Synthesis, characterization, and copolymerization of a series of novel acid monomers based on sulfonimides for proton conducting membranes

Article abstract of DOI:10.1021/ma0498058

The synthesis and characterization of a series of novel acid monomers based on sulfonimides with different acidities are described. Copolymerization of the acid monomers with acrylonitrile was carried out, and the proton exchange membranes were cast from solutions of the resulting copolymers. Flexible, mechanically strong, and transparent membranes with different equivalent weights could be prepared. The membranes with lower equivalent weights and higher acidities exhibited higher proton conductivity under humidifying conditions compared to those with higher equivalent weights and lower acidities. The membrane electrode assembly was fabricated using a synthesized sulfonimide copolymer. The H₂/O₂ fuel cell performance was tested in a single cell, which revealed potential applicability of the sulfonimide membranes to polymer electrolyte fuel cells.

Full text of DOI:10.1021/ma0498058

5572
Macromolecules 2004, 37, 5572-5577
Synthesis, Characterization, and Copolymerization of a Series of Novel
Acid Monomers Based on Sulfonimides for Proton Conducting
Membranes
Md . Kh a lilu r Ra h m a n ,† Gen ta r o Aiba ,† Md . Abu Bin Ha sa n Su sa n ,† Yu k o Sa sa ya ,‡
Ken -ich ir o Ota ,‡ a n d Ma sa yosh i Wa ta n a be*,†
Department of Chemistry and Biotechnology and Department of Safety and Energy Engineering,
Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, J apan
Received J anuary 30, 2004; Revised Manuscript Received May 7, 2004
ABSTRACT: The synthesis and characterization of a series of novel acid monomers based on sulfonimides
with different acidities are described. Copolymerization of the acid monomers with acrylonitrile was carried
out, and the proton exchange membranes were cast from solutions of the resulting copolymers. Flexible,
mechanically strong, and transparent membranes with different equivalent weights could be prepared.
The membranes with lower equivalent weights and higher acidities exhibited higher proton conductivity
under humidifying conditions compared to those with higher equivalent weights and lower acidities. The
membrane electrode assembly was fabricated using a synthesized sulfonimide copolymer. The H2/O2 fuel
cell performance was tested in a single cell, which revealed potential applicability of the sulfonimide
membranes to polymer electrolyte fuel cells.
In tr od u ction
structure of sulfonimide acid groups. This also allows
the membrane to maintain a more hydrated state at
above 80 °C, giving this membrane a distinct advantage
over Nafion, whose conductivity drops sharply due to
dehydration. Des-Marteau and co-workers described
that the thermal cyclopolymerization of various trifluo-
rovinyl aromatic ether monomers, bearing both pendant
sulfonimide groups and sulfonimide groups incorporated
into the monomer main chain to yield perfluorocyclobu-
tane aromatic polyethers, were quite interesting materi-
als for fuel cell membranes.11-13 A series of novel
polystyrenes with pendant lithium perfluoroalkylsul-
fonate or sulfonimide groups have been synthesized by
Feiring and co-workers and identified as potential
electrolytes for lithium batteries. They proposed that
the ionomers of this variety might also be used as fuel
cell electrolytes.9 Hofmann et al. have reported that the
sulfonimide-functionalized phosphazene membrane is
an excellent proton conductor.10 The membrane and a
blended membrane of the sulfonimide-functionalized
phosphazene polymer with poly(vinylidene fluoride)
appear to be promising candidates for use as proton-
conducting membranes in fuel cell applications.10
New proton exchange membranes for fuel cells are
required for a wide range of demanding applications,
such as portable electronic devices and vehicles as well
as stationary power generation.1-3 Until now, numerous
membrane materials have been claimed to be promising
for fuel cell applications. Among them, perfluorosulfonic
acid membranes, typically Nafion, have been chosen as
a standard membrane because of their high perfor-
mance, including high proton conductivity, high me-
chanical strength, and excellent thermal and chemical
stability.1-3 However, a few drawbacks, such as the high
cost and evaporation of water at elevated temperatures,
limit its wide-scale commercialization.1-3
Extensive efforts have been devoted to innovation of
the cost and performance, including the studies on
nonfluorinated and partially fluorinated membrane
materials.2,3 Sulfonimide membranes are of particular
interest because of their strongest gas-phase supera-
cidity, excellent thermal stability, capacity to promote
oxygen reduction kinetics at cathode, and weak adsorp-
tion on platinum catalyst. Thus, sulfonimide mem-
branes have proved to be potential candidates for fuel
cell applications.4-10 Sumner and co-workers have de-
veloped a new class of ionically conducting polymers
based on a bis[(perfluoroalkyl)sulfon]imide.8 This poly-
mer is structurally similar to Nafion, but the sulfonic
acid group is substituted by a sulfonimide group.
Similarly to Nafion, the conductivity of the sulfonimide
membrane is found to be strongly dependent on the level
of hydration and temperature. The conductivity of this
membrane is reported to be 2 times higher than that of
Nafion 117, measured under the same conditions. They
also reported that the water uptake of this membrane
was high due to the internal pore and/or channel
In addition to the sulfonimide membranes, Guo and
co-workers synthesized a series of novel sulfonated
polyimides with different sulfonation degrees from 9,9-
bis(4-aminophenyl)fluorene-2,7-disulfonic acid, 1,4,5,8-
naphthalenetetracarboxylic dianhydride, and common
nonsulfonated diamines and found that the polyimide
membranes showed proton conductivities similar to or
higher than those of Nafion 117 at 100% relative
humidity.14 They also reported that the polyimide
membranes were good for low operating temperature
fuel cell systems due to the low hydrolysis stability.
