SYNTHESIS AND CHARACTERIZATION OF HYDROPHILIC-HYDROPHOBIC

DISULFONATED POLY(ARYLENE ETHER SULFONE)-DECAFLUORO BIPHENYL

BASED POLY(ARYLENE ETHER) MULTIBLOCK COPOLYMERS FOR PROTON

EXCHANGE MEMBRANES (PEMS)

by

Xiang Yu

Dissertation submitted to the Faculty of

Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

in Macromolecular Science and Engineering

Dr. James E. McGrath, Chairman

Dr. Judy S. Riffle Dr. John G. Dillard

Dr. Richey M. Davis Dr. Scott W. Case

January 25, 2008 Blacksburg, VA

Key words: Fuel cells, Proton exchange membranes, Multiblock copolymers, Fluorinated

copolymers, Poly(arylene ether sulfone)s, Morphology, Nanophase separation

Copyright 2008, Xiang Yu

Synthesis and Characterization of Hydrophilic-Hydrophobic Disulfonated Poly(Arylene

Ether Sulfone)-Decafluoro Biphenyl Based Poly(Arylene Ether) Multiblock Copolymers for

Proton Exchange Membranes (PEMs)

Xiang Yu

Abstract

Hydrophilic/hydrophobic block copolymers as proton exchange membranes (PEMs) has become

an emerging area of research in recent years. Three series of hydrophilic/hydrophobic,

fluorinated/sulfonated multiblock copolymers were synthesized and characterized in this thesis.

These copolymers were obtained through moderate temperature (~100°C) coupling reactions,

which minimize the ether-ether interchanges between hydrophobic and hydrophilic telechelic

oligomers via a nucleophilic aromatic substitution mechanism. The hydrophilic blocks were

based on the nucleophilic step polymerization of 3,3’-disulfonated, 4,4’-dichlorodiphenyl sulfone

with an excess 4,4’-biphenol to afford phenoxide endgroups. The hydrophobic (fluorinated)

blocks were largely based on decafluoro biphenyl (excess) and various bisphenols. The

copolymers were obtained in high molecular weights and were solvent cast into tough

membranes, which had nanophase separated hydrophilic and hydrophobic regions. The

performance and structure-property relationships of these materials were studied and compared

to random copolymer systems. NMR results supported that the multiblock sequence had been

achieved. They displayed superior proton conductivity, due to the ionic proton conducting

channels formed through the self-assembly of the sulfonated blocks. The nano-phase separated

morphologies of the copolymer membranes were studied and confirmed by atomic force

microscopy. Through control of a variety of parameters, including ion exchange capacity and

sequence lengths, performances as high, or even higher than those of the state-of-the-art PEM,

Nafion, were achieved.

iii

Acknowledgments

I would like to express my sincere gratitude to my advisor, Prof. James E. McGrath, for his

guidance, encouragement and inspiration throughout my Ph.D. career. I am extremely lucky to

have been his student, and to have benefited from his knowledge, experience, and exceptional

personality. I would also like to thank the other members of my advisory committee, Dr. Judy S.

Riffle, Dr. John G. Dillard, Dr. Richey M. Davis and Dr. Scott Case, for their great support.

My work could not have been accomplished without the help of the people in our research group.

I particularly want to thank Mr. Abhishek Roy for fuel-cell related testing, Ms. Juan Yang for

molecular weight characterizations, Dr. Anand Badami for morphological characterizations, and

Ms. Ozma Lane for mechanical and dynamic mechanical analyses. Dr. William Harrison, Dr.

Brian Einsla, Dr. Melinda Hill, and Dr. Kent Wiles are all acknowledged for familiarizing me

with chemicals, equipments, procedures and basic research techniques in the lab. I have also

benefited much from valuable discussions with my lab mates including Dr. Yanxiang Li, Dr.

Guangyu Fan, Dr. Zhongbiao Zhang, Dr. Hang Wang, Mr. Harry Lee, Mr. Yu Chen, Ms. Rachael

Hopp, and Ms. Natalie Arnett.

I am also very grateful to the staff in the Macromolecules and Interfaces Institute, particularly

Mrs. Laurie Good, Mrs. Millie Ryan, and Mrs. Angie Flynn—all of whom have been of

significant assistance with all the little details that go hand-in-hand with being a graduate student, iv

and who helped make this department such a great environment in which to work.

Finally, I would like to thank my fiancé, Yi Hou, for her love and understanding, and for being so

patient, and my parents for their love and support through all these years.

v

Table of Contents

ABSTRACT................................................................................................................................... ii ACKNOWLEDGMENTS ........................................................................................................... iv LIST OF FIGURES ..................................................................................................................... ix LIST OF TABLES....................................................................................................................... xv INTRODUCTION......................................................................................................................... 1 CHAPTER 1. LITERATURE REVIEW .................................................................................... 3 1.1. INTRODUCTION........................................................................................................................ 3 1.1.1. FUNDAMENTALS OF PROTON EXCHANGE MEMBRANES (PEMS).............................................. 3 1.1.2. BASIC CRITERIA FOR A PEM................................................................................................... 5 1.1.3. PEMS BASED ON HYDROPHOBIC-HYDROPHILIC BLOCK COPOLYMERS .................................... 7 1. 2. PEMS BASED ON HIGH PERFORMANCE ENGINEERING MATERIALS ...................................... 8 1.2.1. SYNTHESIS OF COPOLYMER BACKBONES ................................................................................ 8 1.2.1.1. Poly(arylene ether)s ........................................................................................................... 9 1.2.1.2. Aromatic Poly(imide)s ..................................................................................................... 13 1.2.1.3. Aromatic 5-membered-ring heterocyclic polymers ......................................................... 15 1.2.1.4. Poly(p-phenylene) derivatives ......................................................................................... 17 1.2.2. FABRICATION OF PEMS: INTRODUCTION OF PROTON-CONDUCTING MOIETIES...................... 19 1.2.2.1. Post sulfonation of poly(arylene ether)s .......................................................................... 19 1.2.2.2. Direct copolymerization of sulfonated monomers: preparation of poly(arylene ether) random copolymers....................................................................................................................... 20 1.2.2.3. Sulfonated poly(imide)s: hydrolytic stability issues........................................................ 24 1.2.2.4. Sulfonated poly(benzimidazole)s, poly(benzoxazole)s and poly(benzthiazole)s............ 26 1.2.2.5. Sulfonation of poly(2,5-benzophenone)s......................................................................... 30 1.3. FLUORINATED AROMATIC HIGH PERFORMANCE COPOLYMERS FOR PEMS ........................ 32 1.3.1. MODERATELY FLUORINATED COPOLYMERS CONTAINING HEXAFLUOROISOPROPYLIDENE UNITS

....................................................................................................................................................... 32 1.3.2. HIGHLY FLUORINATED POLY(ARYLENE ETHER)S CONTAINING PERFLUOROPHENYLENE UNITS

....................................................................................................................................................... 34 1.4. HYDROPHILIC-HYDROPHOBIC BLOCK COPOLYMER SYSTEMS AS PEMS............................. 41 1.4.1. BLOCK COPOLYMER PEMS BASED ON SULFONATED STYRENICS AND HYDROGENATED

vi

POLYDIENES ................................................................................................................................... 42 1.4.2. MULTIBLOCK COPOLYMERS SYNTHESIZED BY STEP OR CONDENSATION POLYMERIZATION.... 47 1.4.2.1. Partially aromatic multiblock copolymers ....................................................................... 47 1.4.2.2. Wholly aromatic multiblock copolymers......................................................................... 51 1.4.3. COMPARISONS BETWEEN RANDOM AND BLOCK COPOLYMER PEMS ..................................... 65 CHAPTER 2. SYNTHESIS AND CHARACTERIZATION OF BISAF-BPSH HYDROPHOBIC-HYDROPHILIC MULTIBLOCK COPOLYMERS ............................... 75 2.1. EXPERIMENTAL ..................................................................................................................... 76 2.1.1.. SOLVENTS ........................................................................................................................... 76 2.1.2. MONOMERS ......................................................................................................................... 77 2.1.3. MONOMER SYNTHESIS......................................................................................................... 79 2.1.4: POLYMER SYNTHESIS ........................................................................................................... 80 2.1.5. NMR SPECTROSCOPY, GEL PERMEATION CHROMATOGRAPHY, INTRINSIC VISCOSITY AND

ATOMIC FORCE MICROSCOPY CHARACTERIZATION ....................................................................... 82 2.1.6. CHARACTERIZATION OF FUEL CELL RELATED PROPERTIES .................................................. 83 2.2. RESULTS AND DISCUSSION..................................................................................................... 86 2.2.1. SYNTHESIS AND CHARACTERIZATION ................................................................................... 86 2.2.1.1. Synthesis of fluorinated Oligomers ................................................................................. 86 2.2.1.2. Synthesis of fully disulfonated hydrophilic oligomers .................................................... 90 2.2.1.3. Synthesis of Multiblock Copolymers............................................................................... 93 2.2.1.4. Fundamental characterizations of BisAF-BPSH multiblock copolymers ....................... 94 2.2.2. FUEL CELL RELATED CHARACTERIZATIONS OF MULTIBLOCK COPOLYMERS...................... 100 2.2.2.1. Proton conductivity under fully hydrated conditions..................................................... 100 2.2.2.2. Proton conductivity under partially hydrated conditions............................................... 104 2.2.2.3. Diffusion coefficients..................................................................................................... 107 2.2.2.4. Methanol Permeability................................................................................................... 108 CHAPTER 3. SYNTHESIS AND CHARACTERIZATION OF 6FBISAF-BPSH MULTIBLOCK COPOLYMERS.............................................................................................111 3.1. EXPERIMENTAL ....................................................................................................................111 3.1.1. MATERIALS. ........................................................................................................................111 3.1.2. POLYMER SYNTHESIS......................................................................................................... 112 3.1.3. POLYMER ISOLATION AND CHARACTERIZATION .................................................................. 114 3.2. RESULTS AND DISCUSSION ....................................................................................................115

vii

3.2.1. POLYMER SYNTHESIS AND CHARACTERIZATION ................................................................. 115 3.2.2. FUNDAMENTAL CHARACTERIZATIONS……………………………………………………………..118 3.2.3. CHARACTERIZATION OF FUEL CELL RELATED PROPERTIES .................................................. 126 CHAPTER 4. SYNTHESIS AND CHARACTERIZATION OF BISSF-BPSH MULTIBLOCK COPOLYMERS............................................................................................ 132 4.1. EXPERIMENTAL ................................................................................................................... 133 4.1.1. MATERIALS ........................................................................................................................ 133 4.2. RESULTS AND DISCUSSION ................................................................................................... 133 4.2.1. POLYMER SYNTHESIS AND CHARACTERIZATION ................................................................. 133 4.2.1.1. Synthesis of fluorinated oligomers ................................................................................ 133 4.2.1.2. Synthesis of Fully disulfonated hydrophilic oligomers ................................................. 139 4.2.1.3. Synthesis of BisSF-BPSH multiblock copolymers........................................................ 141 4.2.1.4. Characterizations of molecular weights of BisSF-BPSH copolymers........................... 144 4.2.1.5. Thermal analysis ............................................................................................................ 152 4.2.1.6. Mechanical Properties.................................................................................................... 154 4.2.1.7. Surface morphological features ..................................................................................... 155 4.2.2. STUDY OF FUEL CELL-RELATED PROPERTIES....................................................................... 158 4.2.2.1. Effects of block lengths on proton conductivity and water uptake................................ 158 4.2.2.2. Effects of composition (IEC) on proton conductivity and water uptake ....................... 162 4.2.2.3. Effects of hydrophobic block length (hydrophobic/hydrophilic block length ratio) ..... 165 4.2.2.4. Swelling-deswelling properties of multiblock copolymers ........................................... 168 4.2.3. SOME COMPARISONS OF BISAF-BPSH, 6FBISAF-BPSH AND BISSF-BPSH MULTIBLOCK

COPOLYMERS................................................................................................................................ 173 CHAPTER 5. CONCLUSIONS............................................................................................... 176 REFERENCES.......................................................................................................................... 180

viii

List of Figures Figure 1.1.1. Electrochemical reactions for a PEMFC and DMFC......................................... 3

Figure 1.1.2. Membrane electrode assembly in a proton exchange fuel cell membrane......... 4

Figure 1.1.3. Proposed structure of Nafion®............................................................................ 5

Figure1.2.1. Some possible structures of poly(arylene ether)s................................................ 9

Figure 1.2.2. Mechanism of SNAr nucleophilic aromatic substitution .................................. 10

Figure 1.2.3. Synthesis of bisphenol-A polysulfones ............................................................ 11

Figure 1.2.4. Mechanism of K2CO3-catalyzed synthesis of Bisphenol-A polysulfones........ 12

Figure 1.2.5. Synthesis of poly(arylene ether ketone)s via the ketimine precursor method.. 13

Figure 1.2.6. Synthesis of poly(arylene ether ketone)s from bulky substituted bisphenol.... 13

Figure 1.2.7. Two-stage synthesis of poly(imide)s ................................................................ 14

Figure 1.2.8. One-stage synthesis of poly(imide)s with improved solubility........................ 15

Figure 1.2.9. General scheme for the synthesis of (a) poly(benzimidazole)s, (b) poly(benzoxazole)s and (c) poly(benzthiazole)s ........................................................... 16

Figure 1.2.10. Synthesis of PBI by melt polymerization....................................................... 16

Figure 1.2.11. Synthesis of substituted PPPs via Ni(0) coupling .......................................... 17

Figure 1.2.12. Synthesis of PPP alternating copolymers via Suzuki coupling ...................... 18

Figure 1.2.13. Synthesis and polymerization of 2,5-dichlorobenzophenone monomers....... 18

Figure 1.2.14. Examples of post-sulfonated poly(arylene ether sulfone)s and poly(arylene ether ketone)s ................................................................................................................. 19

Figure 1.2.15. Synthesis of 3,3’-disulfonated 4,4’-dichlorodiphenyl sulfone (SDCDPS) .... 21

Figure 1.2.16. Synthesis of BPSH-xx random copolymers ................................................... 22

Figure 1.2.17. Sodium 5,5'-carbonylbis(2-fluorobenzenesulfonate) (SDFBP) ..................... 23

Figure 1.2.18. Structures of sulfonated poly(arylene ether ketone) random copolymers studied............................................................................................................................ 24

Figure 1.2.19. Structures of (a) phthalic and (b) naphthalenic imide units............................ 25

Figure 1.2.20. Synthesis of naphthalenic sulfonated poly(imide) random copolymers......... 25

Figure 1.2.21. Structures of (a)sulfonated and (b)unsulfonated diamine monomers used .... 26

Figure 1.2.22. Synthesis of sulfonated heterocyclic homopolymers ..................................... 27

Figure 1.2.23. Synthesis of sulfonated poly(benzimidazole) random copolymers................ 27

ix

Figure 1.2.24. Synthesis of sulfonated poly(benzimidazole) random copolymers................ 28

Figure 1.2.25. Sulfonation of (a)PBI and (b)ABPBI ............................................................. 29

Figure 1.2.26. Proposed structures of sulfonated poly(2,5-benzophenone)s......................... 31

Figure 1.2.27. Sulfonation of poly(4-fluoro-2,5-benzophenone)s......................................... 31

Figure 1.3.1. Synthesis of 6F-BPA ........................................................................................ 32

Figure 1.3.2. Structures of partially sulfonated, partially fluorinated poly(arylene ether sulfone)s......................................................................................................................... 33

Figure 1.3.3. Partially sulfonated, partially fluorinated poly(arylene ether sulfone benzonitrile)s.................................................................................................................. 34

Figure 1.3.4. Synthesis of poly(arylene ether)s containing perfluorophenylene units .......... 35

Figure 1.3.5. Synthesis of perfluorinated aromatic compounds by saturation-aromatization 36

Figure 1.3.6. Synthesis of perfluorinated aromatic compounds by nucleophilic aromatic substitution..................................................................................................................... 36

Figure 1.3.7. Reactions of ortho-position fluorine leading to branching and/or gelation ..... 37

Figure 1.3.8. Direct synthesis of sulfonated perfluorinated poly(arylene ether) random copolymers..................................................................................................................... 38

Figure 1.3.9. Post sulfonation of fluorinated poly(arylene ether)s ........................................ 39

Figure 1.3.10. Synthesis and sulfonation of fluorinated poly(arylene ether sulfone) random copolymers..................................................................................................................... 40

Figure 1.3.11. Synthesis and sulfonation of fluorinated poly(arylene ether sulfone) random copolymers..................................................................................................................... 41

Figure 1.4.1. Structure of S-SEBS block copolymers ........................................................... 43

Figure 1.4.2. Modification of S-SEBS surface with plasma treatment followed by hydrolysis........................................................................................................................................ 44

Figure 1.4.3. Structure of S-SIBS block copolymers............................................................. 45

Figure 1.4.4. Synthesis of partially sulfonated (PVDF-ran-PHFP)-b-PS copolymers .......... 46

Figure 1.4.5. Synthesis of sulfonated PAES-PB multiblock copolymers .............................. 48

Figure 1.4.6. Synthesis of sulfonated PAES-PVDF multiblock copolymers......................... 49

Figure 1.4.7. Epoxidation of sulfonated PAES-PB multiblock copolymers.......................... 50

Figure 1.4.8. Synthesis of sulfonated PPP-PAES multiblock copolymers ............................ 52

Figure 1.4.9. Synthesis of sulfonated poly(arylene ether sulfone) multiblock copolymers .. 54

Figure 1.4.10. Synthesis of sulfonated poly(arylene ether ketone) multiblock copolymers.. 54

x

Figure 1.4.11. Synthesis of sulfonated poly(arylene ether ketone) multiblock copolymers.. 56

Figure 1.4.12. Synthesis of (a) phenoxide-teminal; (b) Cl-terminal; (c) NH2-endcapped BPS-100 oligomers ........................................................................................................ 57

Figure 1.4.13. Synthesis of PPP-BPSH100 multiblock copolymers ..................................... 58

Figure 1.4.14. Synthesis of polyimide-BPS100 multiblock copolymers............................... 59

Figure 1.4.15. Synthesis of Perfluoroarylene ether-BPS100 multiblock copolymers ........... 60

Figure 1.4.16. BisAF-BPS100 multiblock copolymers ......................................................... 61

Figure 1.4.17. Synthesis of fluorine-terminal hydrophobic oligomers.................................. 63

Figure 1.4.18. Questionable synthesis of poly(arylene ether sulfone) multiblock copolymers........................................................................................................................................ 64

Figure 1.4.19. Synthesis of sulfonated PAES-PAEK multiblock copolymers....................... 65

Figure 1.4.20. (a) structure of S-SE “pseudo-random” copolymers. (b) proton conductivity vs. water content for S-SEBS, S-SE and Nafion PEMs................................................. 67

Figure 1.4.21. Structures of (a) poly(ether sulfone) and (b) poly(ether ketone) random copolymers..................................................................................................................... 67

Figure 1.4.22. Proton conductivity vs. RH plots for Nafion 117, poly(ether sulfone) random copolymers (HQSH 30), and poly(ether ketone) random copolymers (PB-diketone 50)........................................................................................................................................ 69

Figure 1.4.23. Proton conductivity vs. RH plots for Nafion 117 and BisAF-BPSH multiblock copolymers..................................................................................................................... 70

Figure 1.4.24. Synthesis of BisSF-BPSH multiblock copolymers ........................................ 70

Figure 1.4.25. Proton conductivity vs.RH plots for BisSF-BPSH multiblock copolymers, Nafion 112 and BPSH-35 random copolymers.............................................................. 71

Figure 1.4.26. Tapping mode AFM images of BPSH-xx random copolymer membranes: (a).BPSH-30; (b).BPSH-35; (c).BPSH-40; (d).BPSH-45.............................................. 72

Figure 1.4.27. Tapping mode AFM phase images of BPSH-PI multiblock copolymer membranes with different block lengths: (a).5K:5K; (b).10K:10K; (c).15K:15K. ....... 72

Figure 1.4.28. Proton conductivity vs.RH plots for BPSH-PI multiblock copolymers, Nafion 112 and BPSH-35 random copolymers.......................................................................... 73

Figure 2.1. Structures of fluorinated-sulfonated, hydrophobic-hydrophilic multiblock copolymers..................................................................................................................... 75

Figure 2.2. Pulse sequence schematic for PGSE NMR experiments..................................... 85

Figure 2.3. Synthesis of BisAF oligomers ............................................................................. 86

xi

Figure 2.4. 1H NMR spectrum of a BisAF oligomer ............................................................. 87

Figure 2.5. 19F NMR spectra of a BisAF oligomer................................................................ 87

Figure 2.6. logη vs. logMn plot for BisAF oligomers ............................................................ 88

Figure 2.7. Reaction of decafluorobiphenyl with various bisphenol monomers ................... 90

Figure 2.8. Synthesis of fully sulfonated BPS-100 oligomers............................................... 91

Figure 2.9. 1H NMR spectrum of a BPS-100 oligomer ......................................................... 92

Figure 2.10. Synthesis of BisAF-BPSH multiblock copolymers........................................... 93

Figure 2.11. 1H NMR spectrum of a BisAF-BPSH multiblock copolymer........................... 95

Figure 2.12. 19F NMR spectrum of a BisAF-BPSH multiblock copolymer .......................... 96

Figure 2.13. Monitoring of multiblock copolymer synthesis using 1H NMR spectra: (a) BPS-100 oligomer prior to the reaction; (b) 12 h; (c) 20 h; (d) 36 h............................. 97

Figure 2.14. 1H NMR spectra of BisAF-BPSH multiblock copolymers showing the linkages between blocks............................................................................................................... 98

Figure 2.15. Tapping mode AFM phase images of BisAF-BPSH multiblock copolymers with different block lengths ................................................................................................. 100

Figure 2.16. Structures of partially disulfonated random copolymers. (a) BPSH; (b) HQSH; (c) poly(ether ketone) B and PB series ........................................................................ 101

Figure 2.17. Proton conductivity at 30oC in liquid water for partially disulfonated random copolymers plotted against IEC................................................................................... 102

Figure 2.18. Proton conductivity under fully hydrated conductions for BisAF-BPSH copolymers as a function of temperature ..................................................................... 104

Figure 2.19. Proton conductivity vs. RH plots for Nafion 117, poly(ether sulfone) random copolymers (HQSH-30), and poly(ether ketone) random copolymers (PB-50) .......... 105

Figure 2.20. Activation energy of proton transport for HQSH-30 random copolymers as a function of relative humidity: proton transport barrier increases as RH decreases ..... 106

Figure 2.21. Proton conductivity vs. RH plots for Nafion 117and BisAF-BPSH multiblock copolymers................................................................................................................... 107

Figure 2.22. Self-diffusion coefficient for water as a function of IEC for random and block copolymers and Nafion ................................................................................................ 108

Figure 2.23. Methanol permeability at 80 oC as a function of IEC for random and block copolymer membranes and Nafion .............................................................................. 110

Figure 3.1. Synthesis of 6FBisAF oligomers....................................................................... 115

Figure 3.2. Synthesis of BPS-75 hydrophilic oligomers...................................................... 116

xii

Figure 3.3. Synthesis of 6FBisAF-BPSH100 multiblock copolymers ................................ 117

Figure 3.4. Synthesis of 6FBisAF-BPSH75 multiblock copolymers .................................. 117

Figure 3.5. 19F NMR spectrum of a 6FBisAF oligomer ...................................................... 118

Figure 3.6. 19F NMR spectrum of a 6FBisAF oligomer (aromatic region) showing endgroups in detail......................................................................................................................... 119

Figure 3.7. logη vs. logMn plot for 6FBisAF oligomers ...................................................... 120

Figure 3.8. 1H NMR spectrum of a BPS-75 oligomer ......................................................... 120

Figure 3.9. 1H NMR spectrum of a 6FBisAF-BPSH100 multiblock copolymer................. 121

Figure 3.10. 1H NMR spectra of a partially disulfonated BPS oligomer and the corresponding 6FBisAF-BPS83 multiblock copolymer .............................................. 122

Figure 3.11. DSC trace of a 6FBisAF-BPSH100 (9K:9K) multiblock copolymer.............. 125

Figure 3.12. Tapping mode AFM images for a 6FBisAF-BPSH100 (15K:10K) multiblock copolymer .................................................................................................................... 129

Figure 3.13. Tapping mode AFM images for a 6FBisAF-BPSH75 (15K:9K) multiblock copolymer .................................................................................................................... 129

Figure 3.14. Proton conductivity as a function of RH for 6FBisAF-BPSH multiblock copolymers................................................................................................................... 130

Figure 4.1. Synthesis of BisSF telechelic oligomers ........................................................... 134

Figure 4.2. Reaction at para- positions leading to branching .............................................. 134

Figure 4.3. Synthesis of fluorinated poly(ether sulfone) under mild conditions ................. 135

Figure 4.4. Evolution of intrinsic viscosity for BisSF (17K) oligomer synthesis as a function of reaction time ............................................................................................................ 136

Figure 4.5. 19F NMR of a BisSF telechelic oligomer .......................................................... 137

Figure 4.6. logη vs. logMn plot for BisSF oligomers ........................................................... 138

Figure 4.7. 19F NMR plots of BisSF oligomers with Mn of 1) 5K; 2) 10K; 3) 17K; 4) 25K139

Figure 4.8. Synthesis of BisSF-BPSH multiblock copolymers ........................................... 141

Figure 4.9. 19F NMR spectra of a BisSF-BPSH multiblock copolymer .............................. 143

Figure 4.10. 1H NMR spectra of a BisSF-BPSH multiblock copolymer............................. 143

Figure 4.11. 1H NMR spectra of BisSF-BPSH copolymers with increasing block lengths 144

Figure 4.12. Intrinsic viscosity as a function of block lengths for BisSF-BPSH multiblock copolymers (IEC=1.3) ................................................................................................. 147

Figure 4.13. Schematic plots showing the decrease of endgroup concentration as a function

xiii

of time for the syntheses of multiblock copolymers with (a) low block lengths; (b) high block lengths ................................................................................................................ 147

Figure 4.14. Intrinsic viscosity as a function of block lengths for BisSF-BPSH multiblock copolymers (IEC=1.1) ................................................................................................. 149

Figure 4.15. Intrinsic viscosity as a function of block lengths for BisSF-BPSH multiblock copolymers (IEC=1.5) ................................................................................................. 151

Figure 4.16. DSC trace of a BisSF-BPSH (17K-12K) multiblock copolymer .................... 153

Figure 4.17. TGA traces of BisSF-BPSH (17K-12K) multiblock copolymers.................... 153

Figure 4.18. Stress-strain curves for BisSF-BPSH (17K-12K) (IEC=1.5) copolymers ...... 155

Figure 4.19. Tapping mode AFM height (left) and phase (right) images for BisSF-BPSH (5K-5K), (7K-7K), (17K-12K) and (25K-16K) multiblock copolymer membranes... 157

Figure 4.20. Proton conductivity for BisSF-BPSH copolymers having an IEC of 1.3........ 158

Figure 4.21. Water uptake as a function of Block lengths for BisSF-BPSH (IEC=1.3) copolymers................................................................................................................... 159

Figure 4.22. Water uptake as a function of disulfonation degree for BPSH random copolymers................................................................................................................... 160

Figure 4.23. Proton conductivity at 80 oC as a function of relative humidity for BisSF-BPSH (IEC=1.3) copolymers ................................................................................................. 161

Figure 4.24. Proton conductivity vs. RH: the effect of IEC on the performance of BisSF-BPSH (17K-12K) copolymers .......................................................................... 163

Figure 4.25. Proton conductivity vs. RH: the effect of IEC on the performance of BisSF-BPSH (25K-16K) copolymers .......................................................................... 165

Figure 4.26. Proton conductivity under partially hydrated conditions for BisSF-BPSH (25K-16K) (IEC=1.5) and BisSF-BPSH (15K-15K) (IEC=1.5) ................................. 166

Figure 4.27. Degrees of swelling in x, y and z directions for different copolymer membranes...................................................................................................................................... 169

Figure 4.28. Multiblock copolymers with long blocks or higher IEC show higher z-direction swelling........................................................................................................................ 171

Figure 4.29. Imaginary cross-sectional view of BisSF-BPSH (17K-12K) copolymer membranes ................................................................................................................... 172

Figure 4.30. Comparison of TGA traces for three series of multiblock copolymers........... 175

xiv

List of Tables

Table 2.1. Molecular weight characterizations of BisAF oligomers ..................................... 88

Table 2.2. Some characterizations of BisAF-BPSH multiblock copolymers ........................ 99

Table 2.3. IEC, water uptake and proton conductivity for partially disulfonated random copolymers................................................................................................................... 102

Table 2.4. IEC, water uptake and liquid water proton conductivity for BisAF-BPSH multiblock copolymers................................................................................................. 103

Table 3.1. Molecular weight characterizations of 6FBisAF oligomers ............................... 119

Table 3.2. Comparison of target IEC with experimental values for 6FBisAF-BPSH100 multiblock copolymers................................................................................................. 123

Table 3.3. Solubility of oligomers in DMSO at room temperature ..................................... 123

Table 3.4. Proton conductivity (liquid water) and water uptake for 6FBisAF-BPSH100 multiblock copolymer membranes............................................................................... 127

Table 3.5. Proton conductivity (liquid water) and water uptake for 6FBisAF-BPSH75 and 6FBisAF-BPSH83 multiblock copolymer membranes................................................ 128

Table 4.1. Molecular weight characterizations of BisSF oligomers .................................... 138

Table 4.2. Characterization of BisSF-BPSH copolymers with 1.3 IEC (Series A) ............. 145

Table 4.3. Characterization of BisSF-BPSH copolymers with 1.1 IEC (Series B) ............. 148

Table 4.4. Characterizations of BisSF-BPSH copolymers with 1.5 IEC (Series C)............ 150

Table 4.5. Effects of IEC on the mechanical properties of BisSF-BPSH (17K-12K) multiblock copolymers................................................................................................. 155

Table 4.6. The effect of IEC on the properties of BisSF-BPSH copolymers....................... 162

Table 4.7. Comparison between BisSF-BPSH multiblock copolymers having different hydrophobic block lengths........................................................................................... 166

Table 4.8. Comparison between BisSF-BPSH multiblock copolymers having different hydrophobic block lengths........................................................................................... 167

xv

Introduction

The proton exchange membrane (PEM) is the key component in a proton exchange membrane

fuel cell (PEMFC). Nafion® and comparable perfluorosulfonic acid-based membranes are

currently the state-of-the-art PEMs, but suffer from shortcomings such as high permeability, cost,

and limited operating temperatures.

These drawbacks have collectively sparked an interest in developing novel copolymers as

alternative PEMs, particularly in the last decade or so. Some hydrophilic-hydrophobic block

copolymer membranes, in particular, have shown great promise as potential candidates for PEMs.

Therefore, the goal of this Ph.D. research was to synthesize and characterize multiblock

(segmented) copolymers containing sulfonated and fluorinated blocks.

In Chapter 1, the literature review, research on the synthesis of partially sulfonated high

performance copolymer membranes as alternative PEMs, has been outlined. The main focus is

on wholly aromatic engineering copolymer systems, such as poly(arylene ether)s, poly(imides),

poly(benzimidazole)s and poly(p-phenylene) derivatives. Next, recent progress on nanophase

separated block copolymers of varying structures, which were synthesized by a variety of

methods, is reviewed, and their advantages over random copolymers are discussed.

Specifically, my research involved the investigation of three series of hydrophilic/hydrophobic,

fluorinated/sulfonated multiblock copolymers, which had the same structures in the hydrophilic

1

blocks, but which differed slightly in terms of the hydrophobic (fluorinated) blocks. The results

are shown and discussed in Chapters 2-4, respectively. Their performance and structure-property

relationships were studied and compared to random copolymer systems. These copolymers

displayed superior proton conductivity, due to the ionic proton conducting channels formed

through the self-assembly of the sulfonated blocks. At the end of Chapter 4, the three series are

briefly compared and contrasted. This chapter also includes a discussion of how to develop a

material with good performance and water sorption capabilities. This study concludes with a

summary chapter.

The author is a synthetic chemist who was in charge of the synthesis, structural characterization,

thermal analysis, etc. of the copolymers. Here I would like to again sincerely thank Abhishek

Roy, Juan Yang, Anand Badami, Ozma Lane, and Mark Flynn for their advice and help with

respect to electrochemical testing, molecular weight characterization, morphological studies,

mechanical testing, etc.

2

Chapter 1. Literature Review

1.1. Introduction

1.1.1. Fundamentals of proton exchange membranes (PEMs)

Fuel cells are electrochemical devices that convert chemical energy directly into electrical

energy.1, 2 Proton exchange membrane fuel cells (PEMFCs) have shown promise as alternative

automotive and stationary power sources.2, 3 In a PEMFC, hydrogen is the fuel, oxygen is the

oxidant, and water is produced as the only by-product. A direct methanol fuel cell (DMFC),

which uses dilute methanol as the fuel, is the portable power version of the PEMFC. The basic

electrochemical reactions for PEMFC and DMFC are summarized in Figure 1.1.1.

