Fully aromatic ionomers with precisely sequenced sulfonated moieties for enhanced proton conductivity

Article abstract of DOI:10.1021/ma201599p

A series of six fully aromatic ionomers with precisely sequenced sulfonated sites along the polymer chains have been designed, prepared, and characterized as proton-exchange membranes. Two straightforward and efficient synthetic strategies based on Ullmann ether reactions and a Baeyer-Villiger rearrangement were devised to obtain bisphenol monomers with four or six phenylene units linked exclusively by ether bridges to avoid transetherification reactions. Polycondensations of these bisphenol monomers with mono- or disulfonated dihalide monomers gave high molecular weight poly(arylene ether), poly(arylene ether sulfone), and poly(arylene ether ketone) homopolymers having microblock-like structures with sulfonated moieties separated by monodisperse nonsulfonated oligo(ether) spacers. The nanoscale morphology and properties of solvent cast membranes were closely related to the nature of the oligo(ether) spacers. Small angle X-ray scattering (SAXS) measurements showed intense scattering and very narrow ionomer peaks with second-order features for the polymers with the six-ring spacers. This clearly indicated that the controlled ionic sequencing enabled self-assembly of ionic aggregates with a much higher degree of organization in relation to a corresponding aromatic ionomer with a statistical distribution of the sulfonate groups. At an identical ionic content, the ionomers containing meta ether linkages had lower glass transition temperatures than the all-para materials, leading to a higher water uptake and proton conductivity of the former ionomers. A membrane with an ion exchange capacity (IEC) of 2.05 mequiv g^-1 and containing exclusively para linkages reached the same level of proton conductivity as Nafion at 100% relative humidity (RH), and also had an excellent dimensional stability in boiling water. Under reduced RH, the conductivity of this membrane greatly exceeded that of a membrane based on a statistical copolymer analogue with a similar ionic content.

Full text of DOI:10.1021/ma201599p

Article
pubs.acs.org/Macromolecules
Fully Aromatic Ionomers with Precisely Sequenced Sulfonated
Moieties for Enhanced Proton Conductivity
Xue-Feng Li,†,‡ Francois P. V. Paoloni,† E. Annika Weiber,† Zhen-Hua Jiang,‡ and Patric Jannasch†,*
̧
†Polymer & Materials Chemistry, Department of Chemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden
‡Alan G. MacDiarmid Institute, Department of Chemistry, Jilin University, Qianjin Street 2699, Changchun 130012,
People’s Republic of China
S
* Supporting Information
ABSTRACT: A series of six fully aromatic ionomers with precisely sequenced sulfonated sites along the polymer chains have
been designed, prepared, and characterized as proton-exchange membranes. Two straightforward and efficient synthetic
strategies based on Ullmann ether reactions and a Baeyer−Villiger rearrangement were devised to obtain bisphenol monomers
with four or six phenylene units linked exclusively by ether bridges to avoid transetherification reactions. Polycondensations of
these bisphenol monomers with mono- or disulfonated dihalide monomers gave high molecular weight poly(arylene ether),
poly(arylene ether sulfone), and poly(arylene ether ketone) homopolymers having microblock-like structures with sulfonated
moieties separated by monodisperse nonsulfonated oligo(ether) spacers. The nanoscale morphology and properties of solvent
cast membranes were closely related to the nature of the oligo(ether) spacers. Small angle X-ray scattering (SAXS) measure-
ments showed intense scattering and very narrow ionomer peaks with second-order features for the polymers with the six-ring
spacers. This clearly indicated that the controlled ionic sequencing enabled self-assembly of ionic aggregates with a much higher
degree of organization in relation to a corresponding aromatic ionomer with a statistical distribution of the sulfonate groups. At
an identical ionic content, the ionomers containing meta ether linkages had lower glass transition temperatures than the all-para
materials, leading to a higher water uptake and proton conductivity of the former ionomers. A membrane with an ion exchange
capacity (IEC) of 2.05 mequiv g−1 and containing exclusively para linkages reached the same level of proton conductivity as
Nafion at 100% relative humidity (RH), and also had an excellent dimensional stability in boiling water. Under reduced RH, the
conductivity of this membrane greatly exceeded that of a membrane based on a statistical copolymer analogue with a similar ionic
content.
membranes based on these polymers show excessive swelling in
water, and even water solubility, at high acid contents, which
limits their applicability. Partial sulfonations result in statistical
distributions of acid groups along the polymer backbones.
Unlike perfluorosulfonic acid membranes, these hydrocarbon
membranes do not usually form well-connected water-filled
channels that promote efficient transport of protons across the
membrane.7 Similar issues arise when sulfonated poly(aryl
ether)s are prepared using a presulfonated monomer. If for
example polymers are prepared by polycondensations of a
presulfonated dihalide monomer with a conventional similar-
sized nonsulfonated diol monomer, the acid groups will be
INTRODUCTION
■
Sulfonated aromatic hydrocarbon polymers have attracted
attention as potential candidates for membranes applied to
the production of electrical power and clean water.1 For
example, sulfonated poly(aryl ether)s represent an attractive
alternative to perfluorinated sulfonic acid membranes for high
temperature proton-exchange membrane fuel cells (PEMFCs)
because of their excellent stability and high glass transition
temperature (Tg).2 An increase of the operating temperature
to above 100 °C promises crucial benefits concerning the com-
plexity, cost and performance of automotive PEMFC systems.3,4
However, the performance and durability of fuel cell core
components such as the membrane need be further improved
to meet the demands of the industry.
Received: July 13, 2011
Revised: January 4, 2012
Published: January 23, 2012
Early examples of sulfonated poly(aryl ether)s were primarily
obtained by postpolymerization sulfonations.5,6 Characteristically,
© 2012 American Chemical Society
1447
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
regularly spaced but the acid content is likely to be too high
(Scheme 1). The common strategy is then to use a second non-
The corresponding membranes had higher proton conductivity
than typically observed for randomly sulfonated hydrocarbon
membranes or multiblock copolymers with similar acid content.
Colquhoun and co-workers have recently prepared and studied
microblock ionomers, a novel class of sulfonated poly(aryl
ether)s membranes.27 In these polymers, densely sulfonated
moieties are regularly spaced by monodispersed nonsulfonated
segments. Essentially, these microblock copolymers are
homopolymers with a very large repeat unit. The monomers
were designed to yield a repeat unit composed of a long
sequence of electron-poor aromatic rings, strictly alternating
with a short segment composed of electron-rich phenylene
units. Transetherification was avoided by excluding ether
linkages activated by a sulfone or ketone bridge. Postpolyme-
rization sulfonation under mild conditions was then selective
toward electron-rich aromatic rings. The authors described a
series of membranes with an ion exchange capacity (IEC, milli-
equivalents of H+ per gram dry membrane) of approximately
2.1 mequiv g−1.27 Although these ionomers show attractive
properties, they still suffer from two main drawbacks. First, it is
often acknowledged that sulfonic acid groups located next to
ether linkages are activated for desulfonation under fuel cell
operation.28,29 Second, the synthesis of the monomers used for
the microblock copolymers is limited and quite complex. The
monomers must be designed to eliminate the risk of trans-
etherification reactions during the polymerization, and the bis-
phenol monomers must thus exclusively contain ether linkages.
The number of phenylene units in these monomers will
essentially determine the degree of sulfonation of the corres-
ponding polymer. The bishalide monomers exclusively contain
ketone or sulfone bridges and are often associated with poor
solubility because of the high rigidity of these bonds. In addi-
tion, high local acid concentrations in the repeat unit require
long dihalide monomers which are not straightforward to synthe-
size and purify.
In the present study, we have followed an alternative
approach to sulfonated microblock poly(aryl ether)s by
employing nucleophilic polycondensations of long bisphenols
with sulfonated dihalide monomers. The resulting fully aromatic
ionomers were composed of long hydrophobic segments,
strictly alternating with singlets or doublets of sulfonic acid
groups, as illustrated in Scheme 1. First, efficient methods to
prepare various novel oligomeric bisphenol monomers were
developed. In these monomers, the phenylene rings were
exclusively connected by ether bridges in the meta or para
position. Their polymerization with three different sulfonated
monomers afforded six homopolymers. Two of these were
composed of singlets of sulfonic acid precisely sequenced by a
monodisperse oligo(ether) segment of exactly four phenylene
units. The remaining four homopolymers were composed of
doublets of sulfonic acid precisely sequenced by a mono-
disperse oligo(ether) segment of exactly six phenylene units.
Solvent cast membranes were characterized with regard to ionic
distribution, water uptake, thermal stability, and proton
conductivity. Special attention was given to the polymer
backbone structure, acid group distribution, nanostructure, and
influence on the membrane properties.
Scheme 1. Schematic Ionomer Structures Illustrating the
Distribution of Sulfonic Acid Groups along the Backbone in
Regular Homopolymers (a), in Statistical Copolymers (b),
and in Homopolymers with Precisely Sequenced Singlet (c)
and Doublet (d) Acid Groups
sulfonated comonomer to form statistical copolymers with reduced
acid contents. However, this also leads to an irregular statistical
distribution of sulfonic acid along the polymer backbone.
