Sulfonated polybenzothiazoles: A novel Candidate for proton exchange membranes

Article abstract of DOI:10.1021/cm9019217

Sulfonated polybenzothiazoles (sPBT) with high molecular weight as well as excellent solubility were synthesized for the first time, which was achieved by attaching the phenylsulfonyl pendant groups or incorporating the hexafiuoroisopropylidene moieties t

Full text of DOI:10.1021/cm9019217

1
022 Chem. Mater. 2010, 22, 1022–1031
DOI:10.1021/cm9019217
Sulfonated Polybenzothiazoles: A Novel Candidate for Proton Exchange
Membranes
†
Ning Tan, Guyu Xiao,* and Deyue Yan*
College of Chemistry and Chemical Engineering, The State Key Laboratory of Metal Matrix Composites,
Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, P. R. China
Received July 1, 2009. Revised Manuscript Received November 15, 2009
Sulfonated polybenzothiazoles (sPBT) with high molecular weight as well as excellent solubility
were synthesized for the first time, which was achieved by attaching the phenylsulfonyl pendant
groups or incorporating the hexafluoroisopropylidene moieties to the polymer backbone. Such
sulfonated polybenzothiazoles thus could be cast into homogeneous membranes and could be further
evaluated as proton exchange membranes. These sPBTs showed high thermal stability and high
proton conductivity as well as low swelling. For instance, the hexafluoroisopropylidene-containing
sPBT with a disulfonation degree of 65% exhibited a Td5 of 380 °C, a proton conductivity of 0.11 S/cm,
and a swelling of 15.5% at 80 °C. In addition, the sPBT membranes showed excellent oxidative
and hydrolytic stability. In comparison with the phenylsulfonyl pendant-group-containing sPBT
membranes, the hexafluoroisopropylidene-containing sPBT membranes with an equivalent ion
exchange capacity showed much narrower ionic channels because of the hydrophobic hexafluor-
oisopropylidene moieties. This made the latter display better dimensional stability, oxidative
stability, and hydrolytic stability than the former. This investigation illustrated that sulfonated
polybenzothiazoles are a novel candidate for proton exchange membranes.
Introduction
and anticipated stability in the fuel cell environments.
For example, poly(arylene ether)s,
9-11
poly(arylene
16-18
As a green power device, proton exchange membrane
12-15
thioether)s,
19-22
poly(phthalazinone ether)s,
23,24
poly-
and their deriva-
fuel cell (PEMFC) has attracted increasing attention in
recent years. A proton exchange membrane (PEM) is one
of the key components of PEMFC, and perfluorinated
membranes such as Nafion are the reference membrane
for PEMFC. However, the drawbacks such as low oper-
ating temperature, high methanol permeability, and high
cost seriously limited their application in fuel cells.
Therefore, intensive efforts have been made to develop
aromatic proton exchange membranes and composite
imides,
polyphosphazenes,
1
tives are the focus of recent investigations on PEM.
More and more aromatic polymers were considered to
be used as the bulk materials for PEM.
Aromatic polybenzazoles such as polybenzimidazoles
PBI), polybenzoxazoles (PBO), and polybenzothiazoles
(
(
1
25-27
PBT) are also high performance polymers,
thus they
(
12) Bai, Z. W.; Dang, T. D. Macromol. Rapid Commun. 2006, 27, 1271.
1
-8
membranes.
Wholly aromatic polymers are regarded
as one of the promising matrix for PEM because of their
availability, wide variety of chemical compositions,
(13) Lee, J. K.; Kerres, J. J. Membr. Sci. 2007, 294, 75.
(
14) Ma, X. H.; Shen, L. P.; Zhang, C. J.; Xiao, G. Y.; Yan, D. Y.; Sun,
G. M. J. Membr. Sci. 2008, 310, 303.
(15) Matsumura, S.; Hlil, A. R.; Lepiller, C.; Gaudet, J.; Guay, D.; Hay,
A. S. Macromolecules 2008, 41, 277.
16) Xiao, G. Y.; Sun, G. M.; Yan, D. Y.; Zhu, P. F.; Tao, P. Polymer
2002, 43, 5335.
17) Zhu, X. L.; Liang, Y. F.; Pan, H. Y.; Jian, X. G.; Zhang, Y. X.
J. Membr. Sci. 2008, 312, 59.
(
(
†
Accepted as part of the 2010 “Materials Chemistry of Energy Conversion
Special Issue”.
*
Corresponding author. E-mail: gyxiao@sjtu.edu.cn (G.X.) ; dyyan@
sjtu.edu.cn (D.Y.). Tel.: þ86-21-54742664. Fax: þ86-21-54741297.
(18) Gao, Y.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.; Li,
X.; Kaliaguine, S. Polymer 2006, 47, 808.
(19) Asano, N.; Miyatake, K.; Watanabe, M. Chem. Mater. 2004, 16,
2841.
(
1) Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; McGrath,
J. E. Chem. Rev. 2004, 104, 4587.
(
2) Li, Q. F.; He, R. H.; Jensen, J. O.; Bjerrum, N. J. Chem. Mater.
2
003, 15, 4896.
(20) Marestin, C.; Gebel, G.; Diat, O.; Mercier, R. Adv. Polym. Sci.
2008, 216, 185.
(21) Guo, X. X.; Fang, J. H.; Watari, T.; Tanaka, K.; Kita, H.;
Okamoto, K. I. Macromolecules 2002, 35, 6707.
(22) Li, N. W.; Cui, Z. M.; Zhang, S. B.; Li, S. H. J. Polym. Sci., Part A:
Polym. Chem. 2008, 46, 2820.
(23) Wycisk, R.; Pintauro, P. N. Adv. Polym. Sci. 2008, 216, 157.
(24) Allcock, H. R.; Wood, R. M. J. Polym. Sci., Part B: Polym. Phys.
2006, 44, 2358.
(25) Yang, H. H. Aromatic High-Strength Fibers; Wiley: New York,
1989.
(
(
(
(
(
(
(
3) Kreuer, K. D. J. Membr. Sci. 2001, 185, 29.
4) Savadogo, O. J. Power Sources 2004, 127, 135.
5) Kerres, J. A. J. Membr. Sci. 2001, 185, 3.
6) Jones, D. J.; Rozi ꢀe re, J. Adv. Polym. Sci. 2008, 215, 219.
7) Alberti, G.; Casciola, M. Annu. Rev. Mater. Res. 2003, 33, 129.
8) Jones, D. J.; Rozi ꢀe re, J. J. Membr. Sci. 2001, 185, 41.
9) Kim, Y. S.; Wang, F.; Hickner, M.; Mccartney, S.; Hong, Y. T.;
Harrison, W.; Zawodzinski, T. A.; McGrath, J. E. J. Polym. Sci.,
Part B: Polym. Phys. 2003, 41, 2816.
(
10) Wang, Z.; Ni, H. Z.; Zhao, C. J.; Li, X. F.; Zhang, G.; Shao, K.; Na,
H. J. Membr. Sci. 2006, 285, 239.
11) Lee, C. H.; Min, K. A.; Park, H. B.; Hong, Y. T.; Jung, B. O.; Lee,
Y. M. J. Membr. Sci. 2007, 303, 258.
(26) Hu, X. D.; Jenkins, S. E.; Min, B. G.; Polk, M. B.; Kumar, S.