Anode: 2H2 4H+ + 4e-

Cathode: 4H+ + O2 + 4e- 2H2O

Overall: 2H2 + O2 2H2O

Anode: CH3OH + H2O CO2 + 6H+ + 6e-

Cathode: 3/2 O2 + 6H+ + 6e- 3H2O

Overall: CH3OH + H2O + 3/2 O2 CO2 + 3H2O

DMFC

PEMFC

Figure 1.1.1. Electrochemical reactions for a PEMFC and DMFC

PEMFCs operate through a membrane electrode assembly (MEA), the basic structure of which is

shown in Figure 1.1.2.1, 3 It is composed of an anode, cathode, and a proton exchange membrane

(PEM) sandwiched in between. The PEM, which is the electrolyte that transfers protons from the

3

anode to the cathode, is the key component of the system. It also serves as a separator to prevent

mixing of the fuel and oxygen.2

Figure 1.1.2. Membrane electrode assembly in a proton exchange fuel cell membrane

The current state-of-the-art PEMs are perfluorosulfonic acid membranes such as Nafion®,

developed by Dupont in the late 1960s. The reported structure of Nafion®, which is shown in

Figure 1.1.3, is based on a crystallizable tetrafluoroethylene backbone and contains pendant side

chains of perfluorinated vinyl ethers terminated by perfluorosulfonic acid groups. Nafion® is

believed to be synthesized by free radical polymerization. The sulfonated comonomer cannot

easily self-propagate and thus its sequence length should not be more than one. The highly acidic

perfluorosulfonic acid groups impart high proton conductivity under both fully hydrated and

partially hydrated conditions, while the semicrystalline backbone provides excellent chemical

and electrochemical stability. However, Nafion® and other perfluorosulfonated PEMs suffer from

disadvantages including high cost, limited operating temperature (80°C), and high fuel

permeability (in DMFCs).4, 5 Unfortunately,, the use of thicker membrane in DMFC applications

in order to reduce methanol permeability results in resistance losses of the cell.

4

CF2 CF2 CF CF2

OCF2CF

O(CF2)2SO3H

CF3

x y

Figure 1.1.3. Proposed structure of Nafion®

Therefore, researchers have attempted to develop alternative proton exchange membranes that

can withstand the harsh fuel cell operating conditions that severely compromise their

performance.6, 7 However,it has been a challenge to achieve performance levels comparable to

perfluorosulfonic acid PEMs—especially while keeping costs low. In general, a successful PEM

should have high proton conductivity, low electronic conductivity, good mechanical strength,

high oxidative and hydrolytic stability, low fuel permeability, ease of fabrication into MEA, and

controlled swelling-deswelling behavior as a function of relative humidity.3

1.1.2. Basic criteria for a PEM

Among the most important properties of a PEM are proton conductivity and water uptake, both

of which are closely related to the concentration of ion-conducting units in the membrane. This

is typically known as its ion exchange capacity (IEC). Los Alamos National Laboratory (LANL)

developed a simple method for determining the proton conductivity of PEMs using

electrochemical impedance spectroscopy.8 The conductivity is measured in the plane, because

measuring it normal to the plane is difficult due to large interfacial resistances.9 Water uptake is

most frequently reported in mass percent. Since most proton exchange membranes rely on water

to facilitate proton transport, a PEM must undergo the required water uptake to perform 5

efficiently. Despite the fact that proton conductivity generally increases with water uptake,

excessive water uptake may cause the membrane to lose its mechanical strength in the

water-swollen state. In addition, in the membrane electrode assembly, too much swelling will

lead to stress in the swelling-deswelling cycles in a fuel cell environment and the membrane

and/or MEA may fail. For H2-O2 fuel cell applications, the water uptake generally should not

exceed 50 weight %, and ideally it should be in the 20~30 weight % range.

Due to harsh fuel cell operating conditions, oxidative stability and hydrolytic stability are critical

to the long-term durability of a fuel cell membrane. PEMs based on partially aliphatic backbones,

such as polystyrene, are often subjected to oxidative degradation and thus are generally utilized

in low temperature fuel cells, such as those intended for use in portable power devices.10-12 In

contrast, polymers with high performance aromatic backbones, such as poly (arylene ether)s and

polyimides13-16, are generally much more oxidatively stable. However, certain functional groups

in these systems may be susceptible to hydrolytic degradation, as will be discussed later.

Like other polymeric materials, a PEM must have sufficient mechanical strength, both in the dry

state and water-swollen state, to survive the stress of electrode attachment. The membrane must

also be tough and flexible. Therefore, high molecular weight is always desired to enhance

intermolecular forces and chain entanglements. In contrast, weak and brittle membranes are often

formed from low molecular weight materials and/or result from the degradation of the polymer.

In addition to molecular weight, a membrane’s chemical structure also affects its mechanical

properties. For instance, even substituted poly(1,4-phenylene) derivatives are incapable of 6

forming flexible membranes due to their extremely rigid rod-like chains.17 As will be shown later,

some PEMs have been synthesized by block-copolymerizing poly(1,4-phenylene) derivatives

with other species to improve their film-forming ability.18

Of course, a PEM must be capable of being fabricated into a membrane electrode assembly

(MEA). Therefore, associated research efforts have investigated membrane-electrode interfaces.

Not only must the membrane itself possess long-term stability, but it must also be compatible

with the electrodes, which usually contain highly fluorinated, Nafion-based binders (catalyst

layer).19 Thus, PEMs based on partially fluorinated copolymer backbones have been an emerging

area of research.

1.1.3. PEMs based on hydrophobic-hydrophilic block copolymers

Block copolymers are macromolecules made up of two or more usually multiphase blocks that

are chemically conjoined in the same chain. Unlike random copolymers, in which the monomers

are arranged statistically, in block copolymers different chemical components exist in ordered

sequences. Therefore, they have the potential to display interesting physico-chemical properties.

Various morphological features of block copolymers can be obtained by tailoring the chemical

composition, molecular weight, and/or volume fraction of the blocks.20

Hydrophobic/hydrophilic (amphiphilic) block copolymers can be obtained when one or more of

the blocks is water-soluble, or is fully or partially modified with hydrophilic functional groups.

Due to their phase behavior, these materials can be utilized for a variety of purposes, including as

7

biomaterials, as stabilizers in suspensions and emulsions, for pharmaceutical applications, and in

the synthesis of advanced materials, adhesives, and coatings.21-23

The development of hydrophobic-sulfonated partially ionic block copolymers as PEMs has been

of great interest. These materials contain sequences of sulfonated and nonsulfonated segments

which result in interesting structural and morphological features.24-26 Therefore, various block

copolymer ionomers containing sulfonic acid groups have been synthesized. The systematic

evaluation of their potential as PEMs has, however, largely been ignored. A review of some

recent research on the development of PEMs based on low-cost, high-performance engineering

materials for use as alternatives to perfluorosulfonic acid-based PEMs—with particular emphasis

on hydrophilic-hydrophobic block copolymer membranes—will be provided. At the end of this

overview, current progress on the comparative properties of random and block copolymers (and

especially the advantages of the latter) will be discussed.

1. 2. PEMs based on high performance engineering materials

1.2.1. Synthesis of copolymer backbones

High performance engineering thermoplastics, based on wholly aromatic polymers and

copolymers, are important in a wide variety of applications including automotive, structural, and

microelectronic components. In particular, poly(arylene ether)-based ionomers are a family of

promising candidates for novel PEMs due to their low cost, high glass transition temperatures,

good mechanical properties, excellent oxidative and hydrolytic stability, as well as the ease by

8

which protein conducting moieties can be incorporated. 27 Furthermore, their structures can be

easily modified by varying the linkages between the phenyl rings. As shown in Figure 1.2.1, the

X variable, which can be either a sulfone group, a ketone group, or a phenyl phosphine oxide

group, determines whether the polymer will be a poly(arylene ether sulfone), a poly(arylene

ether ketone), or a poly(arylene ether phosphine oxide). There is also flexibility in the Y variable,

which in principle can be a bond, a sulfone group, an isopropylidene linkage, etc.

1.2.1.1. Poly(arylene ether)s

nO X O Y

S

O

O

C

O P

O

, , C

CH3

CH3

C

CF3

CF3

S

O

O

P

O

, , ,X = Y = a bond,

Figure1.2.1. Some possible structures of poly(arylene ether)s

Although Friedel-Crafts electrophilic sulfonylation and acylation reactions have been used to

synthesize poly(arylene ether sulfone)s and poly(arylene ether ketone)s, respectively, the most

practical method for preparing poly(arylene ether)s is via nucleophilic aromatic substitution.28-31

Figure 1.2.2 shows the generalized mechanism for an SNAr nucleophilic aromatic substitution. In

the first step, which is the rate-determining step, the carbon atom of the activated C-X bond is

attacked by the nucleophile, and a resonance-stabilized Meisenheimer complex is formed. The

leaving group, X, departs in the second step.

9

Y X Nu-+slow Y

X

NuMeisenheimer complex

fast Y Nu + X-

Figure 1.2.2. Mechanism of SNAr nucleophilic aromatic substitution

The synthesis of poly(arylene ether)s is usually achieved via the step polymerization of a

dihalide monomer and a bisphenol monomer, with a base producing the active phenolate. The

bisphenol is converted into the phenoxide ion, which acts as the nucleophile, after which the

carbon atom adjacent to the halogen is attacked by the phenoxide ion to form an ether linkage.

Since the sulfone, ketone and phosphine oxide groups are all electron-withdrawing groups, they

stabilize the Meisenhheimer complex and facilitate the reaction. For this reason, all three

families of polymers are readily synthesized in this way. For example, with respect to the

catalysis of sodium hydroxide, 4,4’dichlorodiphenyl sulfone (DCDPS) and bisphenol-A can be

polymerized to afford bisphenol-A polysulfone (Udel®) (Figure 1.2.3).28, 31 Even though

difluoride monomers are more reactive than their dichloride counterparts, the latter are more

commonly used due to their lower costs.

10

+S

O

O

Cl Cl C

CH3

CH3

HO OH

NaOH Chlorobenzene, DMSO(dry)160oC / 1 h

C

CH3

CH3

O OS

O

O n

Figure 1.2.3. Synthesis of bisphenol-A polysulfones

The use of a strong base catalyst like sodium hydroxide ensures rapid polymerization at

relatively low temperatures, but requires the addition of an exact stoichiometric amount of the

base. Moreover, the diphenolate must be soluble. The use of weak bases (e.g., potassium

carbonate) to synthesize phenolates has been proposed and studied by McGrath et al.32, 33 The

proposed mechanism for K2CO3-catalyzed Udel® synthesis is shown in Figure 1.2.4. The

reaction mixture must be carefully dehydrated using an azeotroping agent while the phenolate is

formed. A high reaction temperature and polar aprotic solvents, such as dimethyl sulfoxide

(DMSO), N-methyl pyrrolidone (NMP) and N,N-dimethyl acetamide (DMAc), are usually

required. The polymerization is believed to be second order, so the rate of polymerization can be

improved by increasing monomer concentrations, but only to the extent that the solution does not

become too viscous during polymer formation. As a rule, reaction mixtures with 15~20% g/mL

(0.3~0.5 mol/L) monomer concentrations are used.

11

+S

O

O

Cl Cl C

CH3

CH3

HO OK

C

CH3

CH3

O OS

O

O n

C

CH3

CH3

HO OHK2CO3

Aprotic solvent/toluene C

CH3

CH3

HO OK + KHCO3

C

CH3

CH3

O OHS

O

O

Cl

+ nH2O + nCO2 + nKCl

KHCO3

Figure 1.2.4. Mechanism of K2CO3-catalyzed synthesis of Bisphenol-A polysulfones

The mechanism of poly(arylene ether ketone) synthesis is similar to that for poly(sulfone)s.

However, the direct synthesis of high molecular weight poly(ether ketone)s from dihalide

benzophenone monomers and bisphenol monomers such as biphenol and hydroquinone can be

challenging, primarily because the resulting polymers tend to be semicrystalline and display poor

solubility.34 High reaction temperatures are therefore needed, but can give rise to undesirable

side reactions.35 An alternative is the “soluble precursor” method, in which certain functional

groups are introduced to suppress crystallinity and improve solubility. For instance, poly(arylene

ether ketone)s have been synthesized using poly(ether ether ketimine) as the precursor,36, 37 as

shown in Figure 1.2.5. These can then be hydrolyzed under acidic conditions to yield the

semi-crystalline ketone counterpart. Precursors with bulky substituents such as t-butyl groups on

the bisphenol unit have also been reported (Figure 1.2.6).38, 39 The poly(ether ketone) is obtained

by removing the t-butyl group using a Lewis acid catalyst.

12

HO OH C

N

F F+ C

N

OOn

K2CO3

NMP, 160oC

HCl, R.T. COO

O

n

Figure 1.2.5. Synthesis of poly(arylene ether ketone)s via the ketimine precursor method

HO OH C

O

F F+ C

O

OOn

K2CO3

DMSO, 170oC

CF3SO3H, R.T. COO

O

n

Figure 1.2.6. Synthesis of poly(arylene ether ketone)s from bulky substituted bisphenol

1.2.1.2. Aromatic Poly(imide)s

Due to their excellent thermal, chemical and mechanical properties, aromatic poly(imide)s have

been important advanced materials ever since they were first introduced by DuPont. Aromatic

polyimides can be synthesized via the step growth polymerization of diamine monomers with

dianhydride monomers.40, 41 One factor limiting their processability and resulting applications is

that they tend to be insoluble in polar organic solvents. Therefore, the synthesis of aromatic

poly(imide)s requires two distinct stages, as shown in Figure 1.2.7. In the first step a poly(amic

acid) is generated at a relatively low temperature, which is soluble and can be made into a film,

for example. The second stage involves the ring closure of the poly(amic acid) to form the

13

ultimate polyimide structure. This is done by subjecting the poly(amic acid) to high temperatures

(200~300oC).

NN

O

O O

O

O

H2N O NH2 + O N

OO

N

O

HO OH

O

H H

n

DMAc

Heatcyclodehydration

O N

OO

N

O

HO OH

O

H H

n n

OO

O

O O

O

(a)

(b)

Figure 1.2.7. Two-stage synthesis of poly(imide)s

During the imidization stage, the high Tg of the system often leads to low mobility of the chains.

Therefore, the type of conformation that favors a cyclodehydration reaction may be hindered. As

a result, quantitative imidization is often hard to achieve. Moreover, the hydrolytic degradation

of the residual amic acid units can cause chain scission. To improve both the solubility and

processability of aromatic poly(imide)s, researchers have introduced bulky groups and/or

flexible linkages, and have also utilized monomers containing meta linkages.42, 43 This allows the

polymerization to be carried out under moderate conditions in one stage. One such example is

the production of Ultem® by GE (Figure 1.2.8).

14

C

CH3

CH3

OOOO

O

O

O

O

H2N N 2H+

C

CH3

CH3

OONN

O

O

O

On

o-dichlorobenzene180oC

Figure 1.2.8. One-stage synthesis of poly(imide)s with improved solubility

1.2.1.3. Aromatic 5-membered-ring heterocyclic polymers

Aromatic heterocyclic polymers such as poly(benzimidazole)s, poly(benzoxazole)s and

poly(benzthiazole)s are well known to have excellent thermal/chemical stability, as well as good

mechanical properties. As shown in Figure 1.2.9, they are produced from the condensation

reaction of a dicarboxyl (or phenyl ester) with a tetramine, a bis-o-aminophenol, or a

bis-o-aminothiophenol, shown sequentially.44-48 Although fabrication of proton exchange

membranes based on these materials has been attempted, particular attention has focused on the

potential applications of phosphoric acid-doped poly[2,2’(m-phenylene)-5,5’bibenzimidazole]

(PBI) membranes. 49-53

15

n+HOOC COOH

H2N

NH2

OH

HO

HOOC COOHH2N

NH2

SH

HS

O

N O

N

S

N S

N+

n

(b)

(c)

n+HOOC COOH

H2N

NH2

NH2

H2N N

N N

N

H

H

(a)

Figure 1.2.9. General scheme for the synthesis of (a) poly(benzimidazole)s, (b) poly(benzoxazole)s and

(c) poly(benzthiazole)s

The conventional way to synthesize PBI and other poly(benzimidazole)s is via the melt

polymerization of a tetramine with the diphenyl ester of a diacid (Figure 1.2.10). Solution

polymerizations of tetramines with dicarboxylates or diacids in solvents such as polyphosphoric

acid have also been developed. These synthetic techniques can be conducted under less harsh

conditions, but also afford completely soluble polymers.54

H2N

H2N

NH2

NH2

C CO O

O O

N

NN

N

H Hn

+

1. 290oC2. 390oC

Figure 1.2.10. Synthesis of PBI by melt polymerization

16

1.2.1.4. Poly(p-phenylene) derivatives

Poly(p-phenylene)s (PPP) derivatives are important high-performance engineering materials that

have received significant attention—particularly due to their excellent mechanical properties and

thermal oxidative stability.55-57 The synthesis, processing and characterization of unsubstituted

PPPs are all challenging due to the intractability of the rod-like chains. Therefore, a wide variety

of lateral substituents has been introduced into the PPP main chains to improve its solubility. One

of the most widely used methods for preparing substituted PPPs is by the Ni(0)-catalyzed

coupling of dihalide monomers.58-61 In addition to homopolymers, statistical copolymers can also

be synthesized by the copolymerization of substituted and unsubstituted monomers (Figure

1.2.11).

Cl Cl

R

n

RNiCl2, Zn, PPh3

Figure 1.2.11. Synthesis of substituted PPPs via Ni(0) coupling

The palladium-catalyzed cross coupling of aromatic bromides with aromatic boronic acids has

been reported (Suzuki coupling).62, 63 Such reactions are suitable for synthesizing homopolymers

or copolymers bearing alternating arrays of substituted and unsubstituted phenyl rings (Figure

1.2.12).

17

Br Br

R

R

Pd(0), Na2CO3(HO)2B B(OH)2+

Br Br

R

R

B BO

OO

O+

R

R

n

R

Rn

Pd(0), NaHCO3

Figure 1.2.12. Synthesis of PPP alternating copolymers via Suzuki coupling

Substituted poly(2,5-benzophenone)s are an important family of thermal-oxidatively stable PPP

derivatives. The pendent phenyl rings not only enhance the solubility of the polymers, but also

facilitate further modification with various functional groups. The monomers are synthesized by

the Friedel-Crafts acylation of substituted benzene with 2,5-dichlorobenzoyl chloride (Figure

1.2.13).17, 59, 64

Cl Cl

COOH

Cl Cl

COCl

X

Cl Cl

C O

X

AlCl3

SOCl2 C O

X

n

NiCl2, Zn, PPh3

NMP, 80oC

X = H, X, O,

Figure 1.2.13. Synthesis and polymerization of 2,5-dichlorobenzophenone monomers

18

1.2.2. Fabrication of PEMs: introduction of proton-conducting moieties

1.2.2.1. Post sulfonation of poly(arylene ether)s

The most widely used proton-conducting moiety is the sulfonic acid group, primarily because of

its availability, high acidity, and ease of introduction into the polymer backbone. Sulfonated

copolymers using either poly(arylene ether sulfone)s or poly(arylene ether ketone)s are most

commonly obtained via the electrophilic sulfonation of the polymers’ aromatic rings (post

sulfonation).65-70 Common sulfonating agents include concentrated sulfuric acid, fuming

sulfuric acid, sulfur trioxide, chlorosulfonic acid, etc. Electrophilic substitution reactions are

favored by electron-donating substituents on the phenyl ring. Therefore, post sulfonations of

poly(arylene ether)s are generally believed to take place on the activated phenyl rings, rather

than on the phenyl rings directly attached to the deactivating sulfone groups or ketone groups

(Figure 1.2.14). In addition, usually no more than one sulfonic acid can be introduced to each

repeat unit.65

C

CH3

CH3

O S

O

O

O

SO3H

O C

O

O

SO3H(a) (b)

Figure 1.2.14. Examples of post-sulfonated poly(arylene ether sulfone)s and poly(arylene ether ketone)s

19

The choice of sulfonating agent and reaction conditions has been found to significantly influence

the properties of the modified polymer.66-68 Moreover, although higher degrees of sulfonation

have been achieved using stronger sulfonating agents such as fuming sulfuric acid and

chlorosulfonic acid, they also result in unwanted side reactions that can lead to the degradation of

the polymer chains. In contrast, the use of mild sulfonating agent, such as concentrated sulfuric

acid or trimethylsilylchlorosulfonate, has been studied with little or no polymer degradation

reported.67, 68 However, such reactions usually involve longer reaction times and low sulfonation

efficiencies. Therefore, no reliable correlations have been established between the amount of

sulfonating agent used, reaction time, and the degree of sulfonation. In addition, the selection of

solvents or solvating agents has also been shown to be important. Even though a homogeneous

reaction medium is considered to be critical for obtaining high sulfonation efficiency,

reproducibility is generally not good.69

1.2.2.2. Direct copolymerization of sulfonated monomers: preparation of poly(arylene ether)

random copolymers

Another method for generating sulfonated copolymers involves the copolymerization of

sulfonated and unsulfonated monomers to form random copolymers. This method has advantages

relative to post-modification with regard to control of the degree and location of sulfonation.

Also, the acidity of the sulfonic acid group may be improved because this methodology

facilitates its introduction onto the more electron-deficient phenyl rings, i.e., those connected to

the sulfone or ketone groups.

20

Based on a method proposed by Ueda et al.,70 McGrath and coworkers71 sulfonated

4,4’-dichlorodiphenyl sulfone (DCDPS) to synthesize 3,3’-disulfonated 4,4’-dichlorodiphenyl

sulfone (SDCDPS) (Figure 1.2.15). They also investigated the copolymerizations of bisphenols

with SDCDPS and DCDPS.13, 14, 72

S

O

O

Cl ClSO3/H2SO4

110oCS

O

O

Cl Cl

HO3S

SO3H

S

O

O

Cl Cl

NaO3S

SO3Na

NaOH

pH 6~7

Figure 1.2.15. Synthesis of 3,3’-disulfonated 4,4’-dichlorodiphenyl sulfone (SDCDPS)

As shown in Figure 1.2.16, a series of so-called BPSH-xx random copolymers were synthesized

by polymerizing 4,4’-biphenol, DCDPS, and SDCDPS, where xx represents the degree of

disulfonation. We have shown that this variable can be precisely controlled by the molar feed

ratio of sulfonated to unsulfonated monomer. Generally, proton conductivity and water uptake

increase almost linearly with the degree of disulfonation. However, when x exceeds 50%,

water uptake increases dramatically and the membrane swells in water like a hydrogel. Based on

these results, Kim et al. proposed that the ”percolation limit” was exceeded at that point.19, 73

Among the various copolymers we tested, BPSH35 was found to have the best combination of

proton conductivity and water uptake; consequently, it is thought to be one of the most promising

alternatives to the perfluorosulfonic-acid PEMs.

21

HO OH S

O

O

Cl Cl

NaO3S

SO3Na

O OO O S

O

O

S

O

O

KO3S

SO3K

K2CO3NMP

Toluene

150oC/4h190oC/18~36h

+ +

1-x x

O OO O S

O

O

S

O

O

HO3S

SO3H1-x x

S

O

O

Cl Cl

H2SO4

Figure 1.2.16. Synthesis of BPSH-xx random copolymers

As illustrated in Figure 1.2.16, once the potassium salt forms of the copolymers are synthesized,

they are converted into the sulfonic acid form through treatment with sulfuric acid. We

determined that the conditions of the acidification reaction strongly influence the morphology of

the membrane. Typically, a so-called “Method 2” is used,74 in which the membrane is boiled in

0.5M H2SO4 for 2h, then boiled in deionized water for 2h. This contrasts to “Method 1,” where

the polymer film is immersed in 1.5M H2SO4 at room temperature for 24h, then in deionized

water for 24h. “Method 2” results in higher proton conductivity and water uptake, and more

distinct ionic-hydrophobic microphase separation as indicated by AFM phase images.19, 73

The synthesis of sulfonated poly(ether ketone) random copolymers via the direct polymerization

of sulfonated monomers has also been studied. Wang et al. was the first to report the synthesis of

the disulfonated difluoride ketone monomer, sodium 5,5'-carbonylbis(2-fluorobenzenesulfonate)

22

(SDFBP) (Figure 1.2.17),75-77 by sulfonating 4,4’-difluorobenzophenone (DFBP) using a

procedure similar to the synthesis of SDCDPS. The sulfonated and unsulfonated monomers were

copolymerized with bisphenol A to afford sulfonated poly(arylene ether ketone) copolymers. As

reported, these copolymers were amorphous and their solubility was significantly influenced by

the degree of sulfonation.

C

O

F F

NaO3S

SO3Na

Figure 1.2.17. Sodium 5,5'-carbonylbis(2-fluorobenzenesulfonate) (SDFBP)

A variety of sulfonated and unsulfonated difluoride ketone monomers were copolymerized with

different bisphenols to prepare PAEK-based PEMs (Figure 1.2.18).78-80 The changes in

electrochemical properties associated with the degree of disulfonation were similar to those

observed for the poly(arylene ether sulfone) copolymer membranes. However, the PAEK-based

PEMs generally displayed lower proton conductivity at comparable IEC or water uptake values.

23

O X O Y O X O Zm n

C

CF3

CF3

S

F3C

CF3

C

O

C

O

C

O

C

O

C

O

C

OHSO3 SO3H

C

O

C

OHSO3 SO3H

C

O

C

OHSO3 SO3H

X =

Y =

Z =

, ,

, ,

Figure 1.2.18. Structures of sulfonated poly(arylene ether ketone) random copolymers studied

1.2.2.3. Sulfonated poly(imide)s: hydrolytic stability issues

Due to the instability of imide groups under post sulfonation conditions, sulfonated aromatic

poly(imide)s have primarily been synthesized by directly polymerizing sulfonated diamine

monomers. However, phthalic poly(imide)s with five-membered rings have been found to

degrade quickly and become brittle in fuel cell environments. This is attributed to the

hydrolysis of the imide structure under acidic conditions, which leads to chain scission.

Conversely, naphthalenic poly(imide)s are generally considered to be much more stable16, 81

(Figure 1.2.19), even though studies on model compounds have questioned the hydrolytic

stability of these materials. Research on polyimide-based PEMs is ongoing.82

24

NN

O

O O

O

NN

O

O

O

O(a) (b)

Figure 1.2.19. Structures of (a) phthalic and (b) naphthalenic imide units

The most frequently used dianhydride monomer is 1,4,5,8-tetracarboxylic dianhydride (NDA),

which has been copolymerized with sulfonated and unsulfonated diamines to synthesize a variety

of partially-sulfonated poly(imide)s, an example of which is shown in Figure 1.2.20.16 The

sulfonated diamine is usually converted to the triamine salt to improve solubility. After

polymerization, acid form copolymers are obtained by treatment in dilute sulfuric acid.

NH2H2N OH2N NH2

SO3NH(Et)3

(Et)3HNO3S

OO

O

O

O

O

NN

O

O

O

O

NN

O

O

O

O

SO3NH(Et)3

(Et)3HNO3S

Om n

+ +

NDA sulfonated diamine unsulfonated diamine

Figure 1.2.20. Synthesis of naphthalenic sulfonated poly(imide) random copolymers

The structures of the sulfonated and unsulfonated diamines do not have to be the same for

copolymer synthesis. In fact, they were found to be closely related to the solubility and

hydrolytic stability of the copolymer.16, 83-86 As a general rule, the solubility of a copolymer in an 25

organic solvent can be improved by using unsulfonated diamines containing flexible linkages,

such as phenyl-ether bonds, or bulky groups.15 In contrast, introducing such groups into

sulfonated diamine units has been found to lead to better stability in water. Figure 1.2.21 shows

the sulfonated and unsulfonated diamine monomers that have been used in this study.

H2N NH2

SO3H

HO3S

H2N NH2

HO3S SO3H

NH2

HO3S

H2N

SO3H

O

H2N O C

CF3

CF3

O NH2

HO3S

SO3H

H2N O S

O

O

O NH2

HO3S

SO3H

H2N NH2 NH2H2N O

H2N NH2

H2N NH2

H2N O S

O

O

O NH2

(a)

(b)

Figure 1.2.21. Structures of (a)sulfonated and (b)unsulfonated diamine monomers used

1.2.2.4. Sulfonated poly(benzimidazole)s, poly(benzoxazole)s and poly(benzthiazole)s

Sulfonated poly(benzimidazole), poly(benzoxazole) and poly(benzothiazole) homopolymers

have been synthesized by the direct polymerizations of sulfonated diacid monomers (Figure

1.2.22).87-90

26

Cl-+H3N SH

HS NH3+Cl-

HOOC COOH

SO3H

H2N NH2

H2N NH2

H2N NH2

HO OH

+

+

+

HOOC COOH

SO3H

HOOC COOH

SO3H

N

O

O

N

SO3H

n

N

S

S

N

SO3H

n

N

N

N

N

SO3H

H

Hn

nn

n

n n

n

Figure 1.2.22. Synthesis of sulfonated heterocyclic homopolymers

Qing et al.91, 92 reported the synthesis of partially sulfonated poly(benzimidazole) random

copolymers by the copolymerization of sulfonated and unsulfonated diacid monomers, as shown

in Figures 1.2.23 and 1.2.24. The PEM properties of these materials have not yet been studied in

detail.

+ +

PPA

COOH

HOOC

SO3H

SO3H

C

CF3

CF3

HOOC COOH

H2N

H2N

NH2

NH2

N

N

N

N

H

H

SO3H

SO3H

N

N

N

N

H

H

C

CF3

CF3m n

Figure 1.2.23. Synthesis of sulfonated poly(benzimidazole) random copolymers

27

S

O

O

HOOC COOH

H2N

H2N

NH2

NH2

m n

N

N

N

N

H

H

S

O

O

N

N

N

N

H

H

SO3H

+ +

SO3H

HOOC COOH

PPA

Figure 1.2.24. Synthesis of sulfonated poly(benzimidazole) random copolymers

The post-sulfonation of poly(benzimidazole)s have also been reported. Both PBI (produced from

A-A and B-B monomers) and AB-PBI copolymers can be sulfonated by immersing the

membrane in sulfuric acid, then heating it to about 400oC.93-95 The proposed mechanism is

shown in Figure 1.2.25. It was found that a sulfonation degree of about 0.6 mole H2SO4 per

repeat unit was usually achieved, regardless of the concentration of the sulfuric acid. However,

these materials showed low proton conductivity, presumably due to the protonation of the

nitrogen in the imidazolium ring. Therefore, while these membranes are not suitable for use as

PEMs themselves, they have been found to be useful in fabricating phosphoric acid-doped

PEMs.

28

N

N

N

N

H

H

H

+

HSO4-

N

N

N

N

H

H

H

+

-O3S

H2SO4

n

n

N

N

H

Hn

+

HSO4-

N

N

H

Hn

+

-O3S

H2SO4N

NH

n

N

NN

N

H Hn

(a)

(b)

Figure 1.2.25. Sulfonation of (a)PBI and (b)ABPBI

When a poly(benzimidazole) membrane is imbibed in phosphoric acid, the basic polymer

absorbs up to 75% of the acid. These acid-doped BPI and ABPBI membranes have sparked the

interest of many research groups as potential PEM candidates.49, 51-53 These membranes do not

require water for proton conduction, can tolerate working temperature as high as 200oC, and are

impermeable to gases and methanol. Their conductivity was found to increase with an increase in

the doping level (acid content).49, 52, 53 One of the problems for systems with higher acid contents

is the issue of acid molecule loss over time. In comparison to the membranes we prepared by

soaking them in acid, membranes cast directly from a solution of PBI and phosphoric acid are

thought to possess better long-term durability.93 This method, however, is only applicable to

polymers that do not precipitate in the presence of phosphoric acid.

29

Sulfonated poly(benzimidazole)s prepared by post sulfonation have also been used as hosts for

phosphoric acid doping.87, 94 Sulfonated polymers that have been subjected to the same acid

doping conditions are reported to have much higher conductivity than their unsulfonated

counterparts. While some researchers have attributed this to the possibility that the presence of

sulfonic acid groups in the polymer facilitates phosphoric acid doping,94 others have suggested

that excess sulfonic acid groups themselves might also contribute to proton conduction.87

1.2.2.5. Sulfonation of poly(2,5-benzophenone)s

The fabrication of potential PEMs obtained via the sulfonation of substituted (Figure 1.2.26, a,b)

and unsubstituted (Figure 1.2.26, c) poly(2,5-benzophenone)s has been studied.17, 18, 96, 97 The

sulfonation reactions of poly(4-phenoxybenzoyl-1,4-phenylene) (Figure 1.2.26, a) and

poly(4-phenyl-2,5-benzophenone) (Figure 1.2.26, b) are generally believed to take place on the

pendent phenyl rings, which have the highest electron density. For poly(2,5-benzophenone)

(Figure 1.2.26, c), however, it was argued that the phenyl rings in the main chains, rather than the

pendant ones, are the ones subjected to sulfonation.98 The synthesis of sulfonated

poly(4-phenoxybenzoyl-1,4-phenylene)s was also achieved by the condensation reaction of

poly(4-fluoro-2,5-benzophenone) with 4-hydroxybenzenesulfonic acid (Figure 1.2.27).

