Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02527445 2012-06-27
Ether Nitrile Co-polymers Containing Sulfonic
Acid Groups for PEM Application
BACKGROUND OF THE INVENTION
[0011 During the past several years, proton conducting polymers have attracted
much
attention due to their considerable promise for applications in some
electrochemical
devices, such as displays or sensors, and which is most important, as proton
exchange
membranes (PEM) in PEM fuel cells (PEMFC) and direct methanol fuel cells
(DMFC).
In PF-MFCs and DMFCs, PEMs serve as separators for the reactants, catalysts
support
and provide the required ionic pathway between the anode and the cathode.
Therefore,
their properties such as proton conductivity, water maintenance, permeability
for fuel and
chemical stability are crucial for the fuel cells performance. Although
perfluorosulfonic
acid ionomers such as Nafion , developed by DuPont, are considered state-of-
the art,
their high cost, difficulty in preparation, high methanol crossover and
dramatic decrease
in proton conductivity at temperatures over 80 C due to the dehydration of
membranes
limit their further applications. As a response to the commercial need for
less expensive
and more versatile polymer electrolytes, the synthesis and characterization of
new
membrane materials has become an active research area'.
[0021 High performance polymers are an important category of alternative
candidates
for PEMs. Many kinds of high performance polymers, such as poly(aryl ether
sulfone)s,
poly(aryl ether ketones, poly(ether imide)s, polybenzimidazole, poly(phenylene
oxide),
poly(phenylene sulfide), etc, are well known for their excellent thermal,
mechanical and
dielectric properties and good oxidative resistance, After modification they
show rather
high proton conductivities2 23 and become promising PEM materials. Wang and
McGrath8 for example, reported the synthesis of biphenyl-based poly(arylene
ether
sulfone)s containing sulfonic acid groups by direct polymerization reactions
of
dipotassium 3,3'-disulfonate-4,4'-
1
CA 02527445 2005-11-18
dichlorodiphenylsulfone (SDCDPS), 4,4'-dichlorodiphenylsulfone and 4,4'-
biphenol.
The proton conductivity values at 30 C of 0.11 S/cm for 40% SDCDPS copolymer
and 0.17 S/cm for the 60% SDCDPS copolymer were measured. Our group17-19 and
Xiao et al20,2 1 also reported the synthesis and conductivities of
poly(phthalazinone
ether ketone)s and poly(phthalazinone ether sulfone)s containing sulfonic acid
groups,
prepared by both, post-synthesis sulfonation reactions and by direct
polymerization
reactions. Both methods gave polymers with conductivities higher than 10-2
S/cm at
around SC 1Ø However, this category of polymers has a tendency to swell at
high
humidity and elevated temperature, especially the polymers with high sulfonic
acid
content. As a consequence the membranes lose the mechanical strength and their
ability to function under FC conditions becomes questionable. Aromatic
poly(aryl
ether nitrile)s are a new class of high performance thermoplastic polymers
that exhibit
good mechanical properties, high chemical and thermal resistance and have
already
been used as matrices in advanced composites in aerospace industries 24-38.
Aromatic
poly(aryl ether nitrile)s have been prepared by Kricheldorf, McGrath and other
researchers24-38 via nucleophilic substitution polycondensation reactions of
bisphenols
and dihalobenzonitriles or dinitrobenzonitriles in dipolar solvents. Unlike
many other
poly(aryl ether)s, poly(aryl ether nitrile)s have strongly polar nitrile
groups, pendant
on aromatic rings, which will most probably promote adhesion of the polymers
to
many substrates via interaction with other polar chemical groups. It is
believed that
for PEM applications, the enhanced adhesive ability of aromatic poly(aryl
ether
nitrile)s to inorganic compounds is beneficial for adhesion of catalyst to the
PEM.
Recently, it was reported 39, 39,40 that nitrile groups were introduced into
poly(aryl ether
sulfone)s containing sulfonic acid groups with the aim to decrease the
swelling of
membrane films via enhanced intermolecular interaction and potentially promote
adhesion of the polymers to heteropolyacids in the composite membrane or to
electrodes in order to improve the quality of membrane electrolyte assemblies
(MEA)s.