Macromol. Mater. Eng. 2003, 288, 823.
(27) So, Y. H. Prog. Polym. Sci. 2000, 25, 137.
(
pubs.acs.org/cm
Published on Web 12/10/2009
r 2009 American Chemical Society

Article
Chem. Mater., Vol. 22, No. 3, 2010 1023
2
6,49
are of great interest to be used as the matrix of PEM. PBI-
based proton exchange membranes have been intensively
intermolecular interaction like polybenzothiazoles,
and thus sPBT could not form homogeneous membrane
by extrusion or by blow molding. Accordingly, the sPBT
membranes for PEM applications might be prepared only
by casting polymer solution. This demands that sulfo-
nated polybenzothiazoles exhibit good solubility as well
as high molecular weight in order to obtain good proces-
sability and mechanical properties. As a result, it is
significant to synthesize such sulfonated polybenzothia-
zoles and to evaluate whether they could be employed as
PEM materials or not.
2
5,28-36
developed.
have also been investigated for PEM applications.
Polybenzoxazole based membranes
7-39
3
However, few polybenzothiazoles have been studied
as proton exchange membrane materials because of their
4
0-43
poor processability.
Great attempts have been put
forth to enhance their solubility by incorporating
the bulky pendant groups or flexible linkage into the
resulting polybenzothiazoles, but the result was not
4
4-47
satisfactory.
On the other hand, though polybenzothiazoles have
high tensile strength, tensile modulus, and thermal stabi-
lity, they showed poor compressive properties in addition
In this article, sulfonated polybenzothiazoles were prepared
by polycondensation of 2,5-diamino-1,4-benzenedithiol
dihydrochloride and bis(3-sulfonate-4-carboxyphenyl)
sulfone with bis(4-carboxyphenyl) sulfone or diphenyl
sulphone-2,5-dicarboxylic acid or 2,2-bis(4-carboxyphe-
nyl)hexafluoropropane. The effects of the structural moi-
eties such as diphenyl sulfone, hexafluoroisopropylidene,
and phenylsulfonyl pendant groups on the properties of
the resulting polymers were investigated. The incorpora-
tion of the phenylsulfonyl pendant groups or the flexible
hexafluoroisopropylidene moieties to the resulting sulfo-
nated polybenzothiazoles was expected to enhance the
solubility and to achieve high molecular weight products.
In this case, such sulfonated polybenzothiazoles could be
cast into homogeneous membranes and could be further
investigated as PEM. Their structure and properties
as PEM, especially their solubility, swelling, hydrolytic
stability, oxidative stability, and proton conductivity,
were studied in detail so as to evaluate whether they are
a promising PEM material or not.
2
6,27
to poor processability.
These disadvantages heavily
limited their applications. To overcome these drawbacks,
polybenzothiazoles with sulfonic acid groups were pre-
4
8-50
pared by direct polycondensation.
sized polymers were used as structural materials and were
The as-synthe-
4
9,50
not evaluated as proton exchange membranes.
Furthermore, few studies relevant to sulfonated polyben-
zothiazoles were reported. It could also be found that the
characterization of their structure and properties is very
limited and that sPBT with high molecular weight as well
as good solubility have not been reported until now. In
addition, sulfonated polybenzothiazoles exhibit no soft-
ening behavior and glass transition prior to thermal
degradation because of their rigid structure and strong
(
(
(
(
(
(
28) Wainright, J. S.; Wang, J. T.; Weng, D.; Savinell, R. F.; Litt, M.
J. Electrochem. Soc. 1995, 142, L121.
29) Qing, S. B.; Huang, W.; Yan, D. Y. React. Funct. Polym. 2006, 66,
219.
30) Kawahara, M.; Morita, J.; Rikukawa, M.; Sanui, K.; Ogata, N.
Electrochim. Acta 2000, 45, 1395.
Experimental Section
31) Mader, J.; Xiao, L. X.; Schmidt, T. J.; Benicewicz, B. C. Adv.
Polym. Sci. 2008, 216, 63.
32) Peron, J.; Ruiz, E.; Jones, D. J.; Rozi ꢀe re, J. J. Membr. Sci. 2008,
Materials. 2,5-Diamino-1,4-benzenedithiol dihydrochloride
DABDT) was purchased from TCI and used as received. Bis(4-
(
carboxyphenyl) sulfone (BCPS) and 2,2-bis(4-carboxyphe-
nyl)hexafluoropropane (6FA) were obtained from ABCR and
Sigma-Aldrich, respectively. Phosphorus pentoxide (corrosive),
polyphosphoric acid (PPA, 81%), methane sulfonic acid (MSA,
corrosive), benzene sulphonyl chloride (corrosive), p-xylene,
aluminum chloride, triethylamine, and other chemicals were
purchased from Shanghai Chemical Reagents Co. and used
without further purification. Bis(3-sulfonate-4-carboxyphenyl)
sulfone (BSCS) was synthesized according to our previous
314, 247.
33) Staiti, P.; Lufrano, F.; Arico, A. S.; Passalacqua, E.; Antonucci, V.
J. Membr. Sci. 2001, 188, 71.
(
(
34) Jannasch, P. Fuel Cells 2005, 5, 248.
35) Asensio, J. A.; Borros, S.; G oꢁ mez-Romero, P. J. Polym. Sci., Part
A: Polym. Chem. 2002, 40, 3703.
(
(
(
36) Li, Q. F.; Pan, C.; Jensen, J. O.; Noy ꢁe , P.; Bjerrum, N. J. Chem.
Mater. 2007, 19, 350.
37) Sakaguchi, Y.; Kitamura, K.; Nakao, J.; Hamamoto, S.; Tachimori,
H.; Takase, S. Polym. Mater. Sci. Eng. 2001, 84, 899.
38) Einsla, B. R.; Kim, Y. J.; Tchatchoua, C.; McGrath, J. E. Abstr.
Pap. Am. Chem. Soc. 2003, 226, U391.
39) Tan, N.; Xiao, G. Y.; Yan, D. Y. Polym. Bull. 2009, 62, 593.
40) Hwang, W. F.; Wiff, D. R.; Verschoore, C.; Price, G. E.; Helminiak,
T. E.; Adams, W. W. Polym. Eng. Sci. 1983, 23, 784.
41) Hattori, T.; Akita, H.; Kakimoto, M.; Imai, Y. Macromolecules
5
1
report.
(
(
Synthesis of Diphenyl Sulfone-2,5-dicarboxylic Acid. 2,5-Di-
methyl diphenyl sulfone (DMPS) was synthesized by an alumi-
num chloride catalyzed Friedel-Crafts acylation reaction of
(
1992, 25, 3351.
(
(
42) Baek, J. B.; Tan, L. S. Macromolecules 2008, 41, 1196.
43) Park, S. -Y.; Koerner, H.; Putthanarat, S.; Ozisik, R.; Juhl, S.;
Farmer, B. L.; Eby, R. K. Polymer 2004, 45, 49.
benzene sulfonyl chloride with p-xylene following previous
52-54
reports.