30

C O

n

SO3H

C O

O

n

C O

n

C O

n

SO3H

SO3H

(a) (b) (c) (d)

SO3H

Figure 1.2.26. Proposed structures of sulfonated poly(2,5-benzophenone)s

C O

F

n

NMP/Toluene

K2CO3, 180oCHO SO3Na+

H2SO4

C O

O

SO3H

n

Figure 1.2.27. Sulfonation of poly(4-fluoro-2,5-benzophenone)s

The main obstacle to applying sulfonated PPP derivatives as PEMs is their poor

membrane-forming characteristics. Even though pendent substituents generally enhance

solubility, the polymer chains were still so rigid that in most cases we could only synthesize

brittle films with poor mechanical strength. The synthesis of block copolymers bearing

substituted PPP blocks and flexible poly(arylene ether) blocks have been attempted to improve

the film-forming capabilities of PPPs. Unsulfonated PPP-poly(arylene ether ketone) multiblock

copolymers were synthesized by Sheares et al.59, 64 McGrath et al. developed PPP-poly(arylene 31

ether sulfone) multiblock copolymers containing either sulfonated PPP or sulfonated

poly(arylene ether sulfone) segments.18, 99 The synthesis of hydrophilic-hydrophobic multiblock

copolymers will be discussed in greater detail in Chapter 4.

1.3. Fluorinated aromatic high performance copolymers for PEMs

1.3.1. Moderately fluorinated copolymers containing hexafluoroisopropylidene units

Fluorinated copolymers are well known for their thermal and chemical stability and mechanical

strength. Therefore, the use of partially fluorinated copolymers as backbones for PEMs has

been widely studied.13, 15, 100-104 For PEM applications, introducing some level of fluorine content

has two advantages. Firstly, the water uptake of the membrane can be reduced due to the high

hydrophobicity of the fluorine units. Secondly, the membrane may be more compatible with

Nafion-based electrodes in comparison to hydrocarbon-based systems, which is likely to lead to

higher fuel cell performance and better MEA stability.

OH

F3C C CF3

O

+CF3SO3H HO C

CF3

CF3

OH2

Figure 1.3.1. Synthesis of 6F-BPA

As shown earlier, 4,4’-hexafluoroisopropylidene diphenol (6F-BPA) as an alternative to

non-fluorinated bisphenols has been widely used in synthesizing partially sulfonated

poly(arylene ether)s, primarily for the purpose of introduce hexafluoroisopropylidene linkages.

Similar to the synthesis of bisphenol-A, 6F-BPA can be prepared via the condensation reaction of 32

hexafluoroacetone with phenol (Figure 1.3.1).105 Using 6F-BPA, biphenol, and bisphenol-A as

the bisphenol monomers, Harrison et al.13 synthesized partially fluorinated and non-fluorinated,

partially disulfonated poly(arylene ether sulfone)s, as shown in Figure 1.3.2 (a). TGA studies

showed that the fluorinated copolymers had higher thermal stability. At similar degrees of

disulfonation, the partially fluorinated membranes displayed lower water uptake, but similar

proton conductivity.

OC

CF3

CF3

O R OO C

CF3

CF3

S

O

O

HO3S

SO3H1-x x

S

O

O

(a) R = (b) R =

CN

Figure 1.3.2. Structures of partially sulfonated, partially fluorinated poly(arylene ether sulfone)s

Partially fluorinated nitrile-functional poly(arylene ether sulfone)s with varying degrees of

disulfonation were also synthesized (Figure 1.3.2, b).101 We then compared these copolymer

membranes to our BPSH-xx membranes and observed similar improvements in stability and

water uptake. A more systematic study was undertaken by Sankir et al., who synthesized partially

disulfonated poly(arylene ether benzonitrile)s with varying degrees of fluorination (Figure

1.3.3).100 Although a substantial decrease in water uptake was observed as the content of

hexafluoroisopropylidene units increased, the membranes displayed only a minor loss in

conductivity. Thermal oxidative stability was increased as well with the increase in fluorination.

The 5% weight loss temperature reached 400oC for the >75% fluorinated copolymers.

33

O

CN

RO O S

O

O

HO3S

SO3H

RO1-x x

C

CF3

CF3

R = controlled amount of and

Figure 1.3.3. Partially sulfonated, partially fluorinated poly(arylene ether sulfone benzonitrile)s

A major disadvantage of hexafluoroisopropylidene-containing copolymers is that 6F-BPA has

lower reactivity toward activated aryl dichloride monomers in comparison to the

hydrocarbon-based bisphenol monomers, which we attributed to the electron-withdrawing nature

of the hexafluoroisopropylidene linkage. It has, therefore, been difficult to produce high

molecular weight copolymers—although the use of longer polymerization times, higher

temperatures, and higher reaction solution concentrations have been somewhat useful in

overcoming this challenge.

1.3.2. Highly fluorinated poly(arylene ether)s containing perfluorophenylene units

Poly(arylene ether)s bearing perfluorophenylene groups have been synthesized from bisphenol

and perfluorinated difluoride monomers via step growth polymerizations, as shown in Figure

1.3.4. These polymers were initially developed as optical wave guide materials for potential

telecommunication applications 106-110. The vibration absorption overtone of C-H bonds can be

decreased by substituting the hydrogen with fluorine so that the optical propagation losses can be

34

reduced. The most widely used perfluorinated monomer is decafluorobiphenyl, in which two

perfluorophenyl rings are directly attached to each other. Because of decafluorobiphenyl’s rigid

structure, the other monomer (bisphenol) used in the polymerization must have a flexible linkage

between the two phenyl rings, such as an isopropylidene or a sulfone group. If a rigid bisphenol

(e.g., hydroquinone or 4,4’-biphenol) is used, insoluble, it is thought that semicrystalline

materials will precipitate out of the solution at the start of the reaction, thereby inhibiting the

formation of a polymer.

+XF F

F F

F F

F F

F F

YHO OH

YO OX

F F

F F

F F

F Fn

X = a bond, Y =S

O

O

C

O

S ,,, ,, C

CH3

CH3

C

CF3

CF3

S S

O

O

DMAc, K2CO3

Figure 1.3.4. Synthesis of poly(arylene ether)s containing perfluorophenylene units

Decafluorobiphenyl and other perfluorinated aromatic compounds have been synthesized by a

saturation-rearomatization method.111-113 As shown in Figure 1.3.5, the perchlorinated precursors

are treated with VF5 or BrF3-SbF5 to form halogenated cycloalkanes, which are then

dechloronated by zinc. Using cobalt trifluoride as the fluorinating agent, the fluorination of

unsubstituted aromatic compounds has also been reported.111

35

Cl Cl Cl Cl

Cl

ClClCl

Cl

Cl BrFCl BrFCl

F

F F

F F

F

F F

F F

BrF3-SbF5

CFCl2CF2Cl

Zn powderdioxane

Cl Cl

Cl

Cl

ClCl

Cl

Cl

F F

F

F

FF

F

F

BrFCl BrFCl

Cl

Cl

Cl

Cl

Cl

Cl

F

F

F

F

F

F

ClF

BrF3-SbF5 Zn

VF5 Zn

C6H3F9, etcKOHCoF3

F

F

F

F

F

F

Figure 1.3.5. Synthesis of perfluorinated aromatic compounds by saturation-aromatization

Perfluorinated compounds can also be obtained by treating perchlorinated counterparts with KF

via a nucleophilic aromatic substitution mechanism (Figure 1.3.6).111-113

Cl Cl Cl Cl

Cl

ClClCl

Cl

Cl

F

F F

F F

F

F F

F F

Cl

Cl

Cl

Cl

Cl

Cl

F

F

F

F

F

FKF

KFSulfolane, 230oC

400oC

Figure 1.3.6. Synthesis of perfluorinated aromatic compounds by nucleophilic aromatic substitution

36

O

F F

F F

F

O F

F

O C

CH3

CH3

OC

CH3

CH3

CH3C

CH3

Figure 1.3.7. Reactions of ortho-position fluorine leading to branching and/or gelation

Due to the large number of fluorine substituents on the phenyl rings, the C-F bonds at the

para-positions are highly activated and the monomer is much more reactive toward the attack of

phenolate ions than other dihalide monomers. Thus, high molecular weight can be achieved at

much lower temperatures (100~120oC) and within shorter reaction times (4~5 h).106, 114, 115

Actually, these polymerizations must be conducted at lower temperatures because the fluorines at

ortho-positions are only slightly less reactive than those at para-positions, which may also

participate in the reaction.109, 116 This undesirable side reaction seems to become even more

pronounced with increasing reaction temperature, and may produce branched or even crosslinked

structures (Figure 1.3.7).116

Decafluorobiphenyl has been polymerized with various bisphenols routinely in DMAc at 120oC

and no gel formation has been reported.106, 114, 115 More reactive monomers such as

decafluorobenzophenone and bis(pentafluorophenyl)sulfone, however, are thought to have

poorer selectivity and have been reported to produce insoluble microgels when polymerized

under similar conditions.108, 117, 118 Zhou and coworkers109, 117 developed mild reaction conditions

37

to polymerize decafluorobenzophenone and bis(pentafluorophenyl)sulfone linearly with 6F-BPA,

where molecular sieves were used during the vapor phase to dehydrate the system.

Polymerizations were conducted at lower temperatures and were completed within 1h.

Kim et al. reported the direct copolymerization of 2,8-dihydroxynaphthalene-6-sulfonated

sodium salt (2,8-DHNS-6) with 6F-BPA and decafluorobiphenyl to synthesize sulfonated random

copolymers for use as PEMs (Figure 1.3.8).104 The degree of sulfonation ranged from 20 to 80

mol%. Since the copolymers had only one sulfonic acid group per repeat unit, they showed low

water uptake and proton conductivity compared to Nafion 117. High molecular weight

copolymers with higher degrees of sulfonation could not be obtained due to the restricted

mobility of the bulky, rigid DHNS monomer.

+F

F F

F F

F

F F

F F

CHO OH

CF3

CF3

O

F F

F F

F F

F F

O

F F

F F

CO O

F F

F F

CF3

CF3

NaSO3

m n

OH

OH

NaSO3

+

K2CO3DMSO

Figure 1.3.8. Direct synthesis of sulfonated perfluorinated poly(arylene ether) random copolymers

Due to limited availability of sulfonated bisphenols, sulfonated copolymers containing

perfluorophenylene units have been mostly synthesized by post-sulfonation methods. Lee et al.114

studied the sulfonation of two series of highly fluorinated poly(arylene ether)s by fuming sulfuric

38

acid (Figure 1.3.9). The IEC values were controlled by varying the amount of sulfonic acid

and/or the reaction time. The copolymer membranes showed higher water uptake and lower

conductivity than Nafion 117 at similar IEC values. Nevertheless, while the more highly

fluorinated series, i.e. copolymers containing hexafluoroisopropylidene units (Figure 1.3.9, a),

displayed much lower water uptake than the relatively lower fluorinated series (Figure 1.3.9, b),

they also achieved comparable proton conductivity. It is likely that the electron-negative

hexafluoroisopropylidene groups both increased the hydrophobicity and enhanced the acidity of

the sulfonic acid groups.

F F

F F

CO O

F F

F F

CF3

CF3

F F

F F

CO O

F F

F F

CF3

CF3

F F

F F

CO O

F F

F F

CH3

CH3

F F

F F

CO O

F F

F F

CH3

CH3

Fuming H2SO4

Chloroform

Fuming H2SO4

Chloroform(a)

(b)

n

n

n

n

SO3H SO3H

SO3HSO3H

Figure 1.3.9. Post sulfonation of fluorinated poly(arylene ether)s

Some bulky bisphenol monomers, although rigid, have also been used for the synthesis of high

molecular weight fluorinated copolymers. The good solubility of these polymers may be due to

the increased inter-chain spacing, which suppresses crystallinity. Hay et al.102 synthesized

unsulfonated fluorinated random copolymers (Figures 1.3.10 and 1.3.11). The bisphenols were

isocynate-masked to improve the monomers’ solubility in the reaction medium. The copolymers

were then sulfonated using chlorosulfonic acid to yield the sulfonated materials. Because the

other possible sulfonation locations were either sterically hindered or masked by methyl groups,

39

it was proposed that the sulfonation only occurred at positions para to the pendant phenyl rings

and thus was quantitative. This selective sulfonation technique may be superior to conventional

direct copolymerization methods because it yields random copolymer ionomers, while enabling

the degree of sulfonation to be easily controlled by the molar feed ratio of the comonomers.

PrNHCOO OCONHPr SPrNHCOO OCONHPr

O

O

F

F F

F F

F

F F

F F

+ +

SO O

O

O

O O

F F

F F

F F

F F

F F

F F

F F

F Fm n

SO O

O

O

O O

F F

F F

F F

F F

F F

F F

F F

F F

SO3HHO3S

SO3HHO3S

m n

K2CO3DMAc

1. ClSO3H, CH2Cl22. KOH, DMSO3. HCl

Figure 1.3.10. Synthesis and sulfonation of fluorinated poly(arylene ether sulfone) random copolymers

40

PrNHCOO OCONHPr PrNHCOO OCONHPr

O OO O

F F

F F

F F

F F

F F

F F

F F

F Fm n

O OO O

F F

F F

F F

F F

F F

F F

F F

F F

SO3H SO3H

SO3H

m n

F

F F

F F

F

F F

F F

1. ClSO3H, CH2Cl22. KOH, DMSO3. HCl

K2CO3DMAc

+ +

Figure 1.3.11. Synthesis and sulfonation of fluorinated poly(arylene ether sulfone) random copolymers

1.4. Hydrophilic-hydrophobic block copolymer systems as PEMs

The proton conductivity of PEMs based on sulfonated random copolymers has been shown to be

strongly dependent on humidity. Although the conductivity of BPSH and other random

copolymer membranes under fully hydrated conditions (i.e. in liquid water) is comparable to that

of Nafion®, it tends to decrease markedly as the level of hydration decreases.26 This trend may be

because a substantial number of water molecules are required to establish connectivity among

the randomly distributed sulfonic acid groups. Conversely, Nafion® is a modestly crystalline

copolymer which may be above its Tg at room temperature. It has also been reported to show a

nanophase separated morphology in the dry state, featuring ionic clusters that are interconnected

by narrow ionic channels.4 The challenge, therefore, lies in developing alternative PEMs which

41

feature associated ionic domains at even low hydration levels. Nanophase separated

hydrophilic/hydrophobic block copolymer ionomers are desirable for this purpose. In contrast to

what occurs in random copolymers, in block copolymer membranes the ionic groups are

selectively incorporated into one or more blocks and may exist in ordered sequences. Continuous

proton conducting channels may thus be formed even at low hydration levels. It is postulated that

high proton conductivity can therefore be sustained under partially hydrated conditions.

1.4.1. Block copolymer PEMs based on sulfonated styrenics and hydrogenated polydienes

Polystyrene-based triblock copolymer PEMs have been developed from commercially available

Kraton G as low-cost alternatives to Nafion®, and some of them have been commercialized. The

most widely studied materials are sulfonated styrene-ethylene/butylenes-styrene (S-SEBS)

membranes developed by Dais Analytics (Figure 1.4.1).24, 119 The unsulfonated precursor, SEBS,

is a well-known commercial polymer, which has been synthesized by the sequential anionic

polymerization of styrene and butadiene, followed by the hydrogenation of the unsaturated

double bonds resulting from both the 1,2- and 1,4-addition of butadiene.11 The polystyrene

blocks are partially sulfonated to produce S-SEBS. The properties of a Dais S-SEBS PEM with

18% polystyrene and 55 mol% sulfonation of the PS blocks were studied by Wnek et al.24, 119

The membrane showed a proton conductivity of 0.085 S/cm when fully hydrated, as compared to

0.079 S/cm for Nafion 117. Its wt% water absorption (water uptake), however, was four times

higher than that of Nafion. X-ray and neutron scattering studies of the wet membrane indicated

co-continuous hydrophobic and hydrophilic domains.

42

In addition to commercially available S-SEBS membranes, several research groups have also

synthesized S-SEBS samples via the sulfonation of commercially-available SEBS block

copolymers that contain about 20~30 mol% styrene.120-124 The sulfonation reagents used include

chlorosulfonic acid, acetyl sulfate, etc., and membranes with up to 47 mol% sulfonation of the

PS block were studied. Generally, both the liquid water proton conductivity and water uptake

were found to increase with the increased level of sulfonation for these ionomer membranes.

CH2 CH CH2 CH CH2 CH2 CH2 CH

CH2

CH3

CH2 CH CH2 CH

SO3H SO3H

m n m

Figure 1.4.1. Structure of S-SEBS block copolymers

Similar to Nafion, the S-SEBS membranes showed high methanol permeability, which makes

them undesirable for DMFC applications. Won et al.125 studied S-SEBS membranes that were

modified by the addition of a thin layer, which was introduced to the top of the membrane by

plasma treatment in the presence of maleic anhydride (Figure 1.4.2). This modification resulted

in a decrease in methanol crossover, as well as an undesirable loss of proton conductivity.

However, hydrolysis of the anhydride layer produced carboxylic acid groups which partially

recovered the diminished proton conductivity.

43

Figure 1.4.2. Modification of S-SEBS surface with plasma treatment followed by hydrolysis

Another important type of styrene-based block ionomers for PEMs is sulfonated

poly(styrene-isobutylene-styrene) (S-SIBS) (Figure 1.4.3).10, 12, 25, 126 These copolymers possess a

completely saturated mid-block, which thus eliminates the need for post-polymerization

hydrogenation. Like S-SEBS membranes, they are obtained via the partial sulfonation of the

polystyrene blocks of commercially-available SIBS triblock copolymers. It is not clear how the

commercial precursors are synthesized, but some authors have reported the synthesis of SIBS

copolymers by living cationic polymerization at -80°C using a

1,3-di(2-chloro-2-propyl)-5-tert-butylbenzene (t-Bu-DCC)-TiCl4 initiating system.127

44

CH2 CH CH2 CH CH2 C

CH3

CH3

SO3H

CH2 CH CH2 CH

SO3H

m n m

Figure 1.4.3. Structure of S-SIBS block copolymers

Elabd et al.12, 25 studied the electrochemical properties of S-SIBS membranes with IECs of up to

1.0. Sulfonation of the block copolymer was achieved using acetyl sulfate in methyl chloride,

after which proton conductivity and methanol permeability were measured and compared to

those of Nafion 117. Interestingly, very different property characteristics were observed in the

plane of the membrane as compared to normal to the plane. Specifically, the selectivities of the

membranes (i.e. proton conductivity/methanol permeability) were higher than those of Nafion

117 at both high and low temperatures when conductivity was measured in the plane. When it

was measured normal to the plane, however, the selectivities were much lower, leading to much

poorer DMFC performance than Nafion. This difference was explained in terms of a lamellar

morphology with a preferred orientation in the plane, as revealed by small angle X-ray scattering

(SAXS) studies. Elabd el al. also synthesized S-SIBS copolymers with high levels of sulfonation

(up to 82 mol% sulfonation and 2.04 IEC).10 These highly sulfonated membranes were found to

be elastic hydrogels that swelled dramatically in water.

Shi et al.128, 129 described the synthesis of partially fluorinated styrenic block ionomer membranes

where the hydrophobic blocks were the random copolymers of vinylidene difluoride (VDF) and

45

hexafluoropropylene (HFP). The block copolymers were prepared via the atom transfer radical

polymerization of styrene initiated by Cl-terminal poly(VDF-co-HFP) macroinitiators, which

were synthesized using an emulsion polymerization of VDF and HFP in the presence of CHCl3

(Figure 1.4.4). The copolymers were partially sulfonated by acetyl sulfate to generate ionomers

bearing up to ~50 mol% sulfonation. The membranes showed comparable proton conductivity

and water uptake to those of S-SEBS PEMs. The conductivity increased with the degree of

sulfonation (DS) of the PS block, but reached a maximum at a DS of 40 mol%. Although the

membrane with 49 mol% DS had 388% water uptake, its conductivity similar to the membrane

with 40 mol% DS. This was likely the result of the percolation limit being exceeded and the

membrane becoming a hydrogel. A similar trend (maximum conductivity at a certain IEC) has

been observed for other PEMs.

CF2 CH2 CF2 CF CF3+CHCl3

chain transferpolymerization

H CH2 CF2 CF2 CF

CF3

CCl3x yyx

styreneCuCl/bpy

H CH2 CF2 CF2 CF

CF3

CCl2 CH2 CH Clyx

H CH2 CF2 CF2 CF

CF3

CH2 CH CH2 CH

SO3H

yx

CH3COOSO3H

Figure 1.4.4. Synthesis of partially sulfonated (PVDF-ran-PHFP)-b-PS copolymers

46

1.4.2. Multiblock copolymers synthesized by step or condensation polymerization

Unlike aliphatic block copolymers based on styrenic and other vinyl monomers, the synthesis of

block copolymers having thermally stable aromatic moieties cannot be achieved by chain-growth

living copolymerization. Instead, telechelic oligomers with appropriate end groups and molecular

weights can be copolymerized by step-growth condensation polymerization to form multiblock,

or segmented, copolymers. The wholly aromatic oligomers themselves also have to be prepared

through condensation polymerizations. First, some multiblock copolymers containing aliphatic

blocks will be reviewed.

1.4.2.1. Partially aromatic multiblock copolymers

Zhang et al.130, 131 synthesized a series of multiblock copolymers containing poly(arylene ether

sulfone) (PAES) and partially sulfonated polybutadiene (PB) blocks. As shown in Figure 1.4.5,

the copolymers were obtained by coupling the telechelic PEAS and PB oligomers. Commercially

available carboxyl-terminated polybutadiene was treated with SOCl2 to give the Cl-terminal

oligomer. The PAES oligomers were obtained by the base catalyzed condensation polymerization

of 4,4’-dichlorodiphenyl sulfone (DCDPS) with 4,4’-isopropylidenediphenol (Bisphenol-A),

which had previously been reported by McGrath et al.18 The products were then reacted with

aminophenol to afford the amino-terminated oligomers (PAES-NH2). A linkage was established

between the two blocks by reacting the endgroups. The PB block was then partially sulfonated

by acetyl sulfate. The flexibility of the sulfonated PB blocks was thought to facilitate the

47

aggregation of the sulfonic acid groups into ion-rich channels, whereas the relatively rigid PAES

blocks provided mechanical strength. However, copolymers with only low degrees of sulfonation

(up to 11.5 mol% of PB) were synthesized with ion exchange capacities of only up to 0.62; their

proton conductivity at 25°C was no more than 0.03 S/cm. The water uptake values, on the other

hand, were unacceptably high considering the low IEC and proton conductivity values.

HO C

CH3

CH3

OH Cl S Cl

O

O

NMP/Toluene K2CO3160oC, 4h

190oC, 16h

+

O C OCH3

CH3

S

O

O

S

O

O

Cl Cl

HO NH250oC

O OCH3

CH3

S

O

O

S

O

O

O OH2N NH2n

O OCH3

CH3

S

O

O

S

O

O

O OHN NH

O

O

n m

CH3COOSO3H 75oC

NH

O

O

PAES NH PAES NH

O

O

OH

SO3H

NHm1 m2

n

ClCl

O

O

m

Figure 1.4.5. Synthesis of sulfonated PAES-PB multiblock copolymers

Yang et al.132, 133 investigated sulfonated PAES-PVDF multiblock ionomer membranes (Figure

1.4.6). The а,ω-dibromo PVDF oligomers were prepared via the radical telomerization of

VDF, using di-t-butylperoxide as the initiator and 1,2-dibromotetrafluoroethylene as the telogen.

48

The copolymers were synthesized by coupling hydroxy-terminated PAES oligomers with PVDF

telechelics. Partial sulfonation of the PAES block was carried out using (CH3)3SiSO3Cl. The IEC

of the sulfonated copolymers ranged from 0.78 to 2.18, and both proton conductivity and water

uptake increased with IEC. Partially sulfonated PAES homopolymers were also synthesized and

compared with the block copolymer PEMs. At low IEC, the latter were found to have higher

proton conductivity than the homopolymers. Because the degree of sulfonation of the polymers

was not clearly defined, and no information was available with respect to the block length or

weight ratio of the block copolymers, it was difficult to determine proton conductivity. Moreover,

the block copolymers displayed low molecular weight and it was unclear whether or not tough

membranes had been obtained.

HO C

CH3

CH3

OH Cl S Cl

O

O

DMAc/Toluene K2CO3 160oC, 16h

+

O S OO

OC C

CH3

CH3

HO OHCH3

CH3

Br CF2 CF2 CH2 CF2 Brm

m

NaH, RT 6h

O S OO

OC C

CH3

CH3

O OCH3

CH3

CF2 CF2 CH2 CF2n

1. (CH3)3SiSO3Cl2. CH3ONa3. HCl

mO S O

O

OC C

CH3

CH3

O OCH3

CH3

CF2 CF2 CH2 CF2n

SO3H

n

Figure 1.4.6. Synthesis of sulfonated PAES-PVDF multiblock copolymers

49

The major disadvantage of proton exchange membranes based on wholly aliphatic or partially

aliphatic backbones is their poor oxidative stability compared to wholly aromatic systems. This

makes them useful only in portable devices at low temperatures and low power densities. To

improve the thermal oxidative stability of their PAES-PB multiblock copolymers, Zhang et al.134

modified the membranes via the epoxidation of the unsulfonated butadiene units (Figure 1.4.7),

which was conducted using a mixture of formic acid and hydrogen peroxide. The modified

copolymer membranes showed only slight improvements in thermal stability, as suggested by the

higher degradation temperature as shown by thermal gravimetric analysis. The proton

conductivity of the membranes was more or less unaffected by the epoxidation. However, the

modification increased both water uptake and the swelling ratio in water.

NH

O

O

PAES NH PAES NH

O

O

OH

SO3H

NHm1 m2

NH

O

O

PAES NH PAES NH

O

O

OH

SO3H

NHO m1 m2

HCOOH

Figure 1.4.7. Epoxidation of sulfonated PAES-PB multiblock copolymers

50

1.4.2.2. Wholly aromatic multiblock copolymers

For the past few years there has been growing interest in developing block copolymer ionomers

having thermally and chemically stable aromatic backbones. As mentioned above, such

multiblock copolymers can be synthesized via the step growth polymerizations of hydrophobic

and hydrophilic telechelic oligomers. The copolymerization between the oligomers takes place

by coupling the end groups via a nucleophilic aromatic substitution mechanism, which is usually

carried out in polar solvents such as N-methyl-2-pyrrolidinone (NMP) and

N,N-dimethylacetamide (DMAc) in the presence of a salt.

Ghassemi et al.18 synthesized multiblock ionomers composed of poly(arylene ether sulfone) and

substituted poly(p-phenylene) blocks. As shown in Figure 1.4.8,

poly(4’-phenyl-2,5-benzophenone) oligomers, precursors for the hydrophilic blocks, were

prepared by a Ni(0) catalytic coupling of 2,5-dichloro-4’-phenylbenzophenone.

4-chloro-4’-fluorobenzophenone, which can participate in the reaction only on the chlorine end,

was used as the end-capper to afford fluorine-terminated telechelic oligomers. These oligomers

were sulfonated by H2SO4 to introduce one sulfonic acid group to each repeat unit. The

sulfonated oligomers were then coupled with phenoxide-terminated PAES oligomers to form

multiblock copolymers.

51

O

Cl Cl

O

Cl F+NiCl2, Zn, PPh3, bpy

NMP, 80oC

O

F

O

F

O

nn

O

F

O

F

O

n

SO3-Na+

H2SO4 50oC

NaCl

O O S O O

O

O m +

HO OH Cl S Cl

O

O+

DMActoluene K2CO3

DMActoluene K2CO3

O O

O O S O O

O

O

O

n m

SO3-Na+

hydrophobic oligomer hydrophilic oligomer

Figure 1.4.8. Synthesis of sulfonated PPP-PAES multiblock copolymers

The copolymers have been reported to be capable of forming strong and flexible membranes.

Our GPC analysis, however, showed that they had relatively low molecular weights (Mn<17,300

g/mol). The potassium salt forms of the copolymer membranes were converted to their acid

forms through treatment with H2SO4 at room temperature. Although their IEC values were

comparable to that of Nafion 117, these multiblock systems showed lower proton conductivity at

room temperature in liquid water.

As mentioned earlier, introducing sulfonic acid groups by post-sulfonation has some

disadvantages. These include unwanted side reactions, poor control in the reaction sites, and/or

the degree of sulfonation in the product. In contrast, direct copolymerizations of sulfonated

52

monomers offer a high quantitative degree of sulfonation, precise control of the sulfonation site,

and favorable ion exchange capacity. In a study by Taeger et al.,135 sulfonated poly(arylene ether

sulfone) telechelics, synthesized from the direct polymerization of hydroquinone 2-potassium

sulfate with bis(4-fluorophenyl)sulfone (DFDPS), were used as hydrophilic oligomers for the

preparation of multiblock copolymers for use as PEMs (Figure 1.4.9). The proton conductivity

and water uptake of these copolymer membranes were both higher than what Ghassemi et al.

obtained in their studies, i.e. they exhibited low proton conductivity at IECs (weight-based)

similar to Nafion. To compare the properties of block copolymers with those of random

copolymers with the same composition, random copolymers were also prepared from the

copolymerizations of DFDPS with sulfonated and unsulfonated hydroquinone. No significant

differences in either water uptake or proton conductivity were observed between the random and

multiblock copolymers having the same IEC. However, the multiblock copolymers showed

higher proton diffusion coefficients than the random copolymers with comparable IECs.

Therefore, the existence of a particular block sequence length might have played a role in the

establishment of nano-sized, ion-rich channels. Unfortunately, no further characterization studies,

which might have included assessing proton conductivity measurements at reduced hydration

levels, were undertaken.

53

F S O O S F

O

O

O

O

SO3-K+

nO O S O O

O

O m+

DMAcK2CO3

50oC

mS O O S

O

O

O

O

SO3-K+

O O S O O

O

On

hydrophilic oligomer hydrophobic oligomer

Figure 1.4.9. Synthesis of sulfonated poly(arylene ether sulfone) multiblock copolymers

The synthesis and characterization of a series of poly(arylene ether ketone) multiblock

copolymers were carried out by Shin et al.136 As shown in Figure 1.4.10,

1,3-bis-(4-fluorobenzoyl)benzene was polymerized with bisphenol-A (in excess) to form the

phenoxide-terminated hydrophobic oligomers. The hydrophilic oligomer can be prepared by the

direct polymerization of hydroquinone 2-potassium with 4,4’-difluorobenzophenone. However,

because a common solvent for both oligomers could not be found, the multiblock

copolymerizations were carried out in such a way that the hydrophilic blocks were generated

in-situ. Thus, it is questionable whether long sequences of sulfonated units had really been

formed in the copolymer.

n

180oC

C C

OO

F F

CH3

CH3

HO OH+

NMP/benzeneK2CO3

CH3

CH3

O O C C

OO CH3

CH3

O O + C

O

F F HO OH

SO3-K+

+

180oCDMSO/benzeneK2CO3

mn

CH3

CH3

O O C C

OO CH3

CH3

O O C

O

O

SO3-K+

C

O

O

hydrophobic oligomer

monomers used to produce hydrophilic oligomers in-situ

Figure 1.4.10. Synthesis of sulfonated poly(arylene ether ketone) multiblock copolymers

54

To improve the copolymers’ solubility in the film-casting solvent, they were treated with sulfonyl

chloride to convert the sulfonic salts into the SO2Cl form. The cast membranes were hydrolyzed

in a NaOH solution and then acidified with 1N HCl to regenerate the sulfonic acid groups. One

of these block copolymers displayed high conductivity and low water uptake (0.081 S/cm and

15%, respectively) at room temperature. However, the data presented in such a way that no clear

trend could be observed in the molecular weight of the hydrophobic and hydrophilic blocks,

which seemed to have been randomly selected. Further systematic studies on the influence of

block length was not performed for this study.

A similar approach was taken by Na et al. in the synthesis of multiblock copolymers containing

sulfonated poly(ether ketone) blocks (Figure 1.4.11). Here, the 3,3’-5,5’-

tetramethyl-4,4’-biphenol monomer was presumably used to suppress the crystallinity of the

poly(ether ketone) oligomers and copolymers. Again, the hydrophilic oligomers were formed in

situ, which led to poor control of the copolymers’ structure and composition. The resulting

copolymers had lower intrinsic viscosities than the corresponding random copolymers, and the

IEC decreased with increasing block length. Nevertheless, the proton conductivity and water

uptake values, which both increased as a function of the targeted block length, do suggest the

existence of sulfonated blocks, although no other proof was manifest. Due to their low IEC,

however, the membranes generally displayed lower proton conductivity compared to Nafion.