SUMMARY OF THE INVENTION
[003] According to one aspect of the invention, we provide a new class of
poly(aryl
ether ether nitrile) copolymers containing various amounts of sulfonic acid
groups
(SPAEEN)s, a method for preparing same and their application in the fuel cell
domain
2
CA 02527445 2005-11-18
as proton exchange membrane (PEM) materials.
[004] According to another aspect of the invention, we provide a new class of
poly(phthalazinone ether ketone nitrile) copolymers containing sulfonic acid
groups
(SPPEKN), a method for their preparation via nucleophilic polycondensation
reactions and their use as PEMs in fuel cells.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 illustrates FT-IR spectra of SPAEEN copolymers
Figure 2 illustrates FT-IR spectra of SPAEEN-100 and SPAEENH -100
Figure 3 illustrates 1H NMR spectrum of SPAEEN-100 in DMSO-d6;
Figure 4 illustrates 1H NMR spectra of SPAEEN copolymers in DMSO-d6; left
column: SPAEEN-Q90 (top), SPAEEN-Q70 (middle) and SPAEEN-Q50 (bottom),
right column: SPAEEN-B90 (top), SPAEEN-B70 (middle) and SPAEEN-B50
(bottom).
Figure 5 illustrates TGA traces of SPAEEN copolymers
Figure 6 illustrates DSC curves of SPAEEN copolymers in potassium sulfonate
form
Figure 7 illustrates Proton Conductivities of SPAEEN-Q copolymers
Figure 8 illustrates Proton Conductivities of SPAEEN-B copolymers
Figure 9 illustrates FT-IR spectra of SPPEKN copolymers in sodium form
Figure 10 illustrates 1H NMR stacked spectra of PPEKN (top), SPPEK(bottom) and
SPPEKN with SC of 30,45 and 60%
Figure 11 illustrates TGA traces of copolymers
Figure 12 illustrates Proton conductivities measured longitudinally of SPPEKNH
copolymers compared with Nafion 117
Figure 13 illustrates Proton conductivities measured transversely of SPPEKH
copolymers compared with Nafion 117
Figure 14-16 illustrates results of water uptake, swelling and proton
conductivity of
the various SPAEENs.
Figure 17 illustrates FT-IR spectra of SPAEEN copolymers
Figure 18 illustrates `H NMR spectrum of m-SPAEEN in DMSO-d6.
Figure 19 illustrates TGA traces of m-SPAEEN copolymers in nitrogen and air
3
CA 02527445 2005-11-18
Figure 20 illustrates water uptake (a) and swelling (b) of m-SPAEENH
copolymers
Figure 21 illustrates proton conductivities of m-SPAEENH copolymers after
hydration in hot water for 36h
Figure 22 illustrates SC dependence of proton conductivity of m-SPAEENH
copolymers
Figure 23 illustrates temperature dependence of proton conductivity of m-
SPAEENH
copolymers with or without hot water treatment
Figure 24 illustratesFT-IR spectra of SPAEEN copolymers
Figure 25 illustrates 'H NMR spectrum of SPAEEN in DMSO-d6
Figure 26 illustrates TGA traces of SPAEEN copolymers in air
Figure 27 illustrates DSC curves of P-SPAEEN copolymers
Figure 28 illustrates water uptake and swelling of SPAEENH copolymers
Figure 29 illustrates molecular models of SPAEENH copolymers obtained from
ACD/ChemSketch
Figure 30 illustrates proton conductivity of SPAEENH copolymers
DETAILED DESCRIPTION OF THE INVENTION
Experimental Part (SPAEEN)s
Materials.
[005] SHQ was recrystallized from water and ethanol. NMP was vacuum distilled
and 4,4'-biphenol was purified by sublimation before use. All other chemicals
(obtained from Aldrich) were reagent grade and used as received.
Copolymerization.