Quite recently, it has been reported that the sulfonated
and phosphonated poly[(aryloxy)phosphazene] mem-
branes are good proton conductors with low methanol
permeation, as compared with Nafion.15
† Department of Chemistry and Biotechnology.
‡ Department of Safety and Energy Engineering.
* Corresponding author: phone/fax +81-45-339-3955; e-mail
mwatanab@ynu.ac.jp.
Although numerous studies on the synthesis and
proton conductive behavior have been conducted on
10.1021/ma0498058 CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/29/2004

Macromolecules, Vol. 37, No. 15, 2004
Novel Acid Monomers 5573
Sch em e 1. Syn th esis of Novel Acid Mon om er s
cooled to room temperature. Then 100 mL of 5 wt % Na2CO3
aqueous solution was added to dissolve the resulting solid.
After filtration of the solution, the filtrate was extracted three
times with benzene (50 mL) to remove phenol. In a 500 mL
round-bottom flask concentrated HCl (17 mL) with water (180
mL) was taken, and the filtrate was added dropwise under
stirring. The resulting mixture was filtered off to discard the
filtrate and washed with ether. The solid white compound was
collected, recrystallized from ethanol, and dried under vacuum
for several hours to obtain 8.20 g of the solid compound. The
yield was 49%.
Syn t h esis of N-(4-Ben zoylim id esu lfon yl)p h en yl-2-
m eth yla cr yla m id e (BP MA). BPMA was synthesized from
BPA and methacryloyl chloride in a similar way to APMA. The
crude product was recrystallized from ethanol and dried under
vacuum for several hours to obtain a solid compound. The yield
was 41%. 1H NMR (270 MHz, DMSO-d6) δ (TMS, ppm): 12.43
(s, 1H), 10.20 (s, 1H), 7.90 (m, 6H), 7.55 (m, 3H), 5.84 (s, 1H),
5.58 (s, 1H), 1.94 (s, 3H). EI mass spectrum: Calculated for
these sulfonimide and sulfonated membranes, at present
there is little available fuel cell performance data in the
literature on these materials. In the present research,
we synthesized a series of novel acid monomers based
on sulfonimides (Scheme 1) and copolymerized them
with a vinyl monomer to realize proton exchange
membranes with high proton conductivity by changing
the acidity, water uptake, and equivalent weight of the
membranes.16-19 Thermal properties of the dry mem-
branes as well as the hydrated membranes were inves-
tigated to know the state of water in the membranes.20
The hydrated membranes showed lower glass transition
temperatures than the dry membranes. The proton
conductivity and water uptake with different equivalent
weights were investigated. Finally, a sulfonimide-based
membrane electrode assembly was fabricated and tested
in an H2/O2 fuel cell.
C
17H16N2O4S: m/e 344.38. Found: m/e 344. Elemental analysis
(%): Calculated for C17H16N2O4S: C, 59.28; N, 8.13; S, 9.30.
Found: C, 59.12; N, 8.25; S, 9.09.
Syn th esis of N-[4-Meth ylbis(su lfon yl)im id e]p h en yl-
a m in e (MBP A). MBPA was synthesized in a similar way to
BPA. The crude product was purified on a silica gel column
using ethanol/benzene (1/2 by volume) as the eluent. The
solvent was evaporated, and the solid product was recrystal-
lized from ethanol to obtain a white solid of MBPA (yield:
26%).
Syn t h esis of N-[4-Met h yl-b is(su lfon yl)im id e]p h en yl-
2-m eth yla cr yla m id e (MBP MA). MBPMA was synthesized
in a similar way to APMA. The monomer was recrystallized
from ethanol and dried under vacuum to obtain a white
crystalline powder. The yield was 45%. 1H NMR (270 MHz,
DMSO-d6) δ (TMS, ppm): 9.99 (s, 1H), 7.72 (m, 4H), 5.84 (s,
1H), 5.54 (s, 1H), 2.87 (s, 3H), 1.94 (s, 3H). EI mass spectrum:
Calculated for C11H14N2O5S2: m/e 318.36. Found: m/e 318.
Elemental analysis (%): Calculated for C11H14N2O5S2: C,
41.50; N, 8.79; S, 20.14. Found: C, 41.46; N, 8.76; S, 19.76.
Syn th esis of N-[4-Tr iflu or om eth ylbis(su lfon yl)im id e]-
p h en yla m in e (TBP A). Sulfanilamide (17.22 g, 100 mmol),
K2CO3 (6.91 g, 50 mmol), and 4-nitrophenyl trifluoromethane-
sulfonate (27.11 g, 100 mmol) were completely mixed and put
into a round-bottom flask. Then, TBPA was synthesized in a
similar way to BPA. The crude product was purified by silica
gel column chromatography eluted with ethanol/benzene (1:
2). The solvent was removed with a rotary evaporator, and
the product was recrystallized from ethyl acetate to obtain
TBPA (yield 15%) as a white powder.