55

+

DMSO, K2CO3, 170oC

C

O

OF O C

O

F

H3C

H3C

CH3

CH3n

F C

O

F

SO3Na

NaO3S

HO OH

H3C

H3C

CH3

CH3

C

O

O O

H3C

H3C

CH3

CH3

b C

O

O

H3C C 3

H3C C 3

O

SO3Na

NaO3S

H

Hn m

Figure 1.4.11. Synthesis of sulfonated poly(arylene ether ketone) multiblock copolymers.

The McGrath research group has recently made significant progress in developing multiblock

copolymer membranes that feature our so-called “BPS-100 hydrophilic blocks,” i.e.

biphenol-based 100% disulfonated poly(arylene ether sulfone)s.17, 26, 99, 137-143 These BPS-100

telechelic oligomers have been synthesized by the polymerization between

3,3’-disulfonated-4,4’-dichlorodiphenylsulfone (SDCDPS) and 4,4’-biphenol (BP). As

mentioned earlier, such direct polymerization of sulfonated monomers enables precise and

consistent control of the degree of sulfonation. Furthermore, the use of disulfonated monomers

ensures that each repeat unit contains two sulfonic acid groups. In comparison to Taeger and

Shin’s work,135, 136 where only one ionic group was introduced to each repeat unit, this increased

population of proton conducting sites in the hydrophilic block may be superior in terms of the

56

formation of proton conducting channels. Both stoichiometry imbalance (one monomer in excess)

and end-capper chemistry have been used by McGrath et al. to prepare oligomers with the

desired molecular weights and endgroups (Figure 1.4.12).

HO OH Cl S Cl

SO3Na

NaO3S

O

O+

O O S O O

O

O

SO3M

MO3S

n

biphenol SDCDPS

(a)

Cl S O O

O

O

SO3M

MO3S

S Cl

SO3Na

NaO3S

O

O n(b)

nO S O O

O

O

SO3M

MO3S

S O

SO3Na

NaO3S

O

OH2N NH2 (c)

excessbiphenol

excessSDCDPS

HO NH2

end-capping

Figure 1.4.12. Synthesis of (a) phenoxide-teminal; (b) Cl-terminal; (c) NH2-endcapped BPS-100

oligomers

The McGrath Group has also attempted to synthesize hydrophobic/hydrophilic multiblock

copolymers by copolymerizing these oligomers with various hydrophobic telechelic groups.

Wang et al.99, 138 prepared substituted poly(p-phenylene)s, i.e. poly(4’-phenyl-2,5-benzophenone)

oligomers with fluorine terminal groups and varying molecular weights, and coupled them with

phenoxide-terminated BPS-100 oligomers (Figure 1.4.12, a) in an attempt to synthesize

multiblock copolymers (Figure 1.4.13). Using a procedure similar to that reported by Ghassemi

et al., substituted poly(p-phenylene) oligomers were obtained by the Ni(0) catalytic coupling of

2,5-dichloro-4’-phenylbenzophenone. 4-Chloro-4’-fluorobenzophenone was monofunctional in

57

this reaction and therefore was used as the end-capper to control the molecular weight and

endgroups. However, block copolymerization between the hydrophobic and hydrophilic

oligomers could not be achieved in a consistently reproducible way. The copolymer membranes

were often brittle and had poor mechanical strength. We attributed this problem to our inability

to control the stoichiometry, which might have originated from the difficulty in obtaining

quantitatively end-capped poly(4’-phenyl-2,5-benzophenone) oligomers. Another possible cause

might be that although the copolymers were only partially composed of poly(p-phenylene)

segments, their rigid, rod-like components dominated the copolymer structure, resulting in the

poor flexibility of the membranes.

C

O

F C

O

Fm

O

O O S O O

O

O

SO3M

MO3S

n

DMAc/TolueneK2CO3

165oC

+

C

OC

O

O O S O O

O

O

SO3M

MO3S

m

O

n

Figure 1.4.13. Synthesis of PPP-BPSH100 multiblock copolymers

In a study by Lee et al.,142-144 amine-terminated BPS-100 telechelics (Figure 1.4.12,c) were

copolymerized with polyimide oligomers to make multiblock copolymers. As shown in Figure

1.4.14, the NDA-terminated polyimide oligomers were prepared by the polymerization of

58

1,4,5,8-naphthalene tetracarboxylic dianhydride (NDA) with

bis[4-(3-aminophenoxy)phenyl]sulfone] (DADPS) in m-cresol. Because these hydrophobic

oligomers were insoluble in NMP, the block copolymerizations were also carried out in m-cresol.

However, only low conversion was observed, resulting in copolymer membranes with low

proton conductivity, high water uptake and poor mechanical strength. Based on these results, we

concluded that m-cresol was not a good solvent for the BPS-100 oligomers. Indeed,

copolymerizations using an m-cresol/NMP solvent mix yielded multiblock copolymers with

much higher molecular weight. These copolymer membranes displayed high proton conductivity

and atomic force micrographs showed an orientated nanophase separated morphology.142, 143, 145

180oC

O O

O O

O O

O S O

O

O

H2N NH2+

m-cresol

nO S O O

O

O

SO3M

MO3S

S O

SO3Na

NaO3S

O

OH2N NH2

O S O

O

O

N

O O

O O

NO

O O

O O

N O S O

O

O

N O

O O

O Om

180oCm-cresol +

Polyimide b BPS-100 multiblock copolymer

NDADADPS

Polyimide oligomer

BPS-100 oligomer

Figure 1.4.14. Synthesis of polyimide-BPS100 multiblock copolymers

Multiblock copolymers composed of perfluoro poly(arylene ether) blocks and

fluorine-terminated hydrophobic blocks were synthesized by Ghassemi et al.146 The hydrophobic

perfluoro oligomers were synthesized by the polymerization of decafluorobiphenyl and 6F-BPA

(Figure 1.4.15). Due to the strong electron-withdrawing effect imposed by the fluorine

59

substituents, the C-F bonds at the 4,4’- positions were very electron-deficient and highly

activated toward nucleophilic aromatic substitution. This allowed the copolymerization to be

carried out at much lower temperatures compared to those used for the less activated dihalide

monomers such as DCDPS (~100°C vs. ~180°C). The block copolymerization of perfluoro

oligomers and hydrophilic oligomers was also conducted at a similar temperature range.

However, our initial attempts to synthesize multiblock copolymers yielded high-viscosity,

gel-like mixtures, and the water uptake values of these membranes were unacceptably high. This

problem might have been due to the fact that perfluoro oligomers have reduced solubility in

DMSO, making it a poor solvent for block copolymers as well. Also, as mentioned earlier, the

fluorine groups at the ortho- positions may have participated in the copolymerization, leading to

branched or even crosslinked structures.

F

F F

F F

F

F

F

FF HO OH

CF3

CF3

DMAc / benzeneK2CO3

120oC

F

F F

F F

F

F

F

F O O

CF3

CF3

F

F F

F F

F

F

F

Fm

+

O O S O O

O

O

SO3M

MO3S

n

n

O S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O O

CF3

CF3

F

F F

F F

F

F

Fm

DMSO / benzeneK2CO3

110oC

Figure 1.4.15. Synthesis of Perfluoroarylene ether-BPS100 multiblock copolymers

60

Using slightly modified procedures, in which the block copolymerizations were conducted in

NMP instead of DMSO and at lower reaction temperatures (95~110°C instead of 120°C), we

successfully synthesized multiblock copolymer membranes with block lengths up to 5 kg/mol.137

While the proton conductivity of the membranes was comparable to that of Nafion®, the water

uptake values were generally still higher than 100%.

Based on Ghassemi’s work, Yu et al.26, 139 reported the synthesis and characterization of a series

of BisAF-BPS100 copolymers for use as PEMs, as shown in Figure 1.4.16. The only difference

in the structures is that Bisphenol-A was used in the place of 6F-BPA. With the hydrophobic

block still heavily fluorinated, the interaction parameter, Х, between the hydrophobic and

hydrophilic blocks was still high enough to promote nanophase separation.

n

O S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O O

CH3

CH3

F

F F

F F

F

F

Fm

Figure 1.4.16. BisAF-BPS100 multiblock copolymers

A number of properties for the copolymers with block lengths ranging from 3 to 8 kg/mol are

listed in Table 1.1. In general, the copolymers displayed high proton conductivity in liquid water.

Moreover, they showed much lower water uptake than the fluorinated multiblock copolymers

synthesized by Ghassemi et al., even though they possessed lower fluorine contents per repeat

unit. Conductivity measurements were also carried out under partially hydrated conditions and

the results will be discussed later.

61

Table 1.1. Properties of BisAF-BPSH100 multiblock copolymers

Polymer Block Lengthsa

(g/mol)

Molar

Feed

Ratio

IECc

Water

Uptake

(mass %)

Proton

Conductivityd

(mS/cm)

BisAF-BPSH-1 3000:3000 1 : 0.93 1.6 71 130

BisAF-BPSH-2 5000:5000 1 : 0.82 1.4 58 104

BisAF-BPSH-3 8000:8000 1 : 0.70 1.1 42 90

a: Block lengths are expressed in the form hydrophobic:hydrophilic

b: The hydrophobic:hydrophilic molar charge ratio in the reaction

c: Measured from 1H NMR

d: Measured at 30oC in liquid water

Due to the high cost of the perfluorinated monomer, decafluorobiphenyl, it would be practical to

reduce its use in multiblock copolymer synthesis. In recent studies, Lee et al.147 used

decafluorobiphenyl as an end-capping agent to produce fluoro-terminal hydrophobic

poly(arylene ether sulfone) oligomers, as opposed to polymerizing it with a bisphenol monomer

(Figure 1.4.17). A three-fold excess of decafluorobiphenyl was used to ensure the quantitative

end-capping of the oligomers. These end-capped oligomers, bearing the same endgroups as the

highly fluorinated oligomers, were equally reactive and readily polymerized with BPS-100

oligomers to obtain the largely hydrocarbon-based, high molecular weight multiblock

copolymers.

62

MO O S O OM

O

On

F

F F

F F

F

F

F

FF (excess)

O O S O O

O

O

F

F F

F F

F

F

F

F

F

F F

F F

F

F

F

Fn

+

Figure 1.4.17. Synthesis of fluorine-terminal hydrophobic oligomers

It is worth mentioning that at high temperatures and in the presence of salts such as K2CO3,

ether-ether interchange reactions can occur and, in the synthesis of multiblock copolymers, may

randomize the block sequences. Nevertheless, in ways that aren’t entirely clear to us, such side

reactions may have been minimized or avoided, as documented in studies by Wang, Lee and

Ghassemi. 137, 138, 142 We surmise that either low polymerization temperatures were used, or the

polymerizations took place via a mechanism that ruled out the possibility of an ether-ether

interchange. Conversely, if both the hydrophobic and hydrophilic oligomers had low-reactivity

chain-ends, then coupling reactions via nucleophilic aromatic substitution to synthesize block

copolymers may have been difficult (Figure 1.4.18).

63

MO O S O OM

O

O

SO3M

MO3S

Cl S O O

O

OS C

O

O

l

+

BPS-00 BPS-100

x

Sulfonated-unsulfonatedpoly(arylene ether sulfone) multiblock copolymer

m

n

randomization?

Figure 1.4.18. Questionable synthesis of poly(arylene ether sulfone) multiblock copolymers

Poly(arylene ether ketone)s or poly(arylene ether phosphine oxide)s, however, are thought to be

less susceptible to ether-ether interchanges. Li et al. reported some preliminary studies of

copolymers containing poly(arylene ether ketone) hydrophobic blocks (Figure 1.4.19).141 A

multiblock copolymer with 4K:4K block lengths was obtained, and its block sequences were

evidenced by 13C NMR. However, higher block length copolymers were not reported,

presumably due to the difficulties in achieving high conversion in the coupling reactions between

the less reactive hydrophilic and hydrophobic oligomers.

64

MO O S O OM

O

O

SO3M

MO3S

+

m

C

O

O C

CF3

CF3

O S O O

O

O

SO3M

MO3S

NMP, 180-190oC

m nx

C

O

O C

CF3

CF3

OF C

O

Fn

Figure 1.4.19. Synthesis of sulfonated PAES-PAEK multiblock copolymers

1.4.3. Comparisons between random and block copolymer PEMs

As mentioned earlier, in contrast to random copolymer ionomers—where ionic groups are

randomly distributed along the polymer chain—block ionomers contain highly ionic sequences

which may aggregate to form ion-rich channels, thereby facilitating the transport of protons. This

is particularly important when the level of hydration is low. For random copolymers, at a certain

temperature the diffusion coefficient of protons (D) should only be related to the level of

hydration. For block ionomers, however, the sequence length may also play a role. Although

only a few studies have explored this field, the advantages of using block copolymer membranes

as PEMs have been demonstrated by comparing the diffusion coefficient and/or proton

conductivity of block ionomer membranes to those of random systems under partially hydrated

conditions.

65

Serpico et al.119 studied proton conductivity under partially hydrated conditions of several types

of proton exchange membranes, including Nafion 117, a Dais Analytic S-SEBS block copolymer,

and a sulfonated styrene-ethylene “pseudo-random” copolymer (S-SE), also developed by Dais

Analytic (Figure.1.4.20, a). On the one hand, ethylene-styrene copolymer membranes were

synthesized by metallocene-catalyzed polymerizations in such a way that no two styrene repeat

units were adjacent to each other. The sulfonated forms, therefore, had every sulfonated styrene

unit separated from each other by at least one ethylene repeat unit. On the other hand, in the

S-SEBS copolymers the styrene units exist within the blocks. Therefore, many of the sulfonated

styrene units should be adjacent to each other. Figure 1.4.20 (b) shows the plots of proton

conductivity vs. water content (humidity) for the S-SEBS, S-SE, and Nafion 117 membranes. All

three PEMs displayed similar proton conductivity at high levels of hydration. With the decrease

in water content, however, the S-SE membrane showed a much more rapid decrease compared to

the Nafion 117. This suggests that the proton conducting channels in the S-SE membranes were

poorly developed (in contrast to Nafion 117) and that protons could transport readily through the

membrane only at high hydration levels. The dependence of proton conductivity on humidity as

observed for the S-SEBS membranes was stronger than that of Nafion 117, but lower than

observed for the S-SE membranes. Therefore, the ion-rich channels may be more established in

block copolymers than in the “pseudo-random” copolymers due to the difference in the sequence

length of the sulfonated units.

66

(a) SO3H S 3HO

x x

(b)

Figure 1.4.20. (a) structure of S-SE “pseudo-random” copolymers. (b) proton conductivity vs. water

content for S-SEBS, S-SE and Nafion PEMs

Roy et al.26 investigated the transport properties of highly fluorinated block copolymers

synthesized by Yu and coworkers. Because no highly fluorinated random copolymers were

available for comparison, two series of poly(ether sulfone) and poly(ether ketone) random

copolymer membranes were studied. Their structures are shown in Figure 1.4.21.

1-xxO O S

O

O

SO3H

HO3S

O O S

O

O

1-xxCF3

CF3

O O C

SO3H

HO3S

OCF3

CF3

O Y

O

(a)

(b)

Figure 1.4.21. Structures of (a) poly(ether sulfone) and (b) poly(ether ketone) random copolymers

67

The plots of proton conductivity vs. relative humidity (RH) for the random and block

copolymers, as well as Nafion 117, are displayed in Figures 1.4.22 and 1.4.23. As shown in

Figure 1.4.22, the conductivity of both random copolymer membranes is comparable to that of

Nafion 117 at high RH. As expected, with a decrease in RH, both decrease more rapidly as

compared to Nafion. The proton conductivity of the multiblock copolymers (Figure 1.4.23),

however, increased with increasing sequence length. Moreover, the conductivity of the

copolymer with the highest block length (8K:8K) was analogous to that of Nafion 117 at all RH

levels tested—in spite of the fact that the 8K:8K copolymer had the lowest IEC and water uptake

values among all the multiblock copolymers. Similar trends were observed for the multiblock

copolymers containing polyimide hydrophobic blocks, as well as the ones containing

fluoro-end-capped hydrophobic blocks, which were synthesized by Lee and coworkers. These

copolymers, however, displayed lower conductivity at similar IEC and/or water uptake values.

The high performance of the BisAF-BPSH copolymers may be attributed to their high content of

highly hydrophobic fluorinated units, which may have resulted in more distinct nanophase

separation. Although these studies were performed using different copolymers under a variety of

experimental conditions, these results suggest that among all the block copolymers we tested, the

ionomers have the most potential for use as alternative PEMs

68

1.0E-01

1.0E+00

1.0E+01

1.0E+02

1.0E+03

20 30 40 50 60 70 80 90 100Relative Humidity (%)

Prot

on c

ondu

ctiv

ity (m

S/cm

)

1 .Nafion117

2. HQSH 30

3. PB-diketone 50

1

2

3

Figure 1.4.22. Proton conductivity vs. RH plots for Nafion 117, poly(ether sulfone) random copolymers

(HQSH 30), and poly(ether ketone) random copolymers (PB-diketone 50)

Furthermore, the polystyrene blocks in the S-SEBS and other styrene-based block systems were

only partially sulfonated, and the sulfonated poly(arylene ether) block copolymers prepared by

Taeger and Shin135, 136 had only one ionic group in each repeat unit of the hydrophilic blocks. The

BPSH-100-based multiblock copolymers, therefore, may be superior in that their hydrophilic

blocks are fully sulfonated and possess two sulfonic acid groups in each repeat unit. As a result,

ion-rich channels may be formed more easily due to highly concentrated ionic groups, thereby

enhancing proton conductivity, as demonstrated in Figure 1.4.23.

69

1.0E+00

1.0E+01

1.0E+02

1.0E+03

20 40 60 80 100Relative Humidity (%)

Prot

on c

ondu

ctiv

ity (m

S/cm

)

1.BisAF-BPSH(8:8)K

2.Nafion117

3.BisAF-BPSH(5:5)K

4.BisAF-BPSH(3:3)K

1

3 2

4

Figure 1.4.23. Proton conductivity vs. RH plots for Nafion 117 and BisAF-BPSH multiblock

copolymers

F

F F

F F

F

F

F

F O S O

F

F F

F F

F

F

F

FO

O

m

n

O S O OMO

O

SO3M

MO3S

MO

+

m

O S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O S O

F

F F

F F

F

F

F

O

O n

x

NMP

K2CO3

100 ~ 110oC

4~5 days

Figure 1.4.24. Synthesis of BisSF-BPSH multiblock copolymers

70

Another variation of the BisAF-BPSH multiblock copolymers are the BisSF-BPSH copolymers,

synthesized by Yu et al. (Figure 1.4.24).140 A similar trend was observed for their proton

conductivity vs. RH performance; i.e., conductivity increased as a function of block length, as

illustrated in Figure 1.4.25.

Figure 1.4.25. Proton conductivity vs.RH plots for BisSF-BPSH multiblock copolymers, Nafion 112 and

BPSH-35 random copolymers

The high performance of multiblock copolymer membranes are believed to have morphological

origins. Figure 1.4.26 shows tapping mode AFM images of partially disulfonated BPSH-xx

random copolymers, with the degree of disulfonation ranging from 30 to 45 mol%. As shown,

the increase in the degree of sulfonation is accompanied by an increase in the size of the

hydrophilic (darker) domains. At high degrees of sulfonation, the hydrophilic regions begin to

display some connectivity. In general, however, all four images show a dispersed morphology,

which is typical of random copolymers.

71

Figure 1.4.26. Tapping mode AFM images of BPSH-xx random copolymer membranes: (a).BPSH-30;

(a) (b)

(c) (d)

500 nm

(a) (b)

(c) (d)

(a) (b)

(c) (d)

500 nm

(b).BPSH-35; (c).BPSH-40; (d).BPSH-45.

(a) (b) (c)

400 nm

(a) (b) (c)

400 nm

Figure 1.4.27. Tapping mode AFM phase images of BPSH-PI multiblock copolymer membranes with

different block lengths: (a).5K:5K; (b).10K:10K; (c).15K:15K.

72

Block copolymers have been reported to show a lamellar or cylindrical-type morphology under

optimized volume fraction and interaction parameters. Most studies, however, have only

investigated copolymers with aliphatic backbones synthesized by living polymerization. Recently,

however, the work by Lee et al. demonstrated that the same morphological features can be

observed in block copolymers with aromatic backbones synthesized via step growth

copolymerization. Figure 1.4.25 shows AFM phase images of BPSH-PI multiblock copolymer

membranes.142, 143 Here, in contrast to the BPSH random copolymers, all images show nanoscale

phase separation. More importantly, as block lengths increased from 5 to 15 Kg/mol, the

hydrophilic regions gradually transformed from dispersed domains to co-continuous channels.

This morphological trend is shown in the performance curves of these membranes under partially

hydrated conditions, wherein proton conductivity increased with increases in block length

(Figure 1.4.28).142, 143

Figure 1.4.28. Proton conductivity vs.RH plots for BPSH-PI multiblock copolymers, Nafion 112 and

BPSH-35 random copolymers

73

.

In summary, although Nafion® is still referred to as the state-of-the-art proton exchange

membrane, new synthetic techniques and a growing understanding of complex structure-property

relationships have resulted in the development of alternative hydrocarbon-based membranes,

which feature optimized combinations of proton conductivity, water sorption, and long-term

durability. Partially sulfonated poly(arylene ether) random copolymers, especially those obtained

through the direct polymerization of sulfonated comonomers, have shown comparable proton

conductivity to Nafion® under fully hydrated conditions, even though their performance tends to

degrade faster than Nafion® as the water content decreases. Although high conductivity at low

humidity, as seen in the block copolymer membranes, is particularly encouraging, the challenge

remains in how to control water uptake and swelling-deswelling without sacrificing performance.

74

Chapter 2. Synthesis and characterization of BisAF-BPSH

hydrophobic-hydrophilic multiblock copolymers

We have synthesized multiblock (segmented) copolymers via coupling reactions of hydrophobic

and hydrophilic telechelic oligomers. Such materials contain fully disulfonated poly(arylene

ether sulfone) (BPSH-100) blocks and highly fluorinated poly(arylene ether) blocks. The

generalized structure of the copolymers studied is shown in Figure 2.1, where X can be an

isopropylidene group, a hexafluoroisopropylidene group, or a sulfone group. Accordingly, the

multiblock copolymers are termed BisAF-BPSH, 6FBisAF-BPSH, and BisSF-BPSH,

respectively. In the following sections, the synthesis and characterization of these materials will

be illustrated in detail and the influence of their structures on resulting properties will be

discussed.

X = CCH3

CH3

CCF3

CF3

SO2X = X = , BisAF-BPSH100 series; , 6FBisAF-BPSH100 series; , BisSF-BPSH100 series

m

O S O O

O

O

SO3H

HO3S

O

F

F F

F F

F

F

F

O X O

F

F F

F F

F

F

Fn

x

(a) (b) (c)

Figure 2.1. Structures of fluorinated-sulfonated, hydrophobic-hydrophilic multiblock copolymers

75

2.1. Experimental

2.1.1.. Solvents

N-Methyl-2-Pyrrolidone (NMP)

N O

CH3 Source: Fisher Scientific Molecular weight: 99.13 g/mol Purification: Dried over CaH2 for 12 h and then distilled at 120 oC under distilled pressure and stored over molecular sieves.

N,N-dimethylacetamide (DMAc)

CH3 C

O

NCH3

CH3

Source: Ficher Scientific Molecular weight: 89.13 g/mol Purification: Dried over CaH2 for 12 h and then distilled at 80 oC under distilled pressure and stored over molecular sieves.

Toluene

CH3

Source: Fisher Scientific Molecular weight: 92.14 g/mol Purification: Used as received.

76

Isopropyl alcohol

CH3CHCH3

OH

Source: Fisher Scientific Molecular weight: 60.01 g/mol Purification: Used as received.

Methanol

CH3OH

Source: Fisher Scientific Molecular weight: 32.04 g/mol Purification: Used as received.

Fuming Sulfuric Acid

H2SO4 SO3

Source: Aldrich Molecular weight: 98.08 g/mol Purification: Used as received.

2.1.2. Monomers

4.4’-Biphenol

HO OH

Source: Molecular weight: 186.21 g/mol Melting point: 282-284 oC Purification: Dried under vacuum at 60 oC for 24 h.

77

4,4-Dichlorodiphenyl Sulfone (DCDPS)

S

O

O

Cl Cl

Source: Solvay Molecular weight: 287.13 g/mol Melting point: 145-147 oC Purification: For sulfonated monomer synthesis, it was used as received.

4,4’-isopropylidenediphenol (Bisphenol A)

C

CH3

CH3

HO OH

Source: Aldrich Molecular weight: 228.29 g/mol Melting point: 152-153 oC Purification: Recrystallized from a 25% (w/v) solution in toluene, and then dried under vacuum at 60 oC for 24 h.

Decafluorobiphenyl

F

F

F

F

F

F F

F

FF

Source: Aldrich Molecular weight: 334.11 g/mol Melting point: Purification: Dried under vacuum at 50 oC for 24 h before use.

78

Potassium Carbonate

K2CO3

Source: Aldrich Molecular weight: 138.21 g/mol Purification: Dried under vacuum at 60oC for 12 h before use.

2.1.3. Monomer Synthesis

3,3’-Disulfonated-4,4’-dichlorodiphenyl sulfone (SDCDPS)

S

O

O

Cl Cl

NaO3S S 3NaO

To a 100 mL round bottom flask equipped with a magnetic stirrer and a nitrogen inlet was added

30 g of DCDPS and 60 mL of fuming sulfuric acid (30% SO3). The mixture was stirred until

DCDPS dissolved and the solution was reacted at 110 oC for 6 h. The reaction was cooled to

room temperature and added to 500 mL of ice-water. Then NaCl was added to salt out the white

powder-like product, which was filtered and redissolved into 400mL of deionized water. The

solution was treated with 2N NaOH aqueous solution to a pH of 6, and then NaCl was added to

salt out the sulfonated monomer again. The product was filtered and recrystallized twice from

methanol.

Molecular weight: 491.24 g/mol

Purification: Dried under vacuum at 160 oC for 3 days before use.

79

2.1.4: Polymer synthesis

Highly Fluorinated Hydrophobic Oligomers (BisAF)

F

F F

F F

F

F

F

F O C O

F

F F

F F

F

F

F

FCH3

CH3n

The synthesis of a fluorine terminated BisAF oligomer with 3 Kg/mol molecular weight, for

example, was carried out as follows: Bisphenol-A (1.174 g, 5.142 mmol) was added to a

three-necked round bottom flask equipped with a mechanical stirrer, condenser, nitrogen inlet,

and a Dean-Stark trap. DMAc (10 mL) was added and the mixture was stirred until dissolved.

K2CO3 (1.183 g, 7.20 mmol) and toluene (5 mL) were added and the system was dehydrated at

150 oC. Then the reaction bath was cooled to 50 oC and decafluorobiphenyl (2.046 g, 6.124

mmol) was added. The polymerization was allowed to proceed at 110 oC for 5 h. The reaction

mixture was isolated by precipitation into a H2O/methanol (50/50 v/v) mixture. The white

polymer was washed with H2O and methanol, and then dried under vacuum at 80 oC before use.

Fully Disulfonated BPS-100 oligomers

MO O S O OM

O

O

SO3M

MO3Sn

The procedures for the synthesis of 3 Kg/mol phenoxide terminated BPS-100 oligomers are as

follows: Biphenol (0.412 g, 2.213 mmol), SDCDPS (0.912 g, 1.856 mmol), and NMP (10 mL)

80

were charged to a three-necked round bottom flask. The mixture was stirred until dissolved, then

K2CO3 (0.430 g, 3.12 mmol) and toluene (5 mL) were added. The system was dehydrated at 150

oC for 4 h and then the temperature was raised to 190 oC with controlled removal of toluene. The

polymerization was allowed to proceed at 190 oC for at least 30 h. 1H NMR was run to study the

molecular weight of the oligomers. Later, the BisAF oligomer was added to the same flask to

originate the formation of multiblock copolymers. Before that, a small BPS-100 sample was

taken from the reaction mixture and precipitated from isopropal alcohol for NMR use.

BisAF-BPSH Hydrophobic-Hydrophilic Multiblock Copolymers

m

O S O O

SO3M

MO3S

O

F

F F

F F

F

F

F

O C O

F

F F

F F

F

F

F

O

O

CH3

CH3n

x

To the flask of BPS-100 oligomers under synthesis was added the corresponding BisAF

oligomers to carry out the synthesis of a multiblock copolymer. The reaction was cooled to 80oC,

and the BisAF oligomer (1.050 g, 0.350 mmol) was dissolved in NMP and slowly added to the

flask. The coupling reactions were run at 95 oC , which was gradually increased to 110 oC, for up

to 6 days, until there was no significant change in viscosity if allowed to proceed for another 12

h. The reaction mixture was then precipitated into isopropal alcohol to obtain a brownish fibrous

polymer. The product was washed twice in deionized water at 60 oC, then washed twice in

acetone. It was then dried and redissolved in NMP to afford a 5% (w/v) solution. The solution

was cast onto a glass substrate and dried with a IR lamp at ~35 oC to obtain a polymer membrane.

81

The membrane was then dried under vacuum at 110 oC for 24 h. The salt form membrane

(SO3-M+) was converted to its acid form (SO3H) by boiling in deionized water for 2 h, and was

then boiled in deionized water for 2 h. It was then stored in water until it was used for

measurements.

2.1.5. NMR Spectroscopy, Gel Permeation Chromatography, Intrinsic Viscosity and Atomic

Force Microscopy Characterization

1H and 19F NMR analysis were conducted on a Varian Unity 400 spectrometer. The spectra of

BPS-100 oligomers and BisAF-BPSH multiblock copolymers were obtained from a 10%

solution (w/v) in a DMSO.d6 solution at room temperature. The spectra of BisAF hydrophobic

oligomers were obtained from a solution in CDCl3. Gel permeation chromatography (GPC)

experiments for BisAF-BPSH copolymers were performed on a liquid chromatograph equipped

with a Waters 1515 isocratic HPLC pump, Waters Autosampler, Waters 2414 refractive index

detector, and Viscotek 270 RALLS/ viscometric dual detector. NMP (containing 0.05M LiBr)

was used as the mobile phase. For BisAF oligomers THF was used as the solvent. The column

temperature was maintained at 60 oC because of the viscous nature of NMP. Both the mobile

phase solvent and sample solution were filtered before introduction to the GPC system.

Molecular weights were determined from universal calibration plot using polystyrene as a

standard. Intrinsic viscosities were determined in 0.05M LiBr NMP at 25 oC using a Cannon

Ubbelholde viscometer. Atomic force microscopy characterization (AFM) images were taken

82

using Digital Instruments Dimension 3000 with a microfabricated cantilever. The force constant

was 40 N/m.

2.1.6. Characterization of Fuel Cell Related Properties

Proton Conductivity

Proton conductivity at 30 oC at full hydration (in liquid water) was determined in a window cell

geometry using a Solartron 1252 + 1287 Impedance/Gain-Phase Analyzer over the frequency

range of 10 Hz to 1 MHz following procedures reported in the literature. In determining proton

conductivity in liquid water, membranes were equilibrated at 30 oC in DI water for 24 h prior to

testing. The temperature range chosen for calculation of activation energy for proton transport

was from 30 to 80 oC. For determining proton conductivity under partially hydrated conditions,

membranes were equilibrated in a humidity-temperature oven (ESPEC, SH-240) at the specified

RH and 80 oC for 24 h before each measurement.

Water Uptake

The water uptake of all membranes was determined gravimetrically. First, the membranes were

soaked in water at 30 °C for 2 days after acidification. Wet membranes were removed from the

liquid water, blotted dry to remove surface droplets, and quickly weighed. The membranes were

then dried at 120 oC under vacuum for at least 24 h and weighed again. The water uptake of the

membranes was calculated according to Equation 2.1 where massdry and masswet refer to the mass

of the dry membrane and the wet membrane, respectively.

83

100mass

massmassuptake%water

dry

drywet ×−

= ……..(2.1)

The hydration number (λ), namely, the number of water molecules absorbed per sulfonic acid,

can be calculated from the mass water uptake and the ion content of the dry copolymer as shown

in Equation 2.2, where MWH2O is the molecular weight of water (18.01 g/mol) and IEC is the ion

exchange capacity of the dry copolymer in equivalents per gram.

dry

OHdrywet

massIEC)/MWmass(mass

λ 2

×

−= ……..(2.2)

Pulsed-Field Gradient Spin Echo Nuclear Magnetic Resonance

The self-diffusion coefficient of water was measured using a Varian Inova 400 MHz (for protons)

nuclear magnetic resonance spectrometer with a 60 G/cm gradient diffusion probe. A total of 16

points were collected across the range of gradient strength and the signal-to-noise ratio was

enhanced by coadding 4 scans. The standard stimulated echo NMR pulse sequence is shown in

Figure 2.2.

84

Figure 2.2. Pulse sequence schematic for PGSE NMR experiments

Measurements were conducted by observing the echo signal intensity (A) as a function of

gradient strength. The diffusion coefficient (D) was determined by fitting the data to Equation

2.3, where A is the NMR signal intensity (A) as a function of gradient strength, γ is the

gyromagnetic ratio (26,752 rad G-1 s-1 for protons), δ is length of the gradient pulse, ∆ is the time

between gradient pulse.

)]3/(exp[)()( 222 δδγ −∆−= gg DoAA ……..(2.3)

Membrane samples of approximately 5 mm × 15 mm × 150 µm were equilibrated in liquid water

for at least 24 h. The samples were removed from the liquid water, blotted to remove droplets,

quickly inserted into the NMR tube, and immediately measured over a span of about 5 min.

Measurements were repeated by reimmersing the sample in DI water, waiting at least 30 min,

and then repeating the transfer and measurement process. Separate measurements were collected

with different times between the gradient pulses.

85

2.2. Results and Discussion

2.2.1. Synthesis and characterization

2.2.1.1. Synthesis of fluorinated Oligomers

The synthetic scheme for the synthesis of BisAF oligomers is shown in Figure 2.3.

Decafluorobiphenyl was used in excess to obtain fluorine terminated oligomers as well as to

control the molecular weight. Figure 2.4 shows the 1H NMR spectrum of a BisAF oligomer. Its

19F NMR spectrum is shown in Figure 2.5.

F

F F

F F

F

F

F

FF HO C OHCH3

CH3

DMAc / cyclohexaneK2CO3

110oC, 5h

F

F F

F F

F

F

F

F O C O

F

F F

F F

F

F

F

FCH3

CH3n

+

(excess)

Figure 2.3. Synthesis of BisAF oligomers

86

F

F F

F F

F

F

F

F O O

CH3

CH3

F

F F

F F

F

F

F

Fn

a

a

b b ccF

F F

F F

F

F

F

F O O

CH3

CH3

F

F F

F F

F

F

F

Fn

a

a

b b cc

Figure 2.4. 1H NMR spectrum of a BisAF oligomer

F

F F

F F

F

F

F

F O O

CH3

CH3

F

F F

F F

F

F

F

Fn

a b b a a b c d

eF

F F

F F

F

F

F

F O O

CH3

CH3

F

F F

F F

F

F

F

Fn

a b b a a b c d

e

Figure 2.5. 19F NMR spectra of a BisAF oligomer

87

The peaks due to the main chain, as well as to the endgroup fluorine moieties, were labeled and

their integrals were used to calculate the oligomers’ experimental Mn. GPC was also used to

verify the molecular weights of the oligomers having target Mn values of 3, 4, 5 and 8 Kg/mol.

Figure 2.6 displays the plot of logη ~ log (Mn) for these oligomers. The experimental and

theoretical Mn values for the oligomers are listed in Table 2.1.

Table 2.1. Molecular weight characterizations of BisAF oligomers

Target Mn Mn from 19F NMR Mn from GPC Intrinsic Viscosity

(dL/g)

3K 3.2K 4.4K 0.08

5K 5.5K 5.2K 0.13

8K 8.3K 8.5K 0.17

-1.2

-1.1

-1

-0.9

-0.8

-0.7

-0.6

3.4 3.5 3.6 3.7 3.8 3.9 4

log(M n)

log(IV)

Figure 2.6. logη vs. logMn plot for BisAF oligomers

88

We have found that not all bisphenol monomers could be polymerized with decafluorobiphenyl

(Figure 2.7). Polymerizations of biphenol with the perfluoro monomer were attempted in DMAc

or NMP 110 oC, but a white powder would precipitate out of the solution in 30 min. Our DSC

experiments using these materials showed a trace with a large endotherm at ~350 oC and a glass

transition at 180 oC. This indicates that the polymerization product was semicrystalline and

precipitated out at low molecular weight due to its poor solubility. The same results were

observed for polymerizations of hydroquinone or 4,4’-dihydroxybenzophenone with

decafluorobiphenyl. The possible crystallinity of the polymers may be attributed to the

presumable planar structures of all these monomers. Conversely, when bisphenol monomers

containing flexible groups in between the phenyl rings (i.e. Bisphenol-A, 6F Bisphenol-A, and

Bisphenol-S) were used, the polymerizations invariably proceeded readily and were complete

within 4~5 h at 110 oC.

89

HO OH

HO S OHO

O

HO C OHO

F

F F

F F

F

F

F

FF

Precipitated

Precipitated

Polymerized successfully

HO C OHCH3

CH3

HO C OHCF3

CF3

OHHO

Precipitated

Polymerized successfully

Polymerized successfully

Figure 2.7. Reaction of decafluorobiphenyl with various bisphenol monomers

2.2.1.2. Synthesis of fully disulfonated hydrophilic oligomers

As in the synthesis of partially disulfonated BPSH-xx random copolymers, the direct

polymerization of the sulfonated monomer, SDCDPS, was used in the synthesis of the fully

disulfonated polymers. Biphenol was polymerized with SDCDPS as shown in Figure 2.8. Excess

biphenol was used to control the molecular weight in order to produce phenoxide-terminated

oligomers.

90

HO OH Cl S Cl

SO3Na

NaO3S

O

O

NMP/TolueneK2CO3

150oC, 4h190oC, 30h

MO O S O OM

O

O

SO3M

MO3Sn

+

(excess)

Figure 2.8. Synthesis of fully sulfonated BPS-100 oligomers

The 1H NMR spectrum of an oligomer is shown in Figure 2.9. The integral of the peaks due to

the endgroups and the main chain were used to estimate the experimental Mn. In the early

attempts, however, the oligomers thus obtained usually showed lower measured molecular

weights than the values predicted based on the Carothers equation:

r-1r1Dp

+=

Where Dp is the degree of polymerization and r is the molar feed ratio of the monomers. It is

now generally accepted that this discrepancy is due to residual NaCl left in the synthesis of

SDCDPS, and, to a much less extent, to moisture absorbed by the monomer during weighing, etc.

The SDCDPS/biphenol stoichiometry (r) is therefore overestimated, as is the product's molecular

weight.

91

MO O S O OM

O

O

SO3M

MO3Sn

a ab b c d

e

e

d c

fghi

Figure 2.9. 1H NMR spectrum of a BPS-100 oligomer

In this work, the problem was solved by using a convenient engineering approach. We

calculated the "effective" molar ratio of SDCDPS to biphenol from the Mn, then adjusted the feed

ratio accordingly, i.e. added more SDCDPS than calculated for the next experiment. Normally,

the use of an extra 1~1.5 mol % of SDCDPS was found to be sufficient to compensate for the

monomer's impurity. Still, it was of fundamental importance to measure the purity of the

monomer. Studies of NaCl content using UV-Vis spectroscopy are underway in our laboratory.

The purity of the solvent also seemed to affect the polymerization. As the NMP aged, the color of

the reaction solution became darker. Moreover, the difference between experimental and target

molecular weights tended to become greater, and even more SDCDPS was needed to maintain

the desired stoichiometry. This mechanism is yet unknown, but we postulated that, although

sealed, the NMP have have absorbed oxygen slowly from the air which caused its oxidative

decomposition under the polymerization conditions. There is no evidence, however, of the

chemical integrity of the oligomer being affect by the decomposition of the solvent.

92

2.2.1.3. Synthesis of Multiblock Copolymers

The synthesis of BisAF-BPSH multiblock copolymers is shown in Figure 2.10. The reaction was

carried out in the same flask, directly after the BPS-100 oligomer synthesis was complete. This

was done in order to simplify experimental procedures, since additional steps would have been

needed to remove the salt from the isolated BPS-100 oligomer for weighing purpose. Also, the

complexities associated with the uncertainties in stoichiometry caused by the isolation of both

oligomers was therefore reduced.

F

F F

F F

F

F

F

F O O

CH3

CH3

F

F F

F F

F

F

F

F

m

n

O S O OM

O

O

SO3M

MO3S

MO

m

+

O S O

O

O

SO3M

MO3S

O

CH3

CH3

F

F F

F F

F

F

F

On

95~110oC

NMP K2CO3

Figure 2.10. Synthesis of BisAF-BPSH multiblock copolymers

The viscosity of the reaction solution increased slowly during the polymerization. In general,

the reactions were deemed to be complete and generally stopped if there were no significant

increases in viscosity after they were allowed to run for another ~12 h. It should be noted that

reactions tended to proceed more slowly for higher molecular weight oligomers, ranging from 24

93

h for BisAF-BPSH (3K:3K) to 6 days for BisAF-BPSH (8K:8K). These longer reaction times

might have resulted from the differences in solubility of the hydrophobic and hydrophilic

oligomers in NMP. Specifically, we speculated that the solubility differences might have led to

some degree of phase separation, causing the chain-ends to migrate to the

hydrophobic-hydrophilic interfaces in order to meet and react with each other. Nevertheless, the

isolated products were completely soluble in DMAc, NMP or DMSO without any gelation

observed. Thus, the reactions were thought to have resulted in linear copolymers, and

contributions from side reactions associated with pendent fluorine moieties along the chains were

disregarded.

2.2.1.4. Fundamental characterizations of BisAF-BPSH multiblock copolymers

An 1H NMR spectrum for a BisAF-BPSH (8K:8K) copolymer is shown in Figure 2.11. The

integral of the peaks due to hydrophilic and hydrophobic moieties were utilized to calculate the

experimental IEC values of the materials, which were found to agree well with theoretical

values.

94

e

d

bf

a,g

c

12000 12500 13000 13500 14000 14500 15000 pts

mO S O

O

O

SO3M

MO3S

OCH3

CH3F

F F

F F

F

F

F

On

a ab b c d e f fgg

x

e

d

bf

a,g

c

12000 12500 13000 13500 14000 14500 15000 pts

mO S O

O

O

SO3M

MO3S

OCH3

CH3F

F F

F F

F

F

F

On

a ab b c d e f fgg

x

Figure 2.11. 1H NMR spectrum of a BisAF-BPSH multiblock copolymer

An 19F NMR spectrum of the copolymer is shown in Figure 2.12. Only the peaks due to the main

chain can be observed in both 1H and 19F spectra, thus indicating the high conversion of the

reaction.

95

i

h

-140 -145 -150 -155 PPM

mO S O

O

O

SO3M

MO3S

OCH3

CH3F

F F

F F

F

F

F

On

x

h i i h

i

h

-140 -145 -150 -155 PPM

mO S O

O

O

SO3M

MO3S

OCH3

CH3F

F F

F F

F

F

F

On

x

h i i h

Figure 2.12. 19F NMR spectrum of a BisAF-BPSH multiblock copolymer

The synthesis of the BisAF-BPSH (4K:4K) copolymer was monitored using 1H NMR

spectroscopy. Figure 2.13 shows the spectra of samples taken at different stages of the coupling

reaction (b ~ c), as well as that of the 4K BPS-100 oligomer prior to the reaction (a). It can be

seen that the small peaks corresponding to the oligomer’s endgroups are the largest in the spectra

of the BPS-100 oligomer, then gradually become smaller as the reaction proceeds, and eventually

disappear. This largely confirms the pathway of the polymerization, which occurs through the

reactions between phenoxide and fluorine endgroups.

96

8.2 8.0 7.8 7.6 7.4 7.2 7.0 6.8 PPM

(a)

(b)

(c)

(d)

8.2 8.0 7.8 7.6 7.4 7.2 7.0 6.8 PPM

(a)

(b)

(c)

(d)

Figure 2.13. Monitoring of multiblock copolymer synthesis using 1H NMR spectra: (a) BPS-100

oligomer prior to the reaction; (b) 12 h; (c) 20 h; (d) 36 h.

1H NMR spectra of the BisAF-BPSH multiblock copolymers with block lengths of 3K, 4K and

8K are overlaid in Figure 2.14. Here, the small peaks at around 7.3 ppm, though partially

overlapping with the other peaks, were thought to be due to one of the four types of protons at

the linkages between blocks. As shown in this figure, with the intensity of the other peaks more

or less normalized, the peak for the 4K:4K copolymer is slightly smaller than that of the 3K:3K

copolymer. In comparing the 4K:4K to the 8K:8K copolymers, however, the intensity of the peak

significantly decreased. It would seem, therefore, that for the multiblock copolymers, the higher

the block length, the smaller the number of linkages between hydrophobic and hydrophilic

blocks. This is consistent with observations from NMR spectra that verify the existence of block

sequences in these materials. 97

7.8 7.6 7.4 7.2 7.0 PPM

m

O S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O C O

F

F F

F F

F

F

F

CH3

CHe n

x

Possible causes for the peak

4K:4K

3K:3K

8K:8K7.8 7.6 7.4 7.2 7.0 PPM

m

O S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O C O

F

F F

F F

F

F

F

CH3

CHe n

x

Possible causes for the peak

4K:4K

3K:3K

8K:8K

Figure 2.14. 1H NMR spectra of BisAF-BPSH multiblock copolymers showing the linkages between

blocks

Characterization data for a series of the multiblock copolymers are listed in Table 2.2. It is worth

noting that the copolymers have different IEC values due to differences in the molar charge

ratios for the hydrophilic:hydrophobic oligomers. Although the feed ratios are generally lower

than 1:1, all the polymerizations still yielded materials with high intrinsic viscosities that are

capable of being cast into tough membranes. Thus, unlike polymerizations of smaller monomers,

in coupling reactions of telechelic oligomers, a perfect stoichiometry is not necessary. Therefore,

the molar feed ratio (which does not have to be 1:1) can be adjusted to control a copolymer’s

IEC. This type of approach was more systematically used in the synthesis of BisSF-BPSH

multiblock copolymers, which will be discussed in more detail in a subsequent chapter.

98

Table 2.2. Some characterizations of BisAF-BPSH multiblock copolymers

Copolymer Block Lengthsa (g/mol)

Molar Feed Ratiob IECc Intrinsic Viscosity

(dL/g)d

BisAF-BPSH-1 3000:3000 1 : 0.93 1.6 0.60

BisAF-BPSH-2 4000:4000 1 : 0.87 1.5 0.64

BisAF-BPSH-3 5000:5000 1 : 0.82 1.4 0.54

BisAF-BPSH-4 8000:8000 1 : 0.70 1.1 0.53

a. Block lengths are expressed in the form hydrophobic:hydrophilic b. Molar ratios are expressed in the form hydrophobic:hydrophilic c. Measured from 1H NMR d. Measured at 25 oC in 0.05M LiBr NMP solution

Tapping mode atomic force microscopy (AFM) phase images for the copolymers with 3K, 4K

and 5K molecular weights are displayed in Figure 2.15. In these images, the dark regions are

believed to be due to the hydrophilic domains (BPSH), whereas the bright regions correspond to

the hydrophobic ones (BisAF). Although we noted the absence of an oriented morphology, we

observed from the phase images that the phase separation between the hydrophilic and

hydrophobic regions was sharper as block lengths increased. Furthermore, copolymers with

longer blocks featured more connectivity among the hydrophilic domains. In other words, there

tended to be increasing numbers of ionic channels in the membranes, which are thought be

important to proton conductivity at low humidity, as block lengths increased—even though this

was accompanied by a decrease in IEC. Thus, this study demonstrated that the morphological

features of these copolymers can be controlled by synthesis. As will be discussed below, by

controlling morphology, we were also able to control how these materials performed under fuel

cell conditions.

99

100 nm 100 nm 100 nm

3K:3K 4K:4K 5K:5K

100 nm 100 nm 100 nm

3K:3K 4K:4K 5K:5K

Figure 2.15. Tapping mode AFM phase images of BisAF-BPSH multiblock copolymers with different

block lengths

2.2.2. Fuel Cell Related Characterizations of Multiblock Copolymers

2.2.2.1. Proton conductivity under fully hydrated conditions

It has been reported in the literature that the proton conductivity of sulfonated polymers is a

function of both water uptake and the ion exchange capacity. In addition to IEC and water

sorption on a mass basis, the hydration number (λ) is widely used to compare membranes with

different polymer backbone structures. In this portion of the research, the properties of our

BisAF-BPSH multiblock copolymers will be compared to those of several series of poly(arylene

ether sulfone) and poly(arylene ether ketone) random copolymers, i.e. BPSH, HQSH, PB and B

series, as shown in Figure 2.16.

100

1-xxO O S

O

O

SO3H

HO3S

O O S

O

O

1-xxCF3

CF3

O O Y

SO3H

HO3S

OCF3

CF3

O Y

(a)

(b)

O S

O

O

SO3H

HO3S

O O S

O

OO

x 1-x

(c)

Y = C

OC

O

C

Oor(B series)

HQSH series

(PB series)

BPSH series

Figure 2.16. Structures of partially disulfonated random copolymers. (a) BPSH; (b) HQSH; (c) poly(ether

ketone) B and PB series

Previous studies in our laboratory indicated that proton conductivity scales linearly with IEC for

the BPSH random copolymer membranes under fully hydrated conditions. Here, similar tends

were observed for HQSH, PB, and B series, as shown in Table 2.3. and Figure 2.17.

The increase in conductivity can be explained in terms of an increase in the concentration of

protons and a corresponding increase in water uptake. It has been reported that water uptake for

these random copolymers also increases with increases in the ion exchange capacity.

101

Table 2.3. IEC, water uptake and proton conductivity for partially disulfonated random

copolymers

Copolymer IEC

(mequiv/g)

Water Uptake

(λ)

Proton Conductivity at 30oC in

liquid water (mS/cm)

HQSH-30 1.6 29 100

HQSH-25 1.4 24 80

HQSH-20 1.2 14 70

PB-50 1.4 26 71

PB-40 1.2 12 40

PB-30 0.9 9 10

B-50 1.7 32 90

B-40 1.4 18 73

B-30 1.1 13 23

Figure 2.17. Proton conductivity at 30oC in liquid water for partially disulfonated random copolymers

plotted against IEC

102

The proton conductivity and water uptake for BisAF-BPSH multiblock copolymers are listed in

Table 2.4. Similar to the case of random copolymers, it can be observed that the proton

conductivity for the multiblock copolymers under fully hydrated conditions is generally a

function of the IEC. Although both proton conductivity and water uptake decrease with the

decrease in IEC, there does not seem to be a significant effect of block length on these properties.

Table 2.4. IEC, water uptake and liquid water proton conductivity for BisAF-BPSH multiblock

copolymers

Copolymer Block Lengths

(g/mol)

IEC

(mequiv/g)

Water

Uptake

Proton Conductivity at

30oC (mS/cm)

BisAF-BPSH-1 3K:3K 1.6 71% 130

BisAF-BPSH-3 5K:5K 1.4 58% 104

BisAF-BPSH-4 8K:8K 1.1 42% 90

Thus, under fully hydrated conditions, proton conductivity is determined mainly by the ion

exchange capacity of the membrane, regardless of the polymer architecture or the sequence

length of sulfonic acid groups.

Proton conductivity for the BisAF-BPSH copolymers was also measured at various temperatures

under fully hydrated conditions, and the results are shown in Figure 2.18. Again, within the

entire temperature range, proton conductivity increased with increasing IEC values.

103

0

0.05

0.1

0.15

0.2

0.25

0 20 40 60 80 100

Temperature (deg C)

Prot

onic

con

dutiv

ity(S

/cm

)

1Xu12 Xu23. Xu4

1

2

3

Figure 2.18. Proton conductivity under fully hydrated conductions for BisAF-BPSH copolymers as a

function of temperature

2.2.2.2. Proton conductivity under partially hydrated conditions

Up to this point, we have only discussed proton conductivity under fully hydrated conditions,

with the concentration of ionic groups identified as the primary factor influencing this important

variable. Figure 2.19 shows proton conductivity as a function of relative humidity (RH) for

HQSH-30, PB-50, and Nafion 117. It can be seen that below 50% RH, the proton conductivity

decreases significantly for the random copolymers. Thus, under partially hydrated conditions,

proton conductivity for random copolymer membranes is strongly dependent on the hydration

level. This observation is consistent with the decrease in activation energy for proton transport

with an increase in RH, as shown in Figure 2.20 for the HQSH-30 random copolymers. For

Nafion, however, the decrease in conductivity as RH decreases is not as great. We attributed this

104

to the unique chemical structure of Nafion, which consists of highly flexible side chain

hydrophilic sulfonic acid groups and a hydrophobic fluorinated backbone. These chemical

attributes are believed to promote strong nanophase separation between hydrophilic and

hydrophobic domains, as well as facilitate proton transport among the interconnected hydrophilic

domains—even at low hydration levels.

1.0E-01

1.0E+00

1.0E+01

1.0E+02

1.0E+03

20 30 40 50 60 70 80 90 100Relative Humidity (%)

Prot

on c

ondu

ctiv

ity (m

S/cm

)

1 .Nafion117

2. HQSH 30

3. PB-diketone 50

1

2

3

Figure 2.19. Proton conductivity vs. RH plots for Nafion 117, poly(ether sulfone) random copolymers

(HQSH-30), and poly(ether ketone) random copolymers (PB-50)

105

Figure 2.20. Activation energy of proton transport for HQSH-30 random copolymers as a function of

relative humidity: proton transport barrier increases as RH decreases

The proton conductivity of BisAF-BPSH multiblock copolymers and Nafion is plotted against

RH in Figure 2.21. Here, it can be seen that the conductivity of the block copolymers increased

with increasing block length. The 8K-8K sample displayed the best performance—in fact,

comparable to that of Nafion—despite the fact that it had the lowest IEC and water uptake values

in the series, This contrasts to performance under fully hydrated conditions where IEC, rather

than block length, dominated conductivity. This suggests that with an increase in block length,

the extent of nanophase separation and the connectivity between the hydrophilic domains

increase, which then decreases the barrier for proton transport.

106

1.0E+00

1.0E+01

1.0E+02

1.0E+03

20 40 60 80 100Relative Humidity (%)

Prot

on c

ondu

ctiv

ity (m

S/cm

)

1.BisAF-BPSH(8:8)K

2.Nafion117

3.BisAF-BPSH(5:5)K

4.BisAF-BPSH(3:3)K

1

3 2

4

Figure 2.21. Proton conductivity vs. RH plots for Nafion 117and BisAF-BPSH multiblock copolymers

2.2.2.3. Diffusion coefficients

The self-diffusion coefficient of water was measured under fully hydrated conditions by PGSE

NMR. The obtained values at different IECs for the HQSH random copolymers, as well as for

the BisAF-BPSH multiblock copolymers, are shown in Figure 2.22. As expected, for the random

copolymers the diffusion coefficient increased as a function of the IEC. The diffusion

coefficients for the block copolymers, however, increased with an increase in block length,

irrespective of IEC values. This clearly illustrates the importance of connectivity between the

hydrophilic domains with respect to transport properties.

107

3K:3K

5K:5K

8K:8K

3K:3K

5K:5K

8K:8K

Figure 2.22. Self-diffusion coefficient for water as a function of IEC for random and block copolymers

and Nafion

2.2.2.4. Methanol Permeability

In a direct methanol fuel cell (DMFC), methanol—rather than hydrogen—is used as the fuel. The

use of a DMFC, however, means that not only is there a tradeoff between proton conductivity

and water uptake, but also between the transport properties of proton and methanol, since both

depend to some extent on the diffusion of water molecules. One of the major disadvantages of

using Nafion in a DMFC is the material’s high methanol permeability. Conversely, BPSH and

other random copolymer membranes have generally shown lower methanol permeability

compared to Nafion.

108

Methanol permeability (at 80oC) values as a functional of IEC for Nafion 117, HQSH-xx,

B-ketone-xx and BisAF-BPSH multiblock copolymers are shown in Figure 2.23. As in the case

of proton conductivity, the methanol permeability of random copolymers increased with an

increase in IEC, which was also seen for the multiblock copolymers. However, the block

copolymers were designed in such a way that the sample with the lowest IEC had the longest

hydrophobic and hydrophilic blocks. Thus, while the low-IEC random copolymers may suffer

from low conductivity, this would not be a problem for the BisAF-BPSH (8k:8K) multiblock

copolymers. This membrane, despite its low IEC, showed high proton conductivity under

partially hydrated conditions and a high water self-diffusion coefficient, as shown earlier. This

suggests that the transport of methanol may have a different mechanism than that of water in

hydrophilic-hydrophobic block copolymers, in that both high proton conductivity and low

methanol crossover can be achieved by controlling the morphology. The connectivity among the

long hydrophilic sequences ensures a better pathway for the diffusion of water and protons,

whereas the low methanol permeability we observed may have resulted from the methanol

transport being significantly hampered by the long hydrophobic blocks.

109

Figure 2.23. Methanol permeability at 80 oC as a function of IEC for random and block copolymer

membranes and Nafion

110

Chapter 3. Synthesis and Characterization of 6FBisAF-BPSH Multiblock

Copolymers

3.1. Experimental

The materials and procedures used to synthesize 6FBisAF-BPSH multiblock copolymers are

analogous to those of BisAF-BPSH copolymers, which were discussed earlier. Therefore, only

those that have not been previously described will be discussed here.

3.1.1. Materials.

4,4’-(Hexafluoroisopropylidene) diphenol (6F-Bisphenol A, 6F-BPA)

C

CF3

CF3

HO OH

Source: Ciba

Molecular weight: 336.33 g/mol

Purification: The as-received, slightly pink-colored monomer was purified by sublimation,

yielding a white powder. The product was dried under vacuum at room temperature for at least

24 h.

111

3,3’-Disulfonated-4,4’-dichlorodiphenyl sulfone (SDCDPS)

S

O

O

Cl Cl

NaO3S S 3NaO

SDCDPS was synthesized and purified as described earlier.

3.1.2. Polymer Synthesis

6F-BisAF hydrophobic oligomers

nC O

F

F F

F F

F

F

F

FCF3

CF3

O

F

F

F

F

F F

F F

The synthesis of a 6FBisAF oligomer with 5 Kg/mol molecular weight, for example, was carried

out as follows: 6F-BPA (1.147 g, 3.411 mmol) was added to a three-neck round bottom flask

equipped with a mechanical stirrer, a condenser, a nitrogen inlet and a Dean-Stark trap. DMAc

(10 mL) was added to the flask and the mixture was dissolved. Then K2CO3 (0.644 g, 4.669

mmol) was added, followed by 5 mL of toluene. The reaction bath was heated to 150 oC and kept

at this temperature for 2 h to dehydrate the system. The reaction was cooled to 50 oC and

decafluorobiphenyl (1.200 g, 3.591 mmol) was added. The bath temperature was raised to 110 oC

and the reaction was allowed to proceed at this temperature for 5 h. The mixture was precipitated

into 200 mL of water/methanol (50/50 v/v) and rinsed with water and methanol. The precipitated

polymer was dried under vacuum at 100 oC.

112

BPS-100 hydrophilic oligomers

MO O S O OM

O

O

SO3M

MO3Sn

BPSH-100 oligomers with various molecular weights were synthesized by the same procedures

described earlier.

Partially disulfonated hydrophilic oligomers (BPS-75 and BPS-83)

MO O S

O

O

SO3M

MO3S

O O S

O

OO OM

x 1-x

To study the effect of the degree of disulfonation on the properties of the multiblock copolymers,

some partially disulfonated BPS oligomers—specifically, BPS-75 and BPS-83—were

synthesized and used later in the preparation of multiblock copolymers. The procedures we used

to synthesize a BPS-75 oligomer with 8 Kg/mol molecular weight are as follows: A three-neck

round bottom flask, equipped with a mechanical stirrer, a condenser, a nitrogen inlet and a

Dean-Stark trap, was charged with biphenol (0.27 g, 1.450 mmol), SDCDPS (0.513 g, 1.044

mmol), DCDPS (0.100 g, 0.348 mmol), and 10 mL of NMP. The mixture was dissolved, then

K2CO3 (0.26 g, 1.885 mmol) and 5 mL of toluene was added. The reaction bath was heated to

150 oC to dehydrate the system. The bath temperature was then slowly raised to 190 oC by the

controlled removal of toluene. The polymerization was allowed to proceed at this temperature for

30 h, and the resulting oligomer was used in the block copolymer synthesis without isolation.

113

6FBisAF-BPSH100 multiblock copolymers containing fully disulfonated hydrophilic blocks

m

O S O O

O

O

SO3H

HO3S

O

F

F F

F F

F

F

F

O O

CF3

CF3

F

F F

F F

F

F

Fn

x

The 6FBisAF-BPSH100 multiblock copolymers were synthesized by coupling reactions of

6FBisAF and BPS100 oligomers, using similar procedures described earlier for BisAF-BPSH

series.

6FBisAF-BPSH75 and 6FBisAF-BPSH83 multiblock copolymers containing partially disulfonated hydrophilic blocks

n

O O S

O

O

SO3H

HO3S

O O S

O

OO

O

F

F F

F F

F

F

F

O O

CF3

CF3

F

F F

F F

F

F

F

0.75 0.25

x

The 6FBisAF-BPSH75 and 6FBisAF-BPSH83 multiblock copolymers were synthesized by

coupling reactions of 6FBisAF and the partially disulfonated BPS oligomers, using similar

procedures described earlier for the BisAF-BPSH series.

3.1.3. Polymer isolation and characterization

The multiblock copolymers synthesized were purified and characterized using the same

procedure described earlier for the BisAF-BPSH series.

114

3.2. Results and discussion

3.2.1. Polymer synthesis and characterization

The synthetic scheme for the 6FBisAF oligomers is shown in Figure 3.1. The procedures for the

synthesis of the 6FBisAF oligomers are the same as those used for the BisAF oligomers. It was

found, however, that the polymerization would not proceed well when NMP instead of DMAc

was used as the solvent. The synthetic scheme for BPS-75 oligomers is shown in Figure 3.2 as an

example of the partially disulfonated BPS oligomers. The same procedures used to synthesize the

BPSH random copolymers were used for the BPS-75 hydrophilic oligomers, except that biphenol

was used in excess to control molecular weight.

F

F F

F F

F

F

F

FF HO C OHCF3

CF3

DMAc / cyclohexaneK2CO3

110oC, 5h

F

F F

F F

F

F

F

F O C O

F

F F

F F

F

F

F

FCF3

CF3n

+

(excess)

Figure 3.1. Synthesis of 6FBisAF oligomers

115

HO OH Cl S Cl

SO3Na

NaO3S

O

O

NMP/TolueneK2CO3

150oC, 4h

190oC, 30h

MO O SO

O

SO3M

MO3S

O O SO

OO OM

+

(excess)

0.75 0.25

Cl S Cl

O

O

+

Figure 3.2. Synthesis of BPS-75 hydrophilic oligomers

The block copolymerizations (i.e. coupling reactions) between 6FBisAF and BPS100 oligomers

were slower than those shown earlier with the BisAF oligomers. Two types of oligomers were

used as the hydrophilic block: BPS-100 and the partially disulfonated oligomers, BPS-75 and

BPS-83. This was done in order to investigate the effect of the hydrophilic block’s sulfonation

degree on the properties of the multiblock copolymers. The schemes for the synthesis of

6FBisAF-BPSH100 and 6FBisAF-BPSH75 are shown in Figures 3.3 and 3.4, respectively. When

isolated from isopropanol, the resulting 6FBisAF-BPSH copolymers formed weaker fibers

compared to the BisAF-BPSH series copolymers, although they generally showed acceptable

intrinsic viscosity values.

116

F

F F

F F

F

F

F

F O C O

F

F F

F F

F

F

F

FCF3

CF3

m

n

O S O OMO

O

SO3M

MO3S

MO

+

mO S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O C O

F

F F

F F

F

F

F

CF3

CF3n

x

NMP

K2CO3

105 ~ 115oC

4~5 days

Figure 3.3. Synthesis of 6FBisAF-BPSH100 multiblock copolymers

MO O S

O

O

SO3M

MO3S

O O S

O

OO OM

0.75 0.25

F

F F

F F

F

F

F

F O O

CF3

CF3

F

F F

F F

F

F

F

F+n

O O SO

O

SO3M

MO3S

O O SO

OO O

F

F F

F F

F

F

F

O O

CF3

CF3

F

F F

F F

F

F

F

0.75 0.25

n

K2CO3, NMP

105~115oC , 4~5 days

Figure 3.4. Synthesis of 6FBisAF-BPSH75 multiblock copolymers

117

3.2.2. Fundamental characterizations

The 19F NMR spectrum of a 6FBisAF hydrophobic oligomer is shown in Figure 3.5. A spectrum

of the aromatic region, emphasizing the fluorine endgroups, is shown in Figure 3.6. Again, the

integral of peaks due to the main chain and endgroups were used to obtain the experimental Mn.

The calculated and experimental molecular weights for the oligomers, as well as intrinsic

viscosity values (IV), are listed in Table 3.1. Figure 3.7 shows the plot of lnη as a function of

lgMn.

c

ab

-60 -80 -100 -120 -140 -160 PPM

nC OCF3

CF3

O

F

F

F

F

F F

F F

abbac

c

ab

-60 -80 -100 -120 -140 -160 PPM

nC OCF3

CF3

O

F

F

F

F

F F

F F

abbac

Figure 3.5. 19F NMR spectrum of a 6FBisAF oligomer

118

ab

c de

-135 -140 -145 -150 -155 -160 PPM

n

abba c

nC O

F

F F

F F

F

F

F

FCF3

CF3

O

F

F

F

F

F F

F F

d

e

a b

ab

c de

-135 -140 -145 -150 -155 -160 PPM

n

abba c

nC O

F

F F

F F

F

F

F

FCF3

CF3

O

F

F

F

F

F F

F F

d

e

a b

Figure 3.6. 19F NMR spectrum of a 6FBisAF oligomer (aromatic region) showing endgroups in detail

Table 3.1. Molecular weight characterizations of 6FBisAF oligomers

Target Mn

(g/mol

Mn from 19F NMR

(g/mol)

Mn from GPC

(g/mol)

Intrinsic Viscosity

(dL/g)

6K 6.4K 5.6K 0.12

9K 9.5K 10.1K 0.20

12K 13.0K 13.3K 0.23

15 15.9 15.8 0.27

119

-1

-0.9

-0.8

-0.7

-0.6

-0.5

-0.4

3.7 3.8 3.9 4 4.1 4.2 4.3

log(M n)

Log(IV)

Figure 3.7. logη vs. logMn plot for 6FBisAF oligomers

MO O SO

O

SO3M

MO3S

O O SO

OO OM

0.75 0.25

a b b a c d e i h g f

MO O SO

O

SO3M

MO3S

O O SO

OO OM

0.75 0.25

a b b a c d e i h g f

k l l k m n n m k l l k m n n m

Figure 3.8. 1H NMR spectrum of a BPS-75 oligomer

120

The 1H NMR spectrum of a BPS-75 oligomer is shown in Figure 3.8, where all the peaks are

assigned. Figure 3.9 shows the 1H spectrum of a 6FBisAF-BPSH100 multiblock copolymer.

Figure 3.10 shows the spectra of a 6FBisAF-BPSH83 copolymer, as well as the corresponding

BPS-83 oligomer. The peaks due to the oligomer’s endgroups were not observed in the spectrum

of the copolymer, suggesting high conversion of the coupling reaction.

mO S O

O

O

SO3M

MO3S

O

CF3

CF3

F

F F

F F

F

F

F

On

a ab b c

c

d

de

e f fgg

mO S O

O

O

SO3M

MO3S

O

CF3

CF3

F

F F

F F

F

F

F

On

a ab b c

c

d

de

e f fgg

Figure 3.9. 1H NMR spectrum of a 6FBisAF-BPSH100 multiblock copolymer

121

MO O SO

O

SO3M

MO3S

O O SO

OO OM

0.83 0.17

a b b a c d e k l l k m n n m i h g f

n

F

F F

F F

F

F

F

O OCF3

CF3F

F F

F F

F

F

F

BPSH83

O P P O

MO O SO

O

SO3M

MO3S

O O SO

OO OM

0.83 0.17

a b b a c d e k l l k m n n m i h g f

n

F

F F

F F

F

F

F

O OCF3

CF3F

F F

F F

F

F

F

BPSH83

O P P O

Figure 3.10. 1H NMR spectra of a partially disulfonated BPS oligomer and the corresponding

6FBisAF-BPS83 multiblock copolymer

From Figures 3.9 and 3.10, it can be observed that the two peaks that were anticipated for the

6FBisAF blocks merged into one broad peak in these spectra. Therefore, it is possible that the

copolymers might not have formed true solutions in the NMR solvent, DMSO-d6, but rather

tended to form micelle-like aggregates. Furthermore, as shown in Table 3.2, the copolymers’

experimental IEC, obtained from 1H NMR, did not agree with either theoretical value. The

titration values, on the other hand, showed reasonably good agreement with the target IEC.

122

Table 3.2. Comparison of target IEC with experimental values for 6FBisAF-BPSH100

multiblock copolymers

Copolymer Target IEC IEC from 1H NMR IEC from Titration

6FBisAF-BPSH100 (6K-5K) 1.77 1.84 ---

6FBisAF-BPSH100 (12K-12K) 1.46 2.10 1.37

6FBisAF-BPSH100 (9K-9K) 1.32 1.97 1.39

6FBisAF-BPSH100 (12K-7K) 1.25 1.74 1.26

6FBisAF-BPSH100 (12K-11K) 1.2 1.70 1.15

6FBisAF-BPSH100 (9K-8K) 1.2 1.77 1.24

Such large discrepancies cannot be justified by experimental uncertainties. The reason is not

clear, but some qualitative hypotheses could be suggested on the basis of the solution properties

of the copolymers. In fact, the various hydrophobic oligomers including BisAF, 6FBisAF, and

BisSF (which will be discussed in the following chapter), have different solubility in DMSO, as

shown in Table 3.3.

Table 3.3. Solubility of oligomers in DMSO at room temperature

Hydrophobic Hydrophilic

BisAF 6FBisAF BisSF BPS-100 BPS-83 BPS-75

Yes No Yes Yes Yes Yes

123

The insolubility of the 6FBisAF block was probably due to a decrease in polarity, caused by its

much higher fluorine content. The 6FBisAF-BPSH block copolymer chains could, therefore,

have very different conformations in DMSO, compared to those of BisAF-BPSH and

BisSF-BPSH. This may be the reason why experimental IEC values from NMR were not

consistent with the target ones.

Although the copolymerizations were conducted in NMP, in which both 6FBisAF and BPSH-100

oligomers are soluble, there could be finite differences between their solubility. This, in addition

to the increased difference in hydrophobicity of the blocks due to the hexafluoroisopropylidene

group, may partially explain the longer reaction times for the 6FBisAF-BPSH series compared to

the BisAF-BPSH series. We also propose that there could be more pronounced separation

between the hydrophobic and hydrophilic phases, thereby inhibiting the oligomers from

migrating toward and reacting with each other.

Figure 3.11 shows the DSC trace of a 6FBisAF-BPSH100 multiblock copolymer. Two transitions,

one at around 195 oC and the other 245 oC, can be clearly observed, which we assigned to the

glass transitions for the hydrophobic and hydrophilic oligomers, respectively. Such obvious dual

Tgs are only occasionally seen in block ionomers because the molecular motion tends to be

hindered by electrostatic interactions among ionic groups. The two separate Tgs were never

clearly identified in the DSC traces for the BisAF-BPSH and BisSF-BPSH systems, where only

one single Tg was observed. Again, this is probably due to the more distinct phase separation 124

between the 6FBisAF and BPSH-100 components, i.e. the higher χ parameter caused by the

larger differences in their hydrophobicity, in contrast to the BisAF-BPSH and BisSF-BPSH

systems. As we shall see later, the 6FBisAF-BPSH series copolymers displayed very different

proton conductivity and water uptake than the other two series—possibly due to their peculiar

morphological features.

Figure 3.11. DSC trace of a 6FBisAF-BPSH100 (9K:9K) multiblock copolymer

125

3.2.3. Characterization of fuel cell related properties

The proton conductivity and water uptake values for 6FBisSF-BPSH100 copolymers are listed in

Table 3.4. The proton conductivity values at 30 oC in liquid water were generally lower

compared to the BisAF-BPSH and BisSF-BPSH series, and decreased with the decrease of IEC,

regardless of block length. The water uptake values, conversely, were surprising high given the

low IEC and conductivity values. The high water may indicate a percolated hydrophilic

(sulfonated) nanophase throughout the membrane, possibly due to the very high hydrophobicity

of the fluorinated blocks. Although a certain degree of water sorption is required for good proton

conductivity, excessive water uptake may adversely affect the conductivity. In other words, high

water uptake results in high volume swelling of the membrane, so the concentration of sulfonic

acid groups is greatly lowered, leading to low proton conductivity. The only exception is the

15K:10K sample, which featured low IEC values and long hydrophilic block length. This

copolymer displayed much lower water uptake yet still reasonably good proton conductivity. We

attributed this to the fact that the long hydrophobic blocks may have served as a barrier and

prevented the formation of a percolated morphology.

126

Table 3.4. Proton conductivity (liquid water) and water uptake for 6FBisAF-BPSH100

multiblock copolymer membranes

Copolymer IEC (Target) Proton Conductivity at 30 oC (S/cm) Water Uptake

6FBisAF-BPSH100 (6K:5K) 1.77 0.13 150%

6FBisAF-BPSH100 (12K:12K) 1.46 0.09 260%

6FBisAF-BPSH100 (9K:9K) 1.32 0.09 124%

6FBisAF-BPSH100 (12K:7K) 1.25 0.08 69%

6FBisAF-BPSH100 (12K:11K) 1.15 0.065 50%

6FBisAF-BPSH100 (9K:9K) 1.20 - 78%

6FBisAF-BPSH100 (15K:10K) 1.20 0.09 42%

Table 3.5 shows data for the copolymers having partially sulfonated BPSH blocks; the

6FBisAF-BPSH100 (15K:10K) sample is shown for comparison. They all possessed 15 Kg/mol

fluorinated blocks and increasing BPSH block lengths. The 6FBisSF-BPSH-100 (15K:10K) is

also listed for comparison. Although the copolymers had hydrophilic blocks with lower degrees

of sulfonation (as well as lower IECs), they showed much higher water uptake compared to the

BPSH-100-based block copolymers. Thus, they are unlikely candidates for PEMFC applications.

The reason for the high water uptakes in these materials, although not clear, could lie in their

more complex morphology. 127

Table 3.5. Proton conductivity (liquid water) and water uptake for 6FBisAF-BPSH75 and

6FBisAF-BPSH83 multiblock copolymer membranes

Copolymer IEC (Target) Proton Conductivity at 30 oC (S/cm) Water Uptake

6FBisAF-BPSH75 (15K:8K) 1.09 0.07 72%

6FBisAF-BPSH75 (15K:10K) 1.17 0.09 87%

6FBisAF-BPSH75 (15K:20K) 1.27 0.08 113%

6FBisAF-BPSH83 (15K:10K) 1.15 0.08 89%

6FBisAF-BPSH100 (15K:10K) 1.20 0.09 42%

Figures 3.12 and 3.13 display tapping mode AFM images for the 6FBisSF-BPSH-100 (15K:10K)

and 6FBisSF-BPSH-75 (15K:10K) samples. In the phase image in Figure 3.12, the copolymer

with fully sulfonated BPSH blocks shows sharp nanophase separation between distinct

hydrophobic and hydrophilic domains. The morphology for the latter sample, shown in Figure

3.13, is not well defined. Since there are non-sulfonated moieties in the BPSH blocks, there

might exist phase separation between the sulfonated and non-sulfonated potions in the

hydrophilic regions, in addition to the phase separation between 6FBisSF and BPSH domains.

128

Height Phase

500 nm

Height Phase

500 nm

Figure 3.12. Tapping mode AFM images for a 6FBisAF-BPSH100 (15K:10K) multiblock copolymer

Height Phase

500 nm

Height Phase

500 nm

Figure 3.13. Tapping mode AFM images for a 6FBisAF-BPSH75 (15K:9K) multiblock copolymer

129

The proton conductivity at 80 oC under partially-hydrated conditions for the 6FBisSF-BPSH

copolymers is shown in Figure 3.14. Surprisingly, all the 6FBisSF-BPSH100 samples showed

very similar conductivity as a function of RH, which was comparable to that of Nafion 117,

regardless of their IEC and block lengths. This may be attributed to the possible percolated

morphology in all these membranes, but further evidence from X-ray or neutron scattering

experiments would be helpful in determining the reason. In contrast, the 6FBisSF-BPSH75

membranes (not shown) displayed much lower performance under partially hydrated conditions.

This is consistent with its very different morphological features as shown by AFM (discussed

earlier).

0.1

1

10

100

1000

20 30 40 50 60 70 80 90 100Relative Humidity

Pro

ton

Con

duct

ivity

(mS

/cm

)

15-9

12-11

12-7

Nafion 112

Figure 3.14. Proton conductivity as a function of RH for 6FBisAF-BPSH multiblock copolymers

130

In summary, the 6FBisSF-BPSH100 copolymer membranes generally showed higher water

uptake, which was reduced to a satisfactory level when a 15 Kg/mol fluorinated block was used.

The 6FBisSF-BPSH100 (15K:10K) copolymers, displaying higher proton conductivity under

partially hydrated conditions and lower water uptake, have been shown to be excellent

candidates for use in proton exchange membrane. However, the copolymers with partially

sulfonated BPSH blocks displayed lower proton conductivity and higher water uptake, making

them undesirable as PEMs.

131

Chapter 4. Synthesis and characterization of BisSF-BPSH multiblock

copolymers

Although BisAF-BPSH and 6FBisAF-BPSH copolymers have been successfully synthesized and

have been shown to display high performance under partially hydrated conditions, both materials

have drawbacks. While the former have questionable hydrolytic and thermal stability, the latter

suffer from excessive water sorption and swelling, making them less useful as proton exchange

membranes. In this section, we discuss a BisSF-BPSH series featuring fluorinated poly(arylene

ether sulfone) hydrophobic blocks. This type of copolymer should have much better thermal and

hydrolytic stability than BisAF-BPSH series. Moreover, since they don’t have the extremely

hydrophobic 6FBisAF blocks, the undesirable morphology of the 6FBisAF-BPSH samples may

be avoided. Furthermore, this approach is more economically viable because the

4,4’-dihydroxydiphenyl sulfone (bisphenol-S) monomer, which is polymerized with

decafluorobiphenyl, is very inexpensive and readily produced from phenol and sulfuric acid.

For this series of copolymers, we synthesized and characterized the largest number of samples,

with variations in both ion content and block length, and their structure-property relationships

will be discussed in a more systematic way.

132

4.1. Experimental

4.1.1. Materials

4,4’-Dihydroxydiphenyl sulfone (bisphenol-S)

S

O

O

HO OH

Source: Aldrich

Molecular weight: 250.27 g/mol

Purification: Dried under vacuum at 60 oC for 24 h.

4.2. Results and discussion

4.2.1. Polymer synthesis and characterization

4.2.1.1. Synthesis of fluorinated oligomers

The synthetic scheme for BisSF hydrophobic oligomers is shown in Figure 4.1, which was fairly

straightforward. However, the polymerization of decafluorobiphenyl with bisphenol monomers

can be complicated due to the high reactivity of the decafluorobiphenyl. The fluorine moieties at

its para positions are presumably only slightly more reactive than the ones at the ortho positions,

so careful control of the polymerization temperature was needed to avoid branching or

crosslinking (Figure 4.2). For even more reactive perfluoro monomers, such as

133

bis(pentafluorophenyl) sulfone and decafluorobenzophenone, it was even more difficult to obtain

linear polymers, and novel dehydration methods have been attempted to enable the use of even

lower reaction temperatures (Figure 4.3)

F

F F

F F

F

F

F

FF HO S OHO

O

DMAc / cyclohexaneK2CO3

110oC, 5h

F

F F

F F

F

F

F

F O S O

F

F F

F F

F

F

F

FO

On

+

(excess)

Figure 4.1. Synthesis of BisSF telechelic oligomers

O

F F

F F

F

O F

F

O SOS

S

O

O

O

O

O

O

Figure 4.2. Reaction at para- positions leading to branching

134

C

CH3

CH3

HO OH S

O

O

F

F

F F

F F F

F

F F

+

DMAc/benzene/THFK2CO3Molecular sieves 83 oC

C

CH3

CH3

O O S

O

O

F

F F

F F F

F Fn

Figure 4.3. Synthesis of fluorinated poly(ether sulfone) under mild conditions

We found, however, that the polymerization would produce a variety of results depending on

monomer stoichiometry. In the synthesis of all the fluorine-terminal BisSF oligomers, where

excess decafluorobiphenyl was used, the reactions were allowed to proceed for 5 h. However,

as shown in Figure 4.4, a high IV was attained within 2 h. No gel formation was observed during

the reactions, and NMR and GPC results ruled out the existence of branched or crosslinked

structures. However, when the other monomer, 4,4'-dihydroxydiphenyl sulfone, was used in

excess (phenoxide-terminal BisSF oligomer synthesis) under identical conditions, the gelation

would invariably occur within 1.5~2 h. This time range roughly coincides with the time needed

to achieve maximum IV of the linear chains. Thus, one may reasonably infer that bisphenol

monomers and phenoxide chain ends first tend to react preferentially with the para- fluorine

groups until they are depleted (i.e. when the linear polymerization has completed), after which

135

the free phenoxide endgroups remaining at the chain ends can react with the fluorine moieties

along the chains, leading to crosslinking.

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 1 2 3 4 5 6

Polymerization Time (h)

Intrinsic Viscosity (dL/g)

Figure 4.4. Evolution of intrinsic viscosity for BisSF (17K) oligomer synthesis as a function of reaction

time

Stoichiometry may, therefore, be an alternative parameter one can adjust to control side reactions.

To verify this point, two control polymerizations were conducted using a 1:1 stoichiometry. One

of them resulted in a linear polymer with an IV of 0.86 dL/g, whereas gelation was observed in

the other. The slight inconsistency in weight, causing either monomer to be in excess relative to

the other, may have contributed to the unpredictability of these reactions. To reproducibly

synthesize high molecular weight linear polymers, a slight excess of the fluorinated monomer

may be used to avoid the existence of free phenoxide groups. Fortunately, only the

fluorine-terminated oligomers are of interest here, although phenoxide-terminated oligomers may

also be successfully synthesized through very careful control of reaction temperature and/or

time. 136

The 19F NMR spectrum of a BisSF oligomer is shown in Figure 4.5. The Mn values obtained

from GPC and 19F NMR, as well as the target values, are listed in Table 4.1. Figure 4.6 shows log

η plotted against ln(Mn) for the BisSF oligomers, wherein a linear relationship was observed.

F

F F

F F

F

F

F

F O S O

F

F F

F F

F

F

F

FO

On

a b b a a b c d

e

Figure 4.5. 19F NMR of a BisSF telechelic oligomer

137

Table 4.1. Molecular weight characterizations of BisSF oligomers

Target Mn

(g/mol

Mn from 19F NMR

(g/mol)

Mn from GPC

(g/mol)

Intrinsic Viscosity

(dL/g)

5K 5.3K 5.6K 0.15

10K 10.7K 10.9K 0.22

17K 16.8K 17.6K 0.34

25K 26.3 25.9K 0.43

-0.9

-0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

3.6 3.7 3.8 3.9 4 4.1 4.2 4.3 4.4 4.5

log(M n)

log(IV)

Figure 4.6. logη vs. logMn plot for BisSF oligomers

The 19F NMR spectra of a series of BisSF oligomers, with number average molecular weights

ranging from 5K to 25K, are displayed in Figure 4.7. As expected, the peaks corresponding to the

endgroups become smaller as the molecular weight increases.

138

a

e d

-150 -152 -154 -156 -158 -160 PPM

F

F F

F F

F

F

F

F O S O

F

F F

F F

F

F

F

FO

On

a a a d

e

1)

2)

3)

4)

a

e d

-150 -152 -154 -156 -158 -160 PPM

F

F F

F F

F

F

F

F O S O

F

F F

F F

F

F

F

FO

On

a a a d

e

1)

2)

3)

4)

Figure 4.7. 19F NMR plots of BisSF oligomers with Mn of 1) 5K; 2) 10K; 3) 17K; 4) 25K

4.2.1.2. Synthesis of Fully disulfonated hydrophilic oligomers

Fully disulfonated BPS-100 oligomers were synthesized by the nucleophilic aromatic

substitution polymerization of SDCDPS with biphenol. In our early attempts, however, the

oligomers obtained in this way usually showed lower measured molecular weights than the

values predicted based on the Carothers equation. It is now generally accepted that this

discrepancy is due to the presence of residual NaCl from the synthesis of SDCDPS, and, to a

much lesser extent, to moisture absorbed by the monomer during weighing, etc. The

139

SDCDPS/biphenol stoichiometry is therefore overestimated, which has corresponded to an

overestimation of the product's molecular weight. This problem was solved in our study by

taking a convenient engineering approach, namely, calculating the "effective" molar ratio of

SDCDPS to biphenol from the Mn, then adjusting the feed ratio accordingly, i.e. adding more

SDCDPS than calculated for the next experiment. Normally, the use of an extra 1~1.5 mol % of

SDCDPS was found to be sufficient to compensate for the monomer's impurity. Still, it is of

fundamental importance to measure the purity of the monomer. Studies of NaCl content using

UV-Vis spectroscopy are underway in our laboratory.

The purity of the solvent also seemed to affect the polymerization. As the NMP aged, the color of

the reaction solution became darker. Moreover, the difference between experimental and target

molecular weights tended to become greater, requiring even more SDCDPS to maintain the

desired stoichiometry. The reason for this phenomenon is yet unknown, but it is postulated that,

although sealed, NMP may slowly absorb oxygen from the air which causes its oxidative

decomposition under polymerization conditions. There is no evidence, however, of the chemical

integrity of the oligomer being affect by the decomposition of the solvent.

140

4.2.1.3. Synthesis of BisSF-BPSH multiblock copolymers

As illustrated in Figure 4.8, the multiblock copolymers were synthesized via a sequential

addition method. In other words, the selected BisSF oligomer was added to the same flask after

the BPS-100 oligomer synthesis was complete. The BPS-100 oligomers were not isolated before

the block copolymerization because additional steps would have been needed to remove the

residual salt. Furthermore, oligomers with very low molecular weights may not precipitate out

and this would have affected the stoichiometry

F

F F

F F

F

F

F

F O S O

F

F F

F F

F

F

F

FO

O

m

n

O S O OM

O

O

SO3M

MO3S

MO

+

m

O S O O

O

O

SO3M

MO3S

O

F

F F

F F

F

F

F

O S O

F

F F

F F

F

F

F

O

O n

x

NMP

K2CO3

100 ~ 110oC

4~5 days

Figure 4.8. Synthesis of BisSF-BPSH multiblock copolymers

In fact, in our work a perfect 1:1 stoichiometry of BisSF and BPS-100 oligomers was never

utilized. This enabled us to independently control the three parameters of a multiblock

copolymer: hydrophilic block length, hydrophobic block length, and ion exchange capacity

(IEC). In some studies, a 1:1 stoichiometry has typically been used in coupling reactions, and

control of IEC was usually achieved by changing the ratios of hydrophilic/hydrophobic block

141

lengths. For instance, a 10K-5K (hydrophobic-hydrophilic) copolymer, which has shorter

hydrophilic blocks than a 10K-10K specimen, obviously has a lower ion exchange capacity than

the latter copolymer. The same goal can be achieved by raising the hydrophobic block length to

make it a 20K-10K specimen. Thus, as IEC values change, the length of either block will also no

longer be constant. The study of the properties of multiblock copolymer membranes as a function

of ion exchange capacity is, therefore, complicated.

Previously, we demonstrated that a 1:1 stoichiometry was not necessary to synthesize block

copolymers via oligomeric coupling. This is based on the fact that because the repeat units

(blocks) have molecular weights of at least a few thousand g/mol, the degree of polymerization

(coupling) does not have to be very high to achieve a substantial total molecular weight.

Furthermore, the IEC can be controlled by simply changing the stoichiometry (molar feed ratio)

of the oligomers, while maintaining the length of the blocks. Two series of copolymers were

synthesized, each of which had three different IEC values. Within each IEC value, the

copolymers were varied with respect to block length. For simplicity, the hydrophobic block

length was altered simultaneously as the hydrophilic block became longer, and the former was

the same or slighter longer than the latter. As will be shown, however, when other parameters

were kept constant, the only variable that impacted the properties of the membrane was

hydrophobic block length.

142

19F and 1H NMR spectra of a multiblock copolymer are shown in Figures 4.9 and 4.10,

respectively. Here, the peaks due to the endgroups in the case of the BisSF and BPS100

oligomers were not observed; they either disappeared (Figure 4.9) or shifted (Figure 4.10). Thus,

a high conversion coupling reaction was thought to have been achieved. In Figure 4.10, the small

peak at about 7.3 ppm was tentatively assigned to the junction between the fluorinated and

sulfonated blocks.

mO S O

O

O

SO3M

MO3S

S O

F

F F

F F

F

F

F

OO

O n

a b b a

mO S O

O

O

SO3M

MO3S

S O

F

F F

F F

F

F

F

OO

O n

a b b a

Figure 4.9. 19F NMR spectra of a BisSF-BPSH multiblock copolymer

mO S O

O

O

SO3M

MO3S

S O

F

F F

F F

F

F

F

OO

O n

a ab b c d e f fgg

x

mO S O

O

O

SO3M

MO3S

S O

F

F F

F F

F

F

F

OO

O n

a ab b c d e f fgg

x

fg

fg

Figure 4.10. 1H NMR spectra of a BisSF-BPSH multiblock copolymer

143

An 1H NMR spectra of a series of BisSF-BPSH copolymers with increasing block lengths are

overlaid in Figure 4.11. With the other peaks normalized, it can be observed that, as block length

increased, the intensity of the 7.3 ppm peak gradually became smaller. This is consistent with the

fact that, for multiblock copolymers, the longer the segments—the smaller the number of

linkages in each copolymer chain. Therefore, this finding indicates that the hydrophilic and

hydrophobic sequences were preserved and not randomized.

Figure 4.11. 1H NMR spectra of BisSF-BPSH copolymers with increasing block lengths

4.2.1.4. Characterizations of molecular weights of BisSF-BPSH copolymers

Table 4.2 lists the characterization properties of a series of multiblock copolymers having a target

IEC of 1.3 mmeq/mol (Series A). On the one hand, the hydrophobic/hydrophilic weight charge

ratios, which determines IEC, were all the same. The molar charge ratios, on the other hand,

would depend on the hydrophobic and hydrophilic block lengths. Equation 4.1 illustrates the

relationships among the theoretical molecular weight, feed ratio, and block length.

144

2ic)](hydrophilic)(hydrophob[

r-1r1al)(theoretic nn

nMMM +

×+

= ……..(4.1)

Where Mn (theoretical) is the theoretical total molecular weight of the multiblock copolymer, r is

the hydrophilic/hydrophobic molar feed ratio, and Mn(hydrophobic) and Mn(hydrophilic)

represent the hydrophobic and hydrophilic block lengths, respectively. As expected, the

molecular weight (intrinsic viscosity) generally increased as block length increased.

Table 4.2. Characterization of BisSF-BPSH copolymers with 1.3 IEC (Series A)

Copolymer BisSF-BPSH

(xK:yK)a

Weight Feed Ratiob

Molar Feed Ratioc Target IEC IEC from

1H NMR η

(dL/g)

5K:5K 1 : 0.59 1 : 0.59 1.30 1.31 0.64

9K:7K 1 : 0.59 1 : 0.76 1.30 1.20 0.85

17K:12K 1 : 0.59 1 : 0.84 1.30 1.32 1.00

25K:16K 1 : 0.59 1 : 0.79 1.30 1.29 1.03

28K:20K 1 : 0.59 1 : 0.83 1.30 1.33 1.04

a. x and y represent hydrophobic and hydrophilic block lengths, respectively b. Hydrophobic : hydrophilic weight feed ratio c. Hydrophobic : hydrophilic molar feed ratio

The copolymers’ IV values are graphically displayed in Figure 4.12., where it can be seen that as

block lengths increased, the increase in IV gradually slowed down. Apparently, the average

degree of polymerization appears to be lower for copolymers with higher block lengths. In other

words, with an increase in block length, each block copolymer chain tends to incorporate fewer

145

blocks. This trend can be explained as follows: It is hard to achieve quantitative conversion in

such reactions, possibly due to the solution’s high viscosity near the end of the polymerization,

as well as the low concentration of reactive endgroups. If the initial endgroup concentrations are

lower in reactions of longer oligomers—and the residual concentration at the end of the

polymerization presumably shows little variation from reaction to reaction—then lower

endgroup conversion would be observed in the synthesis of longer block copolymers. This is

schematically illustrated in Figure 4.13. This would lead to lower degrees of polymerization as

governed by the Carothers equation, which takes monomer conversion (p) into account:

2rp-r1r1Xn

++

= ………4.2

2ic)](hydrophilic)(hydrophob[

2rp-r1r1al)(theoretic nn

nMM

M+

×++

= ……..4.3

According to Equation 4.2, the degree of polymerization nX decreases as p decreases. As

shown in Equation 4.3, the contributions from the molecular weights of the repeat units and

endgroup conversions may, to some extent, have cancelled each other out.

146

0

0.2

0.4

0.6

0.8

1

1.2

5K:5K 9K:7K 17K:12K 25K:16K 28K:20K

Figure 4.12. Intrinsic viscosity as a function of block lengths for BisSF-BPSH multiblock copolymers

(IEC=1.3)

End

-gro

up c

once

ntra

tion

End

-gro

up c

once

ntra

tion

Time Time

Highconversion

Lowconversion

(a) (b)

End

-gro

up c

once

ntra

tion

End

-gro

up c

once

ntra

tion

Time Time

Highconversion

Lowconversion

(a) (b)

Figure 4.13. Schematic plots showing the decrease of endgroup concentration as a function of time for

the syntheses of multiblock copolymers with (a) low block lengths; (b) high block lengths

147

The above discussion is highly qualitative and idealized. First, since no reliable molecular weight

data can be obtained from GPC for these block ionomers, intrinsic viscosity may be the only

parameter one can use to assess molecular weight. Secondly, it is impossible to know whether as

many functional groups as possible have reacted at the end of the polymerization. Finally, r, the

molar feed ratio, varies from reaction to reaction, and even small variations can cause big

changes in stoichiometry. Yet, the molar feed ratio may still be of some practical importance in

controlling molecular weight control during block copolymer synthesis.

The properties of another series of copolymers with 1.1 mmeq/mol IEC (Series B) are shown in

Table 4.3. In contrast to Series A, the feed ratios used in this series were much lower in order to

achieve lower IEC values. It can be inferred from the IV values that the copolymers still display

acceptable molecular weights despite the use of such low molar feed ratios.

Table 4.3. Characterization of BisSF-BPSH copolymers with 1.1 IEC (Series B)

Copolymer

BisSF-BPSH

(xK:yK)a

Weight

Feed Ratiob

Molar Feed

Ratioc Target IEC

IEC from 1H NMR

η

(dL/g)

17K:12K 1 : 0.45 1 : 0.64 1.10 1.06 0.78

25K:16K 1 : 0.45 1 : 0.65 1.10 1.08 0.87

28K:20K 1 : 0.45 1 : 0.63 1.10 1.04 0.95

a. x and y represent hydrophobic and hydrophilic block lengths respectively b. Hydrophobic : hydrophilic weight feed ratio c. Hydrophobic : hydrophilic molar feed ratio

148

A plot of intrinsic viscosity vs. block length for BisSF-BPSH multiblock copolymers is shown in

Figure 4.14, which clearly indicates a more linear relationship. Due to the fact that the

hydrophobic oligomers were all used in great excess, high endgroup conversion may have been

achieved without the endgroups being depleted. In other words, there would always be a large

number of residual fluorine endgroups remaining in the system, even at high oligomer

conversion. Therefore, even when low molar feed ratios were used, the target total molecular

weights would remain low, such that the low viscosity of the reaction solution would allow the

polymerizations to proceed until completion. And indeed, the molecular weights of this series of

copolymers show a trend that is consistent with predicted values.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

17K:12K 25K:16K 28K:20K

Figure 4.14. Intrinsic viscosity as a function of block lengths for BisSF-BPSH multiblock copolymers

(IEC=1.1)

Finally, the characterization of copolymers with 1.5 mmeq/mol IEC (Series C) are listed in Table

4.4. For this series of copolymers, in order to obtain high IEC values the

149

hydrophilic/hydrophobic molar feed ratios were kept quite high and close to a 1:1 ratio. The

intrinsic viscosity values for these copolymers, as expected, were lower than the ones in the other

two series.

Table 4.4. Characterizations of BisSF-BPSH copolymers with 1.5 IEC (Series C)

Copolymer BisSF-BPSH

(xK:yK)a

Weight Feed Ratiob

Molar Feed Ratioc Target IEC IEC from

1H NMR η

(dL/g)

7K:5K 1 : 0.72 1 : 1.01 1.50 1.51 1.01

17K:12K 1 : 0.72 1 : 1.02 1.50 1.46 1.12

25K:16K 1 : 0.72 1 : 1.04 1.50 1.47 1.09

28K:20K 1 : 0.72 1 : 1.01 1.50 1.44 1.08

a. x and y represent hydrophobic and hydrophilic block lengths respectively b. Hydrophobic : hydrophilic weight feed ratio c. Hydrophobic : hydrophilic molar feed ratio

Figure 4.15 compares intrinsic viscosity vs. block length for this series of copolymers. Here, the

copolymers all had similar intrinsic viscosity values, regardless of the block length, i.e., the

longer block length materials did not benefit from the high MW of the repeat units. This is due to

the fact that the synthesis of these copolymers is of a type in which quantitative conversion can

never be achieved. The coupling reactions generally stop when the concentration of reactive

endgroups falls below a certain level, and the endgroup concentration is inversely proportional to

the total molecular weight, no matter what the block lengths are.

150

0

0.2

0.4

0.6

0.8

1

7K:5K 17K:12K 25K:16K 28K:20K

Figure 4.15. Intrinsic viscosity as a function of block lengths for BisSF-BPSH multiblock copolymers

(IEC=1.5)

Two conclusions may be drawn from this discussion. First, the molecular weights of multiblock

copolymers, which have been synthesized by block copolymerizations of telechelic oligomers,

are semi-quantitatively predictable from the Carothers equation, but only when the blocks are

short, or when low molar feed ratios are used. For reactions of oligomers with high molecular

weights and/or near perfect feed ratios, the experimental molecular weights tend to deviate from

the predicted values, presumably due to low endgroup conversion. Secondly, since similar

molecular weights are achieved at both high and low feed ratios, there is no advantage in using

high molar ratios, especially for copolymers with high block lengths. This fact, as well as the fact

that acceptable molecular weights were obtained at very low feed ratios (Table 4.3), justifies

molar ratio as another parameter that can be adjusted with some success to control IEC without

adversely affecting molecular weight.

151

4.2.1.5. Thermal analysis

The DSC trace of the BisSF-BPSH (17K-12K) multiblock copolymer is shown in Figure 4.16.

The copolymer was tested in the salt (K+) form because the acid-form copolymer membranes are

known to be thermally and oxidatively unstable above about 240 oC. A sharp transition was

observed at 205 oC and was assigned to the Tg of BisSF blocks. Another very broad transition,

covering a temperature range of 230-270 oC, was linked to the relaxation of the BPSH ionic

blocks. The electrostatic interactions among sulfonic groups in this system probably compound

the intermolecular forces, resulting in the broad thermal transition. Nevertheless, the observation

of the clear Tg for fluorinated blocks indicates the existence of nanophase separated block

structures.

The TGA traces of BisSF-BPSH (17K-12K) copolymers, whose IECs range from 1.1 to 1.5 are

shown in Figure 4.17. Such copolymers generally display a three-step weight loss: desulfonation

(~270 oC), degradation of BPSH blocks (~490 oC), and degradation of BisSF blocks (~550 oC).

The three copolymers in Figure 4.17 displayed very similar weight loss behaviors, which differed

slightly in terms of percentage weight within the range of 270 ~ 490 oC. As expected, this

suggests different degrees of desulfonation, caused by differences in the ion content of the

copolymers (IEC). Similar trends have been observed for BPSH-xx random copolymers.

152

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

160 180 200 220 240 260 280 300

Hea

t Flo

w (W

/g)

Temp (oC)

Tg of BisSF blocks

BPSH blocks

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

160 180 200 220 240 260 280 300

Hea

t Flo

w (W

/g)

Temp (oC)

Tg of BisSF blocks

BPSH blocks

Figure 4.16. DSC trace of a BisSF-BPSH (17K-12K) multiblock copolymer

-20

0

20

40

60

80

100

120

0 100 200 300 400 500 600 700

1. BisSF-BPSH (17K-12K) IEC=1.12. BisSF-BPSH (17K-12K) IEC=1.33. BisSF-BPSH (17K-12K) IEC=1.5

Wei

ght %

Temp (oC)

1

23

Figure 4.17. TGA traces of BisSF-BPSH (17K-12K) multiblock copolymers

153

4.2.1.6. Mechanical Properties

Stress-strain test results for the BisSF-BPSH (17K-12K) (IEC=1.5) copolymer membranes are

illustrated in Figure 4.18. These materials show the typical mechanical response of glassy

amorphous polymers. Copolymer membranes having the same block length—but an IEC of

1.1—were also tested, and Table 4.5 lists the moduli, tensile strength, and elongation-at-break

results for the two samples. The low-IEC copolymer had lower molecular weight, but according

to Table 4.5, it displayed a higher modulus than the high-IEC sample. The two membranes,

however, had very similar tensile strengths. The molecular weight, therefore, has no significant

effect on the moduli of the copolymers. The differences in moduli may have resulted from the

differences in their composition. In addition, the material having an IEC of 1.1 had a higher

content of fluorinated (hard) segments and fewer BPSH (soft) segments. In general, BPSH

blocks are relatively soft because they readily absorb moisture, which acts as a plasticizer. The

only negative effect of low-IEC (low molecular weight) on the copolymer’s mechanical

properties was lower ductility, as reflected in its lower elongation-at-break values compared to

the high-IEC materials.

154

Elongation

Stre

ss (M

pa)

0

10

20

30

40

50

60

0% 10% 20% 30% 40% 50%

Elongation

Stre

ss (M

pa)

0

10

20

30

40

50

60

0% 10% 20% 30% 40% 50%

Figure 4.18. Stress-strain curves for BisSF-BPSH (17K-12K) (IEC=1.5) copolymers

Table 4.5. Effects of IEC on the mechanical properties of BisSF-BPSH (17K-12K) multiblock

copolymers

IEC 1.1 1.5

Modulus (MPa) 1191 1716

Tensile Strength (MPa) 50.4 48.5

Elongation-at-Break (%) 29.4 44.9

4.2.1.7. Surface morphological features

Block copolymers are well known to form a nanophase separated morphology. The

tapping-mode atomic force micrographs of four BisAF-BPSH multiblock copolymers, with

increasing block lengths, are shown in Figure 4.19. Here, the bright regions are due to the

fluorinated hydrophobic component, while the dark regions represent the hydrophilic, sulfonated

155

domains. From the phase images it can be seen that, with an increase in block length, the

nanophase separation between hydrophobic and hydrophilic domains becomes sharper and,

moreover, the hydrophilic domains tend to become increasingly connected. For the copolymer

with the highest block lengths, i.e. BisSF-BPSH (25:16), co-continuous hydrophilic domains can

be observed. This increased connectivity with an increase in sequence length has been identified

for other multiblock copolymer systems as well. In fact, it is thought to be responsible for the

increase in proton conductivity with increasing block length under partially hydrated conditions

(to be documented in a later publication).

156

100 nm

100 nm

100 nm

100 nm

5K-5K

17K-12K

25K-16K

7K-7K

100 nm

100 nm

100 nm

100 nm

100 nm

100 nm

100 nm

100 nm

5K-5K

17K-12K

25K-16K

7K-7K

Figure 4.19. Tapping mode AFM height (left) and phase (right) images for BisSF-BPSH (5K-5K),

(7K-7K), (17K-12K) and (25K-16K) multiblock copolymer membranes

157

4.2.2. Study of fuel cell-related properties

4.2.2.1. Effects of block lengths on proton conductivity and water uptake

The proton conductivity for selected BisSF-BPSH copolymers is shown in Figure 20. These

samples all have a nominal IEC of 1.3 and display gradually increasing block lengths. As shown,

the proton conductivity at 30 oC in liquid water gradually increased with IEC, but reached a

maximum of 0.13 S/cm for the 17K:12K sample.

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

Nafion112 BPSH-35 7K:5K 10K:8K 15K:10K 17K:12K 25K:16K

Proton Conductivity (S/cm)

Figure 4.20. Proton conductivity for BisSF-BPSH copolymers having an IEC of 1.3

The water uptake values for this series of copolymers are shown in Figure 4.21. The data are

shown as a function of the hydrophilic (BPSH) block length. When it was below 16K,

corresponding to the BisSF-BPSH (25K-16K) copolymer, the water uptake gradually increased

up to around 40 wt%, which was well within the acceptable range for the membrane’s

dimensional stability and long-term durability. When the hydrophilic block length was further

158

increased to 20K, corresponding to a BisSF-BPSH (28K-20K) copolymer, the water uptake

experienced a sharp increase to over 130 wt%. At this level, the membrane loses all of its

mechanical strength in the water-swollen state and becomes impractical for use as a PEM.

0

20

40

60

80

100

120

140

160

0 10 20

Hydrophilic block length (Kg/mol)

Wat

er u

ptak

e (w

t%)

30

Figure 4.21. Water uptake as a function of Block lengths for BisSF-BPSH (IEC=1.3) copolymers

A plot of water uptake vs. disulfonation degree for BPSH random copolymers is shown in Figure

4.22. A similar trend was observed here; namely, the water uptake increased almost linearly as a

function of the degree of disulfonation and then rose abruptly to over 180 wt%. This is

commonly referred to as the “phase inversion” (i.e. percolated hydrophilic and isolated

hydrophobic domains) or the point at which the percolation limit has been exceeded. The high

water uptake for BisSF-BPSH (28K-20K) copolymers (Figure 4.21) can therefore be viewed as

the block copolymer equivalent of the percolation limit. Thus, we have found that the same

159

phenomenon that occurs in random copolymers as a result of excessive IEC, can also occur in

multiblock copolymers as a result of excessive block length. It is also clear that when an IEC of

1.3 is utilized, the maximum block length that can be used without losing dimensional stability is

25K-16K.

0

20

40

60

80

100

120

140

160

180

200

0 20 40 60

mol% Disulfonation

Wat

er u

ptak

e (w

t%)

80

0

20

40

60

80

100

120

140

160

180

200

0 20 40 60

mol% Disulfonation

Wat

er u

ptak

e (w

t%)

80

Figure 4.22. Water uptake as a function of disulfonation degree for BPSH random copolymers

The performance characteristics of a multiblock copolymer under partially-hydrated conditions is

extremely important for a PEM. Figure 4.23 shows proton conductivity at 80 oC as a function of

relative humidity (RH) for a series of BisSF-BPSH multiblock copolymers. The data for Nafion

112 and BPSH-35 random copolymer membranes are shown for comparison. As shown, the

conductivity for BPSH-35 decreased very rapidly as RH decreased. The performance of the

block copolymer membranes, however, was much higher and consistently increased with

increases in block length. This is consistent with the more developed nanophase separated

160

morphology illustrated in our AFM characterization studies (Figure 4.19). As our results

indicated, however, Nafion 112 was still the best material we evaluated, outperforming the

BisSF-BPSH (25K-16K) copolymers. Since the 25K-16K materials were the highest block

length copolymers we used (while maintaining reasonable water uptake), the BisSF-BPSH

(25K-16K) materials may represent the upper limit of performance improvements by means of

altering (hydrophilic) sequence lengths.

0.01

0.1

1

10

100

1000

20 30 40 50 60 70 80 90 100

(1) Nafion112(2) 25K-16K(3) 15K-10K(4) 10K-8K(5) 5K-5K(6) BPSH-35 Random

Proton Conductivity (mS/c

m)

Relative Humidity (%)

1

2

34

5

6

Figure 4.23. Proton conductivity at 80 oC as a function of relative humidity for BisSF-BPSH (IEC=1.3)

copolymers

Thus, the BisSF-BPSH series copolymers generally showed high liquid-water proton

conductivity (up to 0.13 S/cm), but at similar IECs and block lengths, their performances under

partially hydrated conditions generally seemed to be lower than those of the BisAF-BPSH and

161

6FBisAF-BPSH systems. This indicates a lower extent of nanophase separation between the

BisSF and BPSH blocks. In other words, the presence of the sulfone groups in the BisSF

blocks may cause some similarity between the hydrophobic and hydrophilic blocks, thus

discouraging them from phase separating.

4.2.2.2. Effects of composition (IEC) on proton conductivity and water uptake

Another parameter that can be tailored in multiblock copolymers, in addition to block length, is

the ion exchange capacity. With the goal of achieving higher performance at low RH,

BisSF-BPSH copolymer membranes having higher IEC values were synthesized and tested.

Table 4.6 shows some characterizations of two series of copolymers: BisSF-BPSH (17K-12K)

and BisSF-BPSH (25K-16K). The IEC values of each series varied from 1.1 to 1.5. As shown,

when the IEC was increased from 1.1 to 1.3, significant changes were observed in both liquid

water conductivity (30 oC) and water uptake. However, there seemed to be little effect on IEC as

documented by the fact that it only increased from 1.3 to 1.5.

Table 4.6. The effect of IEC on the properties of BisSF-BPSH copolymers

Block Lengths (g/mol)

IEC Molar Feed Ratio

Intrinsic Viscosity

(dL/g)

Water Uptake Proton Conductivity

at 30oC (S/cm)

17K-12K 1.1 1 : 0.64 0.78 25% 0.08 17K-12K 1.3 1 : 0.84 1.00 42% 0.13 17K-12K 1.5 1 : 1.02 1.02 43% 0.13 25K-16K 1.1 1 : 0.65 0.95 23% 0.08 25K-16K 1.3 1 : 0.84 1.03 44% 0.13 25K-16K 1.5 1 : 1.04 1.04 50% 0.14

162

The effects of IEC on the performance of copolymers under partially hydrated conditions have

also been studied. Figure 4.24 shows the proton conductivity at 80 oC as a function of RH for

both BisSF-BPSH (17-12) copolymers and Nafion 112. Here, in contrast to the case of fully

hydrated conditions, performance significantly increased as IEC increased from 1.3 to 1.5.

Moreover, the copolymer with an IEC of 1.5 outperformed the Nafion 112. Thus, what could not

be achieved by increasing block length was accomplished by changing the IEC. More

importantly, good performance was obtained without sacrificing durability, since the increase in

IEC did not result in any changes in water uptake.

0.1

1

10

100

1000

20 30 40 50 60 70 80 90 100

(1) 17K-12K (IEC=1.5)(2) 17K-12K (IEC=1.3)(3) 17K-12K (IEC=1.1)(4) Nafion112

Proton Conductivity (mS/cm)

Relative Humidity (%)

1

2

3

4

Figure 4.24. Proton conductivity vs. RH: the effect of IEC on the performance of BisSF-BPSH (17K-12K)

copolymers

163

Figure 4.25 shows the same data for the BisSF-BPSH (25K-16K) copolymers. Again, the

increase in IEC resulted in higher performance. It is worth noting that the performance of the

17K-12K samples exhibited a greater dependence on IEC than did the 25K-16K copolymers,

possibly because the longer block samples displayed a better established morphology, regardless

of ion content. However, it is not clear why the 25K-16K (IEC=1.5) sample performed poorly in

comparison to the 17K-12K (IEC=1.5) sample, even though the former has longer blocks. This

might have been due to the fact that both the hydrophobic (BisSF) and hydrophilic (BPSH)

blocks are higher in the 25K-16K (IEC=1.5) sample. As will be discussed later, while longer

hydrophilic (BPSH) blocks favor higher performance, an increase in hydrophobic block length

tends to lower a membrane’s transport properties. These are contradicting variables and in most

cases, and the hydrophilic block length effect seems to dominate when BisSF and BPSH block

lengths are raised simultaneously. For the 25K-16K copolymer (IEC=1.5), however, it is thought

that the BisSF block is so long that its influence is dominant over the BPSH block.

164

0.1

1

10

100

1000

20 30 40 50 60 70 80 90 100

(1) 25K-16K (IEC=1.5)(2) 25K-16K (IEC=1.3)(3) 25K-16K (IEC=1.1)P

roton Conductivity (mS/cm)

Relative Humidity (%)

1

2

3

Figure 4.25. Proton conductivity vs. RH: the effect of IEC on the performance of BisSF-BPSH (25K-16K)

copolymers

4.2.2.3. Effects of hydrophobic block length (hydrophobic/hydrophilic block length ratio)

Up to this point, the hydrophobic blocks of most of the copolymers we have studied have been

longer than the hydrophilic blocks for the same copolymer—and have increased as the latter

have increased. Therefore, the discrete effects of only the hydrophobic block on a copolymer’s

properties have not been studied. The properties of a BisSF-BPSH (15K-15K) copolymer, with

an IEC of 1.5, are listed in Table 4.7. The BisSF-BPSH (25K-16K) sample is also shown for

comparison. In contrast to the 25K-16K sample, the 15K-15K copolymer, with its shorter

hydrophobic blocks, displayed slightly higher proton conductivity at 30 oC in liquid water and

significantly higher water uptake (78% vs. 42%). We surmised that this sample probably has a 165

percolated morphology simlar to the 28K:20K (IEC=1.3) material. As shown in Figure 4.26, the

BisSF-BPSH (15K-15K) materials exhibited significantly higher performance than the

BisSF-BPSH (25K-16K) copolymers under partially hydrated conditions.

Table 4.7. Comparison between BisSF-BPSH multiblock copolymers having different

hydrophobic block lengths

Block Lengths

(g/mol)

IEC Proton Conductivity

at 30 oC (S/cm)

Water Uptake

15K:15K 1.5 0.14 78%

25K:16K 1.5 0.13 50%

1

10

100

1000

20 30 40 50 60 70 80 90 100

(1) BisSF-BPSH (25K-16K)(2) BisSF-BPSH (15K-15K)

Pro

ton

Con

duct

ivity

(mS

/cm

)

Relative Humidity (%)

2

1

Figure 4.26. Proton conductivity under partially hydrated conditions for BisSF-BPSH (25K-16K)

(IEC=1.5) and BisSF-BPSH (15K-15K) (IEC=1.5)

166

As another example, Table 4.8 shows the properties of BisSF-BPSH (25K:12K) and

BisSF-BPSH (17K:12K) multiblock copolymers with IEC values of 1.3. As shown, these two

materials differed only in their hydrophobic block lengths. Again, the (25K:12K) copolymers

showed much lower water uptake and proton conductivity than the 17K:12: materials. Thus, the

molecular weight of the hydrophobic block alone does have a substantial effect on a copolymer’s

properties.

Table 4.8. Comparison between BisSF-BPSH multiblock copolymers having different

hydrophobic block lengths

Block Lengths

(g/mol)

IEC Proton Conductivity

at 30 oC (S/cm)

Water Uptake

25K:12K 1.3 0.04 23%

17K:12K 1.3 0.13 50%

In summary, given a certain structure of the hydrophobic and hydrophilic blocks, the

performance of a hydrophilic/hydrophobic multiblock copolymer with a certain chemical

structure is a function of three parameters: hydrophilic block length, IEC, and hydrophobic block

length. New synthetic methodologies have enabled us to tailor a material for optimal

performance by controlling all three parameters independently. Accordingly, within a certain

range the proton conductivity of a copolymer at low hydration levels can increase when any of

the following occurs:

167

The hydrophilic block length increases;

Or the hydrophobic block length decreases;

Or the IEC increases.

However, an increase in conductivity performance would generally be accompanied by an

increase in water uptake. But since conductivity and water sorption can change by different

degrees depending on nature of the copolymer, it is critical to determine the optimum balance

between these two variables. Thus far, the BisSF-BPSH (17K-12K) (IEC=1.5) copolymers

have displayed the highest performance/water sorption selectivity of all the multiblock

copolymers we tested based on our careful control of composition and sequence length.

4.2.2.4. Swelling-deswelling properties of multiblock copolymers

The weight-based water sorption of a PEM is an important parameter in determining its

mechanical strength in the water swollen state. However, we have recently been studying the

volume-based water swelling ratio, which may be more relevant to the long-term stability of a

membrane electrode assembly (MEA). When a membrane undergoes significant swelling in

water, considerable stress is created at the interface between the membrane and the electrodes,

causing fatigue of the PEM during the wet-dry cycles. These swelling-deswelling cycles can

eventually to the delamination or failure of the membrane. We have found, however, that the

BisSF-BPSH multiblock copolymer membranes display attractive swelling-deswelling behavior,

making them promising as alternative PEMs.

168

The volume-based swelling ratios for the copolymers were measured in x, y and z axes. Figure

4.27 shows the results for BisSF-BPSH (17:12) (IEC=1.3), BisSF-BPSH (5K:5K) (IEC=1.3)

copolymers, BPSH-35 random copolymer membranes, and Nafion 117. The BPSH-35

copolymer displayed an isotropic swelling behavior, that is, it had similar degrees of swelling in

all x, y and z directions. Nafion, however, exhibited different swelling in different directions.

Specifically, the increase in thickness (z axis) was greater than in x-y plane. This anisotropic

swelling behavior has its origin in Nafion’s unique nanophase separated morphology.

0

5

10

15

20

25

30

1 2 3 4

X

Y

Z

BPSH-35 Nafion112 5K-5K 17K-12K0

5

10

15

20

25

30

1 2 3 4

X

Y

Z

BPSH-35 Nafion112 5K-5K 17K-12K

Figure 4.27. Degrees of swelling in x, y and z directions for different copolymer membranes

The block copolymers with different block lengths showed very different swelling behaviors due

to differences in their morphology. The BisSF-BPSH (5K-5K) sample had slightly more in-plane

swelling than through-plane swelling. The copolymers with long hydrophobic and hydrophilic

blocks, however, showed different swelling patterns. For example, the 17K-12K sample

demonstrated tremendously higher swelling in the z axis than in the x-y plane. This is exactly

169

what is desired for a PEM, because the stability of the membrane-electrode interfaces should be

mostly dependent on the in-plane swelling, and should not be strongly affected by any increase

in the membrane thickness. This strong anisotropic swelling has been found for other types of

hydrophilic-hydrophobic multiblock copolymers, which we believe will result in significant

improvements in the MEA’s long term durability, making these materials promising candidates

as PEMs. Some propose that Nafion, with its sulfonic acid group-bearing side chains, possesses

some block copolymer-like characteristics. The argument is supported by the fact that Nafion

shows a swelling behavior that falls somewhat in between random and block copolymers.

The swelling-deswelling properties of BisSF-BPSH (17K-12K) and BisSF-BPSH (25K-16K)

copolymers are shown in Figure 4.28. Each series contains two copolymers having IECs of 1.3

and 1.5, respectively. As shown in Table 4.6, although the four copolymers possess very similar

weight-based water sorption levels, in this graph one can observe that the z-direction swelling

degree for the multiblock copolymers increases as the block length or IEC is increased. This can

be ascribed to the higher proportion of “free water” (freezable water) in a system with a more

developed nanophase separated morphology due to higher block lengths or IEC.

170

BPSH-350

10

20

30

40

50

60

70

Swel

ling

Rat

io (%

)X

Y

Z

Nafion112 17K-12KIEC=1.3

17K-12KIEC=1.5

25K-16KIEC=1.3

25K-16KIEC=1.5

BPSH-350

10

20

30

40

50

60

70

Swel

ling

Rat

io (%

)X

Y

Z

Nafion112 17K-12KIEC=1.3

17K-12KIEC=1.5

25K-16KIEC=1.3

25K-16KIEC=1.5

Figure 4.28. Multiblock copolymers with long blocks or higher IEC show higher z-direction swelling

It is worth noting here that while the swelling degree in the z direction increased as the block

lengths and/or IEC increased, these parameters had no significant effect on swelling in x and y

directions. Thus, if such swelling behavior persists for copolymers having even higher block

lengths and/or IEC (at least to a certain extent), then one may be able to obtain copolymer

membranes that show higher performance—but that still possess dimensional stability. That said,

there must be a limit to this approach. For example, the BisSF-BPSH (28K-20K) (IEC=1.3)

and BisSF-BPSH (15K-15K) (IEC=1.5) samples demonstrated completely different water

sorption and swelling properties compared to other materials in the series as a result of

qualitative, significant changes in morphology.

171

Although we are uncertain as to why the copolymer membranes tend to swell in the z direction,

with instruments currently available it is impossible to characterize the cross-sectional

morphology of a water-swollen membrane. However, we propose a possible explanation for

the z-direction swelling, as follows. During the casting of a membrane, the block copolymer

chains may be orientated parallel to the membrane plane, but there is no orientation in the x-y

plane. The hydrophilic blocks may be brought together vertically through self-assembly, thereby

creating the hydrophilic domains (Figure 4.29). When the membrane is wet, the ionic domains

swell vertically as water molecules enter the intermolecular spaces, causing an increase in the

membrane’s thickness. Along the x-y plane, however, intermolecular forces and chain

entanglements preclude any significant degree of parallel expansion due to lack of polymer chain

orientation. The reason why the BisSF-BPSH (5K-5K) copolymer demonstrated largely isotropic

swelling may have been due to the less sharp nanophase separation (and thus the absence of

self-assembly) associated with the short sequence lengths in this material.

BisSF block BPSH block

z

BisSF block BPSH block

z

Figure 4.29. Imaginary cross-sectional view of BisSF-BPSH (17K-12K) copolymer membranes

172

If this somewhat idealized scenario is true, then it turns out that the multiblock copolymer PEMs

have another great advantage: the ionic, hydrophilic domains are preferentially orientated in the z

direction. This would cause the through-plane conductivity, which dominates fuel cell

performance, to be much higher than the in-plane conductivity. Therefore, the block copolymer

PEM approach is promising in terms of both performance (in a real-world fuel cell testing

environment) and durability.

4.2.3. Some comparisons of BisAF-BPSH, 6FBisAF-BPSH and BisSF-BPSH multiblock

copolymers

All three series of copolymer membranes studied thus far, BisAF-BPSH (Series 1),

6FBisAF-BPSH (Series 2) and BisSF-BPSH (Series 3), have shown superior performance over

random copolymers under both fully- and partially hydrated conditions. However, both Series 1

and Series 2 multiblock copolymers suffer from disadvantages that prevent them from being

seriously considered for fuel cell applications.

For example, with respect to the Series 1 materials (BisAF-BPSH), the hydrolytic stability of the

isopropylidene units has been questioned. We postulated that by catalyzing the sulfonic acid

groups, a reverse reaction of bisphenol-A synthesis would occur in water at elevated

temperatures, leading to chain scission. However, a BisAF-BPSH membrane that was kept in

water at 80 oC for up to 8 days did not show any decrease in intrinsic viscosity—nor did NMR

173

studies reveal any structural changes. Therefore, the concentration of the sulfonic acid groups is

probably not high enough to catalyze the degradation.

Instead, it may be that thermal oxidative stability is more important. To make a membrane

electrode assembly, the copolymer membrane must undergo a “hot-pressing” procedure at 220 oC

for 10 min to obtain good interfaces between the membrane and the electrodes. After

hot-pressing the BisAF-BPSH copolymer, it would become dark-colored and brittle; whereas

samples from the other two series did not show any visible change. Thus, although the

BisAF-BPSH copolymers may be durable in the fuel cell operation environment, they are

currently unable to survive essential processing procedures due to the existence of thermally

unstable isopropylidene groups—thereby making them impractical.

The BisSF-BPSH copolymers (Series 3), therefore, are the most promising candidates as PEMs.

They do not suffer from thermal oxidative stability issues associated with Series 1 copolymers,

and they do not show the excessive water sorption and swelling-deswelling behaviors that were

observed in the Series 2 membranes. Moreover, the 4,4’-dihydroxydiphenol (Bisphenol-S)

monomer is very inexpensive, which makes this approach also economically more attractive than

Series 1 and 2.

The TGA traces for the copolymers from all three series are shown and compared in Figure 4.30.

As expected, the BisAF-BPSH copolymer displayed much more rapid weight loss than the other

174

two series. Also, the BisSF-BPSH and 6FBisAF-BPSH copolymers showed very similar weight

loss profiles, with the former just as thermally stable as the latter.

-20

0

20

40

60

80

100

120

0 100 200 300 400 500 600 700

1. BisAF-BPSH 8K-8K2. BisSF-BPSH 17K-12K3. 6FBisAF-BPSH 12K-9K

Wei

ght %

Temp (oC)

12

3

Figure 4.30. Comparison of TGA traces for three series of multiblock copolymers

175

Chapter 5. Conclusions

Hydrophilic-hydrophobic, sulfonated-fluorinated multiblock copolymers, were obtained through

moderate temperature (100°C) coupling reactions between fluorinated and fully disulfonated

telechelic oligomers. Three series of hydrophilic-hydrophobic copolymers, BisAF-BPSH (Series

1), 6FBisAF-BPSH (Series 2), and BisSF-BPSH (Series 3), were studied. Their structures were

similar except for the molecular structure of the bisphenol unit in the hydrophobic fluorinated

blocks: isopropylidene for Series 1, hexafluoroisopropylidene for Series 2, and a sulfone group

for Series 3. The highly reactive fluorine endgroups in the hydrophobic oligomers enabled the

coupling reactions to be run at low temperatures (100 ~ 115 oC) at reasonable rates, in order to

minimize the possibility of ether-ether interchange. The composition (ion exchange capacity) of

these copolymers could be conveniently controlled through molar charge ratios of

hydrophilic/hydrophobic telechelic oligomers and verified by 1H NMR studies or titration. The

existence of block sequences in these materials was supported by 1H NMR results, where the

distinct peak associated with the junction of the hydrophobic and hydrophilic blocks could be

observed. Nanophase separated morphologies have been confirmed by atomic force microscopy

studies, and the high molecular weights were confirmed by intrinsic viscosity and indirectly by

mechanical testing results.

All three series of copolymer membranes studied have shown superior performance over random

copolymers under both fully and partially hydrated conditions, which is probably related to

easier water diffusion paths through the co-continuous hydrophobic phase. However, both 176

BisAF-BPSH (Series 1) and 6FBisAF-BPSH (Series 2) copolymers suffer from disadvantages

that make them less suitable as PEMs. The former may be thermally and/or hydrolytically

unstable under fuel cell conditions due to the presence of aliphatic isopropylidene groups; while

the latter ones—although much more stable—show excessive water sorption and swelling that

may reduce their mechanical and dimensional stability. However, the BisSF-BPSH (Series 3)

copolymers are the most promising candidates as PEMs: they possess a wholly aromatic

backbone, which makes them as stable as Series 2 polymers, and they also show controllable

water sorption and swelling properties. The low cost of the 4,4’-dihydroxydiphenol (Bisphenol-S)

monomer also contributes to their economical advantages.

A major discovery was made in the synthesis of all the multiblock copolymers

studied. Specifically, the molar feed ratios of hydrophilic/hydrophobic telechelic moderate

molecular weight oligomers in the coupling reactions do not have to be close to 1:1 and can be as

low as 0.6:1. Another important outcome of this study was the development of an effective,

novel synthetic methodology that was systematically used in the synthesis of the BisSF-BPSH

(Series 3) copolymers. Because the segment repeat units (blocks) in these copolymers had

molecular weights of at least 5Kg/mol, only a low degree of polymerization (coupling) was

needed to achieve a high total molecular weight. Therefore, the hydrophobic block length,

hydrophilic block length, and IEC can be altered independently to some degree to tailor the

properties of copolymer membranes.

177

The electrochemical properties of the BisSF-BPSH copolymer membranes were found to be

affected by their composition (IEC) and sequence lengths. While proton conductivity and water

uptake are both favored by higher IEC and longer hydrophilic (sulfonated) block lengths,

increasing the hydrophobic (fluorinated) block length was found to adversely affect these

properties. The effects of all three parameters on the performance of the membranes were studied

independently. Through careful control of block length and IEC, the optimal trade-off between

performance (proton conductivity under partially hydrated conditions) and water uptake was

obtained with a BisSF-BPSH (17K-12K) (IEC=1.5) copolymer. Although BisSF-BPSH

copolymers generally demonstrated reduced performance compared to Series 1 and Series 2

copolymers at similar IEC and block lengths, these materials were the only ones in the series that

were able to outperform the commercial product, Nafion 112, while still maintaining low water

uptake (~40 wt%) well within the acceptable range.

The BisSF-BPSH multiblock copolymers also displayed interesting swelling-deswelling

behaviors. In contrast to sulfonated poly(ether sulfone) random copolymers, which swell

isotropically in water, these block copolymer membranes swelled much less in the x-y plane than

in the z direction (through-plane). This should significantly reduce the stress at the interface

between the membrane and the electrode, which can result from a significant amount of in-plane

swelling-deswelling during wet-dry cycles. The long-term durability of the MEAs, therefore,

may be greatly improved.

178

Although these sulfonated-fluorinated multiblock copolymers offer improved performance

compared to random copolymer membranes, their practical application may be limited because

the decafluorobiphenyl monomer needed to produce them is not yet commercially available.

However, the high reactivity of the fluorine endgroups is essential because the coupling reactions

must be conducted at low temperatures to avoid an ether-ether interchange. Research in our

group has recently focused on an alternative type of multiblock copolymer, where the

hydrophobic oligomers are mostly hydrocarbon-based and bear fluorinated moieties only at the

chain-ends. In principle, these oligomers have the same reactivity as the fully fluorinated ones,

enabling multiblock copolymers to be obtained through low-temperature coupling reactions as

well.

179

References 1. Whittingham, M. S.; Savinell, R. F.; Zawodzinski, T., Introduction: Batteries and Fuel

Cells. Chem. Rev. 2004, 104, 4243.

2. Winter, M.; Brodd, R. J., What Are Batteries, Fuel Cells, and Supercapacitors? Chem. Rev. 2004, 104, 4245.

3. Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; McGrath, J. E., Chem. Rev. 2004, 104, 4587.

4. Mauritz, K. A.; Moore, R. B., State of Understanding of Nafion. Chem. Rev. 2004, 104, 4535.

5. Yeo, R. S., J. Electrochem. Soc. 1983, 130, 533.

6. Kerres, J. A., Development of ionomer membranes for fuel cells. J. Membr. Sci. 2001, 185, 3.

7. Roziere, J.; Jones, D. J., Non-Fluorinated Polymer Materials For Proton Exchange Membrane Fuel Cells. Annu. Rev. Mater. Res. 2003, 33, 503.

8. Zawodzinski, T. A.; Neeman, M.; Sillerud, L. O.; Gottesfeld, S., Determination of water diffusion coefficients in perfluorosulfonate ionomeric membranes. J. Phys. Chem 1991, 95, 6040.

9. Alberti, G.; Casciola, M.; Palombari, R., Inorgano-organic proton conducting membranes for fuel cells and sensors at medium temperatures. J. Membr. Sci. 2000, 172, 233.

10. Elabd, Y. A.; Napadensky, E., Sulfonation and characterization of poly(styrene-isobutylene-styrene) triblock copolymers at high ion-exchange capacities. Polymer 2004, 45, 3037.

11. Weiss, R. A.; Sen, A.; Willis, C. L.; Pottick, L. A., Block Copolymer Ionomers. Part 1. Synthesis and Physical Properties of Sulfonated Poly(styrene-ethylene/butylene-styrene),. Polymer 1991, 32, 1867.

12. Elabd, Y. A.; Walker, C. W.; Beyer, F. L., Triblock copolymer ionomer membranes Part II. Structure characterization and its effects on transport properties and direct methanol fuel cell performance. J. Membr. Sci. 2004, 231, 181.

13. Harrison, W. L.; Wang, F.; Mecham, J. B.; Bhanu, V. A.; Hill, M.; Kum, Y. S., Influence of the bisphenol structure on the direct synthesis of sulfonated poly(arylene ether) copolymers. J. Polym. Sci. Part A: Polym. Chem. 2003, 41, 2264.

14. Wang, F.; Hickner, M.; Ji, Q.; Harrison, W.; Mecham, J.; Zawodzinski, T. A.; McGrath, J. E., Synthesis of highly sulfonated poly(arylene ether sulfone) random (statistical) copolymers via direct polymerization. Macromol. Symp. 2001, 175, 387.

180

15. Einsla, B. R.; Hong, Y.-T.; Kim, Y. S.; Wang, F.; Gunduz, N.; McGrath, J. E., Sulfonated naphthalene dianhydride based polyimide copolymers for proton-exchange-membrane fuel cells. I. Monomer and copolymer synthesis. J. Polym. Sci. Part A: Polym. Chem. 2004, 42, 862.

16. Genies, C.; Mercier, R.; Sillion, B.; Cornet, N.; Gebel, G.; Pineri, M., Soluble sulfonated naphthalenic polyimides as materials for proton exchange membranes. Polymer 2001, 42, 359.

17. Ghassemi, H.; McGrath, J. E., Synthesis and properties of new sulfonated poly(p-phenylene) derivatives for proton exchange membranes. I. Polymer 2004, 45, 5847.

18. Ghassemi, H.; Ndip, G.; McGrath, J. E., New multiblock copolymers of sulfonated poly(4′-phenyl-2,5-benzophenone) and poly(arylene ether sulfone) for proton exchange membranes. II Polymer 2004, 45, 5855.

19. Kim, Y. S.; Hickner, M. A.; Dong, L.; Pivovar, B. S.; McGrath, J. E., Sulfonated poly(arylene ether sulfone) copolymer proton exchange membranes: composition and morphology effects on the methanol permeability. J. Membr. Sci. 2004, 243, 317.

20. Noshay, A.; McGrath, J. E., Block Copolymers: Overview and Critical Survay. Academic Press: New York, 1977.

21. Petit, F.; Iliopoulos, I.; Audebert, R.; Szonyi, S., Associating Polyelectrolytes with Perfluoroalkyl Side Chains: Aggregation in Aqueous Solution, Association with Surfactants, and Comparison with Hydrogenated Analogues Langmiur 1997, 13, 4229.

22. Kennedy, J. P., Novel designed polyisobutylene-based biopolymers: synthesis, characterization, and biological testing of amphiphilic chameleon networks. J. Macromol. Sci., Pure. Appl. Chem. 1994, A31, 1771.

23. Forster, S.; Antonietti, M., Amphiphilic Block Copolymers in Structure-Controlled Nanomaterial Hybrids. Adv. Mater. 1998, 10, 195.

24. Edmondson, C. A.; Fontanella, J. J.; Chung, S. H.; Greenbaum, S. G.; Wnek, G. E., Complex impedance studies of S-SEBS block polymer proton-conducting membranes. Electrochimica Acta 2001, 46, 1623.

25. Elabd, Y. A.; Napadensky, E.; Sloan, J. M.; Crawford, D. M.; Walker, C. W., Triblock copolymer ionomer membranes Part I. Methanol and proton transport. J. Membr. Sci. 2003, 217, 227.

26. Roy, A.; Hickner, M. A.; Yu, X.; Li, Y.; Glass, T.; McGrath, J. E., Influence of Chemical Composition and Sequence Length on the Transport Properties of Proton Exchange Membranes. J. Polym. Sci. Part B: Polym. Phys. 2006, 44, 2226.

181

27. Wang, S.; McGrath, J. E., Synthesis of Poly(arylene ether)s.Synthetic Methods in Step-Growth Polymers. Wiley: New York.

28. Johnson, R. N., Encyclopedia of Polymer Science and Technology. Wiley: New York, 1969.

29. Attwood, T. E.; Barr, D. A.; Kin, T.; Newton, A. B.; Rose, J. B., Poly(arylene ether sulfones) by polyetherification. 2. Polycondensations. Polymer 1977, 18(4), 359.

30. Attwood, T. E.; Newton, A. B.; Rose, J. B., Kinetic investigation of the synthesis of a polyethersulfone. Br. Polym. J. 1972, 4, 391.

31. Johnson, R. N.; Farnham, A. G.; Clendinning, R. A.; Hale, W. F.; Merriam, C. L., Poly(aryl ethers) by nucleophilic aromatic substitution. I. Synthesis and properties. J. Polym. Sci. Part A: polym. Chem. 1967, 5, 2375.

32. Hedrick, J. L.; Dumais, J. J.; Jelinski, L. W.; Patsiga, R. A.; McGrath, J. E., Synthesis and characterization of deuterated poly(arylene ether sulfones). J. Polym. Sci. Part A: polym. Chem. 1987, 25, 2289.

33. Hedrick, J. L.; Mohanty, D. K.; Johnson, B. C.; Viswanathan, R.; Hinkley, J. A.; McGrath, J. E., Radiation resistant amorphous-all aromatic polyarylene ether sulfones: synthesis, characterization, and mechanical properties. J. Polym. Sci. Part A: polym. Chem. 1986, 23, 287.

34. Attwood, T. E.; Dawson, P. C.; Freeman, J. L.; Hoy, L. R.; Rose, J. B.; Staniland, P. A., Synthesis and properties of polyaryletherketones. Polymer 1981, 20, 1204.

35. Mohanty, D. K.; Sachdeva, Y.; Hedrick, J. L.; Wolfe, J. F.; McGrath, J. E., Synthesis and transformations of tetramethyl bis[phenol] A polyaryl ethers. Polym. Prepr. 1984, 25, 19.

36. Lindfors, B. E.; Mani, R. S.; McGrath, J. E.; Mohanty, D. K., Heterogeneous hydrolysis of poly(aryl ether ketimines) - an alternative synthetic route to semicrystalline poly(aryl ether ketones). Makromol. Chem., Rapid. Commun. 1991, 12, 337.

37. Mohanty, D. K.; Lowery, R. C.; Lyle, G. D.; McGrath, J. E., Ketimine modifications as a route to novel amorphous and derived semicrystalline poly(arylene ether ketone) homo- and copolymers. Int. SAMPE Symp. Exp. 1987, 32, 408.

38. Mohanty, D. K.; Lin, T. S.; Ward, T. C.; McGrath, J. E., Novel synthesis of thermoplastic semicrystalline poly(arylene ether ketone). Int. SAMPE Symp. Exp. 1986, 31, 945.

39. Risse, W.; Sogah, D. Y., Synthesis of soluble high molecular weight poly(aryl ether ketones) containing bulky substituents. Macromolecules 1990, 23, 4029.

40. Johnston, J. C.; MMeador, M. A. B.; Alston, W. B., A mechanistic study of polyimide formation from diester-diacids. J. Polym. Sci. Part A. Polym. Chem. 1987, 25, 2175.

41. de Abajo, J., Modern trends in the synthesis and application of heat-resistant polymers. Makromol. Chem. Macromol. Symp. 1988, 22, 141.

182

42. Sillion, B., Aromatic and heterocyclic polymers - what future? High Perform. Polym. 1999, 11, 417.

43. de Abajo, J.; de la Campa, J. G., Processable Aromatic Polyimides. Adv. Polym. Sci. 1999, 140, 23.

44. Ueda, M.; Sato, M.; Mochizuki, A., Poly(benzimidazole) synthesis by direct reaction of diacids and diamines. Macromolecules 1985, 18, 2723.

45. Chow, A. W.; Bitler, S. P.; Penwell, P. E.; Osborne, D. J.; Wolfe, J. F., Synthesis and solution properties of extended chain poly(2,6-benzothiazole) and poly(2,5-benzoxazole). Macromolecules 1989, 22, 3514.

46. Evers, R. C.; Arnold, F. E.; Helminiak, T. E., Articulated all-para polymers with 2,6-benzobisoxazole, 2,6-benzobisthiazole, and 2,6-benzobisimidazole units in the backbone. Macromolecules 1981, 14, 925.

47. Krause, S. J.; Haddock, T. B.; Vezie, D. L.; Lenhert, P. G.; Hwang, W.-F.; Price, G. E.; Helminiak, T. E.; O'Brien, J. F.; Adams, W. W., Morphology and properties of rigid-rod poly(p-phenylenebenzobisoxazole) (PBO) and stiff-chain poly(2,5(6)-benzoxazole) (ABPBO) fibers. Polymer 1988, 29, 1354.

48. So, Y.-H.; Heeschen, J. P.; Bell, B.; Bonk, P.; Briggs, M.; DeCaire, R., Study of the Mechanism for Poly(p-phenylene)benzoxazole Polymerization-A Remarkable Reaction Pathway To Make Rigid-Rod Polymers. Macromolecules 1998, 31, 5229.

49. Bouchet, R.; Siebert, E., Proton conduction in acid doped polybenzimidazole. Solid State Ionics 1999, 118, 287.

50. Fontanella, J. J.; Wintersgill, M. C.; Wainright, J. S.; Savinell, R. F.; Litt, M., High pressure electrical conductivity studies of acid doped polybenzimidazole. Electrochem. Acta 1998, 43, 1289.

51. Mecerreyes, D.; Grande, H.; Miguel, O.; Ochoteco, E.; Marcilla, R.; Cantero, I., Porous polybenzimidazole membranes doped with phosphoric acid: highly proton-conducting solid electrolytes. Chem. Mater. 2004, 16, 604.

52. Wainright, J. S.; Wang, J. T.; Weng, D.; Savinell, R. F.; Litt, M., Acid-doped polybenzimidazoles: a new polymer electrolyte. J. Electrochem. Soc. 1995, 142(7), L121.

53. Xing, B. Z.; Savadogo, O., The effect of acid doping on the conductivity of polybenzimidazole (PBI). J. New Mater. Electrochem. Syst. 1999, 2, 95.

54. Xiao, L.; Zhang, H.; Scanlon, E.; Ramanathan, L. S.; Choe, E.-W.; Rogers, D.; Apple, T.; Benicewicz, B. C., High-Temperature Polybenzimidazole Fuel Cell Membranes via a Sol-Gel Process. Chem. Mater. 2005, 17(21), 5328.

55. Percec, V.; Okito, S., Synthesis and nickel(0)-catalyzed oligomerization of isomeric 4,4'''-dichloroquaterphenyls. J. Polym. Sci. Part A: Polym. Chem. 1993, 31, 877.

183

56. Kovacic, P.; Jones, M. B., Dehydro coupling of aromatic nuclei by catalyst-oxidant systems: poly(p-phenylene). Chem. Rev. 1987, 87, 357.

57. Ueda, M.; Ichikawa, F., Synthesis of aromatic poly(ether ketone)s by nickel-catalyzed coupling polymerization of aromatic dichlorides. Macromolecules 1990, 23, 926.

58. Colon, I.; Kwiatkowski, G. T., High-molecular-weight aromatic polymers by nickel coupling of aryl polychlorides. J. Polym. Sci. Part A: Polym. Chem. 1990, 28, 367.

59. Pasquale, A. J.; Sheares, V., Alkyl-substituted poly(2,5-benzophenone)s synthesized via Ni(0)-catalyzed coupling of aromatic dichlorides and their miscible blends. J. Polym. Sci. Part A: Polym. Chem. 1998, 36, 2611.

60. Percec, V.; Zhao, M.; Bae, J. Y.; Hill, D. H., Regioregular and Regioirregular Poly(p-phenylene)s via Ni(0)-Catalyzed Homocoupling of Arylene Bismesylates. Macromolecules 1996, 29, 3727.

61. Wang, Y.; Quirk, R. P., Synthesis and Characterization of Poly(benzoyl-1,4-phenylene)s. 2. Catalyst Coligand Effects on Polymer Properties. Macromolecules 1995, 28, 3495.

62. Child, A. D.; Reynolds, J. R., Water-Soluble Rigid-Rod Polyelectrolytes: A New Self-Doped, Electroactive Sulfonatoalkoxy-Substituted Poly(p-phenylene). Macromolecules 1994, 27, 1975.

63. Wallow, T. I.; Novak, B. M., In aqua synthesis of water-soluble poly(p-phenylene) derivatives. J. Am. Chem. Soc. 1991, 113, 7411.

64. Bloom, P. D.; Sheares, V. V., Synthesis of poly(p-phenylene) macromonomers and multiblock copolymers. J. Polym. Sci. Part A: Polym. Chem. 2001, 39, 3505.

65. Johnson, B. C.; Yilgor, I.; Tran, C.; Iqbal, M.; Wightman, J. P.; Lloyd, D. R.; McGrath, J. E., J. Polymj. Sci. Part A: Polym. Chem. 1984, 22, 721.

66. Genova-Dimitrova, P.; Baradie, B.; Foscallo, D.; Poinsignon, C.; Sanchez, J. Y., Ionomeric membranes for proton exchange membrane fuel cell (PEMFC): sulfonated polysulfone associated with phosphatoantimonic acid. J. Membr. Sci. 2001, 185, 59.

67. Bishop, M. T.; Karasz, F. E.; Russo, P. S.; Langley, K. H., Solubility and properties of a poly(aryl ether ketone) in strong acids. Macromolecules 1985, 18, 86.

68. Huang, R. Y. M.; Shao, P.; Burns, C. M.; Feng, X., Sulfonation of poly(ether ether ketone)(PEEK): Kinetic study and characterization. J. Appl. Polym. Sci. 2001, 82, 2651.

69. Bailly, C.; Williams, D. J.; Karasz, F. E.; MacKnight, W. J., The sodium salts of sulphonated poly(aryl-ether-ether-ketone) (PEEK): Preparation and characterization. Polymer 1987, 28, 1009.

70. Ueda, M.; Toyota, H.; Ochi, T.; Sugiyama, J.; Yonetake, K.; Masuko, T.; Teramoto, T., Synthesis and characterization of aromatic poly(ether sulfone)s containing pendant sodium sulfonate groups. J. Polym. Sci. Part A: Polym. Chem. 1993, 31, 853.

184

71. Sankir, M.; Bhanu, V. A.; Harrison, W. L.; Ghassemi, H.; Wiles, K. B.; Glass, T. E.; Brink, A. E.; Brink, M. H.; McGrath, J. E., Synthesis and characterization of 3,3-disulfonated-4,4-dichlorodiphenyl sulfone (SDCDPS) monomer for proton exchange membranes (PEM) in fuel cell applications. J. Appl. Polym. Sci. 2006, 100(6), 4595.

72. Wang, F.; Hickner, M.; Kim, Y. S.; Zawodzinski, T. A.; McGrath, J. E., Direct polymerization of sulfonated poly(arylene ether sulfone) random (statistical) copolymers: candidates for new proton exchange membranes. J. Membr. Sci. 2002, 197, 231.

73. Kim, Y. S.; Dong, L.; Hickner, M. A.; Pivovar, B. S.; McGrath, J. E., Processing induced morphological development in hydrated sulfonated poly(arylene ether sulfone) copolymer membranes. Polymer 2003, 44, 5729.

74. Kim, Y. S.; Wang, F.; Hickner, M.; Mccartney, S.; Hong, Y. T.; Harrison, W.; Zawodzinski, T. A.; Mcgrath, J. E., Effect of acidification treatment and morphological stability of sulfonated poly(arylene ether sulfone) copolymer proton-exchange membranes for fuel-cell use above 100 °C. J. Polym. Sci. Part B. Polym. Phys. 2003, 41, 2816.

75. Wang, F.; Chen, T.; Xu, J., Sodium sulfonate-functionalized poly(ether ether ketone)s. Macromol. Chem. Phys. 1998, 199, 1421.

76. Wang, F.; Li, J.; Chen, T.; Xu, J., Synthesis of poly(ether ether ketone) with high content of sodium sulfonate groups and its membrane characteristics. Polymer 1999, 40, 795.

77. Wang, F.; Chen, T.; Xu, J.; Liu, T.; Jiang, H.; Qi, Y.; Liu, S.; Li, X., Synthesis and characterization of poly(arylene ether ketone) (co)polymers containing sulfonate groups. Polymer 2006, 47, 4148.

78. Xing, P.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.; Kaliaguine, S., Sulfonated poly(aryl ether ketone)s containing naphthalene moieties obtained by direct copolymerization as novel polymers for proton exchange membranes. J. Polym. Sci. Part A: Polym. Chem. 2004, 42, 2866.

79. Liu, B.; Robertson, G. P.; Guiver, M. D.; Sun, Y.-M.; Liu, Y.-L.; Lai, J.-Y.; Mikhailenko, S.; Kaliaguine, S., Sulfonated poly(aryl ether ether ketone ketone)s containing fluorinated moieties as proton exchange membrane materials. J. Polym. Sci. Part B: Polym. Phys. 2006, 44, 2299.

80. Xing, P.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.; Kaliaguine, S., Synthesis and characterization of poly(aryl ether ketone) copolymers containing (hexafluoroisopropylidene)-diphenol moiety as proton exchange membrane materials. Polymer 2005, 46, 3257.

81. Savadogo, O., J. New Mater. Electrochem. Syst. 1998, 1, 47.

82. Einsla, B. R.; Kim, Y. S.; Hickner, M. A.; Hong, Y.-T.; Hill, M. L.; Pivovar, B. S.; McGrath, J. E., Sulfonated naphthalene dianhydride based polyimide copolymers for

185

proton-exchange-membrane fuel cells: II. Membrane properties and fuel cell performance. J. Membr. Sci. 2005, 255, 141.

83. Genies, C.; Mercier, R.; Sillion, B.; Petiaud, R.; Cornet, N.; Gebel, G.; Pineri, M., Stability study of sulfonated phthalic and naphthalenic polyimide structures in aqueous medium. Polymer 2001, 42, 5097.

84. Fang, J.; Guo, X.; Harada, S.; Watari, T.; Tanaka, K.; Kita, H.; Okamoto, K.-I., Macromolecules 2002, 35, 9022.

85. Vallejo, E.; Pourcelly, G.; Gavach, C.; Mercier, R.; Pineri, M., Sulfonated polyimides as proton conductor exchange membranes. Physicochemical properties and separation H+/Mz+ by electrodialysis comparison with a perfluorosulfonic membrane. J. Membr. Sci. 1999, 160, 127.

86. Guo, X.; Fang, J.; Watari, T.; Tanaka, K.; Kita, H.; Okamoto, K.-I., Novel Sulfonated Polyimides as Polyelectrolytes for Fuel Cell Application. 2. Synthesis and Proton Conductivity of Polyimides from 9,9-Bis(4-aminophenyl)fluorene-2,7-disulfonic Acid Macromolecules 2002, 35, 6707.

87. Asensio, J. A.; Borros, S.; Gomez-Romero, P., Proton-conducting polymers based on benzimidazoles and sulfonated benzimidazoles. J. Polym. Sci. Part A: Polym. Chem. 2002, 40, 3703.

88. Dang, T. D.; Bai, S. J.; Heberer, D. P.; Arnold, F. E.; Spry, R. J., Ionic conductivity of conjugated water-soluble rigid-rod polymers. J. Polym. Sci. Part B: Polym. Phys. 1993, 31, 1941.

89. Einsla, B. R.; Kim, Y. J.; Tchatchoua, C.; McGrath, J. E., Polym. Prepr. 2003, 44, 645.

90. Kim, S.; Cameron, D. A.; Lee, Y.; Reynolds, J. R.; Savage, C. R., Aromatic and rigid rod polyelectrolytes based on sulfonated poly(benzobisthiazoles). J. Polym. Sci. Part A: Polym. Chem. 1996, 34, 481.

91. Qing, S.; Huang, W.; Yan, D., Synthesis and characterization of thermally stable sulfonated polybenzimidazoles. Europ. Polym. J. 2005, 41, 1589.

92. Qing, S.; Huang, W.; Yan, D., Synthesis and properties of soluble sulfonated polybenzimidazoles. Reactive & Functional Polymers 2006, 66, 219.

93. Asensio, J. A.; Borrós, S.; Gómez-Romero, P., Proton-conducting membranes based on poly(2,5-benzimidazole) (ABPBI) and phosphoric acid prepared by direct acid casting. J. Membr. Sci. 2004, 241, 89.

94. Asensio, J. A.; Borrós, S.; Gómez-Romero, P., Sulfonated poly(2,5-benzimidazole) (SABPBI) impregnated with phosphoric acid as proton conducting membranes for polymer electrolyte fuel cells. Electrochim. Acta 2004, 49, 4461.

186

95. Staiti, P.; Lufrano, F.; Aric, A. S.; Passalacqua, E.; Antonucci, V., J. Membr. Sci. 2001, 188, 71.

96. Kobayashi, T.; Rikukawa, M.; Sanui, K.; Ogata, N., Proton-conducting polymers derived from poly(ether-etherketone) and poly(4-phenoxybenzoyl-1,4-phenylene). Solid State Ionics 1998, 106, 219.

97. Bae, J.-M.; Honma, I.; Murata, M.; Yamamoto, T.; Rikukawa, M.; Ogata, N., Properties of selected sulfonated polymers as proton-conducting electrolytes for polymer electrolyte fuel cells. Solid State Ionics 2002, 147, 189.

98. Fu, H.; Yun, H.; Kwei, T. K.; Okamoto, Y., Anomalous viscosity behavior of rodlike polyelectrolytes: Partially sulfonated poly(2-benzoyl-1,4-phenylene)s in aqueous solution. j. Polym. Sci. Part A. Polym. Chem. 1998, 36, 1425.

99. Wang, H.; McGrath, J. E., ACS. Div. Fuel Chem. 2005, 50(2), 581.

100. Sankir, M.; Kim, Y. S.; Harrison, W. L.; S.Badami, A.; McGrath, J. E., Am. Chem. Soc., Div. Fuel. Chem. 2005, 50(2), 568.

101. Sumner, M. J.; Harrison, W. L.; Weyers, R. M.; Kim, Y. S.; McGrath, J. E.; Riffle, J. S.; Brink, A.; Brink, M. H., Novel proton conducting sulfonated poly(arylene ether) copolymers containing aromatic nitriles. J. Membr. Sci. 2004, 239, 199.

102. Miyatake, K.; Hay, A. S., Synthesis and properties of poly(arylene ether)s bearing sulfonic acid groups on pendant phenyl rings. J. Polym. Sci. Part A: Polym. Chem. 2001, 39, 3211.

103. Lee, H. C.; Hong, H. S.; Kim, Y.-M.; Choi, S. H.; Hong, M. Z.; Lee, H. S.; Kim, K., Electrochimica Acta 2004, 49, 2315.

104. Kim, D. S.; Robertson, G. P.; Guiver, M. D.; Lee, Y. M., Synthesis of highly fluorinated poly(arylene ether)s copolymers for proton exchange membrane materials. J. Membr. Sci. 2006, 281, 111.

105. Krespan, C. G.; Middleton, W. J., Fluorine Chem. Rev. 1967, 1, 145.

106. Lee, H.-J.; Lee, E.-M.; Lee, M.-H.; Oh, M.-C.; Ahn, J.-H.; Han, S. G.; Kim, H. G., Crosslinkable fluorinated poly(arylene ethers) bearing phenyl ethynyl moiety for low-loss polymer optical waveguide devices. J. Polym. Sci. Part A: Polym. Chem. 1998, 36, 2881.

107. Kim, J.-P.; Lee, W.-Y.; Kang, J.-W.; Kwon, S.-K.; Kim, J.-J.; Lee, J.-S., Fluorinated Poly(arylene ether sulfide) for Polymeric Optical Waveguide Devices Macromolecules 2001, 34, 7817.

108. Lee, H.-J.; Lee, M.-H.; Oh, M.-C.; Ahn, J.-H.; Han, S. G., Crosslinkable polymers for optical waveguide devices. II. Fluorinated ether ketone oligomers bearing ethynyl group at the chain end. J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 2355.

187

109. Ding, J.; Liu, F.; Li, M.; Day, M.; Zhou, M., Fluorinated poly(arylene ether ketone)s bearing pentafluorostyrene moieties prepared by a modified polycondensation. j. Polym. Sci. Part A: Polym. Chem. 2002, 40, 4205.

110. Kim, J.-P.; Kang, J.-W.; Kim, J.-J.; Lee, J.-S., Synthesis and optical properties of fluorinated poly(arylene ether phosphine oxide)s. J. Polym. Sci. Part A: Polym. Chem. 2003, 41, 1497.

111. Knunyants, I. L.; Yakobson, G. G., Syntheses of fluoroorganic compounds. Springer-Verlag: Berlin New York, 1985.

112. Hiyama, T., Organofluorine compounds : chemistry and applications. Springer: Berlin New York, 2000.

113. Chambers, R. D., Fluorine in Organic Chemistry. Wiley-Interscience: New York London, 2004.

114. Lee, H. C.; Hong, H. S.; Kim, Y.-M.; Choi, S. H.; Hong, M. Z.; Lee, H. S.; Kim, K., Preparation and evaluation of sulfonated-fluorinated poly(arylene ether)s membranes for a proton exchange membrane fuel cell (PEMFC). Electrochimica Acta 2004, 49, 2315.

115. Goodwin, A. A.; Mercer, F. W.; McKenzie, M. T., Thermal Behavior of Fluorinated Aromatic Polyethers and Poly(ether ketone)s Macromolecules 1997, 30, 2767.

116. Kimura, K.; Tabuchi, Y.; Yamashita, Y.; Cassidy, P. E.; Fitch, J. W.; Okumura, Y., Synthesis of novel fluorine-containing poly(aryl ether ketone)s. Polym. Adv. Technol. 2000, 11, 757.

117. Liu, F.; Ding, J.; Li, M.; Day, M.; Robertson, G.; Zhou, M., Preparation of Highly Fluorinated Poly(ether sulfone)s under Mild Polycondensation Conditions Using Molecular Sieves. Macromol. Rapid. Commun. 2002, 23, 844.

118. Mercer, F. W.; Fone, M. M.; Reddy, V. N., Synthesis and characterization of fluorinated polyetherketones prepared from decafluorobenzophenone. Polymer 1997, 38, 1989.

119. Serpico, J. M.; Ehrenberg, S. G.; Fontanella, J. J.; Jiao, X.; Perahia, D.; McGrady, K. A.; Sanders, E. H.; Kellogg, G. E.; Wnek, G. E., Transport and Structural Studies of Sulfonated Styrene-Ethylene Copolymer Membranes. Macromolecules 2002, 35, 5916.

120. Kim, B.; Kim, J.; Jung, B., Morphology and transport properties of protons and methanol through partially sulfonated block copolymers. J. Membr. Sci. 2005, 250, 175.

121. Kim, J.; Kim, B.; Jung, B., Proton conductivities and methanol permeabilities of membranes made from partially sulfonated polystyrene-block-poly(ethylene-ran-butylene)-block-polystyrene copolymers. J. Membr. Sci. 2002, 207, 129.

122. Yang, J.-E.; Lee, J.-S., Selective modification of block copolymers as proton exchange membranes. Electrochimica Acta 2004, 50, 617.

188

123. Mokrini, A.; Acosta, J. L., Studies of sulfonated block copolymer and its blends. Polymer 2001, 42, 9.

124. Sangeetha, D., Conductivity and solvent uptake of proton exchange membrane based on polystyrene(ethylene–butylene) polystyrene triblock polymer. European Polym. J. 2005, 41, 2644.

125. Won, J.; Choi, S. W.; Kang, Y. S.; Ha, H. Y.; Oh, I.-H.; Kim, H. S.; Kim, K. T.; Jo, W. H., Structural characterization and surface modification of sulfonated polystyrene–(ethylene–butylene)–styrene triblock proton exchange membranes. J. Membr. Sci. 2003, 214, 245.

126. Storey, R. F.; Chisholm, B. J., Morphology and physical properties of poly(styrene-b-isobutylene-b-styrene) block copolymers. Polymer 1996, 37, 2925.

127. Storey, R. F.; Baugh, D. W.; Choate, K. R., Poly(styrene-b-isobutylene-b-styrene) block copolymers produced by living cationic polymerization I. Compositional analysis. Polymer 1999, 40, 3083.

128. Shi, Z.; Holdcroft, S., Synthesis and Proton Conductivity of Partially Sulfonated Poly([vinylidene difluoride-co-hexafluoropropylene]-b-styrene) Block Copolymers. Macromolecules 2005, 38, 4193.

129. Shi, Z.; Holdcroft, S., Synthesis of Block Copolymers Possessing Fluoropolymer and Non-Fluoropolymer Segments by Radical Polymerization. Macromolecules 2004, 37, 2084.

130. Zhang, X.; Liu, S.; Yin, J., Synthesis and characterization of a new block copolymer for proton exchange membrane. J. Membr. Sci 2005, 258, 78.

131. Zhang, X.; Liua, S.; Liu, L.; Yin, J., Partially sulfonated poly(arylene ether sulfone)-b-polybutadiene for proton exchange membrane. Polymer 2005, 46, 1719.

132. Yang, Y.; Shi, Z.; Holdcroft, S., Synthesis of poly[arylene ether sulfone-b-vinylidene fluoride] block copolymers. European Poly. J. 2004, 40, 531.

133. Yang, Y.; Shi, Z.; Holdcroft, S., Synthesis of Sulfonated Polysulfone-block-PVDF Copolymers: Enhancement of Proton Conductivity in Low Ion Exchange Capacity Membranes. Macromolecules 2004, 37, 1678.

134. Zhang, X.; Liu, S.; Yin, J., Modified sulfonated poly(arylene ether sulfone)-b-polybutadiene (SPAES-b-PB) membrane for fuel cell applications. J. Membr. Sci 2006, 275, 119.

135. Taeger, A.; Vogel, C.; Lehmann, D.; Lenk, W.; Meier-Haack, J., Sulfonated Multiblock Copoly(ether sulfone)s as Membrane Materials for Fuel Cell Applications. Macromol. Symp. 2004, 210, 175.

189

136. Shin, C. K.; Maier, G.; Andreaus, B.; Scherer, G. G., Block copolymer ionomers for ion conductive membranes. J. Membr. Sci 2004, 245, 147.

137. Ghassemi, H.; McGrath, J. E.; Zawodzinski, T. A., Multiblock sulfonated-fluorinated poly(arylene ether)s for a proton exchange membrane fuel cell. Polymer 2006, 47, 4132.

138. Wang, H.; Badami, A. S.; Roy, A.; McGrath, J. E., Multiblock copolymers of poly(2,5-benzophenone) and disulfonated poly(arylene ether sulfone) for proton-exchange membranes. I. Synthesis and characterization. J. Polym. Sci. Part A. Polym. Chem. 2007, 45, 284.

139. Yu, X.; Roy, A.; McGrath, J. E., ACS. Div. Fuel Chem. 2005, 50(2), 577.

140. Yu, X.; Roy, A.; Dunn, S.; Yang, J.; McGrath, J. E., Synthesis and characterization of sulfonated-fluorinated, hydrophilic-hydrophobic multiblock copolymers for proton exchange membranes. Macromol. Symp. 2006, 245-246, 439.

141. Li, Y.; Roy, A.; Badami, A. S.; Hill, M.; Yang, J.; Dunn, S.; McGrath, J. E., Synthesis and characterization of partially fluorinated hydrophobic-hydrophilic multiblock copolymers containing sulfonate groups for proton exchange membrane. J. Power Sources 2007, 172, 308.

142. Lee, H.-S.; Badami, A. S.; Roy, A.; McGrath, J. E., Segmented Sulfonated Poly(arylene ether sulfone)-b-Polyimide Copolymers for Proton Exchange Membrane Fuel Cells. I. Copolymer Synthesis and Fundamental Properties. J. Polym. Sci. Part A. Polym. Chem. 2007, 45, (4879).

143. Lee, H.-S.; Roy, A.; Badami, A. S.; McGrath, J. E., Synthesis and characterization of sulfonated poly(arylene ether) polyimide multiblock copolymers for proton exchange membranes. Macromolecular Research 2007, 15, 160.

144. Lee, H.-S.; Einsla, B.; McGrath, J. E., ACS. Div. Fuel Chem. 2005, 50(2), 579.

145. Lee, H.-S.; Roy, A.; Badami, A. S.; McGrath, J. E., ACS. PMSE Preprints 2006, 95, 210.

146. Ghassemi, H.; Zawodzinski, T. A., ACS. Div. Fuel Chem. 2005, 50(2), 531.

147. Roy, A.; Hickner, M. A.; Lee, H.-S.; Badami, A.; Yu, X.; Li, Y.; Glass, T.; McGrath, J. E., Transport properties of proton exchange membranes. ECS Trans. 2007, 2(24, Direct Methanol Fuel Cells), 45.

190