Polymer electrolyte membranes based on poly(phenylene ether)s with sulfonic acid via long alkyl side chains

Article abstract of DOI:10.1039/c3ta12026k

A series of poly(phenylene ether)s with sulfonic acid groups via long alkyl side chains, SPPEs, were successfully prepared as proton exchange membranes for fuel cells. The new monomer, bis[4-fluoro-3-(p-methoxylbenzoyl)]biphenyl (BFMBP), containing two pe

Full text of DOI:10.1039/c3ta12026k

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Polymer electrolyte membranes based on
poly(phenylene ether)s with sulfonic acid via long alkyl
side chains†
Cite this: J. Mater. Chem. A, 2013, 1,
11389
*
*
Xuan Zhang, Li Sheng, Teruaki Hayakawa, Mitsuru Ueda and Tomoya Higashihara
A series of poly(phenylene ether)s with sulfonic acid groups via long alkyl side chains, SPPEs, were successfully
prepared as proton exchange membranes for fuel cells. The new monomer, bis[4-fluoro-3-(p-
methoxylbenzoyl)]biphenyl (BFMBP), containing two pendent methoxyphenyl groups was synthesized by
the Friedel–Crafts reaction of 5-chloro-2-fluorobenzoyl chloride with anisole, followed by the nickel-
mediated homocoupling reaction. Poly(phenylene ether)s (PPEs) were then successfully obtained by the
aromatic nucleophilic substitution polycondensation of BFMBP with dihydroxy-monomers in the presence
of potassium carbonate. Sulfonated PPEs (SPPEs) (IEC: 1.96–2.45 mequiv. gꢀ1) could be prepared by the
reaction of PPEs containing phenol units with 1,4-butanesultone. By the solution casting method, SPPEs
produced transparent membranes with good mechanical properties, and showed good oxidative and
dimensional stabilities, and high proton conductivities, e.g., 8.60 ꢁ 10ꢀ3 S cmꢀ1 under 30% relative
humidity at 80 ꢂC, which is higher than that of Nafion 117.
Received 23rd May 2013
Accepted 22nd July 2013
DOI: 10.1039/c3ta12026k
www.rsc.org/MaterialsA
(SPEKs),11,12 sulfonated poly(arylene ether sulfone)s (SPAESs),16,17
sulfonated polyimides (SPIs),18 sulfonated poly(phenylene)s
Introduction
(SPPs),19–21 sulfonated poly(phenylene ether)s (SPPEs),22–25 etc.,
have been well developed during the past two decades.
Polymer electrolyte membrane fuel cells (PEMFCs) have drawn
worldwide attention because of their potential applications in
transportation, stationary and portable electronics. PEMFC
technology already provides sufficient performance to be
competitive with alternative technologies in energy conversion
applications. Polymer electrolyte membranes (PEMs) play an
important role in the PEMFC system since they serve as a proton
conductive carrier as well as the separator of the anode/
cathode.1–26
Among the state-of-the-art PEMs, sulfonated per-
uoropolymers (e.g., Dupont's Naonꢀ, 3M, and Aquivionꢀ
membranes), have been widely used in PEFCs due to their high
proton conductivity and high chemical stability. By chemical
modications or stabilizations, their new products (e.g., NRE212,
PFIA, Aꢂquivionꢁ, etc.) could even withstand a high temperature
of 120 C.5 However, some other notable drawbacks, such as a
harsh manufacturing process, high cost, and uoro-release
problem still more or less restrict their practical use. Hence, a
number of non-uorinated acid ionomers, especially the
sulfonated aromatic hydrocarbon polymers, have been widely
studied as a branch of alternate PEMs. The typical materials used
for these PEMs, such as sulfonated poly(ether ketone)s
Typical SPPEs are prepared by the sulfonation of the
precursor poly(phenylene ether)s (PPEs). The phenylene rings in
the main chain provide a high electron density and could be
easily functionalized by sulfuric acid, chlorosulfonic acid, etc.
For instance, Xu's group reported a series of SPPEs.24 By
blending SPPEs with brominated PPE, their membranes
showed a higher fuel cell performance than Naon. However,
the sulfonic acid groups were directly located on the main
chain, ortho to the “–O–” linkages, which might decrease the
oxidative stability of these polymers. Unfortunately, no further
information on their durability was presented by the authors.
Watanabe's group previously prepared a series of SPPEs using
the sulfonated alkoxyl side chains as pendants.25 Unlike the
situation of SPPEs described above, the fairly low polarity of the
main chain with a high polarity side chain produced problems
regarding the solubility for such types of SPPEs. The as-
prepared polymers became insoluble once they were recovered
from the solution. Therefore, the authors treated them with
tetrabutylammonium tetrauoroborate. The resulting polymers
with an ammonium-form could nally dissolve in benzyl
alcohol and produce the membranes.
The polymers with exible ether and rigid phenyl–phenyl
linkages in the main chains are promising candidates for PEM
designs, since they could support polymers with good tough-
ness as well as high chemical stability. The issue lies in where to
functionalize the sulfonic acid groups in order to satisfy the
Department of Organic and Polymeric Materials, Tokyo Institute of Technology,
2-12-1-H120, O-okayama, Meguro-ku, Tokyo 152-8552, Japan. E-mail: mueda@
kanagawa-u.ac.jp; thigashihara@yz-yamagata-u.ac.jp; Fax: +81-3-5734-2127
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3ta12026k
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 11389–11396 | 11389

View Article Online
Journal of Materials Chemistry A
Paper
requirements for solubility and proton conductive capacity. Our dropwise to the mixture through a pressure-equilibrium drop
ꢂ
previous study showed interesting results such that the intro- funnel and the reaction mixture was then heated to 50 C for
duction of exible alkyl sulfonated side chains to the random 3 h. The solution was poured into ice water with a few drops of
polymer main chains improved the proton conductivity at a low HCl, and the product was extracted with chloroform. Aer
relative humidity (RH) due to the phase separation between the drying over MgSO4, evaporation, and recrystallization from
hydrophilic and hydrophobic domains, and the oxidative ethanol, CFMB (12.5 g) was obtained; yield: 76%. IR (KBr): n
stability of membranes was also increased due to the location of 2960–2870 (C–H), 1647 (C]O), 1597 (C]C), 1180 cmꢀ1 (C–O–
the sulfonic acid groups far from the main chains, where oxy C). 1H NMR (CDCl3, ppm): d ¼ 7.81–7.78 (d, J ¼ 8.7 Hz, 2H, Ar–
and peroxy radicals might be preferably formed.20,26 However, H), d ¼ 7.47–7.40 (m, 2H, Ar–H), d ¼ 7.12–7.06 (t, J ¼ 8.4 Hz, 1H,
the mechanical stability of these membranes was not sufficient Ar–H), d ¼ 6.95–6.92 (d, 2H, J ¼ 8.7 Hz, Ar–H), d ¼ 3.87 (s, 3H,
for fuel cell applications. Therefore, in this study, the more –OCH3). 13C NMR (CDCl3, ppm): d ¼ 190.24 (1C, C7), 164.18 (1C,
robust poly(phenylene ether) was selected as the polymer main C11), 159.73, 156.40 (1C, C4), 132.31 (1C, C2), 132.25, 132.14
chain.
(1C, C1), 130.03, 129.99 (2C, C9), 129.59, 129.54 (1C, C5), 129.50
We now report a novel type of PPEs with long alkyl sulfonic (1C, C8), 128.89, 128.65 (1C, C6), 117.76, 117.45 (1C, C3), 113.86
acids, which were synthesized from a new monomer of bis[4- (2C, C10), 55.53 (1C, C12). Anal. calcd for C14H10ClFO2: C: 63.53;
uoro-3-(40-methoxylbenzoyl)]biphenyl (BFMBP) and various H: 3.81. Found: C: 63.58; H: 3.90%.
dihydroxy-monomers via aromatic nucleophilic substitution
Synthesis of bis[4-uoro-3-(40-methoxylbenzoyl)]biphenyl
(BFMBP)
polycondensation and demethylation, followed by the reaction
of the resulting polymers having phenol groups with 1,4-buta-
nesultone. Their properties, such as ion exchange capacity
(IEC), water uptake (WU), dimensional change, mechanical
properties, proton conductivities and morphology, were inves-
tigated in detail.
To a 50 mL two-necked ask equipped with a magnetic stirrer
were charged CFMB (5.00 g, 18.9 mmol), Ni(PPh3)2Cl2 (1.24 g,
1.89 mmol), Et4NI (0.486 g, 1.89 mmol) and zinc dust (3.71 g,
56.7 mmol). Aer being exchanged with argon three times,
38.0 mL of anhydrous THF was added. The reaction mixture
underwent a color change from green to yellowish brown and
nally dark red at 50 ꢂC for 24 h. Aer ltration, the product was
separated by a silica column using CH2Cl2 as an eluent. By
recrystallization from ethanol, BFMBP (2.05 g) was obtained as
white needle-like crystals; yield: 46%. IR (KBr): n 2960–2870
(C–H), 1647 (C]O), 1600 (C]C), 1180 cmꢀ1 (C–O–C). 1H NMR
(CDCl3, ppm): d ¼ 7.83–7.80 (d, J ¼ 9.0 Hz, 4H, Ar–H), d ¼ 7.66–
7.62 (m, 4H, Ar–H), d ¼ 7.22–7.16 (t, J ¼ 9.0 Hz, 2H, Ar–H), d ¼
Experimental section
Materials
5-Chloro-2-uorobenzoic acid, anisole, and AlCl3 were
purchased from TCI and used as received. Thionyl chloride was
purchased from Wako Pure Chemical Industries, Ltd and used
as received. Bis(triphenylphosphino)nickel(II) (Ni(PPh3)2Cl2,
TCI) and tetraethylammonium iodide (Et4NI, TCI) were dried at
100 ꢂC under vacuum prior to use. Zinc powder was purchased
from Aldrich, washed with 2.0 M hydrochloric acid, deionized
water, ethanol, and diethylether, successively, and dried at
100 ꢂC for 6 h and at room temperature for 18 h under vacuum.
Tetrahydrofuran (THF, dehydrated, stabilizer-free, 99.5%,
Wako) was reuxed over sodium benzophenone under N2 for 2
h, and then distilled just before use. N,N-Dimethylacetamide
(DMAc, Wako) was dried over CaH2, distilled under reduced
6.93–6.90 (d, J ¼ 9.0 Hz, 4H, Ar–H), d ¼ 3.84 (s, 6H, –OCH3). 13
C
NMR (CDCl3, ppm): d ¼ 191.51 (2C, C7), 164.05 (2C, C11),
161.03, 157.69 (2C, C4), 135.78, 135.74 (2C, C1), 132.30 (2C, C2),
130.98, 130.86 (4C, C9), 129.91 (2C, C8), 128.87, 128.82 (2C, C6),
128.04, 127.82 (2C, C5), 116.90, 116.61 (2C, C3), 113.81 (4C,
C10), 55.51 (2C, C12). Anal. calcd for C28H20F2O4: C: 73.36; H:
4.40. Found: C: 73.03; H: 4.72%.
˚
pressure and stored over 4 A molecular sieves prior to use. Other
Synthesis of poly(phenylene ether) (2)
solvents and reagents were used as received.
A typical procedure for the preparation of polymer 2b is as
follows. A three-necked ask was charged with BFMBP (1.20 g,
2.62 mmol), bisphenol (0.487 g, 2.62 mmol), K2CO3 (0.416 g,
3.01 mmol), DMAc (8.40 mL) and toluene (8.00 mL). The
Synthesis of 5-chloro-2-uoro-40-methoxylbenzophenone
(CFMB)
To a 50 mL ask equipped with a magnetic stirrer, 5-chloro-2- mixture was purged with nitrogen and then heated to 140 ꢂC for
uorobenzoic acid (12.5 g, 71.6 mmol) and 20.0 mL of thionyl 2 h. Aer toluene was fully removed, the reaction temperature
chloride were added. The solution was kept at 85 ꢂC for 10 h and was gradually risen to 165 ꢂC and kept for another 6 h. The
the excessive thionyl chloride was then evaporated. 5-Chloro-2- resulting viscous solution was cooled down to room tempera-
uorobenzoyl chloride (CFBC) was obtained by evaporating ture, diluted with 8.40 mL of DMAc, and precipitated into
under reduced pressure to afford a colorless liquid; 12.8 g (yield: methanol containing a few drops of concentrated HCl. The
92%). 1H NMR (CDCl3, ppm): d ¼ 8.07–8.04 (m, 1H, Ar–H), d ¼ polymer was reprecipitated from 5 wt% DMF solution into a
7.62–7.57 (m, 1H, Ar–H), d ¼ 7.19–7.13 (t, J ¼ 8.7 Hz, 1H, Ar–H). mixture solution (CHCl3–methanol ¼ 2/8). Aer drying at 100 ꢂC
To a fully dried 100 mL three-necked ask, AlCl3 (9.12 g, under vacuum for 12 h, the brous polymer was obtained
68.4 mmol) and anisole (90.0 mL) were charged. Aer cooling to (1.23 g, yield: 77%). IR (KBr): n 2960–2870 (C–H), 1658 (C]O),
0 ꢂC by an ice-water bath, CFBC (12.0 g, 62.2 mmol) was added 1597 (C]C), 1169 cmꢀ1 (C–O–C). 1H NMR (DMSO-d6, ppm):
11390 | J. Mater. Chem. A, 2013, 1, 11389–11396
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
d ¼ 7.87–7.77 (bm, Ar–H), d ¼ 7.52 (bs, Ar–H), d ¼ 7.14–6.96 (bm, GMHHR-M) eluted with N,N-dimethylformamide (DMF) con-
Ar–H), d ¼ 3.82 (s, –OCH2–).
taining 0.01 M LiBr at a ow rate of 1.0 mL minꢀ1 calibrated by
standard polystyrene samples. Thermogravimetric analysis (TGA)
was measured in N2 on a Seiko EXSTAR 6000 TG/DTA 6300
thermal analyzer at a heating rate of 10 ꢂC minꢀ1. Dynamic
mechanical analysis (DMA) was performed on the lm specimens
(length, 30 mm; width, 10 mm; thickness, 50 mm) by using the
Seiko DMS 6300 at a heating rate of 2 ꢂC minꢀ1 with a load
frequency of 1 Hz under a nitrogen atmosphere. The mechanical
properties of SPPE membranes were also analyzed by tensile
measurement, which was performed with a universal testing
Synthesis of poly(phenylene ether) (3)
A typical synthetic procedure for the preparation of polymer 3b
is as follows. To a 200 mL three-necked ask equipped with a
nitrogen inlet/outlet were charged polymer 2b (1.00 g,
1.65 mmol) and CH2Cl2 (80.0 mL). The solution was cooled to
ꢂ
0 C in an ice bath, and a BBr3/CH2Cl2 solution (12.0 mL) was
added dropwise to the solution. Aer addition, the reaction was
kept at room temperature overnight. The resulting copolymer
was obtained by carefully pouring the solution into water,
ltered, washed thoroughly with water, and methanol, succes-
sively, and dried under vacuum at 80 ꢂC for 12 h. Yield: 0.912 g
(96%). IR (KBr): n 3500–3200 (–OH), 2960–2870 (C–H), 1643 (C]
O), 1597 (C]C), 1169 cmꢀ1 (C–O–C). 1H NMR (DMSO-d6, ppm):
10.36 (s, –OH), d ¼ 7.87–7.67 (bm, Ar–H), d ¼ 7.34–7.26 (bm, Ar–
H), d ¼ 7.10–6.99 (bm, Ar–H), d ¼ 6.86–6.64 (bm, Ar–H).
ꢂ
instrument (AGS-X 349-05489A, Shimadzu, Japan) at 20 C and
around 50% RH at a crosshead speed of 2 mm minꢀ1. Wide-angle
X-ray diffraction (WAXD) measurements were performed at
ambient temperature by using a Rigaku-Denki UltraX-18 X-ray
generator with monochromic Cu Ka radiation (40 kV, 50 mA)
from a graphite crystal of monochromator and a at-plate type of
imaging plate. Tapping mode atomic force microscope (AFM)
images were taken by using a Seiko Instruments SPA-400 with a
stiff cantilever of Seiko Instruments DF-20.
Synthesis of sulfonated poly(phenylene ether) (4)
A typical synthetic procedure for the preparation of polymer 4b
is as follows. To a 20 mL three-necked ask equipped with a
nitrogen inlet/outlet were charged polymer 3b (0.80 g,
1.39 mmol), sodium hydride (0.050 g, 2.08 mmol) and dimethyl
sulfoxide (DMSO) (15.0 mL). The solution was heated to 100 ꢂC
and 1,4-butanesultone (0.756 g, 5.45 mmol) was added dropwise
to the mixture. Aer 12 h, the solution was cooled to room
temperature and slowly poured into ethanol. The sulfonated
copolymer was ltered and then dried in a vacuum oven at 100
ꢂC for 12 h. Yield: 1.14 g (97%). IR (KBr): n 2960–2870 (C–H),
1658 (C]O), 1600 (C]C), 1176 cmꢀ1 (C–O–C), 1119, 1049 cmꢀ1
(O]S]O). 1H NMR (DMSO-d6, ppm): d ¼ 7.77–6.80 (bm, Ar–H),
d ¼ 3.91 (s, –OCH2–), d ¼ 2.54 (s, –CH2–SO3Na), d ¼ 1.84–1.74
(m, –CH2–).
Water uptake and dimensional change
The humidity dependence of water uptake was measured by
placing the membrane in a thermo-controlled humid chamber
for 4 h under each RH condition. The membrane was then
taken out, and quickly weighed on a microbalance. Water
uptake (WU) was calculated according to eqn (1):
WU ¼ (Ws ꢀ Wd)/Wd ꢁ 100 wt%
(1)
where Ws and Wd are the weights of wet and dried membranes,
respectively.
For dimensional change measurements, squared membrane
sheets were placed in a thermo-controlled humid chamber for
2 h under each RH condition. The dimensional change in the
membrane thickness direction (Dtc) and the plane direction
(Dlc) was calculated from eqn (2):
Preparation of polymer 4 membranes
Membranes were prepared by a solution casting method. Poly-
mer 4 was dissolved in DMSO with a concentration of 5% (m/v).
Aer ltration, the solution was cast onto a clean Petri dish,
Dtc ¼ (t ꢀ ts)/ts, Dlc ¼ (l ꢀ ls)/ls
(2)
ꢂ
dried at 80 C for 20 h under atmosphere. The membrane was
where ts and ls refer to the thickness and diameter of the
detached from the dish by immersing into water to remove the
residual solvent, followed by soaking in 2 M H2SO4 at 50 ꢂC for
72 h for proton exchange. Then, the membrane was thoroughly
washed with Milli-Q water. The thickness of the membrane was
controlled to be about 50 mm.
ꢂ
membrane equilibrated at about 30 C/30% RH, respectively; t
and l refer to those of the membrane under each condition.
Ion exchange capacity
IECs of the membranes were determined by a titration method.
The proton-type samples were ion-exchanged by 15 wt% NaCl
solutions and titrated with a 0.02 M NaOH solution with
phenolphthalein as the indicator. The IEC was calculated from
eqn (3):
Characterizations
1H (300 MHz) and 13C (75 MHz) NMR spectra were recorded on a
Bruker DPX300S spectrometer using CDCl3 or DMSO-d6 as the
solvent and tetramethylsilane as the reference. Fourier trans-
form-infrared (FT-IR) spectra were obtained with a Horiba FT-
120 Fourier transform spectrophotometer. Number- and
weight-average molecular weights (Mn and Mw) were measured
IEC ¼ CNaOH ꢁ VNaOH/Wd
(3)
by gel permeation chromatography (GPC) on a Hitachi LC-7000 where CNaOH and VNaOH are the concentration of NaOH solution
system equipped with polystyrene gel columns (TSKgel and the consumed volume of NaOH solution, respectively.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 11389–11396 | 11391

View Article Online
Journal of Materials Chemistry A
Paper
Water sorption (l)
The l value was calculated from eqn (4):
[H2O]/[SO3ꢀ] ¼ WU ꢁ 10/18 ꢁ IEC
(4)
Proton conductivity
Proton conductivity in the plane direction of the membrane was
determined using an electrochemical impedance spectroscopy
technique over the frequency from 5 Hz to 1 MHz (Hioki 3532-80).
A two-point probe conductivity cell with two platinum plate elec-
trodes was fabricated. The cell was placed in a thermo-controlled
humid chamber at 80 ꢂC for 2 h before the measurement. Proton
conductivity (s) was calculated from eqn (5):
s ¼ d/(tswsR)
(5)
where d is the distance between the two electrodes, ts and ws are
the thickness and width of the membrane, and R is the resis-
tance measured.
Oxidative stability
Oxidative stability was determined by soaking a membrane
sample (10 ꢁ 10 mm2) in 50 mL of Fenton's reagent (3% H2O2
containing 2 ppm FeSO4) at 80 ꢂC for 1 h. The residue was then
taken out, washed with water, and dried under vacuum at
Fig. 1 1H and 13C NMR spectra of CFMB in CDCl3.
ꢂ
100 C until a constant weight (remaining weight).
aryl halides,27,28 the relatively negative results in this study might
be attributed to the low reactivity of the precursor monomer
CFMB. A carbonyl group was substituted at the meta-position
toward the chloride group, which made it less reactive than some
other chloride groups at the ortho-/para-positions toward the
electron-withdrawing groups. The same phenomenon was also
found by D. K. Taylor's group.29 The chemical structure of BFMBP
was conrmed by its 1H NMR, 13C NMR (Fig. 2) and IR spectra. On
the one hand, the chemical shis of the two protons ortho to the
chloride (H2 and H5 in Fig. 1a, d ¼ 7.47–7.40 ppm) shied to the
low eld (d ¼ 7.66–7.62 ppm) in Fig. 2a. On the other hand, the
peak assigned to the carbon atom C1 (d ¼ 132.25 and132.14 ppm
in Fig. 1b) also shied to 135.78 and 135.74 ppm, respectively, in
Fig. 2b. These results provide obvious evidence for the phenyl–
phenyl bond formation.
Scheme 2 shows the synthetic route towards polymers 4.
Polymers 2 were rst obtained by the aromatic nucleophilic
substitution polycondensation in the presence of K2CO3.
Although two of the three counterpart (dihydroxy) monomers
chosen in this study have rigid rod structures (a and b), no
precipitation occurred during the polymerization reactions. The
results of the polymerization reactions are summarized in
Table 1. The number-average molecular weights (Mn) and
polydispersity indices (PDIs) of polymers 2 were in the range of
3.1–4.3 ꢁ 104 and 2.0–2.4, respectively. Polymers 2 showed good
solubilities in common organic solvents such as CHCl3, CH2Cl2,
DMSO, DMAc, etc. The chemical structures of polymers 2 were
conrmed by their 1H NMR and FTIR spectra. The relative
Results and discussion
Synthesis of monomers and polymers
Scheme 1 shows the synthetic route towards the monomer
BFMBP. CFMB was rstly prepared by the aluminium chloride
catalyzed Friedel–Cras acylation of anisole with 5-chloro-2-
uorobenzoyl chloride. Due to electronic and steric effects, the
substitution reaction selectively occurred at the para-position to
the methoxyl group of anisole. The chemical structure of CFMB
was conrmed by its 1H NMR, 13C NMR (Fig. 1) and IR spectra.
NiBr2 and triphenylphosphine (PPh3) were used as the catalyst
and ligand for the nickel-catalyzed coupling reaction, respectively.
The reaction proceeded well in DMAc, DMF and THF, however,
the yield for the nal product was only 10–20%, which might be
due to the moisture-sensitive catalyst. Thus, a more stable catalyst
of Ni(PPh3)2Cl2 was instead investigated and the temperature was
maintained under the mild condition of 50 ꢂC. The yield was
improved to about 50% aer recrystallization from ethanol.
Unlike many reported nickel-mediated reactions, which provided
almost quantitative yields of the coupling products from various
Scheme 1 Synthetic route towards monomer BFMBP.
11392 | J. Mater. Chem. A, 2013, 1, 11389–11396
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
intensities of the aromatic-H to methoxy-H were 3.09, 3.78 and
3.67 for polymers 2a, 2b and 2c, respectively, close to the
theoretical values of 3.00, 3.67 and 3.67, respectively (see Fig. S1
in the ESI†). The IR spectra of polymers 2 exhibited character-
istic absorptions around 2960–2870, 1658, 1597, and 1169 cmꢀ1
due to the C–H, C]O, C]C and C–O–C stretchings, respec-
tively. Polymers 2 were then converted to polymers 3 by treat-
ment with BBr3. In the 1H NMR spectra of polymers 2b and 3b,
the characteristic methoxy protons at 3.82 ppm of polymer 2b
have completely disappeared, and hydroxy proton signals
appear at 10.36 ppm, as shown in Fig. S2.† Finally, the SPPEs, 4,
were obtained by the reaction of the resulting polymers 3 with
1,4-butanesultone in the presence of sodium hydride.
The chemical structures of polymers 4 were also conrmed
by FTIR and 1H NMR spectroscopies. The IR spectra of polymers
4 showed characteristic absorptions corresponding to the C]O
and C]C stretchings around 1658 and 1600 cmꢀ1, whereas the
typical O]S]O stretching was observed at around 1115 and
1050 cmꢀ1(Fig. S3†). The 1H NMR spectra of polymers 4 are
shown in Fig. 3. New peaks at around 4.01, 3.91 and 3.89 ppm
were assigned to the a-methylene protons next to the phenoxy
groups for 4a, 4b and 4c, respectively, indicating the introduc-
tion of long sulfoalkyl side chains. Since the typical signals of
the a-methylene protons next to the –SO3Na groups overlapped
with the DMSO-d6 peak, the IECNMR values were calculated by
the relative intensities of the methylene protons to the aromatic
protons. All the data were close to the theoretical values, which
suggested the well-controlled nucleophilic reaction of polymers
3 with 1,4-butanesultone (Table 2).
Fig. 2 1H and 13C NMR spectra of BFMBP in CDCl3.
Thermal properties and oxidative stability
The thermal properties of polymers 4 in the acid form were
evaluated by TGA and DSC. No obvious glass transition
temperatures Tgs could be observed in all the polymers 4. Fig. 4
shows the TGA curves of polymers 4 in N2 and the weight loss
temperatures are listed in Table 2. All the polymers 4 exhibited
ꢂ
Td,5% values over 244 C, which could satisfy the requirement
Scheme 2 Synthetic route towards the sulfonated polymers 4.
Table 1 Synthesis of polymers 2
PDIa
a
Code
Mn (kDa)
Mwa (kDa)
2a
2b
2c
31
43
35
74
85
77
2.4
2.0
2.2
a
Fig. 3 1H NMR spectra of polymers 4a, 4b and 4c in DMSO-d6.
J. Mater. Chem. A, 2013, 1, 11389–11396 | 11393
Determined by GPC eluted with DMF using a polystyrene standard.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Journal of Materials Chemistry A
Paper
modulus (E00) of 4c as a function of temperature. The E0 and E00
values of 2.6 GPa and 300 MPa are maintained from 80 to
180 ꢂC, respectively, indicating the excellent mechanical
strength despite the relatively high IEC value (1.96 mequiv. gꢀ1).
Typical tensile stress–strain curves of all the SPPE membranes
are shown in Fig. 5. They exhibited a reasonably high mechanical
strength with a Young's modulus (M), maximum stress (S) and
elongation at break (Eb) values ranging from 1.01 to 1.30 GPa, 34–
54 MPa and 4.4–21.2%, respectively, which indicated that the
membranes could undergo a fairly large deformation or sustain a
quite high external impact without breaking in practical applica-
tions. Polymer 4c exhibited the highest values of 1.30 GPa, 54 MPa
and 21.2% for the M, S and Eb, respectively, which might indicate
that the appropriate level of rigidity for the polymer backbone can
improve the mechanical strength for these types of PEMs. The free
rotation of ether linkages between aromatic groups of 4c should
cause the higher elasticity compared to others, and thereby its
mechanical properties such as tensile strengths could also be
improved.
Table 2 Physical properties of polymers 4
4a
4b
4c
IECcal
IECNMR
IECtitr
(mequiv. gꢀ1
)
)
)
2.59
2.51
2.45
2.36
2.19
2.08
2.31
2.10
1.96
(mequiv. gꢀ1
(mequiv. gꢀ1
Td,5%
(ꢂC)
(ꢂC)
Dlc
Dtc
Dlc
Dtc
Dlc
Dtc
(%)
257.4
268.0
0.021
0.030
0.060
0.090
0.110
0.130
81
244.2
259.0
0.014
0.020
0.048
0.070
0.102
0.120
90
254.2
265.3
0.010
0.020
0.040
0.050
0.070
0.090
94
Td,10%
50% RHa
70% RHa
95% RHa
RWb
a
Dimensional changes of SPPEs at 80 ꢂC. Dlc and Dtc refer to the
changes in plane and through plane directions, respectively.
b
Remaining weight; measured by soaking the membranes in Fenton's
reagent (3% H2O2 containing 2 ppm FeSO4) at 80 ꢂC for 1 h.
IEC, WU and dimensional change
The IEC values were determined by a back-titration method,
and the results are listed in Table 2. All the titrated data were
close to the calculated ones from the 1H NMR spectra, indi-
cating a complete proton exchange process.
As is well known, the WU of the PEMs has a signicant effect
on the membrane conductivity. Water molecules usually facili-
tate the proton transfer; however, an excessive WU would
inevitably lead to an undesirable membrane swelling, resulting
in a reduced mechanical strength of the membrane. Fig. 6
shows the WU of polymers 4 as a function of the relative
humidity, together with that of Naon 117 for comparison.
Analogous to the common aromatic sulfonated polymers, the
WUs of polymers 4 showed a greater RH dependence than that
of Naon, especially under a high RH condition. Regardless of
the similar IEC values, the WU of 4c is 30% lower than 4b over
the entire RH range, which indicates the better swelling
behavior of the latter.
Fig. 4 TGA curves of SPPEs.
for medium-temperature fuel cells. There is an apparent weight
loss starting from 220 to 300 ꢂC, which is close to the theoretical
values for the decomposition of the sulfonic acid groups in the
sideꢂchains. The second weight loss step is observed from 300 to
550 C, attributable to the cleavage of the alkyl side chains.
The oxidative stability of the SPPE membranes was evaluated
by the remaining weight in Fenton's reagent (3% H2O2 aqueous
ꢂ
solution containing 2 ppm FeSO4) at 80 C for 1 h, and these
results are listed in Table 2. All the membranes showed an
excellent oxidative stability with a remaining weight greater
than 81%. Since it is widely considered that the electron-
donating groups (–O–) in the ortho- and para-positions to the
sulfonic acid group may lead to a lower oxidative stability, the
sulfonated aliphatic side chains would contribute to the high
oxidative stabilities of the SPPEs.20
Mechanical properties
The dynamic mechanical properties were investigated by DMA
measurements. Fig. S4† shows the storage modulus (E0) and loss Fig. 5 Stress–strain curves of SPPEs.
11394 | J. Mater. Chem. A, 2013, 1, 11389–11396
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 7 Humidity dependence of proton conductivities of polymers 4 and Nafion
117 at 80 ꢂC.
Fig. 6 WU of polymers 4 as a function of relative humidity.
The dimensional change data of polymers 4 are also
summarized in Table 2. Compared to some other well-devel-
oped poly(phenylene)s or poly(ether sulfones)s, polymers 4
exhibit much lower dimensional changes.30–32 The phenyl–
phenyl bond increases the rigidity of the main chain, which may
suppress the excessive swelling of the membranes even under
high RH conditions. For example, polymer 4a with an IEC of
2.45 mequiv. gꢀ1 shows only 0.021 and 0.03 in-plane and
through-plane directions under 50% RH, respectively, whereas
0.110 and 0.13 in-plane and through-plane directions under
95% RH, respectively. It is expected that polymer 4c displayed
the best dimensional stability due to its lowest IEC values. Its
Dlc was 0.07 under 80 ꢂC/95% RH, which would meet the
requirement for a single cell test.
Fig. 8 Relationship between hydration number (l) and proton conductivity for
polymers 4 and Nafion 117 at 80 ꢂC.
Proton conductivity
Fig. 7 shows the proton conductivity of polymers 4 as a function
of RH at 80 ꢂC, together with that of Naon 117 for comparison.
All the polymers 4 exhibited a greater humidity dependence
Morphology
than that of Naon. For example, polymer 4a with an IEC value The nanostructure and morphology of polymer 4a were studied
of 2.45 exhibited the highest s value of 237 mS cmꢀ1 under the using WAXD, SAXS and AFM. The same lms in the dried and
95% RH condition. It decreased to 8.5 mS cmꢀ1 with a decrease hydrated states were used for the WAXD and SAXS measurements.
of RH to 30%, however, it was still slightly higher than that of In both cases of the dried and hydrated lms, the WAXD proles
Naon (7.3 mS cmꢀ1), indicating the better proton conductive showed a broadened reection at around 2q ¼ 20ꢂ, which corre-
capacity of the former.
sponds to the intermolecular spacing of the benzene rings and/or
The relationship between the WU and proton conductivity of alkyl chains with sulfonic acid. No reections were observed in the
polymers 4 and Naon 117 was investigated. In Fig. 8, the s SAXS proles. These data suggest that the polymer 4a lm has an
value is plotted versus the hydration number, l, which is the amorphous state. The AFM tapping mode phase image of the 4a
number of water molecules per a sulfonic acid unit. As could be membrane was recorded under ambient conditions on an 800 ꢁ
seen, all the membranes tend to increase their proton 800 nm2 size scale to investigate its surface morphology. As can be
conductivity with increasing l values. At a similar proton seen in Fig. 9, a continuous phase-separated nanostructure was
conductivity level, polymers 4 have higher l values than Naon, found in which the sulfonic acid groups aggregate to form the
indicating that polymers 4 require many more water molecules nano-channels structures (dark region in Fig. 9), although
to facilitate proton transfer than the latter. Polymer 4a exhibits a the direction of the structures is disordered. The formation of
higher proton conductivity compared to Naon 117 over the
a
phase-separated nanostructure, even disordered, may
entire range, probably due to its highest IEC value and well- contribute to the efficient proton transport and excellent proton
connected proton pathways.
conductivity.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 11389–11396 | 11395

View Article Online
Journal of Materials Chemistry A
Paper
2 O. J. Savadoga, J. New Mater. Electrochem. Syst., 1998, 1, 47.
3 M. Rikukawa and K. Sanui, Prog. Polym. Sci., 2000, 25, 1463.
4 A. H. Hickner, H. Ghassemi, Y. S. Kim, B. R. Einsla and
J. E. McGrath, Chem. Rev., 2004, 104, 4587.
5 (a) S. Takamuku and P. Jannasch, Adv. Energy Mater., 2012, 2,
129; (b) The 2011 Progress Report for US DOE Hydrogen and
Fuel Cells Program, http://www.hydrogen.energy.gov/pdfs/
progress11/v_c_1_hamrock_2011.pdf;
(c)
M.
Gebert,
A. Ghielmi, L. Merlo, M. Corasaniti and V. Arcella, ECS
Trans., 2010, 279.
6 T. Higashihara, K. Matsumoto and M. Ueda, Polymer, 2009,
50, 5341.
7 C. H. Park, C. H. Lee, M. D. Guiver and Y. M. Lee, Prog. Polym.
Sci., 2011, 36, 1443.
8 Y. L. Liu, Polym. Chem., 2012, 3, 1373.
9 H. Zhang and P. K. Shen, Chem. Rev., 2012, 112, 2780.
10 H. Zhang and P. K. Shen, Chem. Soc. Rev., 2012, 41, 2382.
11 K. Matsumoto, T. Higashihara and M. Ueda, Macromolecules,
2008, 41, 7560.