Several research groups have reported that synthetic strate-
gies involving a careful control of the distribution and location
of the sulfonic acid groups in the polymer structure result in
well-defined hydrocarbon membrane morphologies.8,9 This
may in turn lead to the formation of well-connected water-
filled channels which gives the membranes superior proton-
conducting properties in relation to statistically sulfonated mem-
branes, particularly at low relative humidity (RH). Mild methods to
synthesize sulfonated multiblock copoly(aryl ether sulfone)s
have been extensively used by McGrath et al.10−13 and
others.14−17 Low-temperature coupling of oligomers mediated
by a perfluorinated compound such as hexafluorobenzene or
decafluorobiphenyl affords high molecular weight multiblock
copolymers while limiting the extent of transetherification
reaction. These multiblock copolymers typically exhibit quite
well-defined morphologies, associated with high proton
conductivity under reduced RH.8,9 The length and acid density
of the sulfonated blocks have a major influence on the
membrane morphology and properties. This concept has been
extended to copolymers which contain densely sulfonated
blocks to further improve the membrane properties.18−23 In
another synthetic strategy, high local concentrations of sulfonic
acid groups have been placed on pendant chains to form
defined morphologies, as shown by small-angle X-ray scattering
(SAXS) measurements.24,25 The results indicated that ionic
clustering was promoted by placing the sulfonic acid groups on
the relatively long spacers. Increasing the number of sulfonic
acid groups located on pendant chains from one to three
resulted in larger ionic clusters and narrower ionomer peak
widths.24,25
Several groups have reported on sulfonated poly(aryl ether)s
composed of long hydrophobic segments combined with
densely sulfonated units. For example, Hay et al. have prepared
polymers containing clusters of up to 12 sulfonic acid groups
statistically distributed along the backbone chain.26 Novel
bisphenols having six pendant phenyl groups were incorporated
in poly(aryl ether)s by polymerization of carefully chosen
comonomers. The bisphenol monomer residues underwent
selective multiple sulfonations under mild conditions, resulting
in statistical copolymers containing highly sulfonated moieties.
RESULTS AND DISCUSSION
■
Synthesis of Sulfonated Monomers. The sulfonated
monomers prepared and used in the present work are shown in
Scheme 2. These monomers can take part in nucleophilic
polycondensation via displacement of the chlorine or fluorine
1448
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
Scheme 2. Sulfonated Monomers Used in the Present Study
efficiently isolated and purified by filtration through Celite
to remove the copper catalyst, followed by precipitation in
methanol. The highest yields were obtained under the conditions
described by Song and co-workers.34 Notably, the reaction of
4,4′-bisphenol and 1-bromo-3-methoxybenzene under identical
conditions gave 4 in poor to moderate yield only. The methoxy
groups of 4 were hydrolyzed by using hydrobromic acid in
acetic acid, and bisphenol monomer 5 was isolated by
precipitation in water and then washed extensively to remove
traces of acids. The product was purified by recrystallization
from hot toluene and isolated in 72% overall yield. Complete
removal of the methyl groups was confirmed by the total
disappearance of the resonance corresponding to the methyl
1
1
group at 3.81 ppm in the H NMR spectrum (the H NMR
spectra of all the bisphenol monomers of the present study, and
their respective intermediates, are shown in the Supporting
Information). The complete removal was further confirmed
by the absence of resonances in the aliphatic region of the
13C NMR spectrum. Formation of the phenol was finally also
supported by the presence of a broad absorption band centered
at 3357 cm−1 in the FT-IR spectra. The synthetic strategy
described above is attractive because of the large variety of
aromatic building blocks available and because of its simplicity.
Therefore, additional bisphenol monomers with varying length
and connectivity can be envisaged. Monomer 7 was synthesized
in good yield by an identical method. Ullmann ether con-
densation of 4,4′-DBBP (4,4′-dibromobiphenyl) and 4-methoxy-
phenol, followed by hydrolysis of methoxy protecting groups,
gave a four-ring bisphenol monomer with para connectivity. As
expected, the connectivity had an influence on the melting
point of these oligomers, and bisphenol 7 had a higher melting
point (249 °C) than bisphenol 5 (168 °C). As a result, the
former monomer was not soluble in chloroform and appeared
to generally have a lower solubility than the latter.
atoms which are activated by an electron-withdrawing ketone
or sulfone link. By employing presulfonated monomers, it is
possible to accurately control the location of the sulfonic acid
groups in the resulting polymers.
The lithium salt of 2,6-difluoro-2′-sulfobenzophenone (1)
was conveniently synthesized in one-pot by reacting 2,6-
difluorophenyllithium with 2-sulfobenzoic acid cyclic anhydride
in tetrahydrofuran (THF) at −70 °C.30 After reaction, the
product readily crystallized out of solution. We have recently
reported that the polymerization of this monosulfonated
monomer with various bisphenols and bisthiophenols proceeds
with a high reaction rate to produce corresponding high mole-
cular weight poly(arylene ether)s.30 3,3′-Disulfonated-4,4′-difluoro-
benzophenone (2)31 and 3,3′-disulfonated-4,4′-dichlorodiphenyl
sulfone (3)32,33 were synthesized via well-known procedures.
The use of sulfonated monomers readily ensures reproduci-
bility of the degree of sulfonation. Moreover, the side reactions
and degradation5 that are sometimes observed for sulfonation
under harsh conditions are avoided. In addition, it is recognized
that sulfonated polymers are most stable toward desulfonation
when the sulfonic acid groups are located on electron-
poor aromatic rings. The IEC of the homopolymers contain-
ing monomer 1, 2, or 3 can conveniently be controlled by
employing monodisperse bisphenol monomers of varying
molecular weight. In the case of monomer 1, this will re-
sult in single sulfonic acid groups regularly spaced along
the polymer backbone (Scheme 1c). The ionomers based
on monomers 2 or 3 will consist of doublets of sulfonic
acid groups, strictly alternating with hydrophobic segments
(Scheme 1d).
Synthesis of Bisphenol Monomers. The bisphenol
monomers will form monodisperse hydrophobic segments once
incorporated in the polymer. This will result in a distribution of
sulfonic acid groups that is controlled with molecular precision.
As mentioned above, it is imperative that the bisphenol mono-
mers do not contain any electron-withdrawing groups in order
to exclude the possibility of transetherification during the
polymerizations. Indeed, such reactions would result in chain-
scrambling and introduction of defects in the sequence of the
polymer backbone. Therefore, the bisphenol monomers were
designed to contain aromatic rings linked exclusively by ether
bonds or direct aryl−aryl bonds.
The two-step method described above was also adapted to
attempt the synthesis of longer bisphenol oligomers. Reaction
of 5 with 4-bromoanisole gave a bismethoxy terminated oligo-
mer 8. Its hydrolysis with HBr afforded a six-ring oligo(ether)
monomer 9 that contained both meta and para linkages.
Monomer 9 was obtained in 63% yield over two steps after
purification by recrystallization from hot toluene. The structure
was confirmed using the same analytical techniques as
mentioned previously. The monomer had a melting point
similar to the shorter meta monomer 5. Unfortunately, the
reaction of 5 with 3-bromoanisole gave a mixture of products.
1
Analysis by H NMR spectroscopy indicated that the main
component was the desired all-meta oligomer. However, it was
not possible to isolate the target compound by precipitation
and filtration. In addition, the reaction of 6 with 4-bromo-
anisole did not afford the targeted six-ring oligomers. Analysis
by 1H NMR spectroscopy suggested that the reaction was mostly
limited to a monosubstitution that produced an oligomer com-
posed of five aromatic units.
An alternative route involving a condensation followed by a
Baeyer−Villiger oxidation35 was utilized in order to synthesize
an all para oligomer with six rings. As shown in Scheme 4, the
reaction of bisphenol 7 with 4-fluoroacetophenone gave the
six-ring oligomer 10. In this case the fluorine atom is activated
toward nucleophilic aromatic substitution by the ketone in para
position, resulting in a higher conversion than for the Ullmann
ether condensation. The oligomer was insoluble in both
chloroform and dimethyl sulfoxide (DMSO). Consequently,
it was characterized by NMR spectroscopy using a mixture of
Two bisphenol monomers composed of four aromatic rings
(5 and 7) were obtained by Ullmann ether condensation
followed by hydrolysis of the methoxy protecting groups, as
shown in Scheme 3. Thus, the reaction of 4,4′-dibromobiphenyl
with 3-methoxyphenol gave the methoxy-terminated four-ring
monomer precursor 4 in good yield. This precursor was
1449
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
Scheme 3. Synthesis of Four-Ring Bisphenol Monomers 5 and 7 and Six-Ring Monomer 9 by Ullmann Ether Condensation
a
Followed by Hydrolysis of the Methoxy End Groups
a
Key: (i) CuCl, Cs2CO3, THMD, NMP, 165 °C; (ii) HBr, AcOH, 150 °C.