Diphenyl sulfone-2,5-dicarboxylic acid (PSCA)
was prepared by oxidation of DMPS with potassium perman-
ganate. The preparation procedure is as follows. Ten grams of
DMPS, 4 g of sodium hydroxide, and 650 mL of deionized water
were placed in a three-necked round-bottom flask. Potassium
permanganate (36 g) was added in small portions over a period
of 1 h. After refluxing for 24 h, the hot mixture was cooled and
(44) Hutzler, R. F.; Meurer, D. L.; Kimura, K.; Cassidy, P. E. High
Perform. Polym. 1992, 4, 161.
45) Kricheldorf, H.; Hengelhardt, J. Makromol. Chem. 1990, 191, 2017.
46) Evers, R. C.; Arnold, F. E.; Helminiak, T. H. Macromolecules
(
(
1981, 14, 925.
(
(
(
(
47) Saegusa, Y.; Horikiri, M.; Sakai, D.; Nakamura, S. J. Polym. Sci.,
Part A: Polym. Chem. 1998, 36, 429.
48) Reynolds, J. R.; Lee, Y.; Kim, S.; Bartling, R. L.; Gieselman, M. B.;
Savage, C. S. Polym. Prep. 1993, 34, 1065.
49) Kim, S.; Cameron, D. A.; Lee, Y.; Reynolds, J. R.; Savage, C. R.
J. Polym. Sci., Part A: Polym. Chem. 1996, 34, 481.
50) Dang, T. D.; Koerner, H.; Dalton, M. J.; Iacobucci, A. M.;
(51) Qing, S. B.; Huang, W.; Yan, D. Y. J. Polym. Sci., Part A: Polym.
Chem. 2005, 43, 4363.
(52) Hodson, H. F.; Batchelor, J. F. U.S. Patent 4 103 015, 1978.
(53) Bennett, O. F. Can. J. Chem. 1965, 43, 1880.
(54) Holt, G.; Pagdin, B. J. Chem. Soc. 1960, 2508.
Venkatasubramanian, N.; Arnold, F. E. Polym. Prep. 2003, 44,
927.

1
024 Chem. Mater., Vol. 22, No. 3, 2010
Tan et al.
filtrated. The filtrate was acidified with 6 M sulfuric acid and
then the precipitation appeared. Finally, the white product was
obtained by filtration followed by washing with deionized
Scheme 1. Synthesis of PSCA
water.
1
Yield: 90%. H NMR (DMSO-d
6
, ppm): 13.87 (s, 2H), 8.53
(
s, 1H), 8.24 (d, J=8, 1H), 7.96 (d, J=7.2, 2H), 7.76 (d, J=8,
1
3
1
d
1
3
H), 7.69 (t, J=7.2, 1H), 7.61 (t, J=6.8, 2H). C NMR (DMSO-
, ppm): 168.39, 165.97, 141.22, 138.67, 138.58, 135.22, 134.57,
with tissue paper before the weight and length were measured.
The results were calculated by the following equations:
6
-
1
33.64, 130.86, 130.15, 130.10, 128.43. FT-IR (KBr, cm ):
450-2550, 1747 (-COOH), 1310, 1146 (-SO -).
Synthesis of sPBT. As a typical procedure, the preparation of
2
water uptake ¼ ðWwet - WdryÞ=Wdry ꢀ 100%,
swelling ¼ ðLwet - LdryÞ=Ldry ꢀ 100%
sPBT-6F65 was depicted as follows. To a 50 mL three-necked
round-bottom flask equipped with a nitrogen inlet/outlet and a
mechanical stirrer, PPA (12.4 g) and DABDT (0.45 g, 1.8352 mmol)
were added. The mixture was stirred at room temperature for
where Wwet and Wdry are the weight of the wet and dry
membrane and Lwet and Ldry are the length of the wet and dry
membrane, respectively.
1
2 h and then held at 70 °C for 36 h until the evolution of
The measurement of proton conductivity was conducted on a
PARSTAT 2273 Advanced Electrochemical System in the
hydrogen chloride ceased. After cooling to room temperature,
BSCS (0.6088 g, 1.1929 mmol) and 6FA (0.2519 g, 0.6423 mmol)
were added to the clear reaction system and stirred at 100 °C for
6
frequency range of 10-10 Hz with an oscillating voltage of
1
0 mV. The proton conductivity (σ) was calculated by the
8
h. Subsequently, an additional phosphorus pentoxide (3.8 g)
formula of σ=L /(RA), where R, L, and A are the resistance,
the distance between two electrodes, and the cross-sectional area
of membranes, respectively.
was added, the mixture was heated as follows: 130 °C for 8 h,
1
50 °C for 10 h, 170 °C for 10 h, 190 °C for 12 h, and 210 °C for
2 h. The viscous solution was cooled to 140 °C and poured into
1
deionized water to give yellow-green fibrous polymer. The
The oxidative stability was investigated following a typical
55
procedure. In the course of test, a small piece of membrane
product was washed with deionized water several times to
remove residual acid, then soaked in 0.5 wt % NaHCO solution
overnight and washed until achieving a neutral pH, followed by
sample (0.5 cm ꢀ 1 cm) with a thickness about 40 μm was soaked
3
in Fenton’s reagent (2 ppm FeSO in 3% H O ) at 80 °C. The
4
2
2
oxidative stability was evaluated by recording the residual
weight of the membrane after treatment in Fenton’s reagent
for 1 h and the elapsed time when the membrane disappeared
completely.
drying in vacuo at 120 °C for 24 h.
Yield: 96%. H NMR (DMSO-d
1
6
): 8.92-8.86, 8.83, 8.50,
-
1
8.32-8.23, 8.14-8.06, 8.05-7.98, 7.65-7.54. FT-IR (KBr, cm ):
1630 (CdN), 1487, 1396, 1317 (benzothiazole ring stretching),
958 (benzothiazole ring breathing), 740 (benzothiazole ring
The hydrolytic stability was investigated by treating the
membrane sample in water at 140 °C for 24 h as a typical
accelerated testing. The hydrolytic stability was evaluated by
changes in mechanical properties, residual weight, molecular
deformation), 1088, 1014, 632 (Ar-SO Na).
55
3
Preparation and Acidification of Membrane. The sPBT poly-
mer in the sodium salt form was dissolved in NMP. The solution
1
weight, FT-IR spectra, and H NMR spectra of the samples.
(
5 wt %) was cast onto a clean glass substrate and dried at 70 °C
Mechanical properties were tested with an Instron 4456 instru-
ment at ambient conditions (25 °C and 50% relative humidity)
at a crosshead speed of 1 mm/min. The wet samples for testing
mechanical properties were equilibrated at ambient conditions
for 24 h before measurement according to a well-established
for 36 h. After cooling, the membrane was peeled off from the
glass plate by immersion in deionized water, then acidified by
soaking in 1 M HCl solution for 48 h and rinsed three times with
deionized water to remove the free acid within the membrane.
Characterization of Polymer and Membrane. FT-IR spectra
were performed on a PE Paragon 1000 IR spectrometer. NMR
spectra were recorded using a Mercury Plus 400 MHz NMR
instrument. Gel permeation chromatography (GPC) was car-
ried out on a PE 200 apparatus with DMF containing 0.05 M
LiBr as eluent. Molecular weight was calibrated with polystyr-
ene standards.
56
procedure.
Atomic force microscopy (AFM) imaging was performed on
a Digital Instruments Nanoscope III in tapping mode. All
samples in the acid form were dried at 70 °C for 24 h. The
samples were imaged under ambient conditions.
