Partially fluorinated sulfonated poly(arylene sulfone)s blended with polybenzimidazole

Article abstract of DOI:10.1002/pola.24624

A partially fluorinated and sulfonated poly(arylene sulfone) (SPSO) was successfully synthesized via nucleophilic polycondensation of 2,2-bis(4-fluorophenyl)hexafluoro-propane with 4,4′-thiobisbenzenethiol (TBBT). In a second step, the prepared poly(aryle

Full text of DOI:10.1002/pola.24624

Partially Fluorinated Sulfonated Poly(arylene sulfone)s
Blended with Polybenzimidazole
ANIKA KATZFUß, KATICA KRAJINOVIC, ANDREAS CHROMIK, JOCHEN KERRES
Institute of Chemical Process Engineering, University of Stuttgart, Boeblingerstrasse 78, Stuttgart 70199, Germany
Received 7 December 2010; accepted 7 February 2011
DOI: 10.1002/pola.24624
Published online 1 March 2011 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: A partially fluorinated and sulfonated poly(arylene
sulfone) (SPSO) was successfully synthesized via nucleophilic
polycondensation of 2,2-bis(4-fluorophenyl)hexafluoro-propane
with 4,40-thiobisbenzenethiol (TBBT). In a second step, the pre-
pared poly(arylene sulfide) was oxidized to SPSO. The polymer
were analyzed in this work. Additionally, the blend membranes
were characterized by gel permeation chromatography. The
novel SPSO blend shows a high molecular weight, and its
blend membrane with PBIOO has an excellent onset of ASO3H
group splitting-off temperature (TSO H,onset) of 334 ꢀC. The pro-
3
R
was blended with the polybenzimidazole PBIOOV to obtain a
ton conductivity amounts to 0.11 S cmꢁ1, and the water uptake
C
V
mechanically stable membrane. This film was compared with
other polymer blends, which were synthesized in our group in
the last years. We were especially interested in the influence of
different bridging groups such as ether, ketone, and sulfone
groups. The affect on properties such as water uptake (WU),
thermal stability, proton conductivity, and oxidative stability
reaches 30%. 2011 Wiley Periodicals, Inc. J Polym Sci Part A:
Polym Chem 49: 1919–1927, 2011
KEYWORDS: blends, crosslinking; gel permeation chromatogra-
phy (GPC); ionical crosslinking; PBIOOV; sulfonated poly(ary-
R
lene sulfone)s
radical. Another positive effect of electron-deficient aromatic
polymer building blocks-bearing sulfonic acid groups is their
higher acidity, which leads to a higher dissociation degree of
the pendent SO3H group, which is associated with high pro-
ton conductivity. From the investigations of the different
polymer types prepared in our group, we found that their
mechanical and radical stabilities were not as good as
desired. Particularly, when having a high content in proton
conducting SO3H groups, the polymers were very brittle and
even dissolved in water at ion-exchange capacities (IECs) of
higher than 2- to 2.5-meq SO3H gꢁ1 polymer. Moreover, the
radical stability, namely, of the nonfluorinated arylene main
chain ionomers was not satisfying.
INTRODUCTION Sulfonated arylene main chain ionomers
show the best chemical stabilities next to the perfluorinated
ionomers. A huge amount of different types of sulfonated
arylene main chain ionomers has been developed in the last
decades as potential substitutes for the perfluorinated
ionomers in electrochemical applications, particularly in
membrane fuel cells, including poly(phenylene)s,1,2 poly
(ether sulfone)s,3–5 poly(ether ketone)s,6–9 poly(phenylene
oxide)s,10,11 partially fluorinated poly(phenylene ether)s,12,13
poly(phenylene sulfide)s,14 poly(phenylene phosphine oxide
ether)s,15 poly(thioether sulfone)s,16 poly(arylene thioether
ketone sulfone)s,17 poly(sulfone)s by oxidation of the poly
(thioether sulfone)s,18 and poly(thioether ketone)s.19 In our
group, we have synthesized many different types of these
polymers in the last years, among them nonfluorinated and
partially fluorinated poly(ether)s, poly(ether sulfone)s, poly
(ether ketone)s, and poly(ether phosphine oxide)s.13,20–22
When comparing the nonfluorinated sulfonated arylene main
chain ionomers with those which are partially fluorinated,
we found that the partially fluorinated ionomer types show
better chemical stabilities compared to the nonfluorinated
ones.23 This can be attributed to the higher bond energy of
CAF bonds, compared to CAH bonds,24,25 and to the strong
-I effect of the F bond to the polymer, leading to a decrease
of the electron density of the aromatic building blocks, which
results in a good polymer stability against ipso desulfonation
and against attack of electrophilic radicals such as the OHꢂ
We achieved further improvement of the chemical and the
mechanical stabilities of arylene main chain ionomer mem-
branes by ionical crosslinking of the sulfonated ionomers via
blending with a stable basic polymer such as polybenzimida-
R
zole (PBIOO)V, in which the imidazole groups of PBIOO are
protonated by acid-base reaction with ASO3H groups, form-
ing [ImidazoleHþ] [SOꢁ3 ] crosslinking sites.26–35 As a matter
of fact, a part of the ASO3H groups of the arylene ionomers
is sacrificed by the ionical crosslinking, leading to a reduc-
tion of the proton conductivity of the formed blend mem-
brane, compared to the pure polymer. Therefore, the sulfo-
nated ionomer must have a sufficient SO3H content to be
suitable for use in ionical crosslinking reactions. By using an
ionical crosslinker such as PBI, the oxidative and thermal
Correspondence to: A. Katzfuß (E-mail: anika.katzfuss@icvt.uni-stuttgart.de)
C
V
Journal of Polymer Science Part A: Polymer Chemistry, Vol. 49, 1919–1927 (2011) 2011 Wiley Periodicals, Inc.