Exp er im en ta l Section
Ma ter ia ls. Sulfacetamide (Lancaster), methyl trifluoroac-
etate (Wako), sulfanilamide (Cica), phenyl trifluoroacetate
(Wako), phenyl methanesulfonate (Aldrich), 4-nitrophenyl
trifluoromethanesulfonate (Aldrich), methyl pentadecafluo-
rooctanoate (Aldrich), pottasium carbonate (Wako), phenyl
benzoate (Cica), methacryloyl chloride (Wako), N-phenyl-2-
naphthylamine (TCI), 2,2′-azobis(2-methylpropionitrile) AIBN
(JUNSEI), dimethyl-d6 sulfoxide (Aldrich), carbon paper (Toray),
and Pt/C (46.7 wt %) (Tanaka Kikinzoku Kogyo) were used as
purchased. Acrylonitrile (Wako) was used after distillation. All
reagents used were of reagent grade.
Syn th esis of N-[4-Tr iflu or om eth ylbis(su lfon yl)im id e]-
p h en yl-2-m eth yla cr yla m id e (TBP MA). The synthesis of
TBPMA followed the same procedure as APMA. The monomer
was recrystallized from ethyl acetate and dried under vacuum
Syn t h esis
of N-(4-Acet ylim id esu lfon yl)p h en yl-2-
m eth yla cr yla m id e (AP MA). Sulfacetamide (10.0 g, 46 mmol)
was dissolved in acetone (50 mL), to which a small amount of
N-phenyl-2-naphthylamine was added, and cooled to 10 °C.
Methacryloyl chloride (4.87 g, 46 mmol) was added dropwise
to the mixture at 10 °C, and the solution was heated for 3 h
under reflux condition with constant stirring. To the reaction
mixture, 30 wt % NaHCO3 aqueous solution was added, and
the precipitated solid was collected by filtration. The solid
product was recrystallized from ethanol and dried under
vacuum to obtain 7.50 g of a white crystalline powder. The
1
to obtain a slightly yellowish powder. The yield was 10%. H
NMR (270 MHz, DMSO-d6) δ (TMS, ppm): 10.01 (s, 1H), 7.72
(m, 4H), 5.84 (s, 1H), 5.54 (s, 1H), 1.94 (s, 3H). Electron
impact ionization (EI) mass spectrum: Calculated for
C
11H11F3N2O5S2: m/e 372.33. Found: m/e 372. Elemental
analysis (%): Calculated for C11H11F3N2O5S2: C, 35.48; N, 7.52;
S, 17.22. Found: C, 35.11; N, 7.48; S, 16.90.
Syn th esis of N-(4-Tr iflu or oacetylim idesu lfon yl)ph en yl-
a m in e (TP A). Sulfanilamide (10.33 g, 60 mmol) and K2CO3
(4.14 g, 30 mmol) were mixed, put into a round-bottom flask,
and heated at 180 °C in an oil bath for 20 min under moisture-
free conditions. The flask was removed from the oil bath and
cooled to room temperature. Methanol was added to dissolve
the resulting product, to which methyl trifluoroacetate (7.68
g, 60 mmol) or phenyl trifluoroacetate (11.4 g, 60 mmol) was
added dropwise, and the mixture was stirred at room temper-
ature for 3 h. The resulting solution was filtered off, and the
solution was evaporated. The solid was dissolved in 100 mL
of 5 wt % Na2CO3 aqueous solution. Then concentrated HCl
with water was added to the solution for the neutralization.
The solution was filtered off, and water was evaporated. The
crude product was dried under vacuum for 12 h. The product
1
yield was 57%. H NMR (270 MHz, DMSO-d6) δ (TMS, ppm):
11.96 (s, 1H), 10.18 (s, 1H), 7.88 (m, 4H), 5.86 (s, 1H), 5.58 (s,
1H), 1.96 (s, 3H), 1.91 (s, 3H). Electron impact ionization (EI)
mass spectrum: Calculated for C12H14N2O4S: m/e 282.31.
Found: m/e 282. Elemental analysis (%): Calculated for
C
12H14N2O4S: C, 51.05; N, 9.92; S, 11.35. Found: C, 50.48; N,
9.86; S, 11.35.
Syn th esis of N-(4-Ben zoylim id esu lfon yl)p h en yla m in e
(BP A). Sulfanilamide (10.33 g, 60 mmol), K2CO3 (4.15 g, 30
mmol), and phenyl benzoate (11.89 g, 60 mmol) were com-
pletely mixed and put into a round-bottom flask. This mixture
was heated at 180 °C in an oil bath for 20 min with a
condenser. The flask was removed from the oil bath and was

5574 Rahman et al.
Macromolecules, Vol. 37, No. 15, 2004
Ta ble 1. Con ver sion s, Molecu la r Weigh ts, Molecu la r Weigh t Distr ibu tion s, Equ iva len t Weigh t, a n d Wa ter Up ta k e a t 25
°C of th e Mem br a n es
water uptake (wt %)
EW in feed
(g/mol H+)
EW of copolymersa
(g/mol H+)
polymer
conversion (wt %)
Mn (×105)
Mw/Mn
RH ) 30%
RH ) 98%
P(AN/APMA)
P(AN/BPMA)
P(AN/MBPMA)
P(AN/TBPMA)
P(AN/TBPMA)
P(AN/TPMA)
P(AN/PPMA)
50
54
47
46
43
36
45
4.08
4.85
2.09
3.06
2.53
2.39
1100
1139
1166
1590
1319
1343
1535
680
594
655
1353
764
666
1.66
1.83
1.25
1.41
1.16
0.64
0.60
9.33
10.82
16.21
10.13
8.60
4.08
3.85
1.64
2.25
3.66
2.16
4.59
2.05
1214
a
Equivalent weight was estimated from the relative peaks of phenylene protons of the acidic site to the R-proton of the nitrile group
in the copolymers in 1H NMR spectra.
was then purified by silica gel column chromatography eluted
with ethanol/benzene (1:2), and the solvent was evaporated.