[006] Synthesis of the polymers by nucleophilic substitution reactions was
based on
the procedure reported by McGrath25. In a typical reaction, 10.1 mmol 2,6-
DFBN, 7
mmol SHQ, 3 mmol 4,4'-biphenol, and 15 mmol K2C03 were added into a three-neck
equipped flask with a magnetic stirrer, a Dean-Stark trap, and an argon gas
inlet.
Then, 13 mL NMP and 15 mL chlorobenzene were charged into the reaction flask
under an argon atmosphere. The reaction mixture was heated to 130 C. After
dehydration and removal of chlorobenzene, the reaction temperature was
increased to
about 160 C. When the solution viscosity had apparently increased, the mixture
was
cooled to 100 C and coagulated into a large excess of ethanol or water with
vigorous
stirring. The resulting polymer was designated SPAEEN-B70, where B denotes
that
4
CA 02527445 2005-11-18
the comonomer is 4,4'-biphenol; while Q denotes hydroquinone as comonomer. n
(70) refers to the SHQ content of aromatic phenol monomers. After washing with
ethanol twice, SPAEENs were washed with water to remove salt. SPAEENs with
high
SC values, that either swell or dissolve in water, were purified by dialysis
for a week
to remove salt, using a membrane-cellulose dialysis tube (Serva
Electrophoresis,
Germany) with a molecular weight cut off value of 3500.
Copolymer Analysis and Measurement.
[007] 'H-NMR spectra were obtained on a Varian Unity Inova NMR spectrometer
operating at a proton frequency of 399.95 MHz. Deuterated dimethylsulfoxide
(DMSO-d6) was the NMR solvent and the DMSO signal at 2.50 ppm was used as the
chemical shift reference. IR spectra were measured on a Nicolet 520 Fourier
transform spectrometer with membrane film samples in a diamond cell.
[008] A TA Instruments thermogravimetric analyser (TGA) instrument model 2950
was used for measuring Td. Polymer samples for TGA analysis were preheated to
150 C at 10 C/min under nitrogen atmosphere and held isothermally for 40 min
for
moisture removal. Samples were then heated from 90 C to 750 C at 10 C/min for
Td
measurement. A TA Instruments differential scanning calorimeter (DSC) model
2920
calibrated with Tin at 231.93 C and Zinc at 419.53 C was used for measuring
Tg.
Samples in potassium form for DSC analysis were initially heated rapidly at a
rate of
20 C/min under nitrogen atmosphere to 20 C higher than their Tg, followed by
quenching in liquid nitrogen. When the DSC cell had cooled to around 50 C, the
samples were replaced in the cell and heated at a rate of 10 C/min to 400 C.
The
procedure for samples in acid form was similar except that the initial heating
rate was
C/min and the end point was below the polymer Td point.
[009] Intrinsic viscosities were determined using an Ubbelohde viscometer for
N,N-
dimethylacetamide (DMAc) solutions of copolymer at 25 C.
Preparation of Membrane Films.
[0010] An amount of 0.7 to 0.8 g copolymer in the potassium salt form was
dissolved
in 20 mL of DMAc and filtered. The filtered solution was poured onto a glass
plate
and dried at about 40 C under a constant purge of nitrogen for about one day.
The
5
CA 02527445 2005-11-18
acid form (SPAEENH-B or SPAEENH-Q) membrane films were obtained by
immersing corresponding potassium form SPAEEN-B or SPAEEN-Q membrane
films in 2 N H2SO4 for 24 h at room temperature, and then in deionized water
for
another 24 h during which water was changed several times. The thickness of
all
membrane films was in the range of 40 to 70 m.
Water Uptake Content Measurement and Swelling Ratio
[0011] The membrane films were dried at 100 C overnight prior to the
measurements.