Results and Discussion
The thermogravimetric analysis (TGA) was measured on a
PE TGA-7 thermogravimetric analyzer in air. Samples in the
acid form were heated at 150 °C for 30 min to remove the
moisture, followed by heating to 800 at 10 °C/min. Differential
scanning calorimeter (DSC) measurements were conducted on a
PE Pyris-1 apparatus. Samples in the acid form were preheated
to 150 °C for 30 min under nitrogen and then cooled to 90 °C.
DSC curves were recorded in a temperature range of 90-400 at
Synthesis of Diphenyl Sulphone-2,5-dicarboxylic Acid.
2,5-Dimethyl phenyl sulfone (DMPS) was synthesized by
Friedel-Crafts acylation reaction of benzene sulfonyl
chloride with p-xylene. In the previous reports, diphenyl
sulphone-2,5-dicarboxylic acid (PSCA) was prepared
by oxidation of 2,5-dimethyl phenyl sulfone with nitric
52-54
acid,
which was carried out in a stainless steel autoclave
1
0 °C/min.
under high temperature (160-190 °C) and high pressure.
A mild alternative route was given in this paper. Diphenyl
sulphone-2,5-dicarboxylic acid was easily synthesized by
oxidation of DMPS using potassium permanganate as
Ion exchange capacity (IEC) was determined by titration. The
membrane in the acid form was immersed in saturated NaCl
þ
solution for two days in order to release H by ion exchange.
þ
The released H ions were titrated with 0.01 M NaOH.
The water uptake and swelling were obtained by changes in
weight and length of the membrane from wet to dry state,
respectively. The sample was immersed in deionized water at
defined temperatures for 24 h, and then taken out and wiped
(
55) Asano, N.; Aoki, M.; Suzuki, S.; Miyatake, K.; Uchida, H.;
Watanabe, M. J. Am. Chem. Soc. 2006, 128, 1762.
56) Li, W. M.; Cui, Z. M.; Zhou, X. C.; Zhang, S. B.; Dai, L.; Wei, W.
J. Membr. Sci. 2008, 315, 172.
(

Article
Chem. Mater., Vol. 22, No. 3, 2010 1025
Scheme 2. Preparation of sPBT
repeat units of sPBT-BC65 and sPBT-PS65, respectively.
This is because the H atoms in the identical position of
sulfonated repeat units have a similar chemical environ-
ment and thus show the approximate chemical shift.
The H NMR spectra of sPBT-6F65, sPBT-BC65,
and sPBT-PS65 confirmed their chemical structure. The
1
assignment of H NMR spectra of these sPBT polymers
1
was analyzed in Figure S1 in the Supporting Information.
The IR absorptions of polybenzothiazoles were exten-
4
5,58
sively investigated in the past.
The IR spectra of all
the products are shown in Figure S2 in the Supporting
Information. As shown, there is no residual absorption
-
1
-1
band at 2600-2500 cm (-SH) and 1750-1650 cm
CdO), suggesting that the cyclization of benzothiazole
units is complete. The absorptions at 1487, 1396, 1317,
(
oxidative reagent in refluxing NaOH solution, as shown
1
in Scheme 1. The structure of PSCA was confirmed by H
-
1
1
NMR, C NMR, and FT-IR spectroscopy.
3
and 958 cm are assigned to the characteristic vibration
-
1
of benzothiazole rings, and the peak at 740 cm is
ascribed to the deformation of benzothiazole rings. Besides,
Synthesis of sPBT. Three series of sPBT with various
degrees of sulfonation were prepared by polycondensa-
tion of 2,5-diamino-1,4-benzenedithiol dihydrochloride
the characteristic bands related to -SO Na are located at
3
-
1 14,16
1
088, 1014, and 632 cm
.
Therefore, the chemical
(
(
DABDT) and bis(3-sulfonate-4-carboxyphenyl) sulfone
BSCS) with one of the unsulfonated monomers such
structure of sPBT was further confirmed by FT-IR
spectroscopy.
Thermal Properties of sPBT. The thermal stability of
the products was determined by TGA technique in air.
The TGA traces of the resulting polymers are presented in
panel a and b of the Figure 2, and the 5% weight loss
as bis(4-carboxyphenyl) sulfone (BCPS), diphenyl Sul-
phone-2,5-dicarboxylic acid (PSCA), and 2,2-bis(4-car-
boxyphenyl)hexafluoropropane (6FA), respectively, as
indicated in Scheme 2. sPBT derived from BCPS, PSCA,
and 6FA are denoted as sPBT-BCxx, sPBT-PSxx, and
sPBT-6Fxx, respectively, where the “xx” is the molar
percentage of BSCS in the dicarboxylic monomers. The
polymerization reaction was carried out in polyphospho-
ric acid at a solid content of 6 wt %. In the initial stage of
polycondensation reaction, the temperature was kept at
temperature (T ) is listed in Table 1. As illustrated in
d5
Figure 2, all the sPBT polymers displayed a typical two-
1
4
step degradation profile. The first weight loss starting
from 350 °C belongs to the decomposition of sulfonic acid
groups, and the second weight loss at 480 °C is related to
the degradation of polymer backbone. All the sPBT
displayed no residue in air at 700 °C. This illustrated that
they were completely converted to the acid form by ion
exchange. As seen in Table 1, all the sPBT polymers
exhibited a Td5 no lower than 380 °C, showing high
thermal stability. For each series of sPBT, the Td5 de-
creased slightly as the degree of sulfonation rises, similar
100-150 °C for several hours to completely remove
hydrogen chloride in the mixture, which is critical for
achieving high molecular weight products as mentioned
in the literature.
46,57
The results of polycondensation
reaction are summarized in Table 1. As shown, all the
copolymers, except for sPBT-BC50, exhibited a number
4
average molecular weight higher than 6 ꢀ 10 g/mol, and
1
4,16
to the previous reports.
This is due to the fact that the
their polydispersity index (PDI) was in the range of
1
first weight loss was caused by degradation of sulfonic
acid groups.
.6-3.0, suggesting that high molecular weight sPBT
polymers were successfully prepared. In addition, the
sPBT-6F and sPBT-PS series could be cast into transpar-
ent, tough, and ductile membranes, further confirming
that they showed high molecular weight. It is noticeable
that sPBT-BC50 showed only partial solubility in DMF,
so its molecular weight and polydispersity index could not
be obtained by GPC, but the number average molecular
The glass transition temperature of sPBT polymers was
measured by DSC. However, they showed no glass tran-
sition, no exothermal peak, and no endothermal peak in
the range of 90-350 °C (see Figure S3 in the Supporting
Information), similar to the rigid rod polybenzothia-
2
6
zoles.
4
Solubility of sPBT. As mentioned above, sulfonated
polybenzothiazoles exhibited no glass transition and
melting behavior prior to thermal degradation, so homo-
geneous sPBT membrane might only be obtained by
solution casting. Consequently, it is necessary for sPBT
to possess high solubility so as to prepare the membranes
used as PEM. The solubility of sPBT polymers was tested
at a concentration of 0.05 g/mL, and the results are
presented in Table 2. As shown, all the sPBT polymers
exhibit no solubility in methanol and water. The sPBT-BC
weight of its soluble part is still higher than 6 ꢀ 10 g/mol,
showing high molecular weight.