Fig. 9 AFM tapping mode phase image of the 4a membrane: scan size is 800 ꢁ
12 X. Zhang, Z. Hu, L. Luo, S. Chen, J. Liu, S. Chen and L. Wang,
Macromol. Rapid Commun., 2011, 32, 1108.
800 nm2.
13 X. Zhang, S. Chen, J. Liu, Z. Hu, S. Chen and L. Wang,
J. Membr. Sci., 2011, 371, 276.
Conclusion
14 K. Miyatake, Y. Chikashige, E. Higuchi and M. Watanabe,
J. Am. Chem. Soc., 2007, 129, 3879.
A series of PPEs with sulfonic acids via long alkyl side chains,
SPPEs (polymers 4), were successfully obtained by the poly-
condensation of BFMBP and various dihydroxy-monomers in
the presence of K2CO3, followed by demethylation of polymers 2
with BBr3, and nally by the reaction of the resulting polymers 3
with 1,4-butanesultone. The titrated IEC values of the SPPEs
were slightly lower than the theoretical and calculated ones
15 K. Matsumoto, T. Higashihara and M. Ueda, Macromolecules,
2008, 42, 1161.
16 B. Latte and P. Jannasch, Adv. Funct. Mater., 2007, 17, 2823.
17 C. Wang, D. W. Shin, S. Y. Lee, N. R. Kang, Y. M. Lee and
M. D. Guiver, J. Membr. Sci., 2012, 405–406, 68.
18 Z. Hu, Y. Yin, K.-I. Okamoto, Y. Moriyama and A. Morikawa,
J. Membr. Sci., 2009, 329, 146.
19 S. Seesukphronrarak and A. Ohira, Chem. Commun., 2009,
4744.
20 X. Zhang, L. Sheng, T. Higashihara and M. Ueda, Polym.
Chem., 2013, 4, 1235.
21 X. Zhang, Z. Hu, Y. Pu, S. Chen, J. Ling, H. Bi, S. Chen,
L. Wang and K.-I. Okamoto, J. Power Sources, 2012, 216, 261.
22 T. Xu, D. Wu and L. Wu, Prog. Polym. Sci., 2008, 33, 894.
23 Z. Zhang, L. Wu and T. Xu, J. Membr. Sci., 2011, 373, 160.
24 T. Xu, D. Wu, S.-J. Seo, J.-J. Woo, L. Wu and S.-H. Moon,
J. Appl. Polym. Sci., 2012, 124, 3511.
25 K. Miyatake, H. Zhou and M. Watanabe, J. Polym. Sci., Part A:
Polym. Chem., 2005, 43, 1741.
26 L. Sheng, T. Higashihara, S. Nakazawa and M. Ueda, Polym.
Chem., 2012, 3, 3289.
27 M. Iyoda, H. Otsuka, K. Sato, N. Nisato and M. Oda, Bull.
Chem. Soc. Jpn., 1990, 63, 80.
28 I. Colon and D. R. Kelsey, J. Org. Chem., 1986, 51, 2627.
29 B. D. Koepnick, J. S. Lipscomb and D. K. Taylor, J. Phys.
Chem. A, 2010, 114, 13228.
30 X. Zhang, Z. Hu, S. Zhang, S. Chen, S. Chen, J. Liu and
L. Wang, J. Appl. Polym. Sci., 2011, 121, 1707.
31 K. Nakabayashi, T. Higashihara and M. Ueda, J. Polym. Sci.,
Part A: Polym. Chem., 2010, 48, 2757.
32 K. Nakabayashi, K. Matsumoto, T. Higashihara and M. Ueda,
Polym. J., 2009, 41, 332.
1
from the H NMR spectra. The SPPEs exhibited high thermal
stabilities despite the fully aliphatic side chains, excellent
oxidative stabilities, and reasonably high mechanical strengths
with a Young's modulus (M), maximum stress (S) and elonga-
tion at break (Eb) values that ranged from 1.01 to 1.30 GPa, 34–
54 MPa and 4.4–21.2%, respectively. The DMA results showed
the high storage modulus of 2.6 GPa for 4c from 80–180 ꢂC,
regardless of its relatively high IEC value of 1.96 mequiv. gꢀ1. All
SPPEs exhibited a greater RH dependence than Naon for the
proton conductivities. Polymer 4a with an IEC value of
2.45 mequiv. gꢀ1 showed a high s value of 8.6 mS cmꢀ1 under
80 ꢂC/30% RH conditions, whereas it was 7.3 mS cmꢀ1 for
Naon 117. Consequently, the novel PPEs with pendant alkyl
sulfonated side chains are promising candidates for fuel cell
applications.
Acknowledgements
This work is supported by the Industrial Technology Research
Grant Program in 2011 (#11B02008c) from New Energy and
Industrial Technology Development Organization (NEDO) of
Japan.
Notes and references
1 K. A. Mauritz and R. B. Moore, Chem. Rev., 2004, 104, 4535.
11396 | J. Mater. Chem. A, 2013, 1, 11389–11396
This journal is ª The Royal Society of Chemistry 2013