After measuring the lengths and weights of dry membranes, the sample films
were
soaked in deionized water for 24 h at predetermined temperatures. Before
measuring
the lengths and weights of hydrated membranes, the water was removed from the
membrane surface by blotting with a paper towel. The water uptake content was
calculated by
Uptake content (%) = per - Miry X100%
Wiry
Where Ojdry and aver are the masses of dried and wet samples respectively. The
swelling ratio was calculated from films 5-10 cm long by:
Swelling ratio (%)= lwet - ldr,, X100%
ldry
Where ld and lwet are the lengths of dry and wet samples respectively.
Tensile test
[0012] Tensile tests were performed on an Instron tensile tester (model 1123)
at a
strain speed of 50 mm/min at room temperature. Membrane films with typical
size of
40 mm x 4mmx 0.05 mm were used for testing.
Proton Conductivity
[0013] The proton conductivity measurements were performed on SPAEENH-B or
SPAEENH-Q membrane films by AC impedance spectroscopy over a frequency
range of 1-107 Hz with oscillating voltage 50-500 mV, using a system based on
a
Solatron 1260 gain phase analyzer. A 20 x10 mm membrane sample was placed in a
temperature controlled cell open to the air by a pinhole where the sample was
equilibrated at 100%RH at ambient atmospheric pressure and clamped between two
stainless steel electrodes. Specimens were soaked in deionized water for 24 to
48 h
prior to the test. The conductivity (a) of the samples in the longitudinal
direction was
6
CA 02527445 2005-11-18
calculated from the impedance data, using the relationship a = d/RS where d
and S are
the thickness and face area of the sample respectively and R was derived from
the low
intersect of the high frequency semi-circle on a complex impedance plane with
the Re
(Z) axis. The impedance data were corrected for the contribution from empty
and
short circuited cell.
Results and Discussion
Synthesis and Characterization of SPAEENs
[00141 High performance polymers containing sulfonic acid groups are typically
prepared either by post-sulfonation reaction or direct polymerization reaction
of
sulfonated monomers. Direct polymerization is susceptible to possible side
reactions
such as degradation and cross-linking, that could occur in strongly acidic
media
usually used for post-sulfonation. As shown in Scheme 1, SPAEENs were prepared
via the nucleophilic polycondensation reactions of SHQ, 2,6-DFBN and the third
monomer, 4,4'-biphenol or hydroquinone.
Scheme 1 Synthesis of SPAEEN copolymers
CN
HO OH + F F + HO-Ar-OH
S03X
K2CO3, NMP
CN CN
O O O-Ar-O
S03X
Ar = or C~__Q
(Q) (B)
wherein X=K, Na or H
7
CA 02527445 2005-11-18
[0015] Since the copolymers were formed by reacting a combined amount of one
mole of sulfonated diol (SHQ) and either 4,4'-biphenol (B) or hydroquinone (Q)
with
one mole of 2,6-DFBN, the SC is expressed as the molar ratio of SHQ units
(bearing
the -SO3Na group) to 1.0 molar 2,6-DFBN unit. For example, the average repeat
unit
of SPAEEN-Q70 is composed of 0.7 unit of SHQ, 0.3 unit of hydroquinone (Q) and
1.0 unit of 2,6-DFBN. Expressed in this way, both the number of -SO3Na groups
per
polymer repeat unit and the ratio of diol monomers (SC:1-SC) can be
conveniently
derived. Equivalent molecular weight (Meq) and ionic exchange capability (IEC)
were also calculated theoretically and listed in Table 1 for comparison.
8
CA 02527445 2005-11-18
a)
a)
7
Y u
z o
O O O O O
QJ
a E
d) O
U '
O
C
C CD O o 0 C C O CD 0
C
O
U
N m C d O m m
O N N O ' M O
cl r N Z:- n
O~ .--~ N a\ V M N O~ OC 00
oo N ~O N N d' N N 0\
N M M 'd CI' M M M d'
O
U
7
U
01) 00 N 00 Q\ V'1 M V') M C 00 00
N v) v) D\ t o0 O N M o0
.b M M =--. .-. .--. N N ^-~ N N
~z~qq
fs o 0 0 0 0 0 o O o 0
C
O
a)
t7a ~,
O o I N M L --N M d- Ln
O
c, O
z
U
C/1 a O O Q
C C 00 N C Vl C 00 N ' v)
c E
U
N
(Z) CD 0 CD 0 CD (D (2) 0 C)
2 Eq cQ \0 Lr) 00 ci~ W 0
V] ~C Ol Ol Ol Ol CIS
z z z z z z z z z z z
~ O W W W W W W W W W W (~
a a a a a a a a w a s
CA 02527445 2005-11-18
[0016] For SPAEENs with lower SC values, the polymerization reaction proceeded
homogenously. However, for SPAEENs with higher SC values, SPAEENs
precipitated to the bottom of flask at the end of polymerization reactions.