The final product was recrystallized from acetone/ethanol (7:
3) and dried under vacuum to obtain TPA (6.60 g, 41% yield)
as a white powder.
Syn th esis of N-(4-(Tr iflu or oacetylim idesu lfon yl)ph en yl-
2-m eth yla cr yla m id e (TP MA). The TPA (6.0 g, 22 mmol) was
dissolved in acetonitrile and reacted with methacryloyl chlo-
ride, following the same procedure as APMA. The monomer
was recrystallized from acetonitrile and dried under vacuum
to obtain 3.10 g of a white crystalline powder. The yield was
41%. 1H NMR (270 MHz, DMSO-d6) δ (TMS, ppm): 9.99 (s,
1H), 7.74 (m, 4H), 5.82 (s, 1H), 5.54 (s, 1H), 1.94 (s, 3H). EI
mass spectrum: Calculated for C12H11F3N2O4S: m/e 336.28.
Found: m/e 336. Elemental analysis (%): Calculated for
impurities, especially unreacted acid monomers. The final
products were dried under vacuum at 80 °C for 24 h. A typical
experimental condition is as follows.
TBPMA (0.465 g, 1.25 mmol) and AN (1.521 g, 28.6 mmol)
were dissolved in DMF (1.5 g), and 0.1 mol % of AIBN was
added to this solution. After purging oxygen from the solution,
the copolymerization was conduced for 10 h at 60 °C under
shaking. The copolymer was isolated by precipitation with
methanol, washed with hot methanol, and dried to yield P(AN/
TBPMA) (0.93 g, conversion 46%).
The conversions for all the prepared copolymers are pre-
sented in Table 1. The copolymers were characterized by 1H
NMR spectra. The EWs of the copolymers were estimated from
the relative peak areas of phenylene protons of the acidic
monomer to the R-proton of acrylonitrile.
Gel P er m ea tion Ch r om a togr a p h y (GP C). The dried
copolymer was dissolved in a DMF/LiBr (2000 mL/1.737 g)
solution at a concentration of 2.5 mg/mL. The solution was
filtered through a syringe filter. GPC analyses were carried
out using Shodex GPC KD-806M columns (two columns
connected in series). A Shimadzu SPD-10AV UV detector,
operated at 254 nm, was used for detection. The DMF/LiBr
solution was used as the eluent at a flow rate of 1 mL/min
driven by a Shimadzu LC-10AD pump. The injection volume
was 100 µL. Narrow distribution polystyrene standards (Showa
Denko) with molecular weights ranging from 580 to 3.16 ×
106 were used for calibration. Since P(AN/TBPMA) (EW ) 1353
g/mol H+) was insoluble in the same solvent, the comparable
GPC result could not be obtained.
P r ep a r a tion of Mem br a n es. The membranes were cast
from copolymer solutions (10 wt %) in DMF.16 The solvent was
evaporated under atmospheric pressure at 50 °C for 12 h,
followed by heating at 100 °C under vacuum for 24 h. The
obtained membranes were stored in a desiccator over P2O5.9
The membranes were mechanically strong, flexible, and
transparent. The membrane, P(AN/PPMA), was a little brittle.
Wa ter Up ta k e Mea su r em en ts. Properly weighed dry
membranes were placed inside an oven to reach equilibrium
at various relative humidities. The membranes were then
weighed on a microbalance immediately after wiping off the
surface water, and the water uptake was calculated.16
P r oton Con d u ctivity Mea su r em en ts. Proton conductiv-
ity measurements were conducted using the two-probe tech-
nique.16 Prior to measurements, the cells with the membranes
were placed inside the humidifying chamber for at least 5 h
to reach equilibrium water uptake. The proton conductivity
was determined using the complex impedance method over the
frequency range from 5 Hz to 13 MHz.
C
12H11F3N2O4S: C, 42.86; N, 8.32; S, 9.53. Found: C, 41.16;
N, 8.20; S, 9.44.
Syn t h esis of N-(4-P en t a d eca flu or oh ep t ylim id esu l-
fon yl)p h en yla m in e (P P A). The synthesis of PPA followed
a similar procedure to that of TPA. The crude product was
purified by silica gel column chromatography eluted with
ethanol/benzene (1:2), and the solvent was evaporated to
obtain PPA (yield 37%) as a white powder.
Syn t h esis of N-(4-P en t a d eca flu or oh ep t ylim id esu l-
fon yl)p h en yl-2-m eth yla cr yla m id e (P P MA). PPMA was
synthesized in a similar way to TPMA. The monomer was
recrystallized from acetone/acetonitrile (1:1) and dried under
vacuum to obtain a white crystalline powder. The yield was
28%. 1H NMR (270 MHz, DMSO-d6) δ (TMS, ppm): 9.95 (s,
1H), 7.69 (m, 4H), 5.82 (s, 1H), 5.52 (s, 1H), 1.94 (s, 3H). EI
mass spectrum: Calculated for C18H11F15N2O4S: m/e 636.33.