1
As typical examples, the H NMR spectra of sPBT-
6
shown in Figure 1. LiCl was used to reduce the inter-
F65, sPBT-BC65, and sPBT-PS65 in DMSO-d /LiCl is
6
3
2
molecular association. As shown, the signal peaks were
well-assigned. The chemical shift of H , H , H , and H
4
1
2
3
atoms in the sulfonated repeat units of sPBT-6F65 is close
to that of H , H , H , and H atoms in the sulfonated
1
2
3
4
(57) Chow, A. W.; Bitler, S. P.; Penwell, P. E.; Osborne, D. J.; Wolfe,
J. F. Macromolecules 1989, 22, 3514.
(58) Osaheni, J. A.; Jenekhe, S. A. Chem. Mater. 1992, 4, 1282.

1
026 Chem. Mater., Vol. 22, No. 3, 2010
Tan et al.
Table 1. Polycondensation Results and Properties of sPBT
GPC
IEC (meq/g)
-
4
-4
c
BSCS/BCPS/PSCA/6FA
(molar ratio)
Mn ꢀ 10
M
w
ꢀ 10
yield
(%)
T
(°C)
d5
the
calculated
the
measured
λ
polymer
(g/mol)
(g/mol)
PDI
(H
2
3
O/SO H)
a
a
a
b
b
b
b
sPBT-BC50
sPBT-BC65
sPBT-PS50
sPBT-PS55
sPBT-PS60
sPBT-PS65
sPBT-PS70
sPBT-6F50
sPBT-6F55
sPBT-6F60
sPBT-6F65
50/50/0/0
65/35/0/0
50/0/50/0
55/0/45/0
60/0/40/0
65/0/35/0
70/0/30/0
50/0/0/50
55/0/0/45
60/0/0/40
65/0/0/35
6.26
12.5
1.99
1.73
1.89
1.63
2.18
2.89
2.50
3.00
2.95
2.91
2.98
90
93
93
95
94
96
95
94
92
94
95
395
385
393
392
389
385
382
395
391
385
380
2.06
2.55
2.06
2.22
2.39
2.55
2.70
1.89
2.06
2.23
2.40
6.01
9.89
8.51
13.8
7.61
6.80
7.32
6.85
6.70
6.69
10.4
18.7
13.8
30.3
21.9
17.0
21.9
20.2
19.5
19.9
2.00
2.15
2.30
2.49
2.60
1.81
2.01
2.15
2.31
11.8
11.5
12.2
13.5
14.2
10.2
10.0
10.0
10.2
a
b
c
2
The values are the results of the soluble part. Difficult to obtain because of its poor solubility. The hydration number (number of H O molecules
per sulfonic acid group) at 80 °C.
1
Figure 1. H NMR spectra of sPBT-PS65, sPBT-BC65, and sPBT-6F65.
series derived from bis(4-carboxyphenyl) sulfone, i.e.,
sPBT-BC50 and sPBT-BC65, exhibited the poorest solu-
bility among all the products, only partially soluble in the
aprotic polar solvents even on heating, and hence they
could not be cast into the homogeneous membrane used
as PEM. Generally, the rigid rod polybenzothiazoles
show no solubility in any of those solvents even on
4
0-42
heating.
However, the sPBT-BC series still exhibited
Figure 2. TGA curves of (a) sPBT-BC and sPBT-PS series and (b) sPBT-
F series.
some solubility, which is ascribed to the steric effect and
ionization of sulfonic acid pendant groups. In contrast,
the sPBT-PS series (sPBT-PS50-PS70) containing phe-
nylsulfonyl side groups displayed excellent solubility in
NMP, DMSO, and DMF. This is because the phenylsul-
fonyl pendant groups of sPBT inhibit the chain packing
and increase intermolecular space and thus enhance the
solubility of products. On the other hand, the sPBT series
with hexafluoroisopropylidene moieties such as sPBT-
6
sPBT-6F series) with excellent solubility have been pre-
pared for the first time. Such sulfonated polybenzothia-
zoles could be cast into homogeneous membranes and
thus could be further evaluated as PEM materials.
IEC, Water Uptake, Swelling, and Proton Conductivity
of sPBT. The IEC value of sPBT membranes is an
3
important parameter when used as PEM. The theoretical
6
F50, -6F55, -6F60, and -6F65 were readily soluble in
value of IEC of sPBT was calculated according to their
molecular formulas, and the experimental value was
obtained by titration. The results are compiled in Table 1.
In all cases, the experimental value of IEC agrees with the
theoretical one, indicating that the sulfonate groups were
successfully incorporated into polymer backbone by
polycondensation. The IEC value of sPBT membranes
was adjusted in a range of 1.8-2.7 meq/g by varying
NMP, DMSO, and DMF. This illustrated that the flex-
ible hexafluoroisopropylidene moieties is favorable to
increase the solubility of the resulting sPBT. In conclu-
sion, the excellent solubility of sPBT polymers has been
achieved by attaching the bulky pendant groups or
incorporating flexible moieties to polymer backbone.
High-molecular-weight sPBT polymers (sPBT-PS and

Article
Chem. Mater., Vol. 22, No. 3, 2010 1027
a
Table 2. Solubility of sPBT
solvents
2
polymer NMP DMSO DMF DMAc sulfolane methanol H O
sPBT-BC50 þ-
sPBT-BC65 þ-
þ-
þ-
þ
þ-
þ-
þ
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
þ-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
sPBT-PS50
sPBT-PS55
sPBT-PS60
sPBT-PS65
sPBT-PS70
sPBT-6F50
sPBT-6F55
sPBT-6F60
sPBT-6F65
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
a
The test was performed at a concentration of 0.05 g/mL. (þ) soluble
on heating; (þ-) partially soluble or swelling on heating; (-) insoluble
even on heating.
Figure 3. Water uptake of sPBT membranes at various temperatures.
the feed ratios in order to ensure that they could exhibit
high proton conductivity. The experimental value of
IEC of sPBT-BC series was not obtained because they
denoted low solubility and thus could not be cast into
membranes.
According to the vehicle mechanism of proton trans-
port, the membrane could not conduct proton unless it
absorbs an appropriate amount of water, but excessive
absorption of water would lead to too much swelling and
3
,14
even loss of mechanical strength.
Therefore, it is
significant to investigate the water uptake and swelling
of sPBT membranes in order to maintain the balance
between the proton conductivity and mechanical proper-
ties. The water uptake of sPBT-6F and sPBT-PS series as
functions of temperature and sulfonation degree is
plotted in Figure 3. As shown, the water uptake of
sPBT-PS membranes increased slowly up to 60 °C and
thereafter increased more evidently with temperature,
especially for those membranes with high sulfonation
degree, wheras the water uptake of sPBT-6F membranes
increased slowly in the whole temperature region. It could
also be found that the water uptake of sPBT-PS mem-
branes is higher than that of sPBT-6F membranes with
an equivalent IEC at the same temperature. For example,
sPBT-PS50 and sPBT-6F55 have an equal IEC
Figure 4. Swelling of sPBT membranes at various temperatures.
(
λ), which was listed in Table 1. The hydration number is
in the range of 10.0-14.2, close to that of other sulfonated
22,59
60
aromatic polymers
and Nafion 117.