Table 1 lists
the resulting polymers and details of the polymerization conditions. For the
purpose
of obtaining polymers with high proton conductivities (preferably higher than
10-2
S/cm), only SPAEENs with high SC values were synthesized. SPAEENs with SC
values from 0.5 to 1.0 were obtained by changing the feed ratio of SHQ to
unsulfonated monomer 4,4'-biphenol or hydroquinone. Both the homopolymer and
copolymers exhibit intrinsic viscosities higher than 1.6 in DMAc at 25 C
indicating
the high molecular weights of resulting polymers. Although much higher
molecular
weight polymers could be obtained by lengthening the reaction time,
polymerization
reactions were stopped when obvious increases in the viscosity of reaction
solutions
were observed. Much longer reaction times resulted in much more viscous
polymer
solutions containing some gel. Such products were difficult to redissolve
completely,
which makes subsequent membrane film casting more complicated.
[0017] All the polymer series were transformed by solution casting into strong
transparent and tough membrane films, which is usually characteristic of the
polymers
with high molecular weights. All obtained SPAEENs had good solubility in
aprotic
solvents such as NMP, DMAc, and dimethylsulfoxide (DMSO).
[0018] Sulfonic acids or sulfonates are considered to be leaving groups that
have a
tendency to dissociate from their parent structure during high temperature
reactions,
as has been previously observed41. FT-IR is a convenient method to analyze the
structures of polymers containing sulfonic or sulfonate groups. It was used in
this
work to verify if partial or complete loss of the sulfonate groups occurs
during
polymerization reactions,. Representative FT-IR spectra of SPAEEN in potassium
form are showed in Figure 1. In the spectra of both series of SPAEENs,
characteristic
bands of the aromatic sulfonate salt symmetric and asymmetric stretching
vibrations
were observed at 1032 and 1090 cm-1 . In both series of SPAEENs the intensity
of two
characteristic absorption bands were observed to increase with SC, which
confirm
successful introduction of sulfonate groups into polymers. The characteristic
symmetric stretching band of nitrile groups was observed at 2245 cm 1. The
absorption bands at 1197 and 1244 cm 1, assigned to phenoxy groups, are
overlapped
CA 02527445 2005-11-18
at high SC. The absorption bands at 1458 and 1498 cm-1 were assigned to phenyl
ring
and a band around 1600 cm-1 is attributed to C=C stretching. The FT-IR
spectrum of
SPAEENH- 100 is shown in Figure 2 together reason with that of potassium form
of
SPAEEN-100 for comparative purpose. Figure 2 shows that the vibration
absorptions
of phenoxy groups at 1197 and 1244 cm 1 of SPAEEN are separate in acid form
compared with potassium form. None of the samples exhibited a decrease in the
intensity of the band at 2245 cm 1, which is a characteristic symmetric
stretching
vibration of nitrile groups. In the FT-IR spectrum of SPAEENH-100, an
ambiguous
absorption appears around 1700 cm-1. This absorption region is specific for
stretching
vibrations of carbonyl or carboxyl groups. The observed broad band is however
believed not to be due to absorption of carboxyl groups since they are
sensitive groups
that show sharp absorption bands between 1650 and 1670 cm 1. Thus, immersion
of
the SPAEEN membrane films in 2N H2SO4 followed by immersion in deionized
water at room temperature did not provoke a noticeable hydrolysis of nitrile
groups to
carboxyl groups.