Found: m/e 636. Elemental analysis (%): Calculated for
C
18H11F15N2O4S: C, 33.97; N, 4.40; S, 5.03. Found: C, 33.61;
N, 4.49; S, 4.85.
Acid Dissocia tion Con sta n t (p Ka ) of Mon om er s. The
acid dissociation constant (pKa) for an aqueous solution of
APMA was determined by the pH titration method. Since
BPMA was insoluble in water even in the salt form, the pKa
value for BPMA was estimated from the comparison of the
chemical shift of the peak corresponding to the acidic proton
1
of BPMA with that of APMA in the H NMR spectra. The pKa
values for MBPMA, TBPMA, and TPMA were calculated from
H+ activities of aqueous solutions of the monomers (1 mM)
determined by a pH meter. The pKa value of PPMA, which
was insoluble in water and also did not show NMR peaks in
1H NMR spectra using common deuterated solvents, was
approximated to be close to that of the structurally similar
acid monomer, TPMA.
P r ep a r a tion of Cop olym er s. Prior to polymerization,
acrylonitrile (AN) was deoxygenated with bubbling N2 gas for
20 min. The acid monomers and AN were dissolved in N,N-
dimethylformamide (DMF), and 0.1 mol % AIBN was added
to the solution as a radical initiator. N2 gas was then bubbled
for 5 min to the mixture to ensure an inert atmosphere. The
copolymerization was carried out at 60 °C with continuous
shaking for 10 h in the case of P(AN/APMA), P(AN/BPMA),
P(AN/MBPMA), P(AN/TBPMA), and P(AN/TPMA) and for 30
h for P(AN/PPMA). The resulting copolymers were diluted with
DMF, precipitated in excess methanol, and filtered. The
copolymers were then washed with hot methanol to remove
Th er m al An alysis. Differential scanning calorimetry (DSC)
measurements were performed for dry and hydrated mem-
branes on a Seiko Instruments differential scanning calorim-
eter (DSC 220C). To hydrate the membranes, 10-12 mg of
the previously dried sample was placed in deionized water for
about 6 h, followed by blotting to remove surface water. The
samples were immediately transferred to aluminum pans and
tightly sealed. The thermograms were recorded during cooling
(+30 to -120 °C) and the following heating (-120 to +300
°C) scans at a cooling or heating rate of 10 °C min-1. High-
temperature stabilities for the samples were measured on a
Seiko Instruments thermogravimetry/differential thermal ana-

Macromolecules, Vol. 37, No. 15, 2004
Novel Acid Monomers 5575
lyzer 6200 (TG/DTA) from 30 to 550 °C at a heating rate of 10
°C min-1 under a N2 atmosphere with open aluminum pans.
Mem br a n e Electr od e Assem bly (MEA) P r ep a r a tion .
Carbon paper was cut into a square shape (2 × 2 cm2). The
catalyst slurry was prepared by mixing Pt/C and a sulfonimide
polymer solution (5 wt %) in DMSO. The resulting solution
was sonicated for 30 min and mixed using a high-speed
homogenizer for 5 min. Then the catalyst slurry was pasted
on the carbon paper. The electrodes were heated under an
atmospheric pressure at 50 °C, followed by heating under
vacuum at 80 °C for 24 h to evaporate the solvent. The catalyst
loading on the electrodes was ca. 1 mg cm-2. The membrane
electrode assembly was fabricated by sandwiching a copolymer
membrane between the two catalyst-loaded carbon papers,
followed by hot-pressing the assembly at 50 °C and 6 MPa for
1 min.
Sch em e 2. Str u ctu r es of Acid Mon om er s a n d
Cop olym er s
F u el Cell Eva lu a tion . The MEA was placed in a 4 cm2
single cell test fixture. The gas pressure, flow rate, humidifi-
cation, and temperature were controlled using an Eiwa MET-S
fuel cell humidification system. Pure H2 and O2 were provided
at the anode and cathode, respectively. During the fuel cell
operation, the test cell was operated at a constant temperature,
and fully humidified H2 and O2 were supplied to the anode
and cathode, respectively, at a rate of 200 mL min-1. The cell
was maintained under open-circuit conditions until reaching
a constant value. The current-voltage characteristics were
then monitored with waiting time of 1 min for each measuring
point.
Resu lts a n d Discu ssion
Syn th esis a n d Ch a r a cter iza tion of th e Mon o-
m er s a n d Cop olym er s. The general synthetic route
of the monomers is outlined in Scheme 1. The sulfanil-
amide has two amino groups in its structure. The
nucleophilicity of the two groups is higher for the group
directly attached to the phenylene group and lower for
that attached to the sulfonyl group. To perform selective
nucleophilic substitution to the lower reactive group, the
reaction was carried out in the molten state at a higher
temperature than the melting point of sulfanilamide
(165 °C) in the presence of K2CO3, which dissolved in
the molten sulfanilamide. The base, K2CO3, reacts with
more acidic sulfonamide to yield the imide anion
(-SO2N-H), which can be confirmed by evolution of
CO2. The nucleophilicity, now, becomes higher than that
of the amino group directly attached to the phenylene
group, which makes the selective nucleophilic substitu-
tion toward the esters and sulfonates. By changing R
of the esters and sulfonates (Scheme 1), six different
aniline derivatives could be obtained, and the successive
reaction with methacryloyl chloride yielded the six
different acid monomers.