CDCl3 and trifluoroacetic acid. This solvent mixture is known
to be a good solvent for the solubilization of fully aromatic
poly(ether ketone)s.36−39
monomer 12 by hydrolysis. The bisphenol monomer was
purified by dissolving the crude product in DMF (dimethyl-
formamide), followed by addition of 2-propanol until a cloudy
mixture was obtained. Then, the mixture was heated to give a
clear solution and bisphenol 12 was recovered as a fine powder
upon cooling. This monomer had the highest melting point of
the series reported in the present study (274 °C).
Polycondensation and Polymer Characterization. The
six sulfonated homopolymers shown in Scheme 5 were synthe-
sized by polycondensation via nucleophilic aromatic substitu-
tions. The monomers used for each polymer are indicated in
Table 1. As seen, the polymers (ionomers) differed with regard
to the connectivity and length of the hydrophobic ether
segment, and the nature of the sulfonated unit.
Aromatic−aliphatic ketones can be converted to arylate
esters by Baeyer−Villiger rearrangements.40 The conversion of
the reaction with mCPBA (m-chloroperbenzoic acid) in
chloroform was limited, even when a large excess of reactant
was used because of the poor solubility of 10. Poly(ether
ketone)s can undergo homogeneous chemical modification in a
mixture of chlorinated solvent and TFA (trifluoroacetic
acid).36−39 In the case of Baeyer−Villiger rearrangements,
addition of a small amount of TFA has been shown to improve
the conversion and to increase the reaction rate. The reaction
was monitored by 1H NMR spectroscopy through the progres-
sive disappearance of the resonance at 2.73 ppm corresponding
to methyl group next to the ketone, and the simultaneous
appearance of a singlet at 3.63 ppm originating from the methyl
ester. Excess mCPBA and TFA were removed by liquid extrac-
tion and oligomer 11 was immediately converted to bisphenol
Polymers 13 and 14 were prepared by condensation of the
monosulfonated monomer 1 with bisphenol monomer 5 and 7,
respectively. These polymers contained singlets of sulfonic acid
groups regularly separated by a four aromatic ring spacer, giving
them a theoretical IEC of 1.5 mequiv g−1. In our previous
1450
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
Scheme 4. Synthesis of Bisphenol Monomer 12 by
Figure 1 shows the 1H NMR spectra of all polymers prepared
in this study. In every case, resonances were observed exclusi-
vely in the aromatic region. The compositions of the polymers
obtained by integrating and comparing the NMR signals were
very close to the theoretical values predicted from the mono-
mer feed ratios. Additional spectroscopic data were consistent
with the proposed structures. The inherent viscosity (ηinh) of
the polymers indicated that moderate to high molecular weights
were systematically obtained.43 Notably, the poly(arylene ether
ketone)s 15 and 17 had significantly higher inherent viscosity
than the other polymers (Table 1). This might suggest that
monomer 3 was more efficient in the polycondensations than
monomer 2.
Nucleophilic Aromatic Substitution, Followed by Baeyer−
a
Villiger Rearrangement
Membrane Preparation. Flexible transparent films with a
thickness between 70 and 90 μm were obtained by casting the
polymers from DMAc solutions in all cases but for polymer 18.
According to the inherent viscosity measurements, this polymer
had a similar molecular weight as the other polymers prepared
in the present work. Attempts to prepare thin films from NMP
solution at different temperatures and using different casting
substrates such as glass and Teflon were unsuccessful. A fragile
film was obtained in every case and thus it was not possible to
further evaluate the membrane properties of polymer 18. As
seen in Table 2, the IEC values of the other membranes were in
very good agreement with the theoretical value calculated from
the monomer feed ratios. This again confirmed the structure of
the polymers prepared.
X-ray Scattering. The morphological features of the
completely amorphous ionomer membranes were studied by
SAXS measurements. Generally, a broad scattering peak is
observed at scattering vector (q) ≈1−5 nm−1 in ionomers and
is referred to as the “ionomer peak”. This feature is widely
attributed to interaggregate scattering and its position (qmax
)
can be correlated to the characteristic separation length (d)
between the ionic aggregates in the polymer-rich matrix phase.44
Because this scattering feature is usually relatively broad and
weak, intraparticle interference have also been proposed to give
rise to the ionomer peak.45 However, there is now a general
consensus in the community that the scattering originates from
interaggregate scattering.
Shown in Figure 2 are the SAXS profiles of ionomers 13 and
14, based on the four-ring bisphenols, and ionomers 15-17,
based on the six-ring bisphenols. Before analysis, the mem-
branes were ion-exchanged to selectively stain the sulfonated
domains of the membranes with Pb2+ to enhance the contrast.
The values of qmax and d are collected in Table 2 for membranes
13−17. For comparison, the corresponding data of Nafion
NRE212 (IEC = 0.91 mequiv g−1) and an archetypical post-
polymerization sulfonated aromatic polymer, polyphenylsulfone
(sPPSU), are included in the Pb2+ form. The latter polymer was
obtained by polycondensation of 4,4′-biphenol and 4,4′-
dichlorodiphenyl sulfone, followed by sulfonation using sulfuric
acid. The fully aromatic sPPSU had an IEC of 1.84 mequiv g−1
and a backbone with similar structural elements as ionomers
15−17, but with the sulfonic acid units statistically distributed
along the polymer backbone in ortho-ether positions (Scheme 6).
As seen in Figure 2, the ionomer peak of the sPPSU membrane
a
Key: (i) K2CO3, DMF, 150 °C; (ii) mCPBA, TFA, CHCl3; (iii)
KOH, MeOH; (iv) HCl, H2O.
reports on monomer 1, polymerizations with bisphenols were
carried out in dimethylacetamide (DMAc).41,42 In the present
case, the polymerizations were carried out in DMSO at 160 °C.
After a 4 h dehydration period, the viscosity of the poly-
merization solution increased significantly after 4 and 6 h,
respectively. This suggested that the polymerization of 1
with bisphenols proceeded at a higher rate in DMSO than in
DMAc. Furthermore, no precipitation of the polymer was
observed.
Next, four polymers were prepared by polycondensations of
the six-ring bisphenol monomers 9 and 12 with the
disulfonated monomers 2 and 3. The resulting polymers were
thus composed of doublets of sulfonic acid groups regularly
separated by spacers containing six nonsulfonated aromatic
units. The poly(arylene ether ketone)s 15 and 17 had a
theoretical IEC slightly higher than the poly(arylene ether
sulfone)s 16 and 18. Indeed, monomer 2 has a lower molecular
weight than monomer 3. There are many examples of statistical
copolymers of similar IEC values prepared from the sulfonated
monomers 2 and 3.8,9 However, to the best of our knowledge,
samples 15-18 are the first examples of such homopolymers
with an IEC of approximately 2 mequiv g−1. Polymers 15 and
17 were prepared in DMSO while polymers 16 and 18 were
prepared in DMAc. Again, no precipitation of the polymers was
observed during the polycondensations.
was much broader and shifted toward a higher q-value (qmax
=
2.4 nm−1), as compared to the profile of the Nafion membrane
(qmax = 1.65 nm−1). The values of qmax corresponded to d = 1.65
nm for sPPSU and 3.8 nm for Nafion. The results thus indicted
a smaller average interaggregate distance with a wider
distribution in the sPPSU membrane. The extremely hydrophobic
1451
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
Scheme 5. Ionomers Having Poly(arylene ether) (13 and 14), Poly(arylene ether ketone) (15 and 17) and Poly(arylene ether
sulfone) (16 and 18) Backbones with Precisely Sequenced Ionic Moieties
the interaggregate distances in relation to the statistical
copolymer.
Table 1. Properties of Sulfonated Homopolymers
sulfonated
monomer
bisphenol
monomer
a
b
ηinh (dL g−1)
Tg (°C)
It is immediately clear from Figure 2 that membranes 15−17
gave rise to dramatically different scattering profiles with
considerably more intense scattering and much narrower
ionomer peaks than both sPPSU and Nafion. The value of
qmax for these ionomer membranes was approximately the same
as for Nafion, approximately 2 nm−1. Apparently, by precisely
controlling the acid group separation, and keeping it sufficiently
large, the ionomer peak became much more intense and well-
defined. A further strong indication of the narrow distribution
in the interaggregate correlation length was the appearance of
clear second order-peaks at 2qmax (indicated by arrows in Figure 2).
These were similar in appearance to those only previously
observed in the scattering profiles of monodisperse halato
telechelic ionomers46 and other well-defined model ion-
omers.47,48 To our knowledge, this is the first report on
second-order features observed for fully aromatic ionomers.