The swelling of sPBT membranes as functions of
temperature and sulfonation degree is displayed in Figure 4.
Similar to the water uptake, the swelling increased as the
temperature and sulfonation degree rise. At 80 °C, all
the sPBT membranes, except for the sPBT-PS70 mem-
brane, exhibited a swelling lower than that (20%) of
1
4,18
Nafion 117,
showing excellent dimensional stability.
As shown in Table 1, the sPBT-PS50-PS65 and sPBT-
(
2.06 meq/g), but the former showed a water uptake of
6
2
F50-6F65 membranes indicate a high IEC of 1.89-
.55 meq/g, but they denote a low swelling. This might be
4
3.6%, higher than that (36.9%) of the latter at 80 °C. The
difference in water uptake between the sPBT-PS and -6F
membranes is attributed to the difference in their struc-
ture. The hydrophobic nature of hexafluoroisopropyli-
dene moieties might endow the sPBT-6F membranes with
low water uptake. The sPBT-6F series showed a water
uptake of 34.8-44.0% at 80 °C, slightly higher than that
ascribed to the strong intermolecular interaction between
the sulfonic acid and thiazole rings.
In short, except for the sPBT-PS65 and -PS70, the
sPBT-PS and sPBT-6F series showed moderate water
uptake as well as low swelling, so they possessed excellent
dimensional stability.
The proton conductivity of sPBT membranes was
measured after they were fully hydrated in deionized
water for 48 h. The proton conductivity of Nafion 117
and sPBT membranes against temperature is displayed in
Figure 5. As expected, the proton conductivity increased
with increasing temperature. At the same temperature,
for each sPBT series, the proton conductivity enhanced
1
4
(
-
30%) of Nafion 117. At the same time, the sPBT-PS50,
PS55, and -PS60 membranes displayed a water uptake of
4
3.6-52.5% at 80 °C, higher than that of Nafion 117 but
3,4
still lower than that (54%) of Dow membrane, whereas
the sPBT-PS65 and -PS70 membranes exhibited a water
uptake of 61.9-69.1%. In addition, the water uptake
values at 80 °C were expressed as the hydration number
(
59) Kim, D. S.; Guiver, M. D. J. Polym. Sci., Part A: Polym. Chem.
(60) Xing, P. X.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.;
Kaliaguine, S. Polymer 2005, 46, 3257.
2008, 46, 989.

1
028 Chem. Mater., Vol. 22, No. 3, 2010
Tan et al.
Table 3. Oxidative Stability of sPBT Membranes in Fenton’s Reagent
a
residue (%)
b
membrane
thickness (μm)
τ (h)
sPBT-PS50
sPBT-PS55
sPBT-PS60
sPBT-PS65
sPBT-PS70
sPBT-6F50
sPBT-6F55
sPBT-6F60
sPBT-6F65
40
43
42
40
41
39
40
42
38
98
98
97
97
96
99
99
98
97
4.0
3.5
3.0
2.5
2.0
5.5
5.0
4.5
3.5
a
Residual weight of membranes after treatment in Fenton’s reagent
b
for 1 h. τ refers to the time when the membrane dissolved completely.
time of 2.0-4.0 h and that the sPBT-6F membranes
showed an elapsed time of 3.5-5.5 h, which are also
comparable to or more longer than the elapsed time of
Figure 5. Proton conductivity of sPBT membranes at various tempera-
tures.
with the increase in sulfonation degree. Nafion 117
showed a proton conductivity comparable to the previous
63
sulfonated poly(arylene ether ketone)s and sulfonated
19,55
polyimides
under the same conditions. Both sPBT
6
1,62
value.
In the entire temperature range studied, all the
series denoted high retained weight and long elapsed time,
indicating high oxidative stability. On the other hand,
their elapsed time and residue after 1 h treatment showed
a decreasing tendency with increasing sulfonation degree
sPBT membranes, except for sPBT-6F50 with the lowest
IEC, showed higher proton conductivity than that of
Nafion 117. At the same time, all the sPBT membranes,
except for the sPBT-PS70 membrane, exhibited a swelling
lower than that (20%) of Nafion 117 at 80 °C. Hence most
of sPBT membranes displayed high proton conductivity
as well as low swelling.
5
5
like the previous report. This illustrated that their
oxidative stability decreased with sulfonation degree. In
addition, the sPBT-PS50, -PS55, and -PS60 membranes
displayed a retained weight of 98, 98, and 97% as well as
an elapsed time of 4.0, 3.5, and 3.0 h, whereas the sPBT-
It could also be found that the sPBT-PS membranes
showed higher conductivity than that of sPBT-6F mem-
branes with an equal IEC. For example, both sPBT-PS50
and sPBT-6F55 showed an IEC of 2.06 meq/g, but the
former possessed a proton conductivity of 0.10 S/cm
while the latter denoted a proton conductivity of 0.08 S/cm
at 80 °C. This might be due to the fact that sPBT-PS50
show higher water uptake as well as wider ionic channels
than sPBT-6F55. The effect of ionic channels on the
proton conductivity would be discussed in the last part.
Oxidative and Hydrolytic Stability of sPBT Mem-
branes. Up to now, aromatic proton exchange mem-
branes generally exhibited lower lifetime than perfluori-
6
F55, -6F60, and -6F65 membranes denoted a residual
weight of 99, 99, and 98% as well as an elapsed time of 5.0,
.5, and 3.5 h, respectively. As shown in Table 1, the IEC
4
value of sPBT-PS50, -PS55, and -PS60 membranes is
equivalent to that of the sPBT-6F55, -6F60, and -6F65
membranes, respectively. Hence, these results (the resi-
dual weight and the elapsed time of membranes) demon-
strated that the sPBT-6F membranes had a better
oxidative stability than the sPBT-PS membranes with
an equivalent IEC. The reasons might be as follows. First,
the oxidative attack by the hydrated HO and HOO
3
3
radicals occurred mainly in water-containing hydro-
5
5
6
3
nated membranes. Therefore, the oxidative stability
of aromatic PEM is especially important for fuel cell
applications. The oxidative stability of sPBT membrane
was evaluated by observing their dissolving behavior in
philic domains. Second, the sPBT-PS membranes
exhibited a higher water uptake than the sPBT-6F mem-
branes with an equal IEC, the absorbed water facilitated
the transport of HO and HOO radicals and thus
3
3
Fenton’s reagent (3% H O containing 2 ppm FeSO ) at
4
2
2
enhanced the oxidative attack. In addition, the micro-
structure of sPBT membranes could affect their oxida-
tive stability, which would be discussed in Microscopic
Morphology.
5
5
8
0 °C according to a typical testing procedure. The
weight loss after treatment for 1 h and the elapsed time (τ)
when the membrane samples dissolved completely are
listed in Table 3. As shown, all the membranes retained
more than 96% of their original weight and still kept their
shape and flexibility after treatment for 1 h. The retained
weight of sPBT polymers is comparable to or more than
The hydrolytic stability of PEM is also important
9
because it works in the hydrothermal environment.
1
Most sulfonated polyimide membranes were characterized
in terms of hydrolytic stability. This is probably because
the imide groups of sulfonated polyimides exhibited lower
hydrolytic stability compared with the functional groups,
such as aromatic ether, phosphazene, and benzimidazole
6
3,64
that of sulfonated poly(arylene ether ketone)s
5
and
5
sulfonated polyimides. Meanwhile, it could also be
found that the sPBT-PS membranes exhibited an elapsed
8
,23,31
groups of other sulfonated polymers.
In contrast,
(
(
(
61) Ding, J. F.; Chuy, C.; Holdcroft, S. Macromolecules 2002, 35, 1348.
62) Tian, S. H.; Meng, Y. Z.; Hay, A. S. Macromolecules 2009, 42, 1153.
63) Xing, P.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.;
Kaliaguine, S. Macromolecules 2004, 37, 7960.
few other sulfonated aromatic PEM have been character-
ized in terms of hydrolytic stability. Because sPBT-based
proton exchange membranes have not been reported until
now, their hydrolytic stability should be investigated in
(64) Chikashige, Y.; Chikyu, Y.; Miyatake, K.; Watanabe, M. Macro-
molecules 2005, 38, 7121.

Article
Chem. Mater., Vol. 22, No. 3, 2010 1029
Table 4. Hydrolytic Stability of sPBT Membranes
a
mechanical property
GPC
-
4
-4
IEC
(meq/g)
time
(h)
tensile strength
(MPa)
Young’s modulus
(GPa)
elongation at break
(%)
Mn ꢀ 10
M
w
ꢀ 10
weight loss
(%)
membrane
sPBT-PS50
(g/mol)
(g/mol)
PDI
2.06
2.22
0
2
75.1
68.1
66.1
74.7
65.1
61.2
79.2b
75.2
71.5
69.9
73.8
67.9
80.6
75.7
1.51
1.41
1.38
1.31
1.29
1.30
1.62
49.4
33.6
32.7
48.5
31.2
28.4
39.3
9.89
10.1
10.7
8.51
8.82
9.64
13.8
15.3
12.1
12.2
11.2
11.6
12.3
12.5
18.7
19.3
20.9
13.8
14.9
16.7
30.3
39.7
29.1
29.9
28.0
29.7
30.1
32.1
1.89
1.91
1.94
1.63
1.69
1.73
2.18
2.61
2.41
2.46
2.50
2.56
2.44
2.57
0
1
2
0
2
3
0
2
4
0
2
4
0
4
0
4
0
4
0
4
sPBT-PS55
2
2
2
2
2
sPBT-PS60
sPBT-6F55
sPBT-6F60
sPBT-6F65
2.39
2.06
2.23
2.40
b
b
c
1.59
30.1
-
1.13
1.15
1.19
1.21
1.34
1.33
27.8
21.3
30.1
25.3
30.5
27.8
0
1
0
1
0
1
a
b
The wet samples were equilibrated at 50% RH and 25 °C for 1 day. The data were obtained by measuring the membrane recast from its residue after
aging test. have not been measured due to excessive swelling.
c
detail so as to determine whether they are suitable for fuel
5
cell applications or not. Following a recent procedure,
low molecular weight in samples might dissolve in deio-
nized water during the testing. Different from the sPBT-
PS50 and -PS55 membranes, the sPBT-PS60 membrane
swelled excessively after aging for 24 h and could not
be characterized in terms of mechanical properties, but
the membrane recast from its residue almost retained the
initial mechanical properties, as shown in Table 4. This
result further illustrated that the sPBT-PS membranes
had excellent hydrolytic stability. In the case of sPBT-6F
series membranes, the amplitude of decrease in tensile
strength, elongation at break, and residual weight is even
less than that of sPBT-PS series membranes with an
equivalent IEC. Moreover, their Young’s modulus and
molecular weight after aging test almost remained intact.
In other words, the sPBT-6F membranes displayed better
resistance to hydrolysis in comparison with the sPBT-PS
membranes containing an equal IEC.
5
the sPBT membranes were investigated in water at 140 °C
for 24 h as an accelerated aging test; the hydrolytic
stability was evaluated by changes in appearance, me-
chanical properties, residual weight, molecular weight,
1
H NMR spectra, and IR spectra before and after the
testing. Three samples of sPBT-PS series (sPBT-PS50,
-
(
PS55, and -PS60) and three samples of sPBT-6F series
sPBT-6F55, -6F60, and -6F65) exhibited high proton
conductivity as well as low swelling, and thus their
hydrolytic stabilities were evaluated as typical examples.
After the aging test, all the membranes except for sPBT-
PS60 maintained their transparency, flexibility, and
toughness in appearance. Table 4 lists the mechanical
properties, residual weight, and molecular weight before
and after the aging test. As shown, sPBT-PS50 and -PS55
exhibited little decrease in tensile strength, Young’s mod-
ulus, and elongation at break after 24 h aging test.
Interestingly, the decrease in mechanical properties took
place mainly in the initial period of aging (2 h). The
amplitude of decrease in the initial aging for 2 h is much
higher than that in the subsequent aging for 22 h. It is
well-known that sulfonated polymers indicated a hydro-
phobic/hydrophilic nanophase separation morphology.
The hydrophilic domains absorbed water and swelled
under the harsh hydrothermal conditions, so the mem-
brane might undergo an irreversible change of micro-
structure in the initial aging as in ref 8, whereas the
possible hydrolysis might increase gradually with time.
Hence the decrease in mechanical properties might
mainly result from the microstructural change of the
membranes. On the other hand, the sPBT-PS50 and -
PS55 membranes lost 2-3% of their original weight,
whereas the aromatic PEM with the longest lifetime
reported still lost 8% of its original weight after the same
As typical examples, a group of sPBT membranes
(sPBT-PS50 and -6F55) after the hydrothermal aging
1
for 24 h were analyzed in terms of H NMR and IR
1
spectroscopy to detect their structural change. The H
NMR and IR spectra of the treated sPBT polymers
hardly exhibited any difference compared with those of
the pristine polymers [see Figure S4-S5 in the Supporting
Information]. These results further demonstrated that the
sPBT-PS and sPBT-6F membranes possessed excellent
hydrolytic stability.
Microscopic Morphology. The macroscopic properties
of proton exchange membranes, such as water uptake,
swelling, proton conductivity, oxidative stability, and
hydrolytic stability, are greatly affected by their micro-
3
,14
structure.
The microscopic morphology of sPBT
membranes was investigated by atomic force microscopy
(AFM). As typical examples, the tapping mode AFM
phase images for three samples of sPBT-PS series and
three samples of sPBT-6F series were recorded under
ambient conditions on a 500 nm ꢀ 500 nm size scale,
as shown in Figure 6. All the AFM images exhibited
a dark/bright nanophase separation morphology. The
dark regions are assigned to hydrophilic ionic clusters
5
5
aging test. Therefore, the sPBT-PS50 and -PS55 mem-
branes possessed high hydrolytic stability. In addition,
their molecular weight after treatment showed a slight
increase, probably due to the fact that a small part with

1
030 Chem. Mater., Vol. 22, No. 3, 2010
Tan et al.
Figure 6. AFM images of (a) sPBT-PS50, (b) sPBT-PS55, (c) sPBT-PS60, (d) sPBT-6F55, (e) sPBT-6F60, and (f) sPBT-6F65.
containing small amounts of water, whereas the bright
areas are attributed to hydrophobic polymer back-
These results are due to the hydrophobic and flexible
hexafluoropropylidene moieties. Similar results were
observed in other sulfonated polymers containing bulky
3
,65
bone.