It is interesting to note that acidities of the sulfon-
imide monomers (Scheme 2) can be controlled by chang-
ing the substituted groups on the sulfonimides. The
sulfonimide monomers show different acidities with a
wide difference in the pKa values, depending on the
electron-withdrawing nature of the substituted
groups.16,17,21 The monomers APMA (pKa ) 4.95) and
BPMA (pKa < 4.95) are weak acids in nature. The acidic
proton of water-insoluble BPMA showed a downfield
chemical shift compared to that of the acidic proton of
APMA in the NMR spectra, which is indicative of the
relatively higher acidity of BPMA compared to that of
APMA. Although the determination of pKa values for
strong acids is experimentally very difficult, and the
estimation in this study is simply based on pH mea-
surements of the 1 mM solutions, MBPMA, TBPMA,
TPMA, and PPMA are found to be strong or superstrong
acids (pKa < 1).16 The bis(sulfonyl) links in the imide
group make the acidic proton of MBPMA and TBPMA
highly dissociable in aqueous solutions, resulting in
strong acidities of the monomers.22 TBPMA seems to
have the highest acidity (lowest pKa) among the six
different monomers.
The acid monomers were copolymerized with AN
monomer (Scheme 2) and characterized by 1H NMR
spectra. The 1H NMR spectra correspond to the expected
molecular structures of the copolymers. The monomer
compositions in the feed for the copolymerization and
the copolymer compositions are shown in Table 1 as EW
in the feed and EW of the copolymers, respectively. It
can be noticed that the acid monomer compositions in
the copolymers become higher than those in the feed
for all the copolymers, depending on monomer reactivity
ratio of the monomers, though the precise determination
has not been conducted. Table 1 also lists the molecular
weights and molecular weight distributions of the
prepared copolymers determined by GPC. The number-
average molecular weights (Mn) of the copolymers are
high, indicating high radical copolymerization reactivity
of the acid monomers.
Th er m a l Beh a vior of th e Mem br a n es. Thermal
properties for the dry membranes and hydrated mem-
branes are tabulated in Table 2. All the membranes
have reasonable thermal stability up to 250 °C. How-
ever, below the degradation temperature (Td), all the
copolymers show an exothermic peak (Tc) in the DSC
thermograms, ranging from 179 to 195 °C. This corre-
sponds to the temperature for the commencement of
cyclization for the acrylonitrile moiety in the copoly-
mers.16 The membranes are amorphous and do not show
any melting temperature in the DSC thermograms.
Instead, the dry membranes form homogeneous glasses
with a single Tg, ranging from 54 to 124 °C, depending

5576 Rahman et al.
Macromolecules, Vol. 37, No. 15, 2004
Ta ble 2. Th er m a l P r op er ties a n d P r oton Con d u ctivity of th e Mem br a n es
Tg (°C)b σ at RH ) 98% (S cm-1
hydrated 80 °C
)
polymer
EWa (g/mol H+)
dry
Tc (°C)c
Td (°C)d
40 °C
P(AN/APMA)
P(AN/BPMA)
P(AN/MBPMA)
P(AN/TBPMA)
P(AN/TBPMA)
P(AN/TPMA)
P(AN/PPMA)
680
594
655
1353
764
666
92
96
54
71
71
66
83
e
35
54
68
96
179
182
184
187
187
195
185
271
279
297
267
316
287
318
1.2 × 10-4
1.7 × 10-3
4.2 × 10-3
1.3 × 10-3
2.3 × 10-2
8.7 × 10-3
8.4 × 10-4
1.2 × 10-4
2.1 × 10-4
2.9 × 10-2
2.3 × 10-3
4.0 × 10-2
3.7 × 10-3
2.1 × 10-5
85
124
1214
a
Equivalent weight was estimated from the relative peaks of phenylene protons of the acidic site to the R-proton of the nitrile group
in the copolymers in 1H NMR spectra. Onset temperatures of a heat capacity change of dry membrane and hydrated membrane,
b
respectively (glass transition temperature, Tg). c An exothermic peak (cyclization temperature, Tc). Temperature of 10% weight loss
d
during heating scans from room temperature by using thermogravimetry. e No Tg was observed.
on the structures. With hydration of the membranes,
the Tg is lowered compared with that of the dry
membranes, ranging from 35 to 96 °C (Table 2). How-
ever, no melting endotherm of water is observed in the
thermograms, indicating that water molecules in the
membranes are highly hydrated to the acidic groups.
The hydrated membrane, P(AN/MBPMA), does not
show any Tg but showed an endothermic peak at ca. 0
°C, corresponding to melting of the free water.20 This
seems to be due to the largest water uptake among the
membranes (vide infra). The fraction of the free water
to the total adsorbed water is found to be 45 wt % from
the DSC result.
Wa ter Up ta k e of th e Mem br a n es. The results of
the adsorption isotherm for water uptake for all of the
membranes at 25 °C are presented in Table 1. The water
uptake by all the synthesized membranes is relatively
F igu r e 1. Temperature dependence of proton conductivities
low compared with the membranes bearing the same
for the membranes, P(AN/MBPMA) and P(AN/TBPMA), with
acid functional groups but different backbone.8,10 This
different equivalent weights (EW) at 98% relative humidity.
is possibly due to the hydrophobicity of the acid mono-
mers. The water uptake was increased for all the
membranes with increasing relative humidity (Table 1).
Water uptake by the nonfluorinated acid membranes,
P(AN/APMA), P(AN/BPMA), and P(AN/MBPMA), be-
comes higher compared with those with higher equiva-
lent weights.16,23 The results suggest that hydrophilicity
of the membrane increases with increasing the acid
monomer compositions. On the contrary, the fluorinated
membranes, P(AN/TBPMA), P(AN/TPMA) and P(AN/
PPMA), adsorbed a lower amount of water compared
with those with higher equivalent weights.16 The lower
water uptake may be due to the increase in the
hydrophobicity. The replacement of the nonfluorinated
substituted groups by the highly electron-withdrawing
fluorinated substituted groups causes an increase in the
acidity, while it simultaneously causes significant in-
crease in the hydrophobicity. The water uptake seems
to be determined by the change in the amphiphilicity
with the composition of the acid monomers. The lowest
water uptake is observed in the membrane, P(AN/
PPMA), due to its long fluorinated side chain. The
highest water uptake of 16 wt % is observed for the
P(AN/MBPMA) membrane at 98% relative humidity.
P r oton Con d u ctivity of th e Mem br a n es. Proton
conductivities of the hydrated membranes were mea-
sured in the temperature range of 40-95 °C. The
conductivity data are tabulated in Table 2. All the
membranes exhibit reasonable proton conductivity un-
der the humidifying conditions, while the conductivity
membranes, P(AN/APMA) and P(AN/BPMA), and the
strongly acidic but hydrophobic membranes, P(AN/
TPMA) and P(AN/PPMA), one can recognize that the
conductivities are comparable. The meager water up-
take of the latter membranes cannot afford high con-
ductivity, even if the acidity is high. Furthermore, the
conductivities of the latter membranes decrease, whereas
those of the former increase, with an increase in the
temperature from 40 to 80 °C. The high proton conduc-
tivity is achieved by the membranes, P(AN/MBPMA)
and P(AN/TBPMA), with high acidities of the acid
monomers as well as the high water uptake. The
conductivities of the membranes become higher than
10-2 S cm-1 at 80 °C.
Figure 1 shows the temperature dependence of the
proton conductivity for P(AN/MBPMA) and P(AN/
TBPMA), having different EWs. Both of the membranes
show higher proton conductivities, when the membrane
has a lower EW. This comparison suggests that the
conductivity correlates with the composition of the acid
monomers in the copolymers. In addition, at a similar
EW, P(AN/TBPMA) exhibits a higher conductivity than
that of P(AN/MBPMA), being attributable to the stron-
ger acidity of TBPMA. The proton conductivity for these
membranes is high, though their water uptake is low,
compared with the membranes having the same acid
group but different backbone, which might be due to
relatively strong interactions with the acid groups or
an internal pore and/or a channel structure that is more
favorable for proton conduction.2,8 Although the tem-
perature dependences of proton conductivity are shown
only for the P(AN/MBPMA) and P(AN/TBPMA) mem-
branes (Figure 1), the proton conductivity of the all of
for dry membranes was 10-9-10-8 S cm-1 16
The proton
.
conductivity is influenced not only by acidity of the acid
monomers but also by the water uptake. If a comparison
of the conductivities is made between the weak acid

Macromolecules, Vol. 37, No. 15, 2004
Novel Acid Monomers 5577
The proton conductivity was measured for the hydrated
membranes with a variation of EWs. The lower EW
membranes show higher proton conductivity than the
higher EW membranes. The MEA fabrication and fuel
cell test were performed using a P(AN/TBPMA) mem-
brane. In a single cell test, electric power generation
was possible at 24 and 49 °C, indicating the possible
applicability to fuel cell electrolyte membranes.
Ack n ow led gm en t. This research was supported in
part by Grant-in-Aid for Scientific Research (#404/
11167234 and #14350452) from the J apanese Ministry
of Education, Science, Sports, and Culture and by
NEDO Technology Research Grant. The authors also
acknowledge the kind experimental support by Sun-
J ung Hwang.
Refer en ces a n d Notes
F igu r e 2. Potential-current polarization curves of an H2/O2
fuel cell using a P(AN/TBPMA) membrane at 24 and 49 °C.
(1) Pourcelly, G.; Gavach, C. In Proton Conductors, Solids,
Membranes and GelssMaterials and Devices; Colomban, P.,
Ed.; Cambridge University Press: Cambridge, 1992.
(2) Rikukawa, M.; Sanui, K. Prog. Polym. Sci. 2000, 25, 1463.
(3) Savadogo, O. J . New Mater. Electrochem. Syst. 1998, 1, 47.
(4) Koppel, I. A.; Taft, R. W.; Anvia, F.; Zhu, S.-Z.; Hu, L.-Q.;
Sung, K.-S.; DesMarteau, D. D.; Yagupolskii, L. M.; Yagupol-
skii, Y. L.; Ignat’ev, N. V.; Kondratenko, N. V.; Volkonskii,
A. Y.; Slasov, V. M.; Notario, R.; Maria, P.-C. J . Am. Chem.
Soc. 1994, 116, 3047.
(5) Razaq, M.; Razaq, A.; Yeager, E.; DesMarteau, D. D.; Singh,
S. J . Appl. Electrochem. 1987, 17, 1057.
(6) Razaq, M.; Razaq, A.; Yeager, E.; DesMarteau, D. D.; Singh,
S. J . Electrochem. Soc. 1989, 136, 385.
(7) Appleby, A. J .; Velev, O. A.; LeHelloco, J .-G.; Parthasarthy,
A.; Srinivasan, S.; DesMarteau, D. D.; Gillette, M. S.; Ghosh,
J . K. J . Electrochem. Soc. 1993, 140, 109.
(8) Sumner, J . J .; Creager, S. E.; Ma, J . J .; DesMarteau, D. D.
J . Electrochem. Soc. 1998, 145, 107.
(9) Feiring, A. E.; Choi, S. K.; Doyle, M.; Wonchoba, E. R.
Macromolecules 2000, 33, 9262.
(10) Hofmann, M. A.; Ambler, C. M.; Maher, A. E.; Chalkova, E.;
Zhou, X. Y.; Lvov, S. N.; Allcock, H. R. Macromolecules 2002,
35, 6490.
(11) DesMarteau, D. D.; Martin, C. W.; Ford, L. A.; Xie, Y.
Sulfonated Perfluorovinyl Functional Monomers, US Patent
6, 268, 532 B1, 2001.
(12) Ford, L. A.; Smith, D. W.; DesMarteau, D. D. Abstr. Pap. Am.
Chem. Soc. 2000, 220, 7.
(13) Ford, L. A.; Smith, D. W.; DesMarteau, D. D. Abstr. Pap. Am.
Chem. Soc. 2000, 220, 208.
(14) Guo, X.; Fang, J .; Watari, T.; Tanaka, K.; Kita, H.; Okamoto,
K. Macromolecules 2002, 35, 6707.
(15) Zhou, X.; Weston, J .; Chalkova, E.; Hofmann, M. A.; Ambler,
C. M.; Allcock, H. R.; Lvov, S. N. Electrochim. Acta 2003, 48,
2173.
(16) Rahman, M. K.; Aiba, G.; Susan, M. A. B. H.; Watanabe, M.
Electrochim. Acta 2004, in press.
(17) Park, S. Y.; Bae, Y. H. Macromol. Rapid Commun. 1999, 20,
269.
membranes in this study greatly decreases at above 60-
90 °C. The weak acid membranes tend to decrease the
conductivity at lower temperatures. This decrease seems
to be related to the adsorbed water property rather than
Tg of the hydrated membranes.
F u el Cell P er for m a n ce. Preliminary experiments
to explore the applicability of the present copolymer
membranes to fuel cells were conducted. Figure 2 shows
the current vs potential curve of an H2/O2 fuel cell using
a P(AN/TBPMA) membrane with EW ) 1353 g/mol H+
and 90 µm thickness at 24 and 49 °C. Although this is
just a preliminary experiment, we can confirm that
electric power generation by the fuel cell using the
present copolymer membrane is possible. At 80 °C, a
stable polarization curve for the fuel cell could not be
obtained. This may be due to weak hydrolysis stability
of the membrane. The hydrolysis problem became more
serious for a P(AN/TBPMA) membrane with the highest
conductivity (a lower EW ) 764 g/mol H+), and it was
impossible to obtain stable polarization curves, espe-
cially at higher temperatures. As evidence, we obtained
1
the H NMR spectrum of the membrane after the fuel
cell test and observed some additional peaks which did
1
not correspond to the H NMR spectrum of the original
structure. It is clarified by a hydrolysis stability test for
the monomers by using 1H NMR in water that the
-SO2-NH-SO2- group has much higher hydrolysis
stability than the -SO2-NH-CO-. However, even for
the monomers having the -SO2-NH-SO2- groups
(MBPMA, TBPMA), the hydrolysis at the amide group,
-CONH-, occurs at a high temperature. The hydrolysis
stability of the monomers should be increased in our
future study.
(18) Solan, A. D. B. J . Appl. Chem. Biotechnol. 1973, 23, 251.
(19) Yoda, N.; Marvel, C. S. J . Polym. Sci., Part A 1965, 3, 2229.
(20) Kim, Y. S.; Dong, L.; Hickner, M. A.; Glass, T. E.; Webb, V.;
McGrath, J . E. Macromolecules 2003, 36, 6281.
(21) Bell, P. H.; Roblin, R. O., J r. J . Am. Chem. Soc. 1942, 64,
2905.
Con clu sion
A series of novel acid monomers with different acid
groups based on sulfonimides were successfully synthe-
sized and characterized. The acid monomers were
copolymerized with acrylonitrile monomer. The mem-
branes were prepared from the resulting copolymers.
(22) Foropoulos, J ., J r.; DesMarteau, D. D. Inorg. Chem. 1984,
23, 3720.
(23) Bae, J .-M.; Honma, I.; Murata, M.; Yamamoto, T.; Rikukawa,
M.; Ogata, N. Solid State Ionics 2002, 147, 189.
MA0498058