Synthesis and characterization of multiblock semi-crystalline hydrophobic poly(ether ether ketone)-hydrophilic disulfonated poly(arylene ether sulfone) copolymers for proton exchange membranes

Article abstract of DOI:10.1016/j.polymer.2012.05.022

Multiblock copolymers based on alternating segments of telechelic phenoxide terminated hydrophilic fully disulfonated poly(arylene ether sulfone) (BPS100) and decafluorobiphenyl (DFBP) terminated hydrophobic poly(arylene ether ketimine) (PEEKt), were synthesized from the hydrophilic and ketimine-protected amorphous hydrophobic telechelic oligomers by nucleophilic coupling reactions. After film formation from DMSO, the copolymer was acidified, which converted the ketimine to semi-crystalline ketone segments and the sulfonate salts to disulfonic acids. A semi-crystalline phase with a T_m of 325 °C was confirmed. The semi-crystalline multiblock copolymer membranes were tough, ductile and solvent resistant. Fundamental properties as proton exchange membranes (PEMs) showed enhanced conductivities under fully hydrated and reduced humidity conditions. These multiblock copolymers exhibited low in-plane anisotropic swelling behavior, in contrast to the random copolymers.

Full text of DOI:10.1016/j.polymer.2012.05.022

Polymer 53 (2012) 3143e3153
Contents lists available at SciVerse ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Synthesis and characterization of multiblock semi-crystalline hydrophobic
poly(ether ether ketone)ehydrophilic disulfonated poly(arylene ether sulfone)
copolymers for proton exchange membranes
*
Yu Chen, Chang Hyun Lee, Jarrett R. Rowlett, James E. McGrath
Macromolecular Science and Engineering Program, Department of Chemistry and Macromolecules and Interfaces Institute, Virginia Polytechnic Institute and State University,
Blacksburg, VA 24061, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Multiblock copolymers based on alternating segments of telechelic phenoxide terminated hydrophilic
fully disulfonated poly(arylene ether sulfone) (BPS100) and decafluorobiphenyl (DFBP) terminated
hydrophobic poly(arylene ether ketimine) (PEEKt), were synthesized from the hydrophilic and ketimine-
protected amorphous hydrophobic telechelic oligomers by nucleophilic coupling reactions. After film
formation from DMSO, the copolymer was acidified, which converted the ketimine to semi-crystalline
ketone segments and the sulfonate salts to disulfonic acids. A semi-crystalline phase with a Tm of
325 ꢀC was confirmed. The semi-crystalline multiblock copolymer membranes were tough, ductile and
solvent resistant. Fundamental properties as proton exchange membranes (PEMs) showed enhanced
conductivities under fully hydrated and reduced humidity conditions. These multiblock copolymers
exhibited low in-plane anisotropic swelling behavior, in contrast to the random copolymers.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 14 February 2012
Received in revised form
7 May 2012
Accepted 12 May 2012
Available online 17 May 2012
Keywords:
Semi-crystalline
Hydrophobicehydrophilic multiblock
copolymers
1. Introduction
A number of sulfonated aromatic statistical copolymers such as;
disulfonated poly(arylene ether sulfone)s (BPSHs), poly(ether ether
Proton exchange membrane fuel cells (PEMFCs) have been
extensively studied as potential alternative power sources for
electric vehicles, portable, and residential power sources because of
their high power density, excellent energy conversion efficiency,
quiet operation and zero pollution emission. Proton exchange
membranes (PEMs), the key component of PEMFCs, are responsible
for transporting the protons from anode to cathode [1,2]. There are
several critical requirements for a successful PEM material such as;
high proton conductivity, good mechanical strength, high oxidative
and hydrolytic stability, low fuel and oxidant permeability, ease of
fabrication, and excellent water management under low relative
humidity (RH) cycling [3,4]. Currently, perfluorinated sulfonic acid
containing ionomers (PFSAs), such as NafionÒ, are widely used as
PEM materials due to their good chemical stability, excellent
mechanical properties, and high proton conductivity [5]. However,
there are several major drawbacks of these materials, such as, high
cost, high methanol permeability, and low ceiling operation
temperature (<80 ꢀC) [6].
ketone)s (SPEEKs), poly(arylene ether nitrile)s (SPAEBs), poly-
imides (SPIs), and polybenzimidazoles have been developed as
potential alternatives to NafionÒ [7e14]. Although some of the
sulfonated statistical copolymers achieved reasonable proton
conductivity under fully hydrated states [15], only limited perfor-
mance was observed under partially humidified conditions. Further
improvement of membrane properties for hydrogen/air fuel cells,
especially the conductivity performance under partially hydrated
states and reduced dimensional changes under hydrated condi-
tions, remains
a primary goal. For portable power, reduced
permeability to fuels such as methanol is also important.
The disulfonated multiblock copolymer membranes have been
demonstrated to exhibit hydrophilic/hydrophobic nanophase
separated morphologies in the presence, or absence of water
[16e18]. This nanophase separation permits the formation of
lamellar or co-continuous hydrophilic domains and significantly
enhances proton transport under partially hydrated conditions
[19e22]. Wholly aromatic hydrophilicehydrophobic multiblock
copolymers based on fully disulfonated poly(arylene ether
sulfone) (BPS100) hydrophilic blocks and various hydrophobic
blocks have been introduced as new PEM materials by our group
[16,23e30].
* Corresponding author.
E-mail address: jmcgrath@vt.edu (J.E. McGrath).
0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2012.05.022

3144
Y. Chen et al. / Polymer 53 (2012) 3143e3153
This paper reports the first synthesis and characterization of
toluene to produce 40.0 g (50% yield) of a yellow crystalline solid
semi-crystalline hydrophobic poly(ether ether ketone)-hydrophilic
disulfonated poly(arylene ether sulfone) (PEEK-BPSH100) multi-
block copolymers. Semi-crystalline PEEK polymers exhibit excel-
lent thermal and mechanical stability and solvent resistance
[31,32], which could enhance the performance of PEMs. However,
the semi-crystalline morphology of the PEEK blocks causes insol-
ubility in most organic solvents at relative lower reaction temper-
ature. This prevents the direct coupling reaction between the PEEK
and BPS100 telechelic oligomers. The strategy to afford synthesis
and processing employed the generation of labile bulky ketimine
groups to synthesize amorphous PEEKt pre-oligomers [33e35]. The
procedure first involved synthesis of hydrophobic poly(ether ether
ketimine)-hydrophilic sulfonated poly(arylene ether sulfone)
(PEEKt-BPS100) multiblock pre-copolymers via coupling reactions
between phenoxide terminated hydrophilic BPS100 and decaf-
luorobiphenyl (DFBP) end-capped hydrophobic PEEKt blocks. The
highly reactive perfluorinated end groups of the PEEKt blocks
allowed for a low temperature coupling reaction, which minimized
the possibility of ethereether interchange reactions, and made it
possible to achieve high yields and molecular weights of the mul-
tiblock copolymers. Amorphous PEEKt-BPS100 copolymers were
then cast into films and simultaneously hydrolyzed and acidified to
produce semi-crystalline PEEK-BPSH100. The proton conductivity,
water uptake and other characteristics of the acidified semi-
crystalline PEEK-BPSH100 membranes were evaluated.
with a melting point of 113e115 ꢀC.
2.3. Synthesis of amorphous phenoxide terminated poly(ether ether
ketimine) oligomer (PEEKt)
Amorphous phenoxide terminated poly(ether ether ketimine)
oligomers (PEEKt) were synthesized with different molecular
weights by offsetting the stoichiometry according to the Carothers
equation. A sample synthesis of 9000 g/mol PEEKt oligomer was as
follows; a 100-mL three-necked round-bottom flask, equipped
with a mechanical stirrer, a nitrogen inlet, a condenser and
a DeaneStark trap was charged with 5.6421 g (0.0192 mol) of DFKt,
2.2064 g (0.0200 mol) of hydroquinone and 40 mL of NMP. The
mixture was stirred until dissolved, and then 3.33 g (0.0241 mol) of
powdered and dried K2CO3, and 20 mL of toluene were added. The
reaction bath was heated to 145 ꢀC for 4 h in order to azeotropically
remove water from the system. The bath temperature was slowly
raised to 170 ꢀC by the controlled removal of toluene and allowed to
proceed at 170 ꢀC for 24 h. The mixture was cooled to room
temperature and filtered to remove the salt byproduct, then coag-
ulated in 1 L of methanol. The precipitated oligomer was stirred in
methanol for 24 h and then dried under vacuum at 140 ꢀC for 48 h.
The yield was 90%.
2.4. End-capping of phenoxide terminated PEEKt hydrophobic
oligomer with DFBP
2. Experimental
Phenoxide terminated PEEKt oligomers were end-capped with
DFBP via a nuclephilic aromatic substitution mechanism. A sample
end-capping reaction of 7000 g/mol PEEKt oligomer is as follows;
2.1. Materials
4,40-Difluorobenzophenone (DFK), was purchased from TCI
America, and purified by recrystallization from ethanol. Decaf-
luorobiphenyl (DFBP) was purchased from Aldrich and used as
received. Hydroquinone was provide by Eastman Chemical
Company, and recrystallized from ethanol. 4, 40-hexa-
fluoroisopropylidenediphenol (6F-BPA), received from Ciba, was
sublimated and then recrystallized twice from toluene. Monomer
grade 4,40-dichlorodiphenylsulfone (DCDPS) and 4,40-biphenol (BP)
were provided by Solvay Advanced Polymers and dried under
vacuum at 120 ꢀC prior to use. 3,30-Disulfonated-4,40-dichlor-
odiphenylsulfone (SDCDPS) was synthesized from DCDPS and
purified according to a procedure developed by us and reported
elsewhere [36,37]. Aniline was purchased from Aldrich and purified
by vacuum distillation from calcium hydride. N-Methyl-2-
pyrrolidinone (NMP), N,N-dimethylacetamide (DMAc), cyclo-
hexane, and toluene were purchased from Aldrich and distilled
from calcium hydride before use. The 3 Å molecular sieves and
potassium carbonate (K2CO3) were purchased from Aldrich and
dried under vacuum at 180 ꢀC prior to use. Dimethyl sulfoxide
(DMSO), chloroform, acetone, methanol and 2-propanol (IPA) were
purchased from Fisher Scientific and used without further
purification.
a
100-mL three-necked round-bottom flask, equipped with
a mechanical stirrer, a nitrogen inlet, a DeaneStark trap, and
a condenser was charged with 7.000 g (1 mmol) of phenoxide
terminated DFKt oligomer, 0.5528 g (4 mmol) of K2CO3 and 70 mL
DMAc. The mixture was stirred until dissolved and 25 mL of toluene
were added. The reaction bath was heated to 145 ꢀC in order to
azeotropically remove water from the system. After 4 h, toluene
was removed and the reaction temperature was set at 105 ꢀC, then
2.0067 g (6 mmol) DFBP was added into the flask. The reaction was
allowed to proceed at 105 ꢀC for 24 h, cooled to room temperature,
filtered to remove the salt byproduct, and then coagulated in 1 L of
methanol. The precipitated oligomer was stirred in methanol for
24 h and then dried under vacuum at 140 ꢀC for 48 h. The yield was
over 90%.
2.5. Synthesis of phenoxide terminated fully disulfonated
hydrophilic oligomer (BPS100)
Phenoxide terminated fully disulfonated hydrophilic oligomers
(BPS100) were synthesized with different molecular weights by
offsetting the stoichiometry according to the Carothers equation. A
sample synthesis of 7000 g/mol BPS100follows; a 100-mL three-
necked round-bottom flask, equipped with a mechanical stirrer,
a nitrogen inlet, a DeaneStark trap, and a condenser was charged
with 4.0230 g (0.0216 mol) of BP, 9.8020 g (0.0200 mol) of SDCDPS,
and 70 mL DMAc. The mixture was stirred until dissolved, then
3.50 g (0.0254 mol) of K2CO3 and 35 mL toluene were added. The
reaction bath was heated to 145 ꢀC for 4 h in order to azeotropically
remove water from the system. The bath temperature was slowly
raised to 180 ꢀC by the controlled removal of toluene and allowed to
proceed at 180 ꢀC for 72 h. The mixture was cooled to room
temperature, filtered to remove most of the salt, and then coagu-
lated in 2 L of acetone. The precipitated oligomer was stirred in
2.2. Synthesis of N-phenyl(4,40-difluorodiphenyl) ketimine (DFKt)
DFK (60 g, 0.275 mol) and aniline (40 mL, 0.44 mol) were added
to a two-necked round-bottom flask equipped with a nitrogen inlet,
DeaneStark trap and a condenser. Toluene (250 mL), along with
150 g of 3 Å molecular sieves were added into the flask. The reac-
tion bath was heated to 140 ꢀC to let toluene reflux over 24 h until
100% conversion to ketimine had occurred as confirmed by 1H
NMR. Toluene and excess aniline were then removed by rotary
evaporation. The ketimine product was recrystallized twice from

Y. Chen et al. / Polymer 53 (2012) 3143e3153
3145
2.7. Film casting, hydrolysis and acidification
The PEEKt-BPS100 copolymers in the potassium salt form were
dissolved in DMSO (7% w/v) and filtered through 0.45 mm PTFE
syringe filters. The solutions were then cast onto clean glass
substrates and dried for 24 h with an infrared lamp at 45e55 ꢀC.
The residual solvent was largely removed by drying in a vacuum
oven at 190 ꢀC for 24 h. The amorphous PEEKt-BPS100 membranes
were then converted into the semi-crystalline acid form PEEK-
BPSH100 in 0.5 M hydrochloric acid (HCl) solution at 80 ꢀC for
24 h, followed by boiling in 0.5 M sulfuric acid (H2SO4) solution for
2 h and in deionized water for 2 h. The initial use of HCl facilitated
the removal of the aniline salt.
2.8. Characterization
1H NMR and 19F NMR analyses were conducted on a Varian
INOVA spectrometer operating at 400 MHz. The spectra of BPS100
oligomers and 6FK-BPS100 multiblock copolymers were obtained
from a 10% solution (w/v) in a DMSO-d6 at room temperature. The
spectra of DFBP end-capped PEEKt hydrophobic oligomers were
obtained from a solution in CDCl3. 13C NMR analyses were con-
ducted on a Varian Unity spectrometer, operating at 100.58 mHz
with DMSO-d6 as the solvent.
Fig. 1. Synthesis of amorphous phenoxide terminated poly(ether ether ketimine)
oligomer (PEEKt).
acetone for 24 h and dried under vacuum at 140 ꢀC for 48 h. The
yield was 90%.
2.6. Synthesis of amorphous hydrophobicehydrophilic multiblock
copolymers (PEEKt-BPS100)
Absorbance mode FT-IR spectra of thin films were obtained
using a Bruker Tensor 27 spectrophotometer in a wave number
range from 600 cmꢁ1 to 4000 cmꢁ1
.
A typical multiblock copolymerization coupling reaction of a 9k-
7k PEEkt-BPS100 multiblock copolymer was conducted as follows:
5.00 g (0.714 mmol) of 7 kg/mol BPS100, 0.25 g (1.81 mmol) of
K2CO3, 50 mL of NMP and 15 mL of cyclohexane were added to
a 250-mL three-necked flask equipped with a mechanical stirrer,
a nitrogen inlet, a DeaneStark trap and a condenser. The reaction
bath was heated to 125 ꢀC for 6 h to dehydrate the system. After
removing cyclohexane, the reaction temperature was lowered to
105 ꢀC, and 6.50 g (0.722 mmol) of 9 kg/mol DFBP end-capped
PEEKt oligomer was added. The coupling reaction was allowed to
proceed for 24 h. The reaction mixture was filtered and precipitated
into 1 L of IPA. The copolymer was purified in a Soxhlet extractor
with methanol for 24 h, then with chloroform for 24 h. The
copolymer was dried under vacuum at 140 ꢀC for 24 h. The yield
was 85%.
Intrinsic viscosity (IV) values for PEEKt, DFBP end-capped PEEKt,
BPS100 oligomers and PEEKt-BPS100 multiblock copolymers were
obtained from a size exclusion chromatography (SEC) equipped
with a Waters 1515 isocratic HPLC pump, a Waters autosampler,
a Waters HR5eHR4eHR3 column set, a Waters 2414 refractive-
index detector, and a Viscotek 270 viscometric detector. NMP
(containing 0.05 M LiBr) was used as the mobile phase [38].
Thermogravimetric analyses (TGA) of the PEEK-BPSH100
membranes were determined with the TA Instrument TGA Q500.
Prior to TGA characterization, all the samples were vacuum-dried
and kept in the TGA furnace at 150 ꢀC for 30 min in order to
remove residual solvent and moisture. The samples were then
evaluated over the temperature range of 50e650 ꢀC at a heating
rate of 10 ꢀC/min under an air atmosphere. Glass transition
temperatures (Tgs) and melting exotherms were determined by
Fig. 2. End-capping of PEEKt oligomer with DFBP.

3146
Y. Chen et al. / Polymer 53 (2012) 3143e3153
Fig. 3. 19F NMR of DFBP end-capped PEEKt with 6F-BPA as a reference.
Table 1
3 MHz to 10 Hz under two different conditions (liquid water at
IV of PEEKt before and after DFBP end-capping.
30 ꢀC and relative humidity (RH) from 30% to 95% at 80 ꢀC). The
resistance value of each membrane was obtained by averaging at
least five measurements to guarantee the water equilibration of the
membranes at certain temperature and humidity. The wiring
system for measuring the ohmic resistance was shielded from the
platinum electrodes to the potentiometer (a combined system of an
electrochemical interface, Solartron 1287, and an impedance/gain-
phase analyzer, Solartron 1252A) to avoid electromagnetic noise.
Target Mn
(g/mol)
Hydrophobic blocks
Mn (g/mol) (measured
IVb (dL/g) (before
end-capping)
IVc (dL/g) (after
end-capping)
a
from 19F NMR)
7k
9k
14k
17k
21k
7300
9200
14,500
17,400
20,800
0.16
0.20
0.27
0.31
0.36
0.17
0.24
0.29
0.38
0.42
The proton conductivity (
s
, S cmꢁ1) was obtained from the
a
Mn were calculated from Eq. (3).
Measured by SEC with 0.05 M LiBr/NMP as mobile phase at 60 ꢀC.
Measured by SEC with 0.05 M LiBr/NMP as mobile phase at 60 ꢀC.
following equation (Eq. (1)):
b
c
L
R$S
s
¼
(1)
differential scanning calorimetry (DSC) with a TA Instruments DSC
Q-1000 at a heating rate of 10 ꢀC/min under a stream of nitrogen.
Where
s
(S/cm) is proton conductivity, L (cm) is the distance
between two electrodes, R (U) is the resistance of the membrane
and S (cm2) is the surface area that protons transport though the
membrane. For measurements of proton conductivity in liquid
water, each membrane was fully hydrated by soaking in DI water at
room temperature for 24 h. For determining proton conductivity
under partially hydrated conditions, membranes were equilibrated
2.9. Determine of proton conductivity
The ohmic resistance (R,
measured using a test cell [39] via four-point probe alternating
current (a.c.) impedance spectroscopy in the frequency range from
U) of the PEM membrane was
Fig. 4. Synthesis of phenoxide terminated fully disulfonated hydrophilic oligomer (BPS100).

Y. Chen et al. / Polymer 53 (2012) 3143e3153
3147
Fig. 5. 1H NMR of phenoxide terminated BPS100 oligomer.
in a humidity-temperature controlled oven (ESPEC, SH-240) at the
desired relative humidity (RH) level and 80 ꢀC for 45 min before
each measurement.
was not possible to use the integrations of 1H NMR spectra to
calculate the number average molecular weight (Mn).
3.2. Synthesis and characterization of DFBP end-capped PEEKt
oligomers
2.10. Determination of water uptake and dimensional swelling
The PEEKt and BPS100 oligomers possessed the same phenoxide
telechelic functionality and were necessary to modify one of the
oligomers to the activated halide functionality produce the multi-
block copolymers. Phenoxide terminated PEEKt oligomers were
end-capped with DFPB to achieve highly reactive fluorine-
terminated PEEKt, as shown in Fig. 2. A large molar excess (6
times the amount of PEEKt) of DFBP was added to minimize the
inter-oligomer coupling reaction of PEEKt oligomers. Due to the
high reactivity of DFBP, a mild reaction temperature (105 ꢀC) and
relatively short reaction time (w12 h) were sufficient to afford the
fluorine-terminated PEEKt.
The measurement of the Mns of the PEEKt oligomers was very
important for the synthesis of multiblock copolymers with proper
block lengths (Xn) and IECs. Analysis of a 19F NMR spectrum was
utilized to determine the Mn of the fluorine-terminated oligomer,
with a small amount 6F-BPA added as a reference. By comparing
the respective integrations of the fluorine groups of a quantitative
amount of PEEKt oligomer and 6F-BPA, the Mn of the oligomer was
determined. Fig. 3 shows a 19F NMR spectrum of precisely weighed
DFBP end-capped PEEKt-9k oligomer and 6F-BPA. The peak
at ꢁ64.2 ppm is attributed to the fluorine in the 6F-BPA. Five peaks
at ꢁ137.3, ꢁ137.9, ꢁ149.9, ꢁ152.9 and ꢁ160.3 ppm are attributed to
The water uptake of all membranes were determined gravi-
metrically. The acidified PEEK-BPSH100 membranes (w25 microns)
were equilibrated in DI water at room temperature for 2 days. Wet
membranes were removed from the DI water, blotted dry to
remove surface droplets, and quickly weighed. The membranes
were then dried at 120 ꢀC under vacuum for 24 h and weighed
again. The water uptake of the membranes was calculated
according to the following equation (Eq. (2)), where Wdry and Wwet
refer to the mass of the dry and wet membrane, respectively.
Wwet ꢁ W
Water Uptakeð%Þ ¼
dry ꢂ 100
(2)
Wdry
The volume swelling ratios of the membranes were determined
from the dimensional changes from wet to dry state. Membranes
were equilibrated in DI water, and dimensions in the wet state were
measured. The dried dimensions were obtained after drying the
wet membrane at 80 ꢀC in a convection oven for 2 h.
3. Results and discussion
3.1. Synthesis of phenoxide terminated PEEKt oligomers
Table 2
Characterization of hydrophilic (BPS100) oligomers.
