed in its monomeric structure. The electronic dipole effects are doubly demonstrated by both the partial rejection of the polymer from size-exclusion chromatography analysis with strong light scattering response and by the evidence of a Trommsdorff-Norrish type of auto-acceleration when polymerized in 1,2-dichloroethane.
Once the properties of the monomer and its respective homopolymer were established, the following chapter details TBASS copolymers with styrene and methylmethacrylate (MMA), copolymerized in a common organic solvent. For both copolymerization systems, the reactivity ratios were estimated, and it appears that TBASS readily copolymerizes with both styrene and MMA to form moderately alternating copolymers with good control over composition. The TBASS monomer imparts drastic deviations in properties of polystyrene, even at very low mole percent. The pyrolysis of polystyrene and polystyrene containing a few mole percent of TBASS, reveal some of these dramatic differences in polymer properties. Results from size-exclusion chromatography also corroborate the large differences.
With better understanding of the copolymer kinetics and copolymer properties of the TBASS and styrene system, 4-chloromethylstyrene (CMS) was introduced into the copolymer system. Optimizations of the terpolymerization procedure are discussed with previous investigations and findings referred to. Blends of poly(vinylidene fluoride) and TBASS copolymers containing CMS were fabricated by direct blending in a common solvent. To our surprise, crosslinked composite films, mixed at the molecular level, were readily fabricated in under an hour. The film fabrication process does not require the use of any additional crosslinking agents or compatibilizers. It was discovered that the crosslinking step proceeds unimpeded by the absence of conventional benzylation catalysts. In the absence of conventional transition metal catalysts, the crosslinking step is both quantified and further explored using small molecular analogues to remove the effects of polymer chains. A possible mechanism of benzylation in the absence of catalyst is proposed. To convert the composite PTBASS-PVDF semi-interpenetrating network (semi-IPN) films, into PSSA-PVDF semi-IPN membranes, an optimized ion-exchange and dialysis process is presented and discussed. Mechanical investigations on the membranes were done and results presented. Relevant properties are tabulated for the PSSA-PVDF semi-IPN and commercial standard Nafion-117 PEMs. Morphological investigation reveals extremely small domain sizes in the PSSA-PVDF PEMs. Proton conductivity and methanol crossover values of PSSA-PVDF semi-IPN PEMs were compared to commercial standard Nafion-117. Calorimetry experiments reveal that proton conduction in the PSSA-PVDF membranes is achieved entirely through proton hopping, since there are no free water molecules to sustain any level of vehicular transport. In addition to calorimetry, x-ray diffraction (XRD), and attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) measurements deduce that the presence of the ionic sulfonate copolymer, in any amount, enhances the polar and piezoelectric polymorphs of semi-crystalline PVDF and suppresses the non-polar polymorphs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Development of polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)

PDF

Development of polymer electrolyte membranes for fuel cell applications

PDF

Blends of polystyrene sulfonic acid copolymers and polyvinylidene fluoride as polyelectrolyte membranes

PDF

Studies on direct oxidation formic acid fuel cells: advantages, limitations and potential

PDF

Integrated capture and conversion of carbon dioxide from air into methanol and other C1 products

PDF

The modification of catalysts and their supports for use in various fuel cells

PDF

Synthesis and characterization of 3-hexylesterthiophene based random and semi-random polymers and their use in ternary blend solar cells

PDF

Electrocatalysts for direct liquid-feed fuel cells and advanced electrolytes for lithium-ion batteries

PDF

Direct arylation polymerization (DARP) for the synthesis of conjugated polymers

PDF

Direct C−H arylation for the synthesis of conjugated polymers

Asset Metadata

Creator Ung, Adam Bruce (author)

Core Title PSSA-PVDF semi-IPN blends for direct methanol fuel cells

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Degree Conferral Date 2022-12

Publication Date 12/08/2022

Defense Date 09/23/2022

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag direct blend,free radical polymerization,fuel cells,ink preparation,membrane electrode assembly,membranes,methanol crossover,Nafion,OAI-PMH Harvest,platinum,poly(vinylidene fluoride),polyelectrolyte,polystyrene,polystyrenesulfonic acid,proton conductance,proton conductivity,semi-IPN,styrene sulfonate,Trommsdorff-Norrish effect

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Prakash, Surya (committee chair), Narayan, Sri (committee member), Nutt, Steven (committee member)

Creator Email adambruceung@gmail.com,adamung@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC112619800

Unique identifier UC112619800

Identifier etd-UngAdamBru-11349.pdf (filename)

Legacy Identifier etd-UngAdamBru-11349

Document Type Dissertation

Format theses (aat)

Rights Ung, Adam Bruce

Internet Media Type application/pdf

Type texts

Source 20221213-usctheses-batch-995 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu