Synthesis and properties of sulfonated copoly(p-phenylene)s containing aliphatic alkyl pendant for fuel cell applications

Article abstract of DOI:10.1016/j.polymer.2009.12.023

A series of novel copoly(p-phenylene)s (PPs) containing an alkyl pendant were successfully synthesized via Ni(0)-catalyzed coupling polymerization. Sulfonated copolymers (SPPs) were achieved by postsulfonation from concentrated H₂SO₄. SPPs showed good solubility in polar aprotic solvents and gave flexible, tough, and transparent free-standing films by solvent casting. The ion exchange capacities (IECs) of the membranes ranged from 2.50 to 2.65?meq/g. All SPP membranes displayed proton conductivity similar to or higher than that of Nafion, especially at high relative humidity (>70% RH) (SPP-1: 0.271 Scm^-1, SPP-2: 0.284 Scm^-1, SPP-3: 0.212?S?cm^-1, Nafion: 0.127 Scm^-1; at 80?°C and 95% RH). They also exhibited acceptable water uptake in the range of 52-56?vol% at 80?°C with little dimensional change. The gas permeability of the SPP membranes was much lower than that of Nafion 112. Therefore, these materials are promising for fuel cell application.

Full text of DOI:10.1016/j.polymer.2009.12.023

Polymer 51 (2010) 623–631
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Synthesis and properties of sulfonated copoly(p-phenylene)s containing
aliphatic alkyl pendant for fuel cell applications
Surasak Seesukphronrarak, Kayo Ohira, Koh Kidena, Naohiko Takimoto,
*
Carlos Seiti Kuroda, Akihiro Ohira
FC-Cubic, National Institute of Advanced Industrial Science and Technology (AIST), Tokyo 135-0064, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel copoly(p-phenylene)s (PPs) containing an alkyl pendant were successfully synthesized via
Ni(0)-catalyzed coupling polymerization. Sulfonated copolymers (SPPs) were achieved by postsulfonation
from concentrated H2SO4. SPPs showed good solubility in polar aprotic solvents and gave flexible, tough,
and transparent free-standing films by solvent casting. The ion exchange capacities (IECs) of the
membranes ranged from 2.50 to 2.65 meq/g. All SPP membranes displayed proton conductivity similar to
or higher than that of Nafion, especially at high relative humidity (>70% RH) (SPP-1: 0.271 Scmꢀ1, SPP-2:
0.284 Scmꢀ1, SPP-3: 0.212 S cmꢀ1, Nafion: 0.127 Scmꢀ1; at 80 ꢁC and 95% RH). They also exhibited
acceptable water uptake in the range of 52–56 vol% at 80 ꢁC with little dimensional change. The gas
permeability of the SPP membranes was much lower than that of Nafion 112. Therefore, these materials are
promising for fuel cell application.
Received 5 October 2009
Received in revised form
18 December 2009
Accepted 21 December 2009
Available online 4 January 2010
Keywords:
Polymer electrolyte membrane
Dimensional stability
Proton conductivity
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
membrane. However, an HC-type ionomer membrane with
a random copolymer system needs to have a high IEC to achieve
In recent years, polymer electrolyte membranes (PEMs) have
played an important role as a key component of polymer electrolyte
fuel cells (PEFCs) for transport, stationary, and portable power
sources. Among PEMs, the perfluorinated sulfonic acid (PFSA)
ionomer introduced by DuPont under the registered trademark
Nafion in 1966 is the current state-of-the-art PEM in commercial
systems owing to its excellent mechanical properties, good chem-
ical stability, and high proton conductivity. However, its high cost,
high gas permeability, and relatively low conductivity at high
temperatures have limited its widespread commercial and practical
use in PEFCs [1–3]. Therefore, considerable effort has been devoted
to developing an alternative proton-conductive membrane. To date,
many types of sulfonated hydrocarbon (HC)-type ionomer
membrane based on high-performance aromatic polymers, such as
sulfonated polyimides [4–7], sulfonated poly(ether ketone)s [8–11],
sulfonated poly(arylene ether sulfone)s [12–15], sulfonated poly-
benzimidazoles [16], and sulfonated poly(p-phenylene)s [17–21],
have been extensively developed as candidate PEM materials. In
general, the HC-type ionomer membrane is considered to offer
greater structural and thermal stability than the PFSA ionomer
high proton conductivity, as high as that of Nafion, resulting in
unfavorable excess water swelling of the membrane and loss of
physical properties. Some approaches that have been applied to
solving these problems include the blending and cross linking of
ionomer membranes [22–24], the radiation grafting of polymer
membranes [25,26], and the development of multiblock copolymer
membranes composed of sulfonated hydrophilic and non-
sulfonated hydrophobic [27–30].
Recently, investigations of sulfonated poly(p-phenylene)s, con-
taining sulfonic acid groups on flexible pendant side chains, and
their derivatives have received much attention as candidate PEFC
materials owing to their excellent thermal and hydrolytic stabili-
ties, and high proton conductivities comparable to those of Nafion.
McGrath et al., for example, reported a synthesis of sulfonated
poly(2,5-benzophenone) and its derivative by nickel(0) catalytic
coupling polymerization and followed by sulfonation with sulfuric
acid or by substitution of activated fluoro groups in the polymers
side chain. The resulting polymers showed good solubility in
dipolar solvents but bad film-forming ability, probably owing to
their rigid-rod backbone, which limit their industrial application in
PEFCs [18]. Sulfonated poly(p-phenylene)s should have a high
molecular weight for good handling and mechanical stabilities;
however, there are only a few reports on sulfonated poly(p-phe-
nylene)s with high molecular weights owing to the difficulty in the
* Corresponding author. Tel.: þ81 3 3599 8550; fax: þ81 3 3599 8554.
E-mail address: a-oohira@aist.go.jp (A. Ohira).
0032-3861/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2009.12.023

624
S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
synthesis of such materials. On the basis of the above results, we
have succeeded in the fabrication of membranes with good
swelling control even with increasing water content, without
obstructing the proton conduction as well as enhance membrane
flexibility by introducing comonomer unit that possesses the long
alkyl side chain in a random copolymer.
In the present work we have designed and prepared a series of
sulfonated copoly(p-phenylene)s containing aliphatic alkyl pendant
side chain by Ni(0) catalyzed coupling polymerization. Sulfonated
poly(p-phenylene)s can be synthesized by postsulfonation from
concentrated H2SO4. It is expected that the introduction of aliphatic
alkyl side chain in copoly(p-phenylene)s backbones will result in
good solubility in organic solvents, improved film-forming ability,
good water stability, and high proton conductivity. The properties of
the synthesized copolymer membranes such as solubility, IEC,
water uptake, thermal stability, dimensional change, proton
conductivity, and gas permeability were investigated.
a three-neck flask. The reaction mixture was cooled in an ice/water
bath. Anhydrous AlCl3 (1 eq) was added gradually to the reaction
mixture whose temperature was kept below 10 ꢁC. After adding all
the AlCl3, the reaction solution was allowed to slowly warm to room
temperature and stirred overnight. The reaction was stopped by
pouring the solution into acidic ice water. The resulting two layers
were separated and the organic layer was washed with 10% aqueous
NaOH and water, and dried over MgSO4. After filtration, the solvent
was removed and the residue was purified by column chromatog-
raphy (CH2Cl2/hexane) or recrystallization.
2.3.1. 2,5-dichloro-4-phenoxybenzophenone (M1)
From 2,5-dichlorobenzoyl chloride and diphenylether: White
crystalline solid (yield 85%) 1H NMR(500 MHz, CDCl3):
d
(ppm) 7.77
(d, 2H), 7.41–7.37 (m, 4H), 7.33 (s, 1H), 7.20 (dd, 1H), 7.08 (dd, 2H),
6.99 (dd, 2H). 13C NMR (100 MHz, CDCl3)
192.2, 163.1, 155.1, 140.2,
d
133.0,132.6,131.3,131.0,130.3,130.2,129.5,128.8,125.0,120.6,117.3.
2. Experimental
2.3.2. 2,5-dichloro-4-propylbenzophenone (M2)
From 2,5-dichlorobenzoyl chloride and n-propylbenzene:
2.1. Materials
Colorless liquid (yield 90%) 1H NMR (500 MHz, CDCl3):
d
(ppm) 7.71
(d, 2H), 7.38 (s, 2H), 7.33 (s, 1H), 7.27 (d, 2H), 2.66 (t, 2H), 1.66 (q,
2H), 0.95 (t, 3H). 13C NMR (100 MHz, CDCl3)
193.3, 149.9, 140.3,
All reagents and solvent were purchased from Aldrich and used
without further purification. Triphenylphosphine was purified by
recrystallization from n-hexane. N-Methyl-2-pyrrolidinone (NMP,
anhydrous grade from Aldrich) was also used as received. 2,5-
Dichlorobenzoyl chloride was prepared from 2,5-Dichlorobenzoic
acid with thionyl chloride (SOCl2). Sulfonation of PEEK was carried
out using 96% H2SO4 for 72 h according to a procedure reported
elsewhere [31]
d
133.6, 132.9, 131.3, 130.9, 130.3, 129.6, 128.9, 128.8, 38.3, 24.3, 13.9.
2.3.3. 2,5-dichloro-4-dodecylbenzophenone (M3)
From 2,5-dichlorobenzoyl chloride and n-dodecylbenzene: light
yellow liquid (yield 92%) 1H NMR (500 MHz, CDCl3):
d
(ppm) 7.70 (d,
2H), 7.38 (s, 2H), 7.33 (s, 1H), 7.28 (d, 2H), 2.68 (t, 2H), 1.63 (t, 2H),
1.25 (m, 18H), 0.87 (t, 3H). 13C NMR (100 MHz, CDCl3)
193.3, 150.2,
d
140.3, 133.6, 132.9, 131.3, 130.9, 130.4, 129.6, 128.9, 36.3, 32.0, 31.1,
29.7, 29.6, 29.4, 22.8, 14.3.
2.2. Characterization
Infrared spectra were recorded on a Varian FTS-7000 FT-IR
spectrometer. 1H and 13C NMR spectra were obtained on JEOL ECA-
500 using CDCl3 as the solvent with tetramethylsilane as the
internal reference. The molecular weight of the synthesized poly-
mers were estimated by gel permeation chromatography (GPC)
equipped with two consecutive columns (GF-7M HQ and GF-310
HQ, Asahipak) connected to a Shimadzu SPD-20AU UV detector at
40 ꢁC using polystyrene standards and dimethylformamide
containing 0.05 M LiBr as the eluent at a flow rate of 0.05 mL/min.
UV–vis absorption spectra were measured with a Spectrometer
(MCPD-7000, Otsuka Electronics Co., Ltd.) equipped with an optical
fiber. Thermogravimetric analysis (TGA) was performed with the
TGA-50 analyzer (Shimadzu). The polymer samples were dried at
100 ꢁC prior to the experiments and heated from room temperature
to 800 ꢁC at a heating rate of 10 ꢁC/min in nitrogen atmosphere
(flow rate, 50 mL/min). Tensile measurement was achieved and
analyzed at the Toray Research Center (Shiga, Japan) using InstronÒ
Model 5848 Micro Tester (Instron Co., Ltd.). The test speed of
2 mm/min, the size of specimen is 15 mm ꢂ 4 mm. For each testing,
three measurements at least were recorded and the average value
was calculated.
2.3.4. 2,5-dichloro-4-octadecylbenzophenone (M4)
From 2,5-dichlorobenzoyl chloride and n-octadecylbenzene:
Pale yellow semi-solid (yield 96%) 1H NMR (500 MHz, CDCl3):
d
(ppm) 7.71 (d, 2H), 7.38 (s, 2H), 7.33 (s, 2H), 7.25 (d, 2H), 2.67 (t,
2H), 1.63 (t, 2H), 1.25 (m, 30H), 0.87 (t, 3H). 13C NMR (100 MHz,
CDCl3) 193.3, 150.2, 140.3, 133.6, 132.9, 131.2, 130.9, 130.3, 129.6,
d
128.9, 36.3, 32.0, 31.1, 29.8, 29.6, 29.5, 29.4, 22.8, 14.2.
2.4. Polymerization of poly(p-phenylene)s (PPs)
As shown in Scheme 1, polymers were synthesized via Ni(0)
catalytic polymerization according to a procedure described in
previous reported with some modification [32,33]
A typical
example of polymerization is as follows. In a 100-mL three-neck
round-bottom flask, Ni(PPh3)2Cl2 (0.122 g, 0.186 mmol), Zn (0.654 g,
10 mmol), PPh3 (0.524 g, 2 mmol), and NaI (0.088 g, 0.058 mmol)
were charged into a flask under argon atmosphere. Dry NMP (6 mL)
was added to the flask via a syringe, and the mixture was stirred at
50 ꢁC for 10–20 min. A deep-red color was observed. A monomer
(5 mmol) dissolved in 3 mL of NMP (M1:M2, M3, or M4 was
4.5:0.5 mmol) was added via a syringe. The mixture was stirred at
75 ꢁC for 4 h. The resulting mixture was cooled, and then poured in
10% HCl/acetone to precipitate a polymer. The polymer was
collected by filtration, washed with acetone, and dried at 80 ꢁC for
24 h under reduced pressure to give a product in 90% yield.
2.3. Monomer synthesis
The monomers M1, M2, M3, and M4 were prepared in one step
by the Friedel-Crafts acylation of diphenylether, n-propylbenzene,
n-dodecylbenzene, and n-octadecylbenzene, respectively, with 2,5-
dichlorobenzoyl chloride according to a previously report with
some modifications [32–34].
2.5. Preparation of sulfonated SPPs
The general procedure for the preparation of monomers is as
follows. A solution of aromatic substrate (1 eq) and 2,5-dichloro-
benzoyl chloride (1.1 eq) in dichloromethane (5 eq) was charged into
The postsulfonation of the synthesized copolymers was carried
out using concentrated H2SO4 (98%) as the sulfonating agent at
room temperature for 72 h.

S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
625
a
O
C
O
O
O
O
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
or
or
M1
M4
M3
M2
(1) Ni(PPh3)2Cl2, Zn
PPh3, NaI
NMP
(2) conc. H2SO4
b
HO3S
HO3S
HO3S
O
O
O
C
O
O
C
O
C
O
O
O
0.1
0.1
0.1
0.9
SPP-3
0.9
SPP-1
Scheme 1. Chemical structures: (a) candidate monomers and (b) sulfonated copoly(p-phenylene)s.
0.9
SPP-2
In a typical reaction, 1.0 g of the synthesized copolymers was
dissolved in 30 mL of concentrated H2SO4 (96%). The solution was
maintained at room temperature for 72 h before inducing precip-
itation in a large amount of water at room temperature. The
sulfonated polymers were filtered, washed thoroughly with
deionized water, and dried at 80 ꢁC for 24 h under vacuum. The
following results were obtained: FT-IR (KBr) 1653 cmꢀ1 (C]O),
1586 and 1491 cmꢀ1 (C]C, aromatic), 1123 and 1031 cmꢀ1
(asymmetric and symmetric stretching of –SO3H).
HCl at room temperature for 24 h, followed by washing thoroughly
with deionized water at room temperature, and dried under
vacuum at 100 ꢁC overnight. The thickness of all the membrane
samples was in the range of 40–70 mm.
2.7. Ion exchange capacity
The ion exchange capacity of the copolymer membranes was
determined by a titration method. Polymer membranes in acid
form were dried overnight at 100 ꢁC under vacuum, weighed, and
immersed in saturated NaCl for 24 h. The amount of Hþ released
from the membrane samples was determined by titration with
0.01 M NaOH solution using phenolphthalein as the indicator.
2.6. Membrane preparation
Polymer membranes were prepared by solution casting from
DMF. The dried polymer was dissolved in DMF to form a 5% solu-
tion, and filtered at room temperature. The membranes were
directly casted on a glass plate and dried on a hot plate at 60 ꢁC for
2 h and then at 80 ꢁC in under vacuum for 24 h to give a tough and
flexible membrane. The membranes were then immersed in 1 M
IEC ¼ ðC ꢂ VÞ=M
Here, C and V are the concentration and volume of NaOH, respec-
tively. M is the weight of the membrane.

626
S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
2.8. Dimensional change
chromatograph (G2700T, Yanaco) by monitoring the amounts of H2
and O2 that permeated across the membrane (dehydrated and stored
in dry N2), as detailed elsewhere [36]. The relative humidity depen-
dences of H2 and O2 permeability coefficients P were measured from
0 to 90% RH at 80 ꢁC using argon and helium as the carrier gases of H2
The dimensional changesof the membranes were measuredinthe
thickness and plane directions by immersing the membrane samples
in deionized water at room temperature and 80 ꢁC for a given time.
The changes in thickness and length were calculated using
and O2, respectively. A 40w70-mm-thick membrane was set in a cell
with a gas inlet and a gas outlet, where temperature was controlled.
H2 and O2 were supplied at a flow rate of 30 mL/min. The gas
permeability coefficient P (cm3 (STP) cm cmꢀ2 sꢀ1 cmHgꢀ1) was
calculated according to the following equation:
D
t ¼ ðt ꢀ tsÞ=ts;
l ¼ ðl ꢀ lsÞ=ls
D
where ts and ls are the thickness and length of the dry sample,
respectively; t and l refer to those of the membrane immersed in
water for 5 h.
273
T
1
A
1
60
B
C
106
P ¼
ꢂ
ꢂ d ꢂ
ꢂ V ꢂ
ꢂ
76 ꢀ PH2O
where T (K) is the absolute temperature of the cell, A (cm2) is the
permeation area, V is the sampling tube volume (cm3; 4.03 for H2
and 4.17 for O2) B (cm3/min) is the flow rate of the test gas, C is the
volume of gas (cm3) that permeated through the membrane as
evaluated by the absolute calibration method, d (cm) is the thick-
ness of the membrane, and PH2O (cmHg) is the water vapor
pressure.
2.9. Membrane morphology
Membrane morphology was investigated by atomic force micros-
copy (JSPM-5400, Nihon Denshi) with a humidity control unit [35]. A
Pt-coated cantilever (NSC05/Pt_20, NT-MDT) with a force constant of
12 Nmꢀ1 and a resonance frequency of 250 kHz was used. Membrane
samples were placed on a gold-plated conductive sample stage with
Nafion dispersion (DE-2020, DuPont) as adhesive. Prior to AFM
observations, samples were placed in a humidity-controlled chamber
forat least 1 h. Bias voltagewas applied tothe samplestageduringAFM.
Height and current-mapping images were simultaneously obtained.
3. Results and discussion
3.1. Monomer and polymer synthesis
The chemical structure of the monomer considered here is
shown in Scheme 1. All monomers prepared in one step by the
Friedel-Crafts catalytic reaction of 2,5-dichlorobenzoyl chloride
with various benzene derivatives were synthesized according to the
literature with some modifications [19,20,32]. The benzene deriva-
tives used as counterparts of 2,5-dichlorobenzoyl chloride in this
study were diphenylether, n-propylbenzene, n-dodecylbenzene, or
n-octadecylbenzene. The yield of the final product was approxi-
mately 85–95% based on benzene derivatives. The structure of
monomers was confirmed by NMR spectroscopy.
2.10. Water uptake and proton conductivity
Water uptake and proton conductivity were measured using an
isothermal absorption measurement system (MSB-AD-V-FC, BEL
Japan, Inc.) equipped with an impedance analyzer (SI 1260, Solar-
tron) with a temperature- and humidity-controlled chamber. Each
membrane sample was dried at 80 ꢁC for 2 h under dry nitrogen
flow and then exposed to a humidified nitrogen environment at
80 ꢁC. When there was no further weight change of each sample,
sample weight and proton conductivity were measured sequen-
tially. Humidity conditions were changed stepwise from 10 to 95%
RH. The system used enabled the simultaneous measurements of
water uptake and proton conductivity in the same chamber.
The water uptake of the membranes was calculated as
As shown in Scheme 1, a series of copolymers with alkyl
pendant chains of different lengths (i.e., propyl, dodecyl, and
octadecyl) were prepared. Copolymers (PPs) were synthesized
by Ni(0)-catalyzed coupling polymerization of 2,5-dichloro-4-
phenoxybenzophenone (M1) with 2,5-dichloro-4-propylbenzo-
phenone (M2), 2,5-dichloro-4-dodecylbenzophenone (M3), or
2,5-dichloro-4-octadecylbenzophenone (M4), respectively, in NMP
in the presence of Ni(PPh3)2Cl2, Zinc powder and triphenylphos-
phine at 75 ꢁC for 4 h in a similar synthetic method as described in
references [32,33]. The molar ratio of M1 to either of the alkyl side-
chain monomers (M2, M3, and M4) was controlled to be 90:10 to
provide a reasonably high degree of sulfonation of the synthesized
copolymers. The resulting copolymers (PPs) were soluble in various
organic solvents, such as dichloromethane, chloroform and DMF.
Polymer PPs were then sulfonated with 96% concentrated H2SO4 at
room temperature for 3 days, as described in experimental section.
After purification and drying, sulfonated copolymers were isolated
in yields higher than 90%. All sulfonated copolymers showed
solubilities differ from those of the parent PPs. They were highly
soluble in polar aprotic solvents such as DMAC, NMP, DMSO, and
DMF at room temperature but insoluble in less polar solvents such
as chloroform and dichloromethane as a result of the increased
hydrophilicity by the introduction of sulfonic acid groups. The
molecular weight of each sulfonated polymer was determined by
GPC using polystyrene as the standard. As shown in Table 1, the
number-average molecular weight (Mn) of SPP-1, SPP-2, and SPP-3
was 3.91 ꢂ104, 5.92 ꢂ 104, and 4.44 ꢂ 104,respectively. Sulfonation
of SPPs was confirmed by FT-IR spectra, a convenient method that
was used to identify the sulfonic acid group. An example of FT-IR
ꢂ
ꢃ.
ꢀ
ꢁ
Water volume fraction VfH O
¼
Wwet ꢀ Wdry
dH O=Vdry
2
2
where Wwet and Wdry are the weights of the wet and dry
membranes, respectively. Vdry is the volume of the dry membranes,
and dH O is the water density (1 g/cm3).
2
Proton conductivity was measured using a four-point probe AC
impedance method from 10 to 100 kHz. The results were collected
by Z-view and analyzed using Z-plot software. Proton conductivity
was calculated from dry membrane thickness, and membrane
resistance was taken at the frequency that produced the minimum
imaginary response. Proton conductivity (
the impedance data according to the following equation:
s) was calculated from
s
¼ d=A$R;
where is the proton conductivity, d is the distance between two
s
electrodes, A is the membrane cross-sectional area, and R is the
membrane resistance.
2.11. Gas permeability
Gas permeability was measured by the equal pressure method
with
a GTR-Tech 30XFST apparatus equipped with a gas

S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
627
Table 1
Molecular weights, IECs, water uptakes, proton conductivities and dimensional changes of SPP membranes.
Ionomer
Mn ꢂ 104 (g/mol)a (Mw/Mn)
IECb (mequiv/g)
WUc (vol%)
lc
s
(S/cm)d
Dimensional change
RT
80 ꢁC
50% RH
95% RH
Dt
D
l
D
t
Dl
SPP-1
SPP-2
SPP-3
SPEEK
3.91(2.8)
5.92(3.8)
4.44(3.8)
–
2.53
2.65
2.50
2.07
0.91
55.7
56.4
52.2
41.8
30.1
8.7
9.4
7.3
7.7
8.1
0.020
0.017
0.017
0.001
0.029
0.271
0.284
0.208
0.130
0.127
0.18
0.26
0.36
0.32
0.008
0.09
0.04
0.07
0.16
0.07
0.37
0.39
0.54
–
0.02
0.12
0.05
0.11
–
0.14
e
e
Nafion112
–
a
Determined by GPC based on polystyrene standards.
Measured by titration with 0.01 M NaOH.
Measured at 80 ꢁC and 95% RH.
At 80 ꢁC.
b
c
d
e
Dissolved.
spectrums of SPPs in comparison with parent polymer (non-
sulfonated polymer) was depicted in Fig. 1. The IR spectra of SPPs
showed characteristic absorptions due to the carbonyl and ether
groups at around 1653 and 1240 cmꢀ1, respectively. In addition, the
new absorption bands at around 1123 and 1031 cmꢀ1 assigned as
the symmetric stretching and asymmetric S]O stretching of the
sulfonic acid groups in SPPs were observed for all copolymers,
indicating the existence of sulfonic acid groups in SPPs. Since SPPs
showed good solubility, thin films could be readily prepared by
casting from DMF solution. SPP membranes were transparent,
flexible, and easy to handle even in their water-swollen or dry form,
as opposed to the non-aliphatic membranes containing poly(p-
phenylene)s which were rather brittle or difficult to be made
flexible by casting [18–20]. We speculate that the long alky chains
in copolymers disrupted the polymer chain–chain interaction
(aggregation) and increased the mobility of polymer chains. The
most noticeable feature of the copolymers was their ability to form
more flexible and ductile films, even in hot water (90 ꢁC), compared
with sulfonated poly(4-phenoxybenzoyl-1,4-phenylene)s, (SPPBPs)
or SPEEK of the same molecular weight range and IEC value, which
broke into pieces or dissolved in hot water at 90 ꢁC.
stability as shown in Fig. 2. PPs showed single step degradation with
5% weight loss temperature above 500 ꢁC. However, in the TGA
curves of sulfonated copolymers (SPPs) (Fig. 3), a two-step weight
loss profile was observed. All polymers displayed a similar weight
loss behavior, which differed slightly in terms of total percentage
weight loss. The first-stage weight loss around 250–400 ꢁC was
caused by the decomposition of the sulfonic acid and alkyl moie-
ties. The second-stage weight loss appearing at around 500 ꢁC was
attributed to the main-chain decomposition. Surprisingly, however,
those polymers exhibited almost the same thermal stability in air as
that in N2 atmosphere in the ranging from 220 to 400 ꢁC caused by
the decomposition of the sulfonic acid. This indicates that degra-
dation of SPPs in air had no significant effect on the thermal stability
of the sulfonic group on the polymer side chain. However, the third-
stage weight loss related to the degradation of the main chain of
SPP was faster under air atmosphere than nitrogen atmosphere in
the temperature range of 410–550 ꢁC.
Moreover, breaking strength and elongation at break point
under humidified condition were also investigated. The breaking
strength of SPP-1 and SPP-2 was 26 and 30 MPa, respectively, at 90%
RH and 80 ꢁC. Although there is no marked difference in breaking
strength between SPP-1 and SPP-2, there was a significant differ-
ence in elongation at break point: 17% for SPP-1 and 40% for SPP-2.
This significant difference indicates that a longer alkyl side chain of
the copolymer system can effectively contribute to structural
stabilization, particularly for membrane flexibility even for
membranes with higher water content. This result implies that
a longer alkyl side chain can participate in the enhancement of
intermolecular interaction resulting in mechanical stability.
3.2. Membrane stability
The thermal stability of non-sulfonated (PPs) and sulfonated
copolymers (SPPs) was determined by thermogravimetric analysis
(TGA) at a heating rate of 10 ꢁC/min in nitrogen or air atmosphere.
The non-sulfonated polymers (PPs) showed excellent thermal
PP-1
(non-sulfonated)
SPP-1
SPP-3
1006
1031
1123
2000
1800
800
1600
1400
1200
1000
Wavelength (cm-1)
Fig. 1. FT-IR spectra of non-sulfonated PP-1 and sulfonated SPP-1 and SPP-3.
Fig. 2. TGA thermograms of PPs (before sulfonation) under N2 atmosphere.

628
S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
Concerning the chain conformation in the membrane, the UV–
vis absorption spectra of SPPs in DMF solution and thin film states
were studied. In Fig. 6, the chain conformation corresponding to the
observed absorbance at 294 nm in solution state markedly changed
to a lower-energy band at approximately 380–395 nm in solid state,
due to a rigid-rod conformation that possesses a much larger and
uniform conjugation length in solid state. The coplanar conforma-
tion of the main chain backbone might be favored, resulting in the
extension of the conjugation length in the membrane. Further-
more, although chain conformation seems to be independent of the
chain length of aliphatic alkyl side chain in solution state, some
differences were seen among SPP-1 (propyl), SPP-2 (dodecyl), and
SPP-3 (octadecyl) membranes. The wavelength (lmax) of SPP-3 was
15 nm shorter than that of SPP-1 and SPP-2. The steric hindrance of
the longest alkyl (C18) side chain of SPP-3 may prevent the main
chain perturbation from forming a uniform conformation, while
the C12 side chain does not readily affect the chain conformation of
the backbone.
Fig. 3. TGA thermograms of SPPs (postsulfonation) under N2 and air atmosphere.
3.3. Membrane morphology
3.4. Water uptake and proton conductivity
The membrane morphology of SPPs was examined by AFM. Fig. 4
shows AFM topographic images of SPP-1 under controlled humidity
(50% RH and 85% RH) at room temperature. In Fig. 4, noodle-like
structures were observed. With increasing relative humidity, the
structures became thicker and clearly distinguished from each
other owing to swelling, as shown in the height image. Fig. 5 shows
topographic and current-mapping AFM images of SPP-1 and SPP-2
taken under 85% RH. A correlation between height and current-
mapping images was hardly found in either membrane. A darker
region, which corresponded to a proton conduction area, was found
to be distributed and each domain size was approximately 10–
20 nm in both membranes. From the height images, it has been
found that SPP-2 has a more microscopic structure than SPP-1.
Several 10-nm-scale particles can be seen in Fig. 5C. It is thought
that the slightly weakened interaction between phenylene main
chains with the introduction of a longer alkyl side chain facilitates
the development of a microscopic structure. In both current-
mapping images in Fig. 5B and D, the size of the conduction areas is
only slightly different; a proton conduction area is well distributed
in each membrane. This morphological feature is consistent with
the result of proton conductivity measurement in bulk, and this is
expected because a random copolymer system with a relatively
higher IEC can more easily form a less phase-segregated and
distributed hydrophilic domain.
The water uptake (vol%) of SPP membranes was measured as
a function of humidity (10–95% RH) at 80 ꢁC and compared with
that of the Nafion membrane. As shown in Fig. 7A, SPP membranes
exhibited higher water uptake values than Nafion in the entire
relative humidity range (10–95% RH) owing to their higher IEC
values. When the relative humidity increased from 50 to 95% RH,
water uptake further increased. This may be because of the
formation of a large and continuous ion network in sulfonated
polymers. Water uptake also results in dimensional changes in
membrane thickness and in-plane direction, as determined by
comparing hydrated membranes with dry membranes. In general,
membranes fabricated from polymers with poor hydrophilic–
hydrophobic balance display a marked increase in lateral dimen-
sions upon hydration, particularly those with
a higher IEC.
However, although their water uptake is relatively high SPP
copolymers that had short or long aliphatic alkyl side chains
showed small in-plane and through-plane dimensional changes.
From the results shown in Table 1, SPPs displayed anisotropic
membrane swelling and exhibited much (2–3 times) larger
swelling in the membrane thickness direction than in the plane
direction. This may have resulted from some effects of the aliphatic
alkyl chain that allowed the relaxation of polymer chain within
a small dimensional change. It should be emphasized here that
Fig. 4. AC-AFM height images of SPP-1 at 50% and 85% relative humidity. z-Scale maximum: (A) 11 nm, (B) 16 nm; scan size: 2
ꢀ1.6 V. For comparison, in both images, the same structure (marked by white arrow) was observed.
mm ꢂ 2
mm; scan rate: 1 Hz; and applied bias voltage:

S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
629
Fig. 5. AC-AFM height images of SPPs at 85% RH: (A, C) Topography: z-scale maximum: (A) SPP-1: 13 nm, (C) SPP-2: 7 nm (B, D) Current-mapping images: z-scale maximum:
(B) SPP-1: 50 pA and (D) SPP-2: 40 pA. Scan size: 500 nm ꢂ 500 nm. Scan rate: 1 Hz. Applied bias voltage: ꢀ1.6 V.
even though SPPs have a high water content, their dimensional
change is significantly small compared with that of SPEEK and
SPPBP. This result indicates that an alkyl side chain in a copolymer
can contribute to the prevention of membrane deformation by
swelling. This behavior is also promising because the low in-plane
swelling may result in a much lower stress at the interface and
better stability of the membrane electrode assembly (MEA) [37].
Proton conductivity at 80 ꢁC and different relative humidity
were also measured. For comparison, the conductivity of Nafion
was obtained under the same conditions. Fig. 7B shows the relative
humidity dependence of proton conductivity for SPP membranes
and Nafion at 80 ꢁC. The conductivity at 50, 70 and 95% RH at 80 ꢁC
is also listed in Table 1. The SPP membranes generally displayed
a stronger RH dependence of conductivity than Nafion. All SPP
membranes exhibited higher conductivity than Nafion in the RH
range of 70–95% RH. In comparison, SPP-2 displayed the highest
1
0.8
0.6
0.4
0.2
0
3
conductivity at 95% RH and the
NafionÒ (0.127 S/cm, 95% RH) under the same measurement
conditions. The number of water molecules per sulfonic group ( ) is
also listed in Table 1. The values of SPP membranes increased as
relative humidity increased. Despite the increase in water content
with IEC, there was no significant difference in between SPPs and
Nafion. For Nafion, continuous water channels can be formed even
with a small amount of water, but for SPEEK, a less phase-separated
structure cannot incorporate sufficient water. Less chain mobility
prevented the formation of a large water domain in the membrane.
s values were around 0.284 S/cm,
SPP-1 solution
SPP-2 solution
SPP-1 in solid
SPP-2 in solid
SPP-3 in solid
2.5
l
l
2
l
1.5
For SPPs, however, their
l values were similar to those of Nafion.
1
This result implies that SPPs can form a larger water domain. Our
AFM current-mapping images also clearly show that SPPs have
conductive domains approximately 10w20 nm in size, whereas
SPEEK has small conductive domains less than 10 nm in size [35].
0.5
0
240 320 400 480 560 640 720 800
3.5. Hydrogen and oxygen permeabilities
Wavelength (nm)
Since the crossover of hydrogen (H2) and oxygen (O2) between the
anode and cathode in MEA has a negative impact on the efficiency of
PEFC, the PEM should be a good gas barrier. Fig. 8 shows the humidity
Fig. 6. UV–vis absorption spectra of SPPs in DMF (with 10 mM LiBr) and solid film at
room temperature.

630
S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
Fig. 8. Permeability of (A) H2 and (B) O2 as functions of relative humidity of SPP
membranes at 80 ꢁC.
Fig. 7. Relative humidity dependence of (A) water uptake and (B) proton conductivity
of SPP membranes compared with those of Nafion 112 at 80 ꢁC.
4. Conclusion
A series of novel copoly(p-phenylene) containing pendant
aliphatic alkyl groups weresynthesized via Ni(0)-catalyzed coupling
polymerization from 2,5-dichloro-4-phenoxybezophenone with
aromatic dichloride containing an aliphatic propyl, n-dodecyl, or n-
octadecyl side chain. The synthesized polymers readily dissolved at
room temperature in aprotic polar solvents such as DMF, DMSO, and
NMP and formed tough, transparent and flexible films by solution
casting. All the polymers showed excellent proton conductivity and
good dimensional stability at high temperatures. The gas perme-
ability of the SPP membranes was lower than that of Nafion 112. The
incorporation of an aliphatic alkyl side chain into copolymers
improved the properties of the polymers such as solubility, flexi-
bility, and film-forming ability and reduced swelling in water at
enhanced temperatures compared to other sulfonated poly(p-phe-
nylene)s. We would like to emphasize here that introducing an
aliphatic alkyl pendant chain confers significant dimensional
stability but does not lead to the loss of other significant properties
such as proton conductivity and the ability to serve as a gas barrier.
Thus, we have demonstrated that significant performance such
as higher proton conductivity and good ability to serve as a gas
barrier with dimensional stability can be optimized by a simple and
easy means, that is, by the introduction of an aliphatic alkyl
pendant chain. The SPP membranes synthesized here can be
considered promising alternative PEM materials for fuel cell
application, although further improvement is required for their
practical use.
dependences of H2 and O2 permeability of SPP membranes and Nafion
at 80 ꢁC. Both H2 and O2 permeability of SPP membranes are smaller
than those of Nafion (PH of SPP-1, SPP-2, SPP-3 and Nafion at 80 ꢁC, 85%
2
RH: 2.17 ꢂ 10ꢀ9, 1.87 ꢂ 10-9, 2.24 ꢂ 10ꢀ9 and 6.88 ꢂ 10ꢀ9 cm3
(STP) cm cmꢀ2 sꢀ1 cmHgꢀ1, respectively, and PO of SPP-1, SPP-2, SPP-3
2
and Nafion at 80 ꢁC, 85% RH: 6.93 ꢂ 10ꢀ10, 4.76 ꢂ 10ꢀ10, 6.27 ꢂ 10ꢀ10
and 2.66 ꢂ 10ꢀ9 cm3 (STP) cm cmꢀ2 sꢀ1 cm Hgꢀ1, respectively.).
Although a small difference in H2 permeability can be seen at a lower
RH, the overall permeability is almost the same among the three
membranes SPP-1, SPP-2 and SPP-3. However, the O2 permeation
behavior of SPP-1 is significantly different from that of SPP-2 and SPP-3
during hydration. The O2 permeability of SPP-1 increased with water
uptake, but in the case of SPP-2 and SPP-3 decreased with increasing
relative humidity from 0 to 10% RH. Above 10% RH, an increase in
permeability was clearly observed. These differences are related to the
length of the aliphatic alkyl side chain in SPP copolymers. The decrease
in permeability with decreasing humidity might have been caused by
water molecules filling free-volume holes as a result of stabilized
intermolecular interaction, and the increase in permeability with
increasing humidity is caused by the plasticizing effect of water
molecules, as can be seen in the case of SPP-2 and SPP-3. Similar
permeation behavior has been reported for ethylene–vinyl alcohol
copolymer (EVOH) by Muramatsu et al. [38]. Since SPP-3 has a larger
free volume due to having the longest alkyl side chain (larger van der
waals radii), its permeability is higher than that of SPP-2, although
a stronger intermolecular interaction may occur in its membrane. For
SPP-1, it is not unambiguous at the present why SPP-1 showed different
permeation behavior at low RH from SPP-2 and SPP-3. Since the
permeability coefficient (P) is expressed by product of diffusivity (D)
and solubility (S) parameters (P ¼ DS) both parameters could be
changed during hydration. Further study will be needed for the
clarification.