[0019] The structural properties of the synthesized polymers were also studied
by
liquid phase 1H NMR spectroscopy with DMSO-d6 as the solvent. Figure 3 shows a
spectrum of the aromatic protons for the highly sulfonated homopolymer SPAEEN-
100 in potassium form prepared from 2,6-DFBN and sulfonated hydroquinone at
1:1
monomer ratio. Also illustrated in Fig. 3 are the chemical structure of the
polymer and
the expected distribution of repeat unit configurations arising from the
structural
asymmetry of the sulfonated hydroquinone monomer. The asymmetric monomer can
be introduced in the chain in two different ways where the -SO3K group will
either be
adjacent or opposite to the benzonitrile phenyl ring. Therefore, every polymer
repeat
unit will have one of the three possible configurations whereby the -SO3K
groups can
be symmetrically arranged (forms 1 and 3) or asymmetrically arranged (form 2,
statistically predominant) about the benzonitrile phenyl ring. The evidence of
this
distribution of three types of repeat units was seen in the 1H NMR spectrum of
Fig. 3,
which can be divided into two regions: one of the higher frequencies (7.15-
7.70 ppm,
4.OOH) and the other of the lower frequencies (6.30-6.85 ppm, 2.OOH). The four
low
frequency signals on the right end of the spectrum were found to be the
benzonitrile
phenyl ring protons Hd and Hf which were spin-coupled (3JH_H, 8 Hz) with the
He
triplets (7.64, 7.50, 7.39 ppm, 8 Hz) on the left side of the spectrum. The
sulfonated
11
CA 02527445 2005-11-18
hydroquinone phenyl ring protons Ha,b,c were found at higher frequency due to
the
effect of the electron withdrawing sulfonic acid salt group. A 2D-ROESY
spectrum
was used to show that Ha was close in space to Hd and Hf for the
configurations with
the -SO3K groups opposite the benzonitrile phenyl ring. Therefore, unambiguous
assignment was achieved for the Hd,e,f signals of all three possible
configurations
shown in Fig. 3. The ROESY spectrum showed that the signal at 7.58 ppm (Ha)
was
close in space to two signals at 6.61 and 6.76 ppm which could only be from Hf
of
form 2 and Hd,f of form 1 respectively. As expected, the signals Hd,e,f from
form 2 had
higher intensities than those of forms 1 or 3 due to the statistical
predominance of
form 2; the distribution of forms 1, 2 and 3 was found to be 1:3:1. Electron
shielding
from the adjacent electron rich -SO3K groups is responsible for the shift of
Hd (form
2) and Hdf (form 3) towards lower frequencies.
[0020] The complete analysis of the NMR spectrum from the homopolymer
SPAEEN-100 greatly simplified the interpretation of the more complicated
copolymer
spectra. Figure 4 shows stacked spectra of three SPAEEN-Qs (left column) and
three
SPAEEN-B copolymers (right column). The spectra clearly show the gradual
decrease
of the far right signal (Hd,f of form 3) for both SPAEEN-Q and SPAEEN-B
polymer
derivatives with decreased SCs. That signal is expected to decrease
statistically as it
originates exclusively from the symmetric Hd,f which are shielded by the
adjacent -
SO3K groups of form 3. On the other hand, the less shielded Hd and Hf signals
around
6.75 ppm grow in intensities as the content of sulfonated monomer is replaced
by
either hydroquinone (SPAEEN-Q) or biphenol (SPAEEN-B) monomers. Similarly,
the strong Ha signal (7.58 ppm) originating from the hydrogen at the ortho -
SO3K
position decreases in intensity as the sulfonated monomers are being replaced
by non-
sulfonated monomers. The trend and regularity of intensity variation for all
of the
previously described signals is obvious. As before"' 23, these 'H NMR spectra
were
used to estimate the SCs by comparison of the intensities of specific signals.