SPSOS BLENDED WITH PBIOO, KATZFUß ET AL.
1919

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
stability of the respective acid-base blend membrane is
increased.13 This enhancement of stability of an ionically
crosslinked blend membrane can be explained as follows:
when individual polymer chains within the blend membrane
are suffering from degradation, they cannot leave the mem-
brane matrix because of the ionical crosslinks, which retains
them in the membrane matrix. Therefore, membranes com-
prising a stable crosslinking network show a better oxidative
and thermal stability than membranes without a network.
Consequently, it requires more time until a membrane, which
contains a crosslinking network, is deteriorated.
tion of a novel monomer, 2,2-bis(4-fluoro-3-sulfonatophenyl)-
hexafluoropropane with TBBT, followed by oxidation using a
H2O2/CH3COOH mixture. The procedure for the polyconden-
sation is presented in Figure 1. The obtained high molecular
partially fluorinated poly(sulfone) was blended in a second
step with PBIOO to fine tailor the membrane properties. The
synthesis and characterization of the novel polymer and
blend membrane are reported in this contribution for the
first time. Furthermore, the polymer was compared to three
different polymers that have been synthesized in the past by
our group. We reported about a partially fluorinated sulfo-
nated poly(arylene ether) (SFS), a sulfonated poly(arylene
Apart from ionical crosslinking with basic polymers, the
chemical stability of arylene main chain ionomers can be
improved by avoidance of AOA, alkyl, and ¼¼C¼¼O bridges in
the arylene main chain, which are preferred sites for radi-
cal,23,36,37 and/or nucleophile attack.23,38 Schuster et al.18
have shown that sulfonated polysulfone, which does not con-
tain any of the aforementioned radical attack-sensitive
groups but solely consists of repeating units A(AS(O)2A)-3-
sulfo-1,4-phenylene-, comprises excellent hydrolysis (e.g.,
high stability against splitting-off of the sulfonic acid group
via ipso reaction) and radical stability.39–41 They have
explained these findings with the electron-attracting capabil-
ity of the sulfone bridge by their -I and the -M effect, which
strongly lowers the electron density in the aromatic building
blocks of the polymer. Moreover, this polymer shows excel-
ether sulfone) (SPSU) and
a
sulfonated poly(arylene
ether ketone) in several publications.13,20,21,26,42 All poly-
mers, which are discussed in this contribution are shown in
Figure 2.
EXPERIMENTAL
Materials
2,2-Bis(4-aminophenyl)hexafluoropropane from ABCR and
N,N-dimethylacetamid (DMAc) from Aldrich were used as
received. Thiobisbenzenethiol (TBBT) and anhydrous potas-
sium carbonate were also purchased at Aldrich and dried at
ꢀ
80 C in a vacuum oven (100 mbar) for 20 h before use. The
sulfonated monomer, which was prepared in our laboratory,
was dried at 50 ꢀC in a vacuum oven (100 mbar) for 20 h
R
before use. FumionV PBIOO was purchased from Fuma-Tech.
R
lent proton conductivities (factor 5–7 higher than NafionV,
even at low relative humidities).41 However, a disadvantage
of this polymer is its brittleness (if every second aromatic
ring of the polymer is sulfonated) or even water solubility (if
every aromatic ring of the polymer is sulfonated), which
makes the application of this polymer to membrane fuel cells
questionable or, in the case of the water-soluble polymer,
impossible. There are different possibilities to overcome the
mentioned disadvantages of the reported sulfonated poly
(sulfone)s: one is to blend the highly sulfonated/water-solu-
ble polymer with a basic polymer such as polybenzimidazole
(PBI) to generate water insoluble ionically crosslinked mem-
branes as mentioned above. The membrane properties are
tunable in a wide range by variation of the amount of the
basic PBI component in the blend membrane. Another possi-
bility to improve, particularly, the mechanical properties
of the membranes is to introduce flexible groups into the
poly(sulfone) backbone, which lower the membrane brittle-
ness and the water affinity of the polymer. One possible
bridging group to fulfill this objective is the extremely hydro-
phobic hexafluoroisopropylidene group AC(CF3)2A, which is
more flexible than the sulfone group and decreases the
polymer’s water affinity due to its higher hydrophobicity.
2,2-Bis(4-fluorophenyl)hexafluoropropane (M1)
The preparation of 2,2-bis(4-fluorophenyl)hexafluoropropane
has already been described earlier by Lau and Dougherty.43
Our group changed some of the reaction conditions, which
resulted in better yields.
2,2-Bis(4-aminophenyl)hexafluoropropane (20 g, 59.83 mmol)
was mixed with a 48 wt % tetꢀrafluoroboric acid solution
(70 mL). After cooling down to 0 C, sodium nitrite (16.51 g,
239.33 mmol) in water (60 mL) wasꢀadded dropwise during
30 min. After stirring for 30 min at 0 C, the light yellow solid
was filtered and washed with cold water. The bisdiazonium
salt was dried in a desiccator at room temperature for 24 h.
The dry component was dissolved in 300-mL anhydrous
ꢀ
ꢀ
toluene, stirred 1 h at 90 C, and, subsequently, 2 h at 120 C.
After cooling down to room temperature, the solution was
washed with water, saturated sodium hydrogen carbonate,
and finally sodium chloride. To isolate the product fractional
vacuum distillation was used.
M1
(3.3 ꢃ 10ꢁ2 mbar, 65–75 C). Yield: 15.22 g (74.76%).
ꢀ
On the basis of this concept, we decided to develop a sulfo-
nated poly(sulfone) in which some of the sulfone bridges are
substituted by hexafluoroisopropylidene bridges to increase
the mechanical flexibility of the ionomer.
We have synthesized a novel sulfonated poly(phenylene sul-
fone) in which every fourth sulfone bridge was substituted
by a hexafluoroisopropylidene group by the polycondensa-
dH (200 MHz; CDCl3) 7.06 (m, H-2), 7.35 (m, H-3); dC (100
MHz; CDCl3) 64.15 (quintet, J ¼ 25.62 Hz, C-5), 115.81 (d, J ¼
1920
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
condenser and a mechanical stirrer. N-methylpyrrolidone
(NMP) (40 mL) was added and after dissolving, potassium car-
bonate (47.97 mmol) was added to the solution. After stirring
at 180 ꢀC for 24 h, the solution was cooled down to room tem-
perature and poured into 1 L of isopropanol to precipitate the
polymer. After filtration and washing with several portions of
isopropanol, the polymerꢀwas redissolved in water, dialyzed for
3 days, and dried at 80 C for three more days. Yield: 4.04 g
(94.96%).
dH (400 MHz; DMSO) 7.86 (s, H-9), 7.52 (d, J ¼ 8.16 Hz, H-3),
7.47 (d, J ¼ 7.93 Hz, H-2), 7.1 (d, J ¼ 7.63 Hz, H-7), 6.82 (d, J ¼
8.55 Hz, H-6); dC (100 MHz; DMSO) 145.56 (C-10), 139.19 (C-
5), 136.81 (C-8), 136.32 (C-3), 132,92 (C-2), 131.1 (C-6), 129.7
(C-9), 128.92 (1 þ 4), 128.28 (C-7); dF (200 MHz; DMSO)
ꢁ58.66 (F-12). C-11 and C-12 could not be detected because of
their splitting and low intensity.
FIGURE 1 Polycondensation of 2, 2-bis(4-fluoro-3-sulfonato-
phenyl)-hexafluoropropane with TBBT, followed by oxidation
with H2O2/CH3COOH.
21.80 Hz, C-2), 124.42 (q, J ¼ 286.82 Hz, C-6), 129.51 (s, C-4),
132.48 (m, C-3), 163.25 (d, J ¼ 250.49 Hz, C-1); dF (200 MHz;
CDCl3) ꢁ112.10 (m, F-1), ꢁ64.38 (s, F-6).
SPSO
SPT (4.22 mmol) was dissolved in acetic acid (60 mL) and
water (30 mL). Hydrogen peroxide (12 mL) and sulfuric acid
(2 mL) werꢀe added, and the brown mixture was stirred for
16 h at 80 C. After cooling down to room temperature, the
M2
The monomer M1 was sulfonated using fuming sulfuric acid
(30% SO3) as described in the following. To M1 (10 g, 29.39
mmol), fuming sulfuric acid (24 mL) were added. The solu-
tion was stirred 4.5 h at 110 ꢀC and 2.5 h at 120 ꢀC. After
cooling down to room temperature, the brown solution was
slowly poured into 500 mL of ice water. Some sodium chlo-
ride was added until all product precipitated. After filtering,
the product was redissolved in water and the pH of the solu-
tion adjusted with sodium hydroxide to pH 7. The product
was precipitated a second time with sodium chloride and
ꢀ
was filtered off. After drying at 50 C in a vacuum oven (100
mbar) there was no more purification needed. Yield: 12.66 g
(86.09%).
dH (400 MHz; CDCl3) 7.31 (m, H-2 þ H-3), 7.83 (d, J ¼ 4.96 Hz,
H-7); dC (100 MHz; CDCl3) 63.59 (quintet, J ¼ 26.00 Hz, C-5),
117.31 (d, J ¼ 23.52 Hz, C-2), 124.17 (q, J ¼ 287.11 Hz, C-6),
127.44 (s, C-8), 130.42 (s, C-4), 132.95 (s, C-7), 135.75 (d, J
17.59 Hz, C-3), 159.22 (d, J ¼ 254.31 Hz, C-1); dF (200 MHz;
CDCl3) ꢁ109.67 (m, F-1), ꢁ63.76 (s, F-6).
SPT
M2 (5.99 mmol) and 4,40-TBBT (5.99 mmol) were loaded into a
100-mL three-neck flask equipped with an argon inlet, a reflux
FIGURE 2 Polymers investigated in this report.
SPSOS BLENDED WITH PBIOO, KATZFUß ET AL.
1921

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
yellow solution was dialyzed 3 days and dried at 60 ꢀC for
three more days. Yield: 1.88 g (55.22%).
(IECdirect). After that, an excess of NaOH was added, and this
solution was then back titrated with 0.1 M HCl to obtain the
IECtotal
.
Specific Resistance
The specific resistance of the membranes was determined
via impedance spectroscopy using a Zahner elektrik IM6 im-
pedance spectrometer. The samples were measured in
through plane mode in a frequency range of 200 KHz to
2 MHz and with amplitude of 5 mV. We used 0.5 N H2SO4
for the measurements to achieve better reproducibility in
the measured values, compared to measurements in water.
dH (400 MHz; DMSO) 8.49 (d, J ¼ 8.39 Hz, H-6), 8.33 (s, H-9),
8.18 (d, J ¼ 8.01 Hz, H-2), 8.10 (d, J ¼ 8.09 Hz, H-3), 7.61 (d, J
¼ 7.71 Hz, H-7); dC (100 MHz; DMSO) 148.46 (C-10), 147.03
(C-4), 143.73 (C-1), 137.49 (C-5), 137.08 (C-8), 132.35 (C-6),
131.87 (C-7), 131,47 (C-9), 129.60 (C-3), 128.27 (C-2); dF (200
MHz; DMSO): ꢁ58.32 (F-12).
Thermal Stability
The thermal stabilities of the blend membranes were deter-
mined by thermogravimetry (TGA, Netzsch, model STA 449C)
with a heating rate of 20 ꢀC minꢁ1 under a 65–70% O2
atmosphere. The sulfonated polymers and blend membranes
have been investigated by TGA-FTIR coupling (Nicolet Nexus
FTIR spectrometer) to ascertain the onset of the SO3H group
Blend Membrane Preparation
The composition of the membrane solution is usually calcu-
lated from the IEC value of the acidic and the basic compo-
nents.22 The acidic polymer was dissolved in DMAc (10 wt
%) and neutralized with n-propylamine. The calculated
amount of PBIOO in DMAc was added and, after stirring, the
solution was casted on a glass plate and dried at 50 ꢀC. To
obtain their SO3H form, all membranes were soaked in 0.5 N
H2SO4 for 2 h at 100 ꢀC and washed with water for further
splitting-off temperature (TSO H,onset).
3
Water Uptake
The water uptake (WU) of the samples was determined after
equilibration in water of defined temperatures. The weight
of the wet and the dry membrane (drying at 90 ꢀC) was
measured, and the WU was calculated using the following
formula:
ꢀ
2 h at 100 C.
Polymer and Membrane Characterization
mwet ꢁ mdry
WU ½%ꢄ ¼
Molecular Weight Determination via GPC (SEC)
The molecular weight distributions (MWDs) of the polymers
and of the blend membranes were determined by GPC,
mdry
Oxidative Stability
ꢀ
which was performed at 50 C on a polymer standards serv-
The oxidative stability of the membranes was determined
with H2O2 oxidation tests. The membrane samples were
dried and weighed, followed by equilibration in 5 wt %
H2O2 solution. After specified times of 6, 24, and 48 h at
60 ꢀC, the samples were removed from the oxidizing solu-
tion, washed with water, dried, and weighed a second time
to calculate the weight loss caused by the H2O2 treatment.
Moreover, the degradation of all samples was measured via
gel permeation chromatography (GPC).
ice (PSS) system equipped with Agilent 1200 series refrac-
tive index detector, PSS SLD 7000 multiangle-lightscattering
detector and a ETA2010 viscometer detector, PSS 30 and
3000 Å columns, and Agilent 1200 series pump using poly-
styrene standards for calibration. Eluent DMAc containing
5 wt % LiBr was used to increase the solubility of the ionic
polymer and to reduce the interaction between solutes and
packing materials. The blend membrane and polymer solu-
tions were injected with a concentration of 2 g Lꢁ1. The
obtained curves were integrated to ascertain the MWD of
the polymers and blend membranes.
RESULTS AND DISCUSSION
Polycondensation and Oxidation Reactions
Structure Analysis
The successful synthesis of SPT was performed via polycon-
densation using 4,40-TBBT and the self produced monomer
M2. In the presence of K2CO3 and NMP, the reaction required
24 h at 180 ꢀC. The monomer concentration was around
10 wt %. An excess of K2CO3 was necessary to react with
water traces, which are present in the reaction mixture. The
reaction of SPT to SPSO succeeded by applying an oxidation
route using hydrogen peroxide, acetic acid and, as a catalytic
component, sulfuric acid. After 16 h, the whole SPT was con-
verted into SPSO. The advancement of the oxidation was
indicated by a color change of the reaction mixture from
brown to yellow. The completeness of the oxidation
was proved via NMR and IR measurements. Moreover, GPC
measurements shows only a slight reduction of the MWD
To characterize the structure of the novel monomers and
polymers, NMR spectra were measured on a Bruker Avance
400 spectrometer, and IR spectra were detected on a Nicolet
6700 FTIR instrument.
Ion-Exchange Capacity
The IEC describes the amount of ion-exchange groups per
weight unit of dry ion-exchange resin in meq gꢁ1 or mmol
gꢁ1. Membranes in protonated form were dried at 90 C for
ꢀ
24 h and weighed. After this, the membranes were immersed
in saturated sodium chloride solution (NaCl) for 24 h to
replace the protons of the sulfonic acid groups with sodium
ions. The exchanged and delivered protons were then deter-
mined by titration with 0.1 M NaOH to the equivalent point
1922
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 4 IR spectra of SPT and SPSO. [Color figure can be
viewed in the online issue, which is available at
wileyonlinelibrary.com.]
FIGURE 3 1H NMR spectra of SPT and SPSO.
FTIR Characterization
during the oxidation from Mw ¼ 30.000 Da (SPT) to Mw
¼
Additional to NMR, FTIR spectra of the SPT, and SPSO poly-
mers, respectively, were recorded. In Figure 4, three bands of
the SPT-FTIR spectrum at 480, 816, and 1472 cmꢁ1 could be
assigned, which are the characteristic vibrations of a poly(p-
phenylene sulfide) structure.44 When the complete oxidation
of SPT to SPSO was reached, these three bands should have
disappeared. The bands marked in Figure 4 cannot be found
in the spectra of SPSO. However, a new absorption band at
1350 cmꢁ1 can be found in the FTIR spectrum of SPSO, which
is characteristic for SO2 group-containing polymers.
29.000 Da (SPSO).
NMR Characterization
1
Typical H NMR spectra of SPT and SPSO are shown in Figure
1
3. The H NMR spectra of SPT clearly show the expected five
peaks. Based on the electron withdrawing effect caused by the
ASO3H group, the signal of proton E in ortho position to the
ASO3H group, can be found downfield at 7.86 ppm with an
intensity of 1. The two doublets at 7.52 and 7.47 ppm with an
intensity of 2 for each can be identified as the protons A and
B. In which order the signals appear is again dependent on
the electron withdrawing character of the ASO3H group. There
is still an influence, which is demonstrated by the downfield
shift of B relative to A. The signal of C appeared at 6.82 ppm
and D at 7.1 ppm. That C is located in the upfield has two
reasons. First, the ASO3H group has a stronger influence onto
the ortho and para positions because of mesomeric effects
and, therefore, D is located downfield. Second, the thioether
linkages, which are electron-donating groups next to C, shift
the C protons to the upfield. These findings were backed up
using 2D-NMR measurements.
The IR spectra, together with the NMR spectra results, led to
the conclusion that the oxidation of SPT to SPSO was complete.
Ion-Exchange Capacity and Specific Resistance
The IEC, sulfonation degree per repeating unit, and specific
resistance values of the polymer blends are listed in Table 1.
All used polymers were water soluble. SPSU has an IECtotal of
3.2 meq gꢁ1, and the other investigated polymers have an
IECtotal of 2.4 meq gꢁ1. The polymer blends SPSU-PBIOO con-
tained 27 wt % PBIOO, and the other polymer blends con-
tained 19 wt % PBIOO. Experimentally determined IECs of
the PBIOO blends ranged between 1.1 and 1.6 meq gꢁ1. It is
important to recognize that the IECdirect values were nearly
the same for all polymer blends. The IECdirect values represent
the directly available and exchangeable protons. All the
For the identification of the signals to the arylene H’s of
SPSO, it was necessary to carry out a 2D-NMR experiment.
The analysis shows that the whole arylene H’s shifted to
downfield. This large shift reflects the very strong electron-
withdrawing character of the SO2 group and proves in turn
that the oxidation of SPT to SPSO was complete. One indica-
tion of the electron-withdrawing character is shown by the
position of the 1H NMR signal of C at 8.49 ppm. This is the
strongest downfield shift, compared to the other protons. A
further effect is the change of the positions of A and B. A
moves to the downfield at 8.18 ppm and B appears at
8.10 ppm. It seems that the influence of the ASO3H group
together with the SO2 linkage has an interesting effect for
the position of B. A satisfying explanation of this 1H NMR
signal interchange cannot be given now.
TABLE 1 Properties of the Polymer Blends
Ionomer
Blends
IECdirect
(meq gꢁ1
IECtotal
(meq gꢁ1
Rsp
(X cmꢁ1
r
)
)
)
(S cmꢁ1
)
SFS-PBIOO
0.8
0.7
0.7
0.7
1.6
1.1
1.1
1.4
24.1
41.7
36.8
9.1
0.042
0.024
0.027
0.11
SPSU-PBIOO
SPEEK-PBIOO
SPSO-PBIOO
Rsp Specific resistance; measured at RT and 100% RH.
Proton conductivity.
r
SPSOS BLENDED WITH PBIOO, KATZFUß ET AL.
1923

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
TABLE 2 Thermal Stabilities of the Polymer Blend Membranes
Ionomer Blends
T5 wt % loss (ꢀC)
TSO H onset (ꢀC)
3
SFS-PBIOO
353
382
362
409
260
290
297
334
SPSU-PBIOO
SPEEK-PBIOO
SPSO-PBIOO
polymer blends showed low specific resistance values. It is
remarkable that the novel polymer SPSO comprises the high-
est proton conductivity among all blend samples. The low
electron density of the aromatic building block of SPSO causes
a higher degree of dissociation of the ASOꢁ3 Hþ group. This
leads to a higher mobility of the protons, ending up in a
higher conductivity, compared to the other blend membranes.
FIGURE 5 TGA curves of the blend membranes. [Color figure
can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
Thermal Stability
The thermal stability of the polymer blends was determined
using TGA-FTIR coupling technique. The analysis of the TGA
traces was started at 200 ꢀC at which all water had evapo-
rated from the sample. As it is summarized in Table 2,
degree of dissociation, compared to the other investigated
arylene main chain ionomers investigated in this study.
Thereþfore, it can be speculated that the stronger separated a
SOꢁ3 H ion pair is, the more water molecules can be taken
up into the hydration spheres of both ions.
ꢀ
the T5wt%loss ranged between 350 and 410 C. The poly(ary-
lene ether) and poly(arylene ether ketone) sample, respec-
tively, started to degrade thermally at lower temperatures
than the other investigated membranes. The samples with
sulfone bridges in their repeating units showed the highest
thermal stabilities and the highest onset of splitting-off tem-
peratures of the sulfonic acid group. One reason for this
finding might be the electron-attracting capability of the sul-
fone bridge by its -I and -M effect,18,39–41 leading to higher
thermal stability of the poly(sulfone), compared to the other
arylene main chain ionomers, which comprise electron-
donating groups in the polymer backbone (AOA, ASA). Fur-
thermore, it is assumed that our SPSO polymer shows a
high-morphological stability comparable with that of the
poly(arylene sulfone)s (SPSOs) Kreuer and coworkers18 pre-
sented.41 This high morphological stability leads to a high
thermal stability of SPSO. In Figure 5, it is clearly shown by
the TGA traces that the novel SPSO blend polymer has the
highest thermal stability among all membranes under com-
parison in this study.
Oxidative Stability
An accelerated degradation test of the SPSO-PBIOO blend
membranes, in which the membranes were exposed to an
oxidative environment (H2O2), has been performed. Two
molecular weight reduction mechanisms are possible as a
matter of principle. The first one is the splitting-off of the
sulfonic acid group and the second mechanism comprises
the reduction of the molecular weight of the polymer chains
by radical attack, followed by radical chain scission, at the
bridging groups. The attack at the main chain is the most
critical one, since it strongly reduces the molecular weight of
the polymer. Therefore, we tried to elucidate the degradation
mechanism of the SPSO-PBIOO blend membrane by GPC
measurements. A clarification of the degradation mechanisms
of fuel cell membranes is crucial for fuel cell membrane
Water Uptake
The water uptake values of the PBIOO blend membranes as
a function of temperature are presented in Figure 6. At room
temperature, the polymer blends absorbed between 19 and
30% water of their own weight. In contrast to the initial
expectations, there was no significant increase of water
uptake when the samples were equilibrated in water at
higher temperatures. The SPSO-PBIOO membrane showed
the highest water uptake values at all investigated tempera-
tures but never more than 35%. The high water uptake of
the SPSO-PBIOO blend membrane is thought to be caused by
its higher hydrophilicity, compared to the other investigated
membranes. Based on the low electron density of the aro-
matic building blocks of the SPSO, its pendent ASO3H groups
have a better charge separation and, therefore, a higher
FIGURE 6 Water uptake of the blend membranes at different
temperatures. [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
1924
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
sample is presented at 0 h. It is of interest, that the degrada-
tion process starts at an immersion time of 6 h. Up to this
time, no significant weight loss could be observed. This
means that the reduction of the molecular weight starts, but
the reduction is not strong enough to be reflected in weight
loss, since the slightly degraded macromolecular chains are
retained by the ionical crosslinking network inside the blend
membrane, as stated earlier in this study.
It is interesting to see that the molecular weight reduction
has reached nearly 50%. However, this value is not as high
as it seems to be, because the blend membrane had an initial
Mn of 39.000 Da. After 48 h of oxidative treatment, the
degradation process has led to a Mn of 21.000 Da. As the
polycondensation procedure for SPSO will be optimized in
ongoing research, ending up in higher molecular weight
SPSO, it is expected that the percentage of molecular weight
reduction by oxidative molecular weight reduction will be
lower and the membrane stability can be retained for a
longer time interval.
FIGURE 7 Weight loss of SPSO-PBIOO in function of immer-
sion time. [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
development. The knowledge of the molecular weight
reduction mechanism gives valuable information for the
development of more stable membranes. A molecular weight
reduction of the macromolecular chains building up the
blend membrane leads to a significant loss in mechanical
membrane stability. If mechanical weakening of a membrane
takes place during fuel cell test by attack of radicals formed
during the fuel cell reaction onto the membrane polymer(s),
pinhole formation in the membrane might be the result,
leading to mixing of the fuel cell reaction components O2
and H2, which can ignite, ending up in combustion of the
membrane and fuel cell failure.
In Figure 9, the reduction of the MWD of the SPSO-PBIOO
blend membrane as a function of W (log M) is depicted. W
(log M) is defined as the intensity of the mass fraction of the
logarithmic molecular mass. From 0 to 6 h and from 6 to
24 h, no significant shift in the MWD profile to lower molec-
ular weight can be observed. As expected, the polydispersity
index value expanded with increasing immersion time, which
can be explained with dissimilar molecular weight reduction
of the polymer chains of the blend membrane. To interpret
this result, it is important to understand the ionical cross-
linking mechanism of sulfonated polymers with PBIOOs. The
proton of the sulfonic acid group interacts with the basic
nitrogen atom of the imidazole group from the PBI. These
ionical interactions are quite strong and lead to a better oxi-
dative stability of the blend membrane, compared to the
pure sulfonated ionomer. The macromolecular chains of the
blend membrane sustain molecular weight reduction/degra-
dation by the radical attack and subsequent breakage of the
macromolecular chains. The reason for this stabilization is
In Figure 7, the weight loss values of SPSO-PBIOO blend
membrane are presented as a function of H2O2 immersion
time. After 6-h treatment, no weight loss can be observed
for the blend membrane. The significant weight loss starts at
24 h and further increases up to 48 h. An explanation for
the weight loss of this membrane can be given on compari-
son of these results with the GPC measurement results.
The GPC measurement results of the blend membrane are
shown in Figure 8. The molecular weight of the untreated
FIGURE 9 GPC curves of SPSO-PBIOO samples after different
immersion times shown as the intensity of mass fraction of the
logarithmic molecular mass against Mn.
FIGURE 8 Molecular weight reduction of SPSO-PBIOO as func-
tion of immersion time. [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
SPSOS BLENDED WITH PBIOO, KATZFUß ET AL.
1925

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
that they are retained in the macromolecular ionically cross-
linked network of the membrane by the ionic interactions as
long as they are not degraded to low molecular compounds.
Another reason for the deceleration of the degradation of
PBI-containing blend membranes might be in the OH radical
scavenging properties of the imidazole group, which has
been proven for biochemical imidazole-containing substances
such as histidine.45
uptake. The novel SPSO has aꢀ proton conductꢀivity of 0.11 S
cmꢁ1 and a TSO H onset of 334 C, which is 40 C higher than
3
those of the other polymer blends taken into account in this
study. The reason for the advantageous properties of the
SPSO blend membrane, compared to the other blend mem-
branes, is the strong electron-withdrawing character of the
sulfone linkages, which markedly improves the properties of
the polymer, compared to polymers comprising ether and ke-
tone linkages. In terms of water uptake, SPSO showed a
higher uptake behavior than the other investigated mem-
branes. Compared with polymers, which have only sulfone
bridges in their backbone, the water uptake always remained
below 50%.
By suitable combinations of sulfonated ionomer/PBIOO, fur-
ther optimization of the blend membranes can be brought
about. The strength of the ionical interactions in acid-base
blend membranes depends (1) on the acidity of the sulfo-
nated ionomer, (2) on the basicity of the imidazole-N of the
used PBI, and (3) on the chemical nature of both the acidic
and the basic polymer. Recently, it was found by Kerres
et al.46 that blend membranes formed by use of the same
sulfonated polymer (SFS, see Fig. 2), but different types of
PBI show different thermal and oxidative stabilities. Each
The optimization of the polycondensation reaction to reach
further increase in molecular weight will be carried out in
ongoing research. The motivation for continuative work is to
obtain polymer blend membranes, which show better me-
chanical and chemical stabilities. Furthermore, it is planned
to prepare novel electron deficient partially fluorinated SPSO
s by polycondensation of other fluorinated arylene mono-
mers with TBBT or related bis(thiophenol)s, followed by
oxidation of the polymeric thioethers to poly(sulfone)s. The
sulfonated poly(sulfone)s will be blended with electron-defi-
cient PBIOOs such as F6-PBI and SO2-PBI, for use in H2-
PEFC, DMFC, and PEM-H2-electrolysis.
R
type of PBI, as for example, PBI CelazolV, PBIOO, SO2-PBI,
and F6-PBI, has a different electron density of the aromatic
polymer chain. PBI Celazol and PBIOO can be regarded as
electron-rich PBIOOs, while SO2-PBI and F6-PBI are PBIOOs
comprising electron-deficient aromatic building blocks due
to their strongly electron-attracting SO2 and hexafluoroiso-
propylidene groups, respectively. It was found that the com-
bination of the electron-deficient SFS with each of the elec-
tron-deficient PBIs SO2-PBI and F6-PBI yielded blend
membranes, which were very stable in Fenton’s test. While
the blending of SFS with the two-mentioned electron-rich
PBIs led to membranes, which lost much more weight when
immersed in Fenton’s reagent for the same retention time.46
The authors thank Inna Kharitonova and Galina Schumski for
carrying out the polymer and membrane characterization.
REFERENCES AND NOTES
1 Ogato, N.; Rikukawa M. WO Patent 94/24717, October 27,
Compared to the other blend membranes presented in this
study, the advantage of SPSO, compared to the other sulfo-
nated ionomers, is based upon the bridging groups. The
novel polymer has no ether linkages, which can easily break
off, because ether linkages are susceptible to ipso attack of
HOꢂ, which is an accepted degradation mechanism.44 Instead
of ether linkages, SO2 and hexafluoroisopropylidene bridging
groups are present in the polymer chain. Both groups have
an electron-withdrawing character, which lowers the electron
density of the polymer chain. This is important in terms of
the radical attack, which is hindered with decreasing elec-
tron density of the radical target polymer. Moreover, the hex-
afluoroisopropylidene-bridging group has a steric hindrance
effect which leads to a hindrance of radical addition. We con-
clude from the obtained results that if these two bridging
groups are present in the polymer backbone, it is possible to
reduce the extent of radical attack and to increase the oxida-
tive stability.
1994.
2 Litt, M. Lecture, Symposium ‘‘Advances in Materials for
Proton Exchange Membrane Fuel Cell Systems,’’ Asilomar:
California, 2005.
3 Noshay, A.; Robeson, L. M. J Appl Polym Sci 1976, 20, 1885–1903.
4 Wang, F.; Chen, T.; Xu, J. Macromol Rapid Commun 1998,
19, 135–137.
5 Kerres, J.; Cui, W.; Reichle, P. J Polym Sci Part A: Polym
Chem 1996, 34, 2421–2438.
6 Helmer-Metzmann, F.; Osan, F.; Schneller, A.; Ritter, H.;
Ledjeff, K.; Nolte, R.; Thorwirth, R. Hoechst AG, EP Patent
0,574,791 A2 Offen, 1993.
7 Rozie`re, J.; Jones, D. J. Annu Rev Mater Res 2003, 33, 503–555.
8 Ulrich, H. H.; Rafler, G. Angew Makromol Chem 1998, 263,
71–78.
9 Xing, P.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.;
Wang, K.; Kaliaguine, S. J Membr Sci 2004, 229, 95–106.
CONCLUSIONS
10 Schauer, J.; Lopour, P.; Vacik, J. J Membr Sci 1986, 29,
A novel sulfonated SPSO was successfully synthesized by an
aromatic nucleophilic polycondensation. The polymer was
blended with PBIOO to form a stable membrane. Different
blend membrane types comprising several sulfonated aryl-
ene main chain ionomers and PBIOO were compared in
terms of specific resistance, thermal stability, and water
169–175.
11 Percec, S.; Li, G. Polym Prepr (Am Chem Soc, Div Polym
Chem) 1986, 27, 19–20.
12 Lee, H. C.; Hong, H. S.; Kim, Y.-M.; Choi, S. H.; Hong, M. Z.;
Lee, H. S.; Kim, K. Electrochim Acta 2004, 49, 2315–2323.
1926
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
13 Kerres, J. A.; Xing, D.; Schonberger, F. J Polym Sci Part B:
30 Kerres, J.; Ullrich, A.; Hein, M. J Polym Sci: Part A: Polym
¨
Polym Phys 2006, 44, 2311–2326.
Chem 2001, 39, 2874–2888.
14 Miyatake, K.; Iyotani, H.; Yamamoto, K.; Tsuchida, E. Macro-
31 Kerres, J.; Ullrich, A.; Haring, T. U.S. Patent 6,590,067, 2003;
¨
molecules 1996, 29, 6969.
EP Patent 1105433B1, 2004.
15 Wiles, K. B. High Performance Disulfonated Poly(arylene sul-
fone) co and ter-polymers for Proton Exchange Membranes for
Fuel Cell and Transducer Applications, Ph.D. Thesis, Virginia
Polytechnic Institute and State University, Blacksburg, 2005.
32 Kerres, J.; Ullrich, A.; Haring, T. EP Patent 1076676, 2004,
¨
U.S. Patent 6,723,757, 2004.
33 Kerres, J.; Ullrich, A.; Ha¨ring, Th.; Baldauf, M.; Gebhardt, U.;
Preidel, W. J New Mat Electrochem Sys 2000, 3, 229–239.
16 Bai, Z.; Durstock, M. F.; Dang, T. D. J Membr Sci 2006, 281,
34 Lee, J.; Kerres, J. J Membr Sci 2007, 294, 75–83.
508–516.
35 Kerres, J.; Ullrich, A.; Meier, F.; Haring, Th. Solid State
¨
17 Shen, L.; Xiao, G.; Yan, D.; Sun, G. e-Polym 2005, no 31.
Ionics 1999, 125, 243–249.
18 Schuster, M.; Kreuer, K.-D.; Andersen, H. T.; Maier, J. Macro-
36 Hubner, G.; Roduner, E. J Mater Chem 1999, 9, 409–418.
¨
molecules 2007, 40, 598–607.
37 Zhang, L.; Mukerjee, S.
J Electrochem Soc 2006, 153,
19 Matsumura, S.; Kihara, N.; Takata, T. Macromolecules 2001,
A1062– A1072.
34, 2848–2853.
38 Grot, W. F. DuPont de Nemours. U.S. Patent 3,718,627, 1973.
20 Xing, D.; Kerres, J. J New Mat Electrochem Sys 2006, 9, 51–60.
21 Xing, D.; Kerres, J. Polym Adv Technol 2006, 17, 1–7.
39 Schuster, M.; Kreuer, K. D.; Andersen, H. T.; Maier, J. DE
Patent Application 10 2005 010 411 A1, application date 07.03.
2005.
22 Schonberger, F.; Hein, M.; Kerres, J. Solid State Ionics 2007,
¨
178, 547–554.
40 de Araujo, C. C.; Kreuer, K. D.; Schuster, M.; Portale, G.;
Mendil-Jakani, H.; Gebel, G.; Maier, J. Phys Chem Chem Phys
2009, 11, 3305–3312.
23 Schonberger, F. Synthese und Charakterisierung teilfluor-
¨
ierter Poly(aryl)-Ionomere als Polymerelektrolytmembranen fur
¨
Brennstoffzellen und ESR-spektroskopische Untersuchungen
der radikalinduzierten Degradation von Modellverbindungen,
Ph.D. Thesis, University of Stuttgart, Germany, 2008.
41 Schuster, M.; de Araujo, C. C.; Atanasov, V.; Andersen, H. T.;
Kreuer, K. D.; Maier, J. Macromolecules 2009, 42, 3129– 3137.
42 Gogel, V.; Jorissen, L.; Chromik, A.; Schonberger, F.; Lee, J.;
¨
¨
24 Li, Q.; He, R.; Jensen, J. O.; Bjerrum, N. J Chem Mater 2003,
Schafer, M.; Krajinovic, K.; Kerres, J. Sep Sci Technol 2008, 43,
¨
15, 4896–4915.
3955–3980.
25 Banerjee, S.; Maier, G.; Burger, M. Macromolecules 1999,
43 Lau, K. S. Y.; Dougherty, T. K. U.S. Patent 4,827,054, 1989.
32, 4279–4289.
44 Zimmerman, D. A.; Koenig, J. L.; Ishida, H. Spectrochim Acta
26 Kerres, J. A. J Membr Sci 2001, 185, 3–27.
Part A 1995, 51, 2397–2409.
27 Kerres, J.; Cui, W. U.S. Patent 6,194,474, 2001; U.S. Patent
45 Kammeyer A.; Eggelte T. A.; Bos J. D.; Teunissen M. B. M.
6,300,381, 2001; EP Patent 1073690, 2004.
Biochim Biophys Acta 1999, 1428, 117–120.
28 Cui, W.; Kerres, J.; Eigenberger, G. Sep Purif Techn 1998,
46 Li, Q. F.; Rudbeck H. C.; Chromik, A.; Jensen, J. O.; Pan, C.;
Steenberg, T.; Calverley, M.; Bjerrum N. J.; Kerres, J. J Membr
Sci 2010, 347, 260–270.
14, 145–154.
29 Kerres, J.; Ullrich, A. Sep Purif Techn 2001, 22, 1–15.
SPSOS BLENDED WITH PBIOO, KATZFUß ET AL.
1927