Phenoxide terminated amorphous PEEKt hydrophobic oligo-
mers were synthesized by copolymerization of DFKt and excess
hydroquinone monomers as shown in Fig. 1. The number average
molecular weight and end group functionality of the oligomers
were precisely controlled by offsetting the molar feed ratios of
monomers according to the Carothers equation. In all cases, the
molar feed ratios of HQ over DFKt were larger than 1 to afford
phenoxide telechelic functionality and target number-average
molecular weights from 4 kg/mol to 21 kg/mol. Unfortunately,
due the overlap of the proton signals in the 1H NMR spectrum, it
Target Mn (g/mol)
Hydrophilic blocks
IVb (dL/g)
a
Mn (g/mol)
5000
7000
5200
7100
0.13
0.18
0.20
0.28
0.32
9000
9100
13,000
15,000
13,700
14,600
a
Measured by 1H NMR.
Measured by SEC with 0.05 M LiBr/NMP as mobile phase at 60 ꢀC.
b

3148
Y. Chen et al. / Polymer 53 (2012) 3143e3153
-1.0
-1.2
-1.4
-1.6
-1.8
-2.0
-2.2
a nucleophilic aromatic substitution reaction. Excess BP monomer
was used to control the telechelic functionality and molecular
weight, as shown in Fig. 4.
The Mns of the hydrophilic oligomers were determined by 1H
NMR. Fig. 5 shows the 1H NMR spectrum of BPS100; protons from
the terminal BP moieties were assigned to four small peaks at 6.80,
7.05, 7.40, and 7.55 ppm, while the peaks at 7.10 and 7.65 were
assigned to the protons of the BP moieties in the middle of BPS100
oligomer backbone. By comparing the integrations of the protons
on the end-group BP to the protons on the SDCDPS moieties, the Mn
of each BPS100 oligomer was determined.
y=0.81x-8.97
The molecular weights (Mn) and IV obtained from size exclusion
chromatography (SEC) are summarized in Table 2. An increase in IV
was correlated with the Mn of BPS100. There is a linear relationship
between the logMn and logIV, which indicates that control of the
molecular weights for the hydrophilic oligomers was successful, as
shown in Fig. 6.
8.6
8.8
9.0
9.2
9.4
9.6
3.4. Synthesis and characterization of PEEKt-BPS100 multiblock
copolymers
ln[Mn(g/mol)]
Fig. 6. ln
h vs. ln Mn plot of BPS100 oligomers.
A series of amorphous hydrophobicehydrophilic multiblock
copolymers was synthesized via the coupling reaction between
phenoxide terminated BPS100 and DFBP end-capped PEEKt oligo-
mers (Fig. 7). The stoichiometry of the reactions was set as 1:1. The
IV data confirmed that high molecular weight PEEKt-BPS100 mul-
tiblock copolymers were obtained.
the fluorines on the end-capping DFBP. The Mn of the DFBP end-
capped PEEKt can be calculated with the following equation (Eq.
(3)), where WPEEKt and W6F-BPA are the weights of PEEKt sample and
6F-BPA reference, MW of 6F-BPA is 336.23 g/mol. Ib, Ic and If are the
integrations of corresponding fluorines in Fig. 3. The calculated Mn
of PEEK includes the molecular weight of end-capping groups.
Fig. 8 shows the 1H NMR of a PEEKt-BPS100 multiblock copol-
ymer. It is difficult to monitor the degree of the coupling reaction
from the disappearance of the phenoxide end group peaks of
BPS100 (peak a in Fig. 5) as described in the literature [25]. At this
chemical shift position on 1H NMR spectrum of PEEKt-BPS100,
there are overlapping peaks from the protons of SDCDPS and
DFKt moieties (peak c and peak k).
The very high reactivity of DFBP end groups allowed the
copolymerization to be accomplished at a relatively low temper-
ature and short reaction time. This minimizes the possible
ethereether interchange scrambling side reactions. Fig. 9 shows
the 13C NMR of PEEKt-BPS100 9k-7k. In the chemical shift range of
150e170 ppm, the peaks can be properly assigned to the carbons
on the backbone of the PEEKt-BPS100 polymer. The two peaks at
156.3 ppm and 158.8 ppm, which are assigned to the carbons on
each end of the ether bond in BPS100 blocks (Fig. 9 peaks b and a),
WPEEKt
Mn of PEEKt
W6FꢁBPA
MW of 6F ꢁ BPA
Ib þ Ic
8
¼
(3)
If
ꢂ 6
The intrinsic viscosity (IV) of PEEKt before and after DFBP end-
capping as well as the Mn calculated from 19F NMR are listed in
Table 1. After end-capping with DFBP, the IV of PEEKt only increased
slightly due to chain length extension by the end-capping reagent
DFBP.
3.3. Synthesis and characterization of phenoxide terminated
BPS100 oligomers
Phenoxide terminated fully disulfonated poly(arylene ether
sulfone) hydrophilic oligomers (BPS100) were synthesized via
Fig. 7. Synthesis of PEEKt-BPS100 multiblock copolymers.

Y. Chen et al. / Polymer 53 (2012) 3143e3153
3149
Fig. 8. 1H NMR of PEEKt-BPS100 9k-7k multiblock copolymer.
are still sharp and narrow. This phenomenon suggests that the
multiblock copolymer maintains an ordered sequence and
confirms the minimization of the ethereether interchange side
reactions.
7k-7k). After this procedure, the semi-crystallinity of the PEEK
blocks prevented the PEEK-BPSH100 multiblock copolymer from
dissolution in any organic solvent, such as DMSO, DMAc or NMP.
Resistance to organic solvents could be important in some
applications.
Fourier transform infrared (FT-IR) spectroscopy was used to
characterize the functional groups. The hydrolysis and acidification
of the PEEKt-BPS100 to PEEK-BPSH100 membranes was deter-
mined by FT-IR. Fig. 11 shows the FT-IR spectra of a PEEKt-BPS100
17k-13k membrane before and after acidification. The absorption
bands at 1247 cmꢁ1, 1095 cmꢁ1 and 1029 cmꢁ1 can be assigned to
asymmetric and symmetric O]S]O stretching vibrations of the
sulfonate groups [8,29]. The absorption band assigned to the SeO
stretching of sulfonate group shifts from 696 cmꢁ1 to 686 cmꢁ1
after acidification. This absorption band shift confirms the acidifi-
cation of the potassium sulfonate groups to the desired sulfonic
acid groups.
3.5. Film casting, hydrolysis and acidification
The PEEKt-BPS100 multiblock copolymer membranes were cast
from DMSO solutions. The membranes were then boiled in aqueous
0.5 M HCl then in 0.5 M H2SO4 solutions in order to hydrolyze the
amorphous PEEKt blocks to the semi-crystalline PEEK blocks and to
simultaneously acidify the salt form BPS100 blocks to the acid form
BPSH100 blocks, as Fig. 10 shows. The reason for using 0.5 M HCl
first was that the aniline chloride byproduct is easier to remove
from the membrane. After acidification, the loss of the ketimine
groups induced a change in the molecular weight of the hydro-
phobic blocks (e.g. PEEKt-BPS100 9k-7k becomes PEEK-BPSH
Fig. 9. 13C NMR of PEEKt-BPS100 9k-7k multiblock copolymer.

3150
Y. Chen et al. / Polymer 53 (2012) 3143e3153
F
F
F
F
F
F
F
KO3S
O
N
C
O
S
O
O
O
O
y
x
F
SO3K
0.5 M HCl 80 oC
0.5 M H2SO4 100 oC
F
F
F
F
F
F
HO3S
O
O
C
O
S
O
O
O
O
y
x
F
F
SO3H
Fig. 10. Hydrolysis and acidification to convert amorphous PEEKt-BPS100 to semi-crystalline PEEK-BPSH100.
A representative FT-IR spectrum of PEEKt-BPS100 17k-13k
The amorphous PEEKt-BPS100 17k-13k was hydrolyzed and
acidified to semi-crystalline PEEK-BPSH100 13k-13k in dilute
hydrochloric acid and then sulfuric acid solution. The DSC ther-
mogram (2nd heating cycle) of PEEK-BPSH100 13k-13k is shown in
Fig. 13. A very broad thermal transition from 150 ꢀC to 230 ꢀC was
observed. This very broad thermal transition suggests the
decreased Tg of BPSH100 blocks after acidification may partially
overlaps with the Tg of PEEK blocks. However a melting endotherm
at 325 ꢀC was observed and attributed to the presence of semi-
crystalline domains in the PEEK-BPSH100 13k-13k membrane.
Thermal and oxidative stability of the PEEK-BPSH100 multiblock
copolymers were investigated by TGA, as shown in Fig. 14. The TGA
was conducted in an air atmosphere to assess the oxidative stability
of the copolymers. Such copolymers displayed a two-step weight
loss; the first was observed at 270 ꢀC assigned to the desulfonation
of BPSH100 blocks, the second degradation began at 450 ꢀC, which
was related to the decomposition of the copolymer backbones. The
(Fig.11, top) exhibits a large peak at 1590 cmꢁ1, which is a overlap of
C]N stretching vibration and CeC vibration in aromatic rings
[33,35]. The decease of the C]N stretching peak and the appear-
ance of the C]O stretching peak at 1650 cmꢁ1 in the PEEK-
BPSH100 13k-13k spectrum (Fig. 11, bottom) confirm the hydro-
lysis of PEEKt-BPS100 to PEEK-BPSH100.
3.6. Thermal characterization
Two thermal transitions at around 165 ꢀC and a broad change at
245 ꢀC were observed in the differential scanning calorimetry (DSC)
thermogram of PEEKt-BPS100 17k-13k, as Fig. 12 shows. These can
be assigned to the glass transition temperature of the hydrophobic
PEEKt blocks and that of hydrophilic BPS100 blocks, respectively.
No melting endotherm transition was observed since the soluble
PEEKt-BPS100 series copolymers are amorphous.
A.
B.
PEEKt-BPS100 17k-13k (top)
PEEK-BPSH100 13k-13k (bottom)
A
696 cm-1
686 cm-1
1590 cm-1
B
1650 cm-1
2000
1800
1600
1400
1200
1000
800
600
400
200
Wave Nmuber (cm-1)
Fig. 11. FT-IR spectra of PEEKt-BPS100 17k-13k (before hydrolysis) (top) and PEEK-BPSH100 13k-13k (after hydrolysis) (bottom).

Y. Chen et al. / Polymer 53 (2012) 3143e3153
3151
100
80
60
40
20
0
PEEKt-BPS100 17k-13k
Broad thermal transition of
BPS100 blocks
Thermal transition of
PEEKt blocks
PEEK-BPSH 5k-5k
PEEK-BPSH 13k-13k
PEEK-BPSH 17k-17k
2nd Heating Cycle
Exo down
100
200
300
400
500
600
700
Temperature (oC)
100
150
200
250
300
Temperature (oC)
Fig. 14. TGA traces of PEEK-BPSH100 multiblock copolymers.
Fig. 12. DSC thermogram of PEEKt-BPS100 17k-13k.
Table 3
Characterization of PEEK-BPSH100 multiblock copolymers.
block length variation in the copolymers did not affect the thermal
and oxidative stability. All the copolymers showed very similar
weight loss behavior indicating the control of IEC (or degree of
disulfonation) was successful.
PEEKt-
BPS100
PEEK-
BPSH100
IV (dL/g)a
IEC
Water
uptake
(wt%)
Proton
(meq/g)b
conductivity
(S/cm)c
7k-5k
12k-9k
17k-13k
21k-17k
NafionÒ 112
BPSH40
5k-5k
9k-9k
13k-13k
17k-17k
0.83
0.96
1.10
1.13
N/A
1.60
1.65
1.70
1.70
0.90
1.52
47
50
53
85
22
50
0.11
0.12
0.12
0.13
0.12
0.11
3.7. Characterization of the membrane properties of PEEK-BPSH100
0.85
Tough, ductile, and transparent PEEKt-BPS100 membranes were
cast from DMSO. Table 3 shows some properties of PEEKt-BPS100
and PEEK-BPSH100 multiblock copolymer membranes. After
hydrolysis and acidification, the PEEK-BPSH100 multiblock copol-
ymer membranes turned to light yellow and were insoluble in
normal organic solvents such as NMP and DMSO, due to the high
crystallinity of the PEEK blocks. The water uptake of PEEK-BPSH100
block copolymers increases with an increase in block length, which
is consistent with other reported block copolymers [25].
a
Measured by SEC with 0.05 M LiBr/NMP as mobile phase at 60 ꢀC, in PEEKt-
BPS100 state.
b
c
Measured from 1H NMR.
Measured in liquid water at 30 ꢀC.
copolymer with 40% of SDCDPS (BPSH40) and a NafionÒ 212 were
also included as controls. The proton conductivity is relatively
constant with the increase in block length for PEEK-BPSH100 block
copolymer membranes. All the PEEK-BPSH100 membranes
exhibited comparable or higher conductivity values than random
copolymer BPSH40 and NafionÒ 212 in liquid water at 30 ꢀC.
The proton conductivity of the PEEK-BPSH100 multiblock
copolymer membranes were measured in liquid water at 30 ꢀC as
shown in Table 3. A random sulfonated poly(arylene ether sulfone)
1
Heating cycle of PEEK-BPSH 13k-13k
0.1
A
0.01
B
C
A.
B.
C.
PEEK-BPSH100 17k-17k
Nafion 212
BPSH40 random copolymer
1E-3
2nd Heating Cycle
Exo down
40
50
60
70
80
90
100
50
100
150
200
250
300
350
Relative Humidity (%)
Temperature (oC)
Fig. 15. Proton conductivity vs. RH of PEEK-BPSH100 17k-17k multiblock copolymers
with NafionÒ and BPSH40 as references (measured at 80 ꢀC).
Fig. 13. DSC thermogram of PEEK-BPSH100 13k-13k.

3152
Y. Chen et al. / Polymer 53 (2012) 3143e3153
X direction
Y direction
Z direction
80
60
40
20
0
NRE112
BPSH35 6FK-BPSH17k-17k 5k-5k
9k-9k
13k-13k
17k-17k
Polymer
Fig. 16. Comparison of dimensional swelling data for PEEK-BPSH100 multiblock, copolymers, BPSH35, 6FK-BPSH 17k-17k, and NafionÒ 112.
Fig. 15 shows proton conductivity at 80 ꢀC as a function of
BPSH100 membranes. The lower in-plane swelling of PEEK-
BPSH100 membranes may be important in fuel cell MEAs to keep
the in-plane interface stress low, and to prevent the failure of the
MEAs under dryewet cycling conditions.
relative humidity (RH) for PEEK-BPSH 17k-17k, NafionÒ 212 and
BPSH40 membranes. The proton conductivity of the randomly
copolymerized BPSH40 dropped significantly at lower RH levels as
expected. Sufficient water molecules provide proton transport
under fully hydrated states; however, the isolated hydrophilic
domains in the random copolymer can’t facilitate proton transport
under partially hydrated conditions. In contrast, under partially
hydrated states, proton species are better transported along the
sulfonic acid groups and water molecules through the long, co-
continuous channels in multiblock copolymers.
4. Conclusions
Multiblock semi-crystalline hydrophobic solvent resistance
poly(ether ether ketone)-hydrophilic disulfonated poly(arylene
ether sulfone) (PEEK-BPSH100) copolymers were synthesized and
characterized. PEEK-BPSH100 multiblock copolymers were
produced by hydrolysis of the amorphous multiblock PEEKt-
BPS100 ketimine protected pre-copolymers. FT-IR spectra
confirmed the hydrolysis of PEEKt-BPS100 to PEEK-BPSH100. DSC
thermograms showed a sharp melting endotherm, confirming the
presence of semi-crystalline domains in the PEEK-BPSH100. The
proton conductivities of the multiblock semi-crystalline PEEK-
BPSH100 copolymers were comparable or higher than that of
Nafion212 and BPSH40 random copolymer. PEEK-BPSH100 17k-17k
shows better conductivity performance under low RH, presumably
due to its more distinct nanophase morphology separation
between semi-crystalline hydrophobic domains and disulfonated
hydrophilic domains.
3.8. Swellingedeswelling behavior of PEEK-BPSH100 multiblock
copolymer
The compatibility between the proton exchange membrane
(PEM) and electrodes is always a major concern under real opera-
tion conditions for a membrane electrode assembly (MEA). When
a
proton exchange membrane shows a significant in-plane
swelling, during the wetedry cycles, considerable stress can be
created at the interface between the membrane and electrodes. The
high stress could lead to premature failure of the MEA.
The dimension swelling behavior of PEEK-BPSH100 multiblock
copolymers with BPSH35, a amorphous partially fluorinated pol-
y(arylene ether ketone sulfone) hydrophilic-hydrophobic multi-
block copolymers (6FK-BPSH100 17k-17k) and NafionÒ used as
comparison is shown in Fig. 16. The BPSH35 and NafionÒ 112
membranes show isotropic swelling behavior. In contrast, PEEK-
BPSH100 multiblock copolymer membranes exhibit anisotropic
swelling behavior. For PEEK-BPSH100 membranes, in-plane
swelling (x- and y-directions) decreases, while through-plane (z-
direction) swelling increases as the block length increases. This
behavior indicates the development of enhanced ordered lamellar
morphology as the block length increases. PEEK-BPSH100 17k-17k
shows a significantly low in-plane swelling ratio compared to its
corresponding 6FK-BPSH 17k-17k amorphous multiblock copol-
ymer membrane, which may be induced by the enhanced semi-
crystalline hydrophobic domains in longer block length PEEK-
Acknowledgments
The authors thank Giner Electrochemical Systems, LLC, the
Department of Energy (DE-EE0000471), and Los Alamos National
Laboratory for their support.
References
[1] Winter M, Brodd RJ. Chem Rev 2004;104(10):4245e69.
[2] Whittingham MS, Zawodzinski T. Chem Rev 2004;104(10):4243e4.
[3] Hickner MA, Ghassemi H, Kim YS, Einsla BR, McGrath JE. Chem Rev 2004;
104(10):4587e611.
[4] Kerres JA. J Membr Sci 2001;185(1):3e27.
[5] Mauritz KA, Moore RB. Chem Rev 2004;104(10):4535e85.
[6] Rozière J, Jones DJ. Annu Rev Mater Res 2003;33.
[7] Higashihara T, Matsumoto K, Ueda M. Polymer 2009;50(23):5341e57.

Y. Chen et al. / Polymer 53 (2012) 3143e3153
3153
[8] Wang F, Hickner M, Kim YS, Zawodzinski TA, McGrath JE. J Membr Sci 2002;
[23] Li YX, Roy A, Badami AS, Hill M, Yang J, Dunn S, et al. J Power Sources 2007;
172(1):30e8.
197(1e2):231e42.
[9] Chikashige Y, Chikyu Y, Miyatake K, Watanabe M. Macromolecules 2005;
[24] Lee HS, Roy A, Badami AS, McGrath JE. Macromol Res 2007;15(2):160e6.
[25] Lee HS, Roy A, Lane O, Dunn S, McGrath JE. Polymer 2008;49(3):715e23.
[26] Badami AS, Lane O, Lee H-S, Roy A, McGrath JE. J Membr Sci 2009;333(1e2):
1e11.
38(16):7121e6.
[10] Shang XY, Tian SH, Kong LH, Meng YZ.
J Membr Sci 2005;266(1e2):
94e101.
[11] Summer MJ, Harrison WL, Weyers RM, Kim YS, McGrath JE, Riffle JS, et al.
J Membr Sci 2004;239(2):199e211.
[27] Wang H, Badami AS, Roy A, McGrath JE. J Polym Sci. Part A Polym Chem 2007;
45(2):284e94.
[12] Miyatake K, Zhou H, Matsuo T, Uchida H, Watanabe M. Macromolecules 2004;
37(13):4961e6.
[28] Yu X, Roy A, Dunn S, Badami AS, Yang J, Good AS, et al. J Polym Sci. Part A
Polym Chem 2009;47(4):1038e51.
[13] Einsla BR, Kim YS, Hickner MA, Hong YT, Hill ML, Pivovar BS, et al. J Membr Sci
2005;255(1e2):141e8.
[14] Asensio JA, Borros S, Gomez-Romero P. J Membr Sci 2004;241(1):89e93.
[15] Wang F, Hickner M, Ji Q, Harrison W, Mecham J, Zawodzinski TA, et al.
Macromol Symp 2001;175:387e95.
[29] Ghassemi H, McGrath JE, Zawodzinski TA. Polymer 2006;47(11):4132e9.
[30] Chen Y, Guo R, Lee CH, Lee M, McGrath JE. Int J Hydrogen Energy 2012;37(7):
6132e9.
[31] Attwood TE, Dawson PC, Freeman JL, Hoy LRJ, Rose JB, Staniland PA. Polymer
1981;22(8):1096e103.
[16] Badami AS, Roy A, Lee H-S, Li Y, McGrath JE. J Membr Sci 2009;328(1e2):
156e64.
[32] Bishop MT, Karasz FE, Russo PS, Langley KH. Macromolecules 1985;18(1):
86e93.
[17] Lee HS, Badami AS, Roy A, McGrath JE. J Polym Sci Part A Polym Chem 2007;
45(21):4879e90.
[33] Lindfors BE, Mani RS, Mcgrath JE, Mohanty DK. Macromol Rapid Commun
1991;12(6):337e45.
[18] Badami AS, Lee H-S, Li Y, Roy A, Wang H, McGrath JE. Fuel cell Chemistry and
operation (ACS Symposium series). Am Chem Soc 2010;1040:65e81.
[19] Einsla ML, Kim YS, Hawley M, Lee H-S, McGrath JE, Liu B, et al. Chem Mater
2008;20(17):5636e42.
[20] Elabd YA, Hickner MA. Macromolecules 2010;44(1):1e11.
[21] Roy A, Hickner MA, Yu X, Li YX, Glass TE, McGrath JE. J Polym Sci. Part B Polym
Phys 2006;44(16):2226e39.
[34] Brink AE, Gutzeit S, Lin T, Marand H, Lyon K, Hua T, et al. Polymer 1993;34(4):
825e9.
[35] Bourgeois Y, Devaux J, Legras R, Parsons IW. Polymer 1996;37(14):3171e6.
[36] Sankir M, Bhanu VA, Harrison WL, Ghassemi H, Wiles KB, Glass TE, et al. J Appl
Polym Sci 2006;100(6):4595e602.
[37] Li Y, VanHouten RA, Brink AE, McGrath JE. Polymer 2008;49(13e14):3014e9.
[38] Yang J, Li Y, Roy A, McGrath JE. Polymer 2008;49(24):5300e6.
[39] Lee CH, Park HB, Lee YM, Lee RD. Ind Eng Chem Res 2005;44(20):7617e26.
[22] Roy A, Lee H-S, McGrath JE. Polymer 2008;49(23):5037e44.