There was seemingly no influence by the connectivity of the
oligo(ether) spacers and the sulfonated moieties (sulfone or
ketone) on the scattering peak position and width of ionomers
13-17. Obviously, the precise ionomers with their evenly and
sufficiently spaced ionic moieties allowed for a much more
uniform self-aggregation of the ionic groups to occur. This in
turn indicated an increased degree of phase separation and
morphological control of the membranes. Previously, Colquhoun
and co-workers have studied the morphology of microblock
polymers with sulfonated moieties precisely spaced by oligo-
(arylene ketones) by SAXS.27 They observed that a polymer
polymer
13
14
15
16
17
18
1
1
2
3
2
3
5
0.35
0.24
0.47
0.20
0.51
0.29
230
259
304
283
348
302
7
9
9
12
12
a
b
0.5 wt % in DMF containing LiBr (0.05 M) at 25 °C. Measured for
polymers in the Na+ form.
polymer backbone combined with the superacidity and
hydrophilicity of the −CF2−SO3H unit in Nafion give rise to
a distinct phase separation and a quite well-defined mor-
phology. Typically, aromatic hydrocarbon backbones are far
less hydrophobic and the arylenesulfonic acid units of these
ionomers are far less acidic, which leads to a morphology with a
less pronounced phase separation. The morphological dif-
ferences between Nafion membranes and statistically sulfonated
aromatic hydrocarbon membranes have previously been extensively
discussed by Kreuer on the basis of SAXS data.7
The SAXS profiles of membranes 13 and 14 showed
ionomer peak positions at higher qmax in comparison with
sPPSU. Moreover, the peaks were significantly narrower, with
membrane 14 showing the narrowest one. Seemingly, the
structure of 13 and 14 with monosulfonated sites spaced by
four-ring oligo(ether) segments decreased the distributions of
1452
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
1
Figure 1. H NMR spectra (DMSO-d6) of the precisely sequenced sulfonated polymers, identified by their numerals.
Table 2. Properties of the Ionomer Membranes
a
a
b
c
−1
−1
polymer theoretical IEC (mequiv g ) titrated IEC (mequiv g ) qmax (nm−1)
d (nm)
T95‑Na (°C)
T95‑H (°C)
water uptake (wt %)
λ
13
14
15
16
17
1.50
1.50
2.14
1.99
2.14
1.50
1.48
2.06
1.84
2.05
2.8
2.6
2.0
1.9
2.0
2.2
2.4
3.2
3.3
3.2
398
404
415
424
442
292
171
305
292
296
22
28
50
68
44
8.1
10.5
13.5
20.5
11.9
a
b
Membranes in the Pb2+ form. Temperature corresponding to a 5% weight loss of membranes in their Na+ form under nitrogen atmosphere at a
c
heating rate of 10 °C min−1. Temperature corresponding to a 5% weight loss of membranes in their protonated form under air at a heating rate
of 1 °C min−1
with six-ring spacers showed a small but well-defined peak,
indicating scattering at the same length-scale as that observed
for Nafion. A polymer with 12-ring spacers showed a very much
more intense peak, at significantly lower q, than for the one
with the six-ring spacers, thus showing similar trends as
observed in the present work. However, Colquhoun et al. found
no second-order scattering features of their membranes.
Possibly, the higher mobility of the oligo(ether) segments in
relation to the oligo(ketone) segments allowed for a more
efficient ionic aggregation in the present case. We have
previously shown that membranes based on poly(arylene
ether sulfone)s with sulfonic acid groups on short side chains
gave SAXS profiles with narrow ionomer peaks at lower qmax
than corresponding polymers with sulfonic acid groups
statistically distributed along the backbone.24,25 However, in
this case no second order features were found.
The nature of the counterion have previously been reported
to influence the morphology of ionomers.49 In order to
1453
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
Figure 2. SAXS data recorded using ionomer membranes ion-
exchanged with lead acetate to enhance the scattering intensity. The
Figure 3. SAXS data recorded of ionomer membranes 13 and 15 in
the H+, Na+, and Pb2+ form, respectively.
arrows indicate second-order peaks at 2qmax
.
Scheme 6. Structure of the sPPSU Reference Ionomer with
Statistically Sequenced Ionic Groups
investigate the effect of the counterion in the present case,
membranes 13 and 15 were ion-exchanged from their Pb2+
form to their H+ and Na+ form, respectively, to study the effect
on the SAXS profiles. As seen in Figure 3, the position of the
ionomer peak was not significantly influenced by the nature of
the counterion. The reason may be that the initial membrane
morphology, which is formed during the casting process, is
largely retained during the ion-exchange process because of the
glassy oligo(ether) segments. Perhaps a larger effect of the
counterion would be observed if the membranes had been cast
directly from ionomers in different ionic forms. In particular,
for the series based on membrane 13, the scattering intensity
decreased with the size of the counterion. In the case of
membrane 15, the intensity of the ionomer peak did not change
appreciably with the nature of the counterion. However, the
intensity of the second order-peak at 2qmax decreased. It was
only very slightly visible for the membrane in the Na+ form and
had disappeared for the membrane in the H+ form.
Figure 4. SAXS data of membranes 13 and 15 in the Pb2+ form
recorded before (as cast) and after 10 min annealing 20 °C above the
respective Tg.
The thermal history of ionomer membranes cast from high-
Tg polymers can be expected to greatly influence the
aggregation of ionic groups.50 To study the effect of annealing
on the membranes in the present case, membranes 13 and 15
in the Pb2+ form were kept at a temperature 20 °C above their
respective Tg (Table 2). The SAXS profiles of the annealed
membranes are shown together with the nonannealed (as cast)
ones in Figure 4. As seen, the ionomer peak position of
membrane 13 increased significantly from qmax = 2.8 to 3.6 nm−1,
corresponding to a decrease in d from 2.2 to 1.7 nm. On the
other hand, membrane 15 was only slightly affected by the
annealing and the ionomer peak position decreased only very
slightly, 0.1 nm−1. This indicated that ionomer 15 with its
doublet ionic moieties and longer oligo(ether) segments
1454
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
reached closer to an equilibrium structure during the casting
process.
1.41,42 For all the polymers, the degradation occurred in two
steps, presumably by desulfonation followed by polymer main
chain degradation.
Thermal Properties. The thermal stability of the
membranes was evaluated by thermogravimetric analysis
(TGA) up to 600 °C. Measurements were performed at
10 °C min−1 under N2 with the membranes in their Na+ form.
In addition, measurements were done at 1 °C min−1 under air
with the membranes in their acid form to investigate the
oxidative stability of the materials under more drastic
conditions. The TGA traces are shown in Figure 5 and the
Glass transitions were studied by differential scanning
calorimetry (DSC) for the polymers in their Na+ form and
the values are found in Table 1 (the DSC traces the membranes
in the Na+ form are shown in the Supporting Information).
Glass transitions were not always detected for the polymers in
their acid form. However, in the cases where it was found, the Tg
values recorded for the two forms were similar. The polymers
containing six-ring ether segments and doublets of acid units
showed higher Tg values than the polymers containing four-ring
ether segments and singlets of acid units. As expected for materials
based on the same sulfonated monomer, the polymer having an
all-para connectivity in the hydrophobic polyether segment had a
higher Tg than the corresponding polymer with meta connectivity.
For example, the Tg of polymer 14 was 30 °C higher than for 13.
Also, for polymers with identical ether segment, ketone mono-
mer 2 gave a higher Tg than polymers derived with sulfone mono-
mer 3. As a result, the polymer with the highest Tg in this study
was 17. In its Na+ form, the value was found at 348 °C, close to
the degradation temperature.
Water Uptake and Proton Conductivity. The water
uptake plays a critical role for proton conducting polymer
membranes. The presence of water is necessary to facilitate
proton transport, and high proton conductivity is often
associated with high IECs and high water uptake. However,
excessive water uptake leads to loss of dimensional stability of
the membrane. The water uptake at room temperature and the
corresponding number of water molecules per sulfonic acid
group (λ) are reported in Table 2. The membranes based on
the polymers with the four-ring ether segments had an IEC of
approximately 1.5 mequiv. g‑1 (13 and 14) and showed lower
water uptake values than the membranes based on polymers
15-17 with higher IECs. The latter membranes had water
uptakes of between 44 and 68%. The membrane based on
polymer 16 had a λ value of 20 which was markedly higher than
for the other membranes, which most probably indicated a
limited ability of this membrane to resist an excessive water
uptake. There was seemingly no clear relationship between
polymer chain connectivity and the water uptake at room
temperature. However, the polymers containing meta linkages
in the ether segments all swelled excessively in water at
temperatures exceeding 80 °C. Substituting the meta linkages
by para links in the ether segment resulted in a greatly
improved dimensional stability of the membrane in hot water.
For example, the all-para poly(arylene ether ketone) 17 with its
high Tg value did not swell excessively in boiling water.
The proton conductivity of the membranes was evaluated
using electrochemical impedance spectroscopy (EIS). Measure-
ments were performed as a function of temperature at 100%
RH using a 2-probe method, and also as a function of RH at
80 °C using a 4-probe method. Figure 6 shows the data
collected as a function of temperature. As seen in Figure 6, the
membranes based on polymers 13 and 14 showed a proton
conductivity that was almost an order of magnitude lower than
the other membranes under 100% RH. This was due to their
lower IEC values and water uptake, as discussed above. Mem-
brane 17 showed a conductivity in level with that of Nafion,
and the conductivity of membrane 15 and 16 exceeded that of
Nafion under these conditions to reach 0.2 S cm−1 at 80 °C.
The polymer connectivity had direct influence on proton
conductivity. Consistently, the meta homopolymers had a
Figure 5. TGA traces of the sulfonated membranes in their Na+ form
recorded at 10 °C min−1 under nitrogen (a), and their acid form
recorded at 1 °C min−1 under air (b).
temperatures corresponding to a 5% weight loss (T95) are
summarized in Table 2. In the Na+ form, the polymers con-
taining ether segments with para connectivity showed slightly
higher T95 values than the corresponding polymers containing
meta connectivity. Moreover, the polymers containing six-ring
ether segments and doublets of acid units had higher T95 values
than the polymers containing four-ring ether segments and
singlets of acid units. The T95 values of the polymers in the Na+
form were found between 400 and 442 °C. As expected, the
thermal stability decreased after converting the membranes to
the acid form. Still, the ionomers had high thermo-oxidative
stability as demonstrated by their high T95 values above 290 °C.
For some reason ionomer 14 exhibited a drastic weight loss
already from 170 °C. A similar degradation has previously been
observed in some cases for membranes based on monomer
1455
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
below that of Nafion in the investigated range between 30 and
90% RH. Membrane 15 had higher proton conductivity than
17 presumably because of the higher water uptake of the
former membrane. The significantly higher proton conductivity
of the perfectly sequenced ionic polymer membranes was likely
due to the more regular ionic cluster distribution which
facilitated the formation of a more efficient water pore system
for proton transport.
EXPERIMENTAL SECTION
■
Materials. Monomer 1,30 2,31 and 332,33 were synthesized and
characterized as reported previously. Cesium carbonate, magnesium
sulfate, Celite, 4,4′-dibromobiphenyl (DBBP), 4-fluoroacetophenone
and 4-bromoanisole were obtained from Aldrich. Dimethylformamide
(DMF), N,N-dimethylacetamide (DMAc), N-methyl-2-pyrrolidone
(NMP), 2,2,6,6-Tetramethyl-3,5-heptanedione (TMHD), copper(I)
chloride, 3-methoxyphenol (mMP), 4-methoxyphenol (pMP), potas-
sium carbonate, Hydrobromic acid 48%, acetic acid, m-chloroperben-
zoic acid (mCPBA) and trifluoroacetic acid (TFA) were purchased
from Acros. Methanol, 2-propanol, diethyl ether, toluene and
chloroform were from Fisher. Potassium carbonate was grounded to
a fine powder and dried at 120 °C prior to use. Nafion NRE212 was
from Alfa Aesar.
Figure 6. Arrhenius conductivity plots of data measured by
electrochemical impedance spectroscopy with the membranes under
100% RH in a sealed cell. Values are given for Nafion for comparison.
Bismethoxy Oligomer 4. A setup consisting of a two-neck round
bottomed flask heated by an oil bath and equipped with a magnetic
stirrer, a N2 inlet, and a condenser was used in the syntheses. A
solution of 4,4′-dibromobiphenyl (1.560 g, 5.00 mmol), Cs2CO3
(1.629 g, 5.00 mmol) 3-methoxyphenol (1.862 g, 15.00 mmol), and
2,2,6,6-tetramethyl-3,5-heptanedione (0.460 g, 2.50 mmol) in NMP
(10 mL) was added to the reactor and stirred at room temperature
under nitrogen atmosphere. Then CuCl (0.250 g, 2.50 mmol) was
added and the mixture was degassed, filled with nitrogen and heated at
165 °C. After 22 h, the solution was cooled, diluted with diethyl ether
and filtered through Celite. The filtrate was washed with water, dried
with magnesium sulfate, filtered and evaporated. The resulting brown
solid was stirred in methanol to afford the desired product 4 that was
isolated by filtration as a white powder (1.55 g, 78% yield).
higher proton conductivity than their para analogues. For
example, membrane 13 had higher proton conductivity than
membrane 14. Similarly, membrane 15 had a higher conducti-
vity than 17. This may be explained by differences in the water
uptake and Tg values. A lower Tg indicates a higher chain
mobility which leads to higher water uptake and conductivity,
especially at increasing temperatures.
Figure 7 shows the proton conductivity of 15 and 17 under
reduced RH at 80 °C. Included are the corresponding data of
1
Mp: 104 °C. H NMR (400 MHz, CDCl3, δ): 3.81 (6H, s, O−
CH3), 6.63 (6H, m, Ar), 7.10 (4H, d, Ar), 7.27 (2H, d, Ar), 7.54
(4H, d, Ar). 13C NMR (100 MHz, CDCl3, δ): 55.4, 104.9, 109.1,
111.1, 119.3, 128.2, 130.2, 135.7, 156.3, 158.4, 161.1. IR (KBr): νC−H
3041 cm−1, νO−CH3 2837 cm−1, νC−O−C 1128 cm−1.
Bisphenol Monomer 5. To a stirred solution of 4 (1.99 g,
5.0 mmol) in acetic acid (50 mL) was added 48% HBr (6 mL). After
16 h at 160 °C under air, the solution was cooled and poured in ice.
The dark precipitate was washed with water, filtered, and dried under
vacuum at 60 °C. Recrystallization in toluene afforded 5 as a brown
solid (1.71 g, 92% yield).
1
Mp: 168 °C. H NMR (400 MHz, CDCl3, δ): 6.41 (6H, m, Ar),
7.07 (4H, d, Ar), 7.16 (2H, d, Ar), 7.65 (4H, d, Ar). 13C NMR
(100 MHz, CDCl3, δ): 105.9, 109.8, 110.8, 119.6, 128.4, 130.8, 135.1,
156.4, 158.2, 159.3. IR (KBr): νC−OH 3357 cm−1, νC−H 3062 cm−1,
νC−O−C 1128 cm−1.
Bismethoxy Oligomer 6. A solution of 4,4′-dibromobiphenyl
(7.800 g, 25.00 mmol), Cs2CO3 (9.774 g, 30.00 mmol) 4-methoxy-
phenol (9.310 g, 75.00 mmol), and 2,2,6,6-tetramethyl-3,5-heptane-
dione (2.302 g, 12.50 mmol) in NMP (50 mL) was stirred at room tem-
perature under nitrogen atmosphere. Then CuCl (1.250 g, 12.50 mmol)
was added. The mixture was degassed, filled with nitrogen and heated
at 170 °C. After 22 h, the solution was cooled and filtered through
Celite. The filtrate was washed with water, dried with MgSO4 and
evaporated. The resulting brown solid was stirred in methanol and the
desired product 6 was isolated as a white powder by filtration (6.97 g,
70% yield).
Figure 7. Conductivity data measured by electrochemical impedance
spectroscopy for membranes 15 and 17 measured at 80 °C under
reduced RH. Values for Nafion (IEC = 0.91 mequiv g−1) and sPPSU
(IEC = 1.84 mequiv g−1), with statistically distributed sulfonic acid
groups, are given for comparison.
Mp: 210 °C. 1H NMR (400 MHz, CDCl3, δ): 3.82 (6H, s,
O−CH3), 6.91 (4H, d, Ar), 7.00 (8H, AA′XX′, Ar), 7.48 (4H, d, Ar);
13C NMR (100 MHz, CDCl3, δ): 55.3, 114.7, 117.9, 120.8, 127.9,
Nafion and sPPSU. As seen, the conductivity of 15 and 17
reached well above that of sPPSU under reduced RH, but fell
1456
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
135.0, 149.9, 157.8. IR (KBr): νC−H 3040 cm−1, νO−CH3 2837 cm−1,
νC−O−C 1132 cm−1.
13C NMR (100 MHz, DMSO-d6, δ): 116.7, 118.5, 119.1, 121.0, 121.2,
134.7, 148.8, 151.3, 154.1, 154.8, 157.3. IR (KBr): νOH 3290 cm−1,
νC−H 3040 cm−1, νC−O−C 1098 cm−1.
Bisphenol Monomer 7. To a stirred solution of 6 (1.394 g,
3.50 mmol) in acetic acid (160 mL) was added 48% HBr (12 mL).
The solution was heated to 160 °C for 16 h. Then the solution was
poured in ice to give a white precipitate that was filtered and washed
with water. The desired product 7 was obtained as a white powder
after purification by recrystallization from toluene (1.16 g, 90% yield).
Mp: 249 °C. 1H NMR (400 MHz, DMSO-d6, δ): 6.79 (4H, d, Ar),
6.91 (8H, AA′XX′, Ar), 7.53 (4H, d, Ar). 13C NMR (100 MHz,
DMSO-d6, δ): 116.6, 117.6, 121.5, 128.2, 134.1, 148.1, 154.3, 158.2. IR
(KBr): νC−OH 3466 cm−1, νC−H 3033 cm−1, νC−O−C 1132 cm−1.
Bismethoxy Oligomer 8. A mixture of 5 (0.926 g, 2.50 mmol),
Cs2CO3 (1.955 g, 6.00 mmol), 4-bromanisole (1.402 g, 7.50 mmol),
and 2,2,6,6-tetramethyl-3,5-heptanedione (0.921 g, 5.00 mmol) were
stirred in NMP (7.5 mL) under nitrogen atmosphere. After 30 min,
CuCl (0.500 g, 5.00 mmol) was added and the mixture was degassed,
filled with nitrogen and heated to 170 °C. After 22 h, the solution was
diluted with diethyl ether and filtered through Celite. The filtrate was
washed with water, dried with magnesium sulfate, filtered and evapo-
rated. The resulting brown solid was stirred in methanol to afford the
desired oligomer 8 that was isolated by filtration (1.01 g, 70% yield).
Mp: 98 °C. 1H NMR (400 MHz, CDCl3, δ): 3.82 (6H, s, O−CH3),
6.68 (6H, m, Ar), 6.90 (4H, d, Ar), 7.02 (4H, d, Ar), 7.09 (4H, d, Ar),
7.24 (2H, d, Ar), 7.52 (4H, d, Ar). 13C NMR (100 MHz, CDCl3, δ):
55.3, 108.0, 112.1, 112.5, 114.9, 119.2, 120.8, 128.1, 130.3, 135.9,
149.3, 156.0, 158.2, 159.8. IR (KBr): νC−H 3039 cm−1, νO−CH3 2834
cm−1, νC−O−C 1122 cm−1.
Polymer 13. A mixture of potassium carbonate (0.345 g, 2.50 mmol),
5 (0.370 g, 1.00 mmol) and 1 (0.304 g, 1.00 mmol) was suspended in
DMSO (3 mL) and toluene (10 mL) in a two-necked flask fitted with
a nitrogen gas inlet, a Dean−Stark trap, a condenser and a calcium
chloride guard. Water was removed by azeotropic distillation with
toluene maintaining a reflux temperature of 140 °C for 4 h. Then,
toluene was removed and the reaction temperature was increased to
160 °C. After 4 h the viscous mixture was cooled and poured into
2-propanol. The precipitate was filtered off, washed with water and
dried. The solid was dissolved in DMSO and precipitated from 2-propanol
to give 13 as a powder that was dried in a vacuum oven at 80 °C.
1H NMR (400 MHz, DMSO-d6, δ): 6.57 (2H, d, Ar), 6.71 (4H, m, Ar),
6.81 (2H, d, Ar), 7.09 (4H, d, Ar), 7.23 (6H, m, Ar), 7.63 (1H, d, Ar), 7.69
(4H, d, Ar). 13C NMR (100 MHz, DMSO-d6, δ): 110.7, 113.5, 113.8,
114.9, 119.5, 124.4, 127.6, 128.6, 129.1, 129.4, 130.8,132.4, 135.4, 140.2,
−1
145.5, 156.2, 156.9, 157.7, 158.2, 193.9. IR (KBr): νCO 1681 cm ,
−1
−1
ν
1089 cm , νSO 1022 cm .
C−O−C
3
Polymer 14. The polymerization of 7 (0.370 g, 1.00 mmol) and 1
(0.304 g, 1.00 mmol) was carried out using an identical procedure as
described for polymer 13. The reaction was stopped after 6 h and 14
was recovered as a white powder.
1H NMR (400 MHz, DMSO-d6, δ): 6.46 (2H, br, Ar), 7.00 (12H,
br, Ar), 7.31 (4H, br, Ar), 7.65 (5H, br, Ar). 13C NMR (100 MHz,
DMSO-d6, δ): 112.1, 118.6, 120.7, 121.8, 123.6, 127.7, 128.4, 128.7,
129.3, 132.2, 134.8, 140.4, 145.5, 152.2, 152.7, 157.2, 157.9, 194.1. IR
(KBr): νCO 1683 cm−1, ν
1089 cm-1, νSO 1019 cm−1.
Bisphenol Monomer 9. To a stirred solution of 8 (1.16 g,
2.0 mmol) in acetic acid (20 mL) was added 48% HBr (5 mL). After
16 h at 160 °C, the solution was poured in ice. The white precipitate
was then washed with water, filtered, and dried under vacuum at
60 °C. Recrystallization in toluene afforded the desired product 9 as a
red powder (0.99 g, 90% yield).
C−O−C
3
Polymer 15. A mixture of potassium carbonate (0.207 g, 1.50 mmol),
9 (0.388 g, 0.70 mmol), and 2 (0.296 g, 0.70 mmol) was suspended in
DMSO (3 mL) and toluene (10 mL) in a two-necked flask fitted with
a nitrogen gas inlet, a Dean−Stark trap, a condenser, and a calcium
chloride guard. Water was removed by azeotropic distillation with
toluene maintaining a reflux temperature of 140 °C for 4 h. Then
toluene was removed and the reaction temperature was increased to
160 °C. After 21 h the viscous mixture was cooled and poured in
2-propanol. The precipitate was filtered off, washed with water and dried.
The solid was dissolved in DMSO and precipitated from 2-propanol to
give 15 as a powder that was dried in a vacuum oven at 80 °C.
Mp: 158−166 °C. 1H NMR (400 MHz, DMSO-d6, δ): 6.54 (6H, m,
Ar), 6.77 (4H, d, Ar), 6.92 (4H, d, Ar), 7.09 (4H, d, Ar), 7.30 (2H, d,
Ar), 7.64 (4H, d, Ar). 13C NMR (100 MHz, DMSO-d6, δ): 107.5,
112.1, 112.6, 116.6, 119.7, 121.7, 128.5, 131.3, 135.4, 147,7, 154.5,
156.1, 158.2, 160.4. IR (KBr): νC−OH 3384 cm−1, νC−H 3068 cm−1,
νC−O−C 1122 cm−1.
Bisacetophenone Oligomer 10. A solution of monomer 7 (1.111 g,
3.00 mmol), 4-fluoroacetophenone (0.967 g, 7.00 mmol), potassium
carbonate (0.967 g, 7.00 mmol) in DMF (10 mL) was stirred at
150 °C under nitrogen atmosphere. After 18 h, the red solution was
cooled and poured in water. The resulting white precipitate was
filtered, and dried under vacuum at 80 °C (1.454 g, 80% yield).
Mp: 263 °C. 1H NMR (400 MHz, CDCl3/TFA 6:1, δ): 2.70 (6H, s,
CO−CH3), 7.05 (16H, AA′XX′, Ar), 7.58 (4H, d, Ar), 8.02 (4H, d,
Ar). 13C NMR (100 MHz, CDCl3/TFA 6:1, δ): 25.6, 116.8, 119.0,
120.5, 122.0, 128.3, 129.9, 131.9, 136.0, 150.2, 154.2, 156.4, 164.1. IR
(KBr): νC−H 3043 cm−1, νC−O 1678 cm−1, νC−C 1600 cm−1, νC−O−C
1112 cm−1.
1H NMR (400 MHz, DMSO-d6, δ): 6.73 (6H, br, Ar), 6.87 (2H, m,
Ar), 7.14 (12H, br, Ar), 7.37 (2H, m, Ar), 7.68 (6H, m, Ar), 8.22 (2H,
br, Ar). 13C NMR (100 MHz, DMSO-d6, δ): 108.7, 109.0, 112.8,
113.2, 118.1, 119.9, 121.5, 121,7, 122.3, 128.5, 131.1, 131.2, 131.5,
132.6, 135.4, 138.2, 152.3, 152.6, 156.0, 158.4, 159.1, 193.6. IR (KBr):
νCO 1653 cm−1, νSO 1028 cm−1.
3
Polymer 16. A mixture of potassium carbonate (0.207 g, 1.50 mmol),
9 (0.388 g, 0.70 mmol) and 3 (0.344 g, 0.70 mmol) was suspended in
DMAc (3 mL) and toluene (10 mL) in a two-necked flask fitted with a
nitrogen gas inlet, a Dean−Stark trap, a condenser and a calcium
chloride guard. Water was removed by azeotropic distillation with
toluene maintaining a reflux temperature of 140 °C for 4 h. Then
toluene was removed and the reaction temperature was increased to
165 °C. After 22 h the viscous mixture was cooled and poured in
2-propanol. After the purification procedure described above, 16 was
recovered as a powder.
Bisphenol Monomer 12. A suspension of 10 (0.848 g, 1.40 mmol)
in chloroform (160 mL) was cooled in an ice bath. Then, mCPBA
(2.390 g, 10.00 mmol) and trifluoroacetic acid (5 mL) were added
successively. The mixture was left to warm up to room temperature
and was vigorously stirred in the dark for 72 h. The reaction was then
quenched by addition of a saturated solution of NaHSO3. The organic
layer was washed with saturated NaHCO3 solution and water. It was
dried with magnesium sulfate, filtered and evaporated under reduced
pressure to give 11 as a white powder. Oligomer 11 was stirred in a
0.1 M solution of KOH in methanol (200 mL) under reflux condition
for 90 min. Then the solvent was removed and the resulting residue
was suspended in 1 M HCl. The desired bisphenol monomer 12 was
isolated by filtration and was purified by adding 2-propanol to a DMF
solution until a precipitate was formed. The solid was dissolved by
heating and recrystallization from DMF and 2-propanol afforded 12 as
a white powder.
1H NMR (400 MHz, DMSO-d6, δ): 6.76 (6H, br, Ar), 6.91 (2H, m,
Ar), 7.12 (12H, m, Ar), 7.36 (2H, m, Ar), 7.68 (4H, m, Ar), 7.82 (2H,
m, Ar), 8.28 (2H, br, Ar). 13C NMR (100 MHz, DMSO-d6, δ): 107.6,
109.0, 112.9, 113.2, 116.7, 119.1, 119.9, 121.3, 121.6, 122.3, 128.5,
130.3, 131,3 131.6, 134.4, 135.5, 139.0, 152.0, 152.7, 155.9, 158.5,
159.0. IR (KBr): νOSO 1163 cm−1, νSO 1027 cm−1.
3
Polymer 17. A mixture of potassium carbonate (0.207 g, 1.50 mmol),
12 (0.388 g, 0.70 mmol) and 2 (0.296 g, 0.70 mmol) was suspended in
DMSO (3 mL) and toluene (10 mL) in a two-necked flask fitted with
a nitrogen gas inlet, a Dean−Stark trap, a condenser and a calcium
chloride guard. Water was removed by azeotropic distillation with
toluene maintaining a reflux temperature of 140 °C for 18 h. The
toluene was removed and the reaction temperature was increased
Mp: 274 °C. 1H NMR (400 MHz, DMSO-d6, δ): 6.77 (4H, d, Ar),
6.88 (8H, AA′XX′, Ar), 7.01 (8H, AA′XX′, Ar), 7.60 (4H, d, Ar).
1457
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
to 160 °C. After 40 min the highly viscous mixture was cooled and
poured in 2-propanol. The precipitate was filtered off, washed with
water and dried. The solid was dissolved in DMSO and precipitated
from 2-propanol to give 17 as a powder that was dried in a vacuum
oven at 80 °C.
solid sample holder at 25 °C. The wave vector (q) was calculated by
the software (SAXSquant) according to:
−1
q = 4πλ sin(θ)
(1)
where 2θ is the scattering angle. The characteristic separation length
(d), i.e., the Bragg spacing, was calculated as:
1H NMR (400 MHz, DMSO-d6, δ): 6.89 (2H, br, Ar), 7.12 (20H,
br, Ar), 7.65 (6H, br, Ar), 8.23 (2H, br, Ar). 13C NMR (100 MHz,
DMSO-d6, δ): 118.1, 118.8, 120.3, 120.5, 121.0, 121.2, 121.3, 122.3,
128.4, 131.1, 132.7, 135.0, 138.2, 152.1, 152.3, 153.3, 153.6, 157.1,
−1
d = 2πq
(2)
158.7, 193.5. IR (KBr): νCO 1652 cm−1, νSO 1027 cm−1.
Proton conductivity was measured by electrochemical impedance
spectroscopy (EIS) using a Novocontrol V 1.01S dielectric analyzer
from −20 to +120 °C under 100% relative humidity. In-plane
conductivity was measured using two stainless steel electrodes.
Impedance data were collected between 10−1 and 107 Hz at a voltage
amplitude of 50 mV. For each temperature, the proton conductivity
was taken as the real component of the conductivity corresponding to
the minimum of the imaginary component of the conductivity. The
humidity dependence of the proton conductivity at 80 °C from 30 to
90% RH was investigated by a four-probe method using a Gamry
potentiostat/galvanostat/ZRA and a Fumatech MK3 conductivity cell,
where the humidity was equilibrated by deionized water in a closed
system.
3
Polymer 18. A mixture of potassium carbonate (0.207 g, 1.50 mmol),
12 (0.388 g, 0.70 mmol) and 3 (0.344 g, 0.70 mmol) was suspended in
DMAc (3 mL) and toluene (10 mL) in a two-necked flask fitted with a
nitrogen gas inlet, a Dean−Stark trap, a condenser and a calcium
chloride guard. Water was removed by azeotropic distillation with
toluene maintaining a reflux temperature of 140 °C for 4 h. Then
toluene was removed and the reaction temperature was increased to
160 °C. After 7 h the mixture was cooled and poured in 2-propanol.
Polymer 18 was recovered as a powder after using the same
purification as described above.
1H NMR (400 MHz, DMSO-d6, δ): 6.77 (4H, m, Ar), 7.10 (16H,
m, Ar), 7.64 (4H, m, Ar), 7.82 (2H, m, Ar), 8.27 (2H, br, Ar). 13C
NMR (100 MHz, DMSO-d6, δ): 116.7, 118.5, 118.8, 119.0, 120.3,
120.6, 121.0, 121.2, 122.4, 128.5, 130.3, 134.3, 134.9, 138.9, 151.4,
CONCLUSIONS
■
152.4, 153.1, 154.0, 157.1, 159.2. IR (KBr): νOSO 1163 cm−1, νSO
The facile synthesis of a series of new aromatic oligo(ether)
bisphenol monomers with four or six phenylene rings enabled
the preparation of new ionic aromatic polymers with precisely
sequenced ionic moieties having either one or two sulfonic acid
groups along the backbone. The fully aromatic polymers thus
had a microblock structure and were prepared through con-
ventional polycondensations involving nucleophilic aromatic
substitution reactions of one sulfonated dihalide and one
bisphenol monomer. The polymers showed quite different
properties and morphologies from analogous polymer with
statistically placed sulfonic acid groups. SAXS measurements of
solvent cast membranes the Pb2+ form showed that the precise
ionomers were able to form ionic aggregates with a much more
uniform distribution than a corresponding statistical ionomer
with a similar backbone structure. Scattering profiles with very
intense scattering, narrow ionomer peaks and second order
peaks were observed for the precisely sequenced ionomers
having disulfonated moieties and six-ring spacers. This
indicated a significantly higher degree of morphological control
in relation with the statistical ionomer. No influence of the
connectivity of the oligo(ether) spacers and the nature of the
sulfonated moieties (sulfone or ketone) on the scattering
peak position and width was observed. Thus, ionomers with
precisely and sufficiently spaced ionic moieties are required
facilitate the highly uniform self-aggregation of the ionic groups.
Still, the nature of the connectivity (meta vs para) of the
oligo(ether) influenced the Tg, leading to higher values for the
ionomers having para connectivity. This in turn led to lower
water uptake and proton conductivities of the membranes
based on these ionomers. Yet, an ionomer with disulfonated
moieties perfectly alternating with a six-ring segment having
exclusively para linkages reached a proton conductivity similar
to Nafion under 100% RH at 80 °C, and had an excellent
stability in boiling water. Under reduced RH, the conductivity
of this membrane greatly exceeded that of a membrane based
on the statistical aromatic copolymer analogue with a similar
IEC, especially at low RHs. This was presumably because the
more uniform distribution of the ionic aggregates in the
membranes based on the precise ionomers facilitated more
efficient water channels for proton transport. Even though
there is increasing interest in the development of organized
3
1027 cm−1.
Membrane Casting. An amount of 0.300 g polymer was dissolved
in DMAC (3 mL). The solution was filtered and transferred to a 1 in.
Petri dish. Slow evaporation of the solvent at 80 °C under a nitrogen
flow afforded transparent membranes that were washed successively
with water. The membranes were ion-exchanged to the protonated
form by immersion in 1 N HCl for 24 h, followed by leaching with
distilled water for 2 days during which time the water was repeatedly
exchanged until a neutral pH was reached.
1
Polymer Characterization. H NMR spectra were recorded on a
Bruker 400 MHz spectrometer using CDCl3 or DMSO-d6 solutions of
the compounds. Resonances were recorded in δ (ppm) and referenced
to residual solvent resonances. Fourier-transform infrared (IR) analysis
was obtained with a Bruker FS66 spectrometer. Dry polymer samples
were ground with potassium bromide and pressed into tablets.
Inherent viscosities (ηinh) were determined at 25 °C for filtered 0.5 wt %
polymer solutions in DMF containing 0.05 M LiBr using an Ubbelohde
capillary viscometer.
Membrane Characterization. Ion exchange capacities (IECs)
were determined as follow: a membrane of known weight in its
protonated form was immersed overnight in an aqueous solution of
sodium chloride. The quantity of protons exchanged was titrated with
potassium hydroxide, using phenolphthalein as an indicator. Water
uptakes were determined by immersion of membranes in their
protonated form in water at room temperature. Differential scanning
calorimetry (DSC) was conducted on a TA Instruments Q1000 DSC
at a heating and cooling rate of 10 °C min−1. Values are reported for
the second heating scan of membranes in the Na+ form.
Thermogravimetric analysis was performed on a TA Instruments
Q500. Samples were dried at 120 °C before analysis. Thermal stability
under nitrogen atmosphere was evaluated for membranes in their Na+
form at a heating rate of 10 °C min−1. Degradation of membranes in
their protonated form was assessed at a scanning rate of 1 °C min−1.
SAXS measurements were carried out on all membranes ion-
exchanged to the Pb2+ form by immersion in a 1 wt % aqueous
solution of lead acetate. In addition, membranes 13 and 15 were ion
exchanged to their Na+ and H+ form to study the effect of the
counterion on the scattering. Furthermore, scattering experiments
were performed on membranes 13 and 15 (Pb2+ form) after annealing
at 20 °C above their respective Tg for 10 min. All scattering experi-
ments were performed on a SAXSess camera (Kratky, Anton Paar)
equipped with a CCD detector. CuKα radiation with a wavelength (λ)
of 0.1542 nm was generated by a PANalytical PW3830 X-ray generator
operating at 40 kV and 50 mA. Dry membranes were analyzed in a
1458
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459

Macromolecules
Article
(22) Nakabayashi, K.; Higashihara, T.; Ueda, M. J. Polym. Sci., Polym.
Chem. 2010, 48, 2757.
(23) Takamuku, S.; Jannasch, P. Macromol. Rapid Commun. 2011, 32,
polymer systems such as block copolymers, statistical copol-
ymers are still the class of polymer primarily being used in fuel
cells, especially by industry. This is due to the larger number of
available synthetic techniques and their ease of synthesis in
comparison with complex methods required for block and graft
copolymers. Thus, developing efficient synthetic strategies to
prepare precisely sequenced fully aromatic ionomers with
sufficiently long sequences and appropriate IEC values may be
a viable alternative for use as fuel cell membranes with con-
trolled morphology.
474.
(24) Lafitte, B.; Jannasch, P. Adv. Funct. Mater. 2007, 17, 2823.
(25) Jutemar, E. P.; Jannasch, P. J. Membr. Sci. 2010, 351, 87.
(26) Tian, S.; Meng, Y.; Hay, A. S. Macromolecules 2009, 42, 1153.
(27) Zhu, Z.; Walsby, N. M.; Colquhoun, H. M.; Thompsett, D.;
Petrucco, E. Fuel Cells 2009, 9, 305.
(28) Schuster, M.; Kreuer, K. D.; Andersen, H. T.; Maier, J.
Macromolecules 2007, 40, 598.
(29) Jannasch, P. Fuel Cells 2005, 5, 248.
(30) Jutemar, E. P.; Takamuku, S.; Jannasch, P. Macromol. Rapid
Commun. 2010, 31, 1348.
(31) Ueda, M.; Toyota, H.; Ouchi, T.; Sugiyama, J. I.; Yonetake, K.;
Masuko, T.; Teramoto, T. J. Polym. Sci., Polym. Chem. 1993, 31, 853.
(32) Wang, F.; Chen, T.; Xu, J. Macromol. Chem. Phys. 1998, 199,
1421.
(33) Wang, F.; Li, J.; Chen, T.; Xu, J. Polymer 1999, 40, 795.
(34) Buck, E.; Song, Z. J.; Tschaen, D.; Dormer, P. G.; Volante, R. P.;
Reider, P. J. Org. Lett. 2002, 4, 1623.
(35) Yeager, G. W.; Schissel, D. N. Synthesis 1991, 1, 63.
(36) Ben-Haida, A.; Colquhoun, H. M.; Hodge, P.; Lewis, D. F.
Polymer 1999, 40, 5173.
(37) Ben-Haida, A.; Colquhoun, H. M.; Hodge, P.; Williams, D. J.
J. Mater. Chem. 2000, 10, 2011.
(38) Colquhoun, H. M.; Paoloni, F. P. V.; Drea, M. G. B.; Hodge, P.
Chem. Commun. 2007, 3365.
(39) Colquhoun, H. M.; Hodge, P.; Paoloni, F. P. V.; McGrail, P. T.;
Cross, P. Macromolecules 2009, 42, 1955.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H NMR spectra of the bisphenol monomers and their
intermediates and DSC traces of the membranes in the Na+
form. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: patric.jannasch@polymat.lth.se. Fax: +46-46-2224012.
Telephone: +46-46-2229860.
■
ACKNOWLEDGMENTS
■
We gratefully acknowledge the Swedish Energy Agency for
their financial support. Dr. Xuefeng Li wishes to thank the
China Scholarship Council. We are also grateful to Shogo
Takamuku for the synthesis of the sPPSU sample.
(40) Ogata, Y.; Sawaki, Y. J. Org. Chem. 1972, 37, 2953. Syper, L.
Synthesis 1989, 3, 167.
(41) Jutemar, E. P.; Jannasch, P. J. Polym. Sci., Polym. Chem. 2011, 49,
734.
(42) Jutemar, E. P.; Jannasch, P. ACS Appl. Mater. Interfaces 2010, 2,
3718.
(43) Daoust, D.; Godard, P.; Devaux, J.; Legras, R.; Strazielle, C.
Polymer 1994, 35, 5491.
(44) Eisenberg, A.; Kim, J. S. In Introduction to Ionomers, 1st ed.;
Wiley: New York, 1998.
(45) Fujimura, M.; Hashimoto, T.; Kawai, H. Macromolecules 1981,
14, 1309.
(46) Williams, C. E.; Russel, T. P.; Jerome, R.; Horrion, J.
Macromolecules 1986, 19, 2877.
(47) Moore, R. B.; Bittencourt, D.; Gauthier, M.; Williams, C. E.;
Eisenberg, A. Macromolecules 1991, 24, 1376.
(48) Seitz, M. E.; Chan, C. D.; Opper, K. L.; Baughman, T. W.;
Wagener, K. B.; Winey, K. I. J. Am. Chem. Soc. 2010, 132, 8165.
(49) Batra, A.; Cohen, C.; Kim, H.; Winey, K. I. Macromolecules
2006, 39, 1630.
REFERENCES
■
(1) Hickner, M. A. Mater. Today 2010, 13, 24.
(2) Iojoiu, C.; Sanchez, J.-Y. High Perform. Polym. 2009, 21, 673.
(3) Zhang, J.; Xie, Z.; Zhang, J.; Tang, Y.; Song, C.; Navessin, T.; Shi,
Z.; Song, D.; Wang, H.; Wilkinson, D. P.; Liu, Z.; Holdcroft, S. J. Power
Sources 2006, 160, 872.
(4) Garland, N.; Kopasz, J. J. Power Sources 2007, 172, 94.
(5) Iojoiu, C.; Marechal, M.; Chabert, F.; Sanchez, J. Y. Fuel Cells
2005, 5, 344.
̀
(6) Jones, D. J.; Roziere, J. J. Membr. Sci. 2001, 185, 41.
(7) Kreuer, K. D. J. Membr. Sci. 2001, 185, 29.
(8) Higashihara, T.; Matsumoto, K.; Ueda, M. Polymer 2009, 50,
5341.
(9) Peckham, T. J.; Holdcroft, S. Adv. Mater. 2010, 22, 4667.
(10) Ghassemi, H.; McGrath, J. E.; Zawodzinski, T. A. Polymer 2006,
47, 4132.
(11) Roy, A.; Lee, H. S.; McGrath, J. E. Polymer 2008, 49, 5037.
(12) Lee, H. S.; Roy, A.; Lane, O.; Dunn, S.; McGrath, J. E. Polymer
2008, 49, 715.
(50) Guo, R. L.; Lane, O.; VanHouten, D.; McGrath, J. E. Ind. Eng.
Chem. Res. 2010, 49, 12125.
(13) Badami, A. S.; Lane, O.; Lee, H.-S.; Roy, A.; McGrath, J. E.
J. Membr. Sci. 2009, 333, 1.
(14) Nakabayashi, K.; Matsumoto, K.; Ueda, M. J. Polym. Sci., Polym.
Chem. 2008, 46, 3947.
(15) Mikami, T.; Miyatake, K.; Watanabe, M. J. Polym. Sci., Polym.
Chem. 2011, 49, 452.
(16) Takamuku, S.; Akizuki, K.; Abe, M.; Kanesaka, H. J. Polym. Sci.,
Polym. Chem. 2009, 47, 700.
(17) Schonberger, F.; Kerres, J. J. Polym. Sci., Polym. Chem. 2007, 45,
̈
5237.
(18) Lee, H. S.; Lane, O.; McGrath, J. E. J. Power Sources 2010, 195,
1772.
(19) Bae, B.; Yoda, T.; Miyatake, K.; Uchida, H.; Watanabe, M.
Angew. Chem., Int. Ed. 2010, 49, 317.
(20) Bae, B.; Miyatake, K.; Watanabe, M. Macromolecules 2010, 43,
2684.
(21) Nakabayashi, K.; Higashihara, T.; Ueda, M. Macromolecules
2010, 43, 5756.
1459
dx.doi.org/10.1021/ma201599p | Macromolecules 2012, 45, 1447−1459