Acknowledgements
This work was financially supported by the Ministry of
Economy, Trade and Industry (METI) and the New Energy and
Industrial Technology Development Organization (NEDO), Japan.

S. Seesukphronrarak et al. / Polymer 51 (2010) 623–631
631
[16] Asensio JA, Gomez-Romeo P. Fuel Cells 2005;5:336.
[17] Kobayashi T, Rikukawa M, Sanui K, Ogata N. Solid State Ionics 1998;106:219.
[18] Ghassemi H, McGrath JE. Polymer 2004;45:5847.
We are grateful to Prof. Rikukawa of Sophia University for providing
the sulfonated poly(4-phenoxybenzoyl-1,4-phenylene)s (SPPBPs).
[19] Hickner MA, Fujimoto CH, Cornelius CJ. Polymer 2006;47:4238.
[20] Cherry BR, Fujimoto CH, Cornelius CJ, Alam TM. Macromolecules
2005;38:1201.
References
[21] Fujimoto CH, Hickner MA, Cornelius CJ, Loy DA. Macromolecules 2005;38:
5010.
[1] Hickner MA, Ghassemi H, Kim Y, Einsla B, McGrath JE. Chem Rev
2004;104:4587.
[22] Li Q, Pan C, Jensen JO, Noy P, Bjerrum N. Chem Mater 2007;19:350.
[23] Jorissen L, Gogel V, Kerres J, Garche J. J Power Sources 2002;105:267.
[24] Fang J, Zhai F, Guo X, Xu H, Okamoto K. J Mater Chem 2007;17:1102.
[25] Buchi FN, Gupta B, Haas O, Scherer GG. Electrochemica Acta 1995;40:345.
[26] Kabanov VY. High Energy Chem 2004;38:57.
[27] Nakabayashi K, Matsumoto K, Ueda M. J Polym Sci Part A Polym Chem
2008;46:3947.
[28] Genies C, Mercier R, Sillion B, Cornet N, Gebel G, Pineri M. Polymer 2001;42:359.
[29] Ghassemi H, Ndip G, McGrath JE. Polymer 2004;45:5855.
[30] Lee HS, Badami AS, Roy A, McGrath JE. J Polym Sci Part A Polym Chem
2007;45:4879.
[31] Chang JH, Park JH, Park GG, Kim CS, Park OO. J Power Sources 2003;124:18.
[32] Wang J, Sheares V. Macromolecules 1998;31:6769–75.
[33] Hagberg E, Olson D, Sheares V. Macromolecules 2004;37:4748–58.
[34] Northop BH, Glockner A, Stang PJ. J Org Chem 2008;73:1787.
[35] Takimoto N, Wu L, Ohira A, Takeoka Y, Rikukawa M. Polymer 2009;50:534.
[36] Asano N, Aoki M, Suzuki S, Miyatake K, Uchida H, Watanabe M. J Am Chem Soc
2006;128:1762.
[37] Roy A, Yu X, Dum S, McGrath JE. J Membr Sci 2009;327:118.
[38] Muramatsu M, Okura M, Kuboyama K, Ougizawa T, Yamamoto T, Nishihara Y,
et al. Radiat Phys Chem 2003;68:561.
[2] Mauritz K, Moor R. Chem Rev 2004;104:4535.
[3] Rikukawa M, Sanui K. Prog Polym Sci 2000;25:1463.
[4] Miyake K, Zhou H, Matsuo T, Uchida H, Watanabe M. Macromolecules
2004;37:4961.
[5] Piroux F, Espuche E, Mercier R. J Membr Sci 2004;232:115.
[6] Zhai F, Guo X, Fang J, Xu H. J Membr Sci 2007;296:102.
[7] Guo X, Fang J, Watari T, Tanaka K, Kita H, Okamoto K. Macromolecules
2002;35:6707.
[8] Linkous CA, Anderson HR, Kopitzke RW, Nelson GL. Int J Hydrogen Energy
1998;23:525.
[9] Xing P, Robertson GP, Guiver MD, Mikhailenko SD, Wang K, Kaliaguine S.
J Membr Sci 2004;229:95.
[10] Zhang G, Fu T, Shao K, Li X, Zhao C, Na H, et al. J Power Sources 2009;189:875.
[11] Matsumura S, Hlil AR, Lepiller C, Gaudet J, Guay D, Shi Z, et al. Macromolecules
2008;41:281.
[12] Miyake K, Chikashige Y, Higuchi E, Watanabe M.
2007;129:3879.
J
Am Chem Soc
[13] Kim DS, Shin KH, Park HB, Chung YS, Nam SY, Lee YM.
2006;278:428.
J
Membr Sci
[14] Nolte R, Ledjeff K, Bauer M, Mulbaupt R. J Membr Sci 1993;83:211.
[15] Roy A, Lee HS, McGrath JE. Polymer 2008;49:5037.