The
advantage of 'H NMR over elemental analysis resides in the fact that residual
solvents
or moisture in the polymers do not appear in the aromatic region of the
spectra hence
have no detrimental effect on SC calculations. The equations were conveniently
derived because the spectra are divided in two distinct regions, Si (7.0-8.0
ppm) and
S2 (6.2-7.0 ppm), for both the copolymer derivatives. The integral (signal
intensity)
values of Si and S2 were used in the calculation of the SCs as follows:
12
CA 02527445 2005-11-18
SPAEEN-Q Sl = (5-n) SPAEEN-B Sl = (9-5n)
S2 (2) S2 (2)
where:
Si of SPAEEN-Q (7.0-8.0 ppm) = n x Hahc + He + (1-n) x HQ = 3n + 1 + (1-n)x4
Si of SPAEEN-B (7.0-8.0 ppm) = n x Hahc + He + (1-n) x HB = 3n + 1 + (1-n)x8
S2 of SPAEEN-Q and S2 of SPAEEN-B (6.2-7.0 ppm) = Hdf = 2
n = number of SHQ groups = SC (maximum=1.00)
[0021] The experimental SC for the SPAEEN-Bs copolymers were found to be
within
0.02 of the calculated SC (listed in Table 1). Unfortunately, the differences
between
experimental and calculated SC values for SPAEEN-Q series were larger,
possibly
due to the presence of smaller signals which have a significant effect on the
integration values. These signals may arise from chain-end groups or from
different
conformations of SPAEEN-Q polymer chains with more restricted chain movement.
Their chemical shifts would be different from the main chain proton signals
and
therefore, for the SC calculation to be accurate, they must all be accounted
for and
their intensity values assigned to the proper integral region S1 or S2. As
these signals
are small, difficult to identify and overlap with other major signals, this is
difficult to
accomplish. The deviation between experimentally derived NMR values and
calculated SC values may be a result of distinctive structural properties for
this
polymer in comparison with SPAEEN-Bs. The experimental NMR results for the SCs
of SPAEEN-Q copolymers are not reported. However, based on the regularity of
intensity variations of aromatic signals, observed and described above, it is
believed
that the SCs are close to the expected values, derived from the reaction feed
ratios.
Thermal properties of SPAEEN
[0022] Thermal stabilities of the SPAEEN copolymers in both potassium and acid
forms were investigated by TGA analysis. Ted measurements were conducted from
90 C to 750 C at a heating rate of 10 C/min and the results are listed in
Table 2.
13
CA 02527445 2005-11-18
Table 2 Thermal properties of SPAEEN copolymers
Polymer Tg ( C) Tds% ( C)
extrapo
Potassium form Acid form Potassium form Acid form Potas
SPAEEN-100 371 ND 451 325
SPAEEN-B90 365 ND 445 337
SPAEEN-B80 353 ND 442 340
SPAEEN-B70 339 ND 454 334
SPAEEN-B60 350 ND 428 328
SPAEEN-B50 345 ND 426 318
SPAEEN-Q90 357 ND 453 337
SPAEEN-Q80 350 ND 446 330
SPAEEN-Q70 333 ND 432 330
SPAEEN-Q60 316 ND 443 342
SPAEEN-Q50 308 ND 434 339
*ND: Not detected
14
CA 02527445 2005-11-18
[0023] Table 2 shows that T,15% and onset weight loss temperatures (T1) of
SPAEEN
copolymers in potassium form are observed between 432 to 453 C, indicating
good
thermal stabilities. In contrast with potassium forms, Tc15%s and T1s of
copolymers in
acid form are observed between 318 to 342 C, more than 100 C lower than
potassium
form. A comparison of potassium and acid forms is also shown in Figure 5. The
TGA
curves are similar to those of poly(aryl ether ketone)s and poly(aryl ether
sulfone)s
containing sulfonic acid groups17-19' 22 prepared before by our group.
Polymers in
potassium form exhibit only one weight-loss step in their TGA curves and in
acid
form exhibit a much earlier decomposition and two distinct transition steps.
[0024] Tgs of copolymers in both potassium and acid forms reported here were