In general, the hydrophilic channels absorb
6
6
water and are responsible for proton transport, whereas
the hydrophobic domains provide the ionomers with
morphological stability and keep them from dissolving
hydrophobic moieties. For the ionomer membranes, the
narrow ionic channels receive more strong binding arising
from the surrounding hydrophobic domains in compari-
son with the wide ones. This led to the result that the
narrow ionic channel-containing ionomer membranes
indicated higher dimensional stability (or lower swelling)
than the wide ionic-channel-containing ionomer mem-
3
,63,66
in water.
and -PS60) showed a typical nanophase separation mor-
The sPBT-PS series (sPBT-PS50, -PS55,
6
7
phology like most of sulfonated aromatic polymers,
i.e., the size and connectivity of hydrophilic ionic channels
exhibited an increasing tendency with degree of sulfona-
tion. It is well-known that the oxidative and hydrolytic
degradation occurred mainly in water-containing hydro-
6
6
branes with an equivalent IEC. For example, the sPBT-
F65 and sPBT-PS60 membranes have an equal IEC, but
6
the former showed narrower ionic channels than the
latter, as indicated in Figure 6. Therefore, the sPBT-
6
3
philic domains. Hence the water uptake, swelling,
and proton conductivity enhanced but the oxidative and
hydrolytic stability decreased with increasing degree of
sulfonation. When the sulfonated repeat unit of sPBT-PS
series is at 60 mol %, the size and connectivity of hydro-
philic channels reached a critical limit, wherein the corres-
ponding membrane (sPBT-PS60) underwent excessive
swelling in water at 140 °C for 24 h. In contrast, the sPBT-
6
7
F65 membrane still denoted a tensile strength of
5.7 MPa after the hydrolytic test, whereas the sPBT-
PS60 membrane swelled excessively and almost lost its
mechanical strength under the same conditions as men-
tioned above. That is to say, the former showed much
better dimensional stability than the latter. Similarly, the
narrow ionic channels-containing sPBT-6F membranes
exhibited higher oxidative and hydrolytic stability than
the wide ionic channels-containing sPBT-PS membranes
with an equivalent IEC, because the oxidative and hydro-
lytic degradation occurred mainly in hydrophilic domains
6F membranes (sPBT-6F55, -6F60, and -6F65) displayed a
particular morphology, much different from that of
sPBT-PS membranes. In other words, they showed much
narrower ionic channels in comparison with the sPBT-PS
membranes with an equivalent IEC. The connectivity of
ionic channels enhanced but their size exhibited a slightly
decreasing tendency increase with degree of sulfonation.
6
3
containing absorbed water. Therefore, the incorpora-
tion of the hydrophobic and flexible hexafluoroisopro-
pylidene moieties to the backbone of sulfonated polymers
could enhance their dimensional, oxidative, and hydro-
lytic stability. This provided an approach for improving
the properties of proton exchange membranes.
(
(
(
65) Lee, K. S.; Jeong, M. H.; Lee, J.; Lee, J. S. Macromolecules 2009, 42,
584.
66) Zhang, C. J.; Kang, S.; Ma, X. H.; Xiao, G. Y.; Yan, D. Y.
J. Membr. Sci. 2009, 329, 99.
As shown in Figure 5, the sPBT-6F membranes exhibited
lower proton conductivity than the sPBT-PS membranes
with an equal IEC. For instance, the sPBT-6F55, -6F60,
67) Wang, F.; Hickner, M.; Kim, Y. S.; Zawodzinski, T. A.; McGrath,
J. E. J. Membr. Sci. 2002, 197, 231.

Article
Chem. Mater., Vol. 22, No. 3, 2010 1031
and -6F65 membranes indicated lower proton conducti-
vity than the sPBT-PS55, -PS60, and -PS65 membranes,
respectively. The reason is that the sPBT-6F membranes
showed much narrower ionic channels along with lower
water uptake than the sPBT-PS membranes with an
equivalent IEC. Even so, the sPBT-6F55, -6F60, and
indicating high dimensional stability. In addition, these
sPBT membranes exhibited high oxidative stability and
excellent hydrolytic stability. AFM images revealed
that the sPBT polymers presented an increasing hydro-
philic/hydrophobic nanophase separation morphology
with degree of sulfonation. Thus their water uptake,
swelling, and proton conductivity increased but their
oxidative and hydrolytic stability decreased with
degree of sulfonation. For the sPBT-PS membranes,
the size and connectivity of hydrophilic ionic channels
enhanced with degree of sulfonation like most of sulfo-
nated polymers. In contrast, the sPBT-6F membranes
denoted much narrower ionic channels than the
sPBT-PS membranes with an equivalent IEC because
of the hexafluoroisopropylidene moieties, so the for-
mer displayed better dimensional stability, oxidative
stability, and hydrolytic stability than the latter. As a
result, sulfonated polybenzothiazoles are a novel can-
didate for PEM materials. More sPBT membranes with
various structures would be prepared in our future
studies.
-6F65 membranes still exhibited a proton conductivity
higher than that of Nafion 117.
Conclusions
Three series of high-molecular-weight sulfonated poly-
benzothiazoles (sPBT) were successfully synthesized by
polycondensation of 2,5-diamino-1,4-benzenedithiol di-
hydrochloride with bis(3-sulfonate-4-carboxyphenyl)
sulfone and bis(4-carboxyphenyl) sulfone or diphenyl
sulphone-2,5-dicarboxylic acid or 2,2-bis(4-carboxyphe-
nyl)hexafluoropropane. The sPBT-BC series showed
poor solubility, while the sPBT-PS and sPBT-6F series
exhibited excellent solubility in organic solvents such as
NMP, DMSO, and DMF. The incorporation of the
phenylsulfonyl pendant groups or hexafluoroisopropyli-
dene moieties to the resulting sPBT greatly enhanced the
solubility. High-molecular-weight sulfonated polyben-
zothiazoles with excellent solubility were synthesized for
the first time. Such sulfonated polybenzothiazoles thus
could be cast into homogeneous membranes and could be
further evaluated as proton exchange membranes. All the
sPBT showed high thermal stability, the Td5 of which is no
lower than 380 °C. Except for the sPBT-PS70 mem-
branes, the sPBT-PS and sPBT-6F series indicated high
proton conductivity as well as low swelling. For example,
sPBT-PS55 and sPBT-6F65 exhibited a proton conduc-
tivity of 0.105 and 0.11 S/cm at 80 °C, respectively, much
higher than that of Nafion 117. However, they showed a
swelling of only 14.3 and 15.5% at 80 °C, respectively,
Acknowledgment. We thank the National Natural Science
Foundation of China (50303010, 50633010, and 50973061),
the National Basic Research Program (2007CB808000 and
2009CB930400), the Shanghai Leading Academic Discipline
Project (B202), and the Science and Technology Committee
of Shanghai municipality (065207065) for supporting this
research.
Supporting Information Available: The assignment analysis of
H NMR spectra of sPBT-6F65, sPBT-BC65, and sPBT-PS65;
FT-IR spectra of all the sPBT polymers; DSC curves of the acid
1
1
form sPBT; H NMR and FT-IR spectra of the membranes
(
sPBT-PS50 and sPBT-6F55) before and after the hydrolytic
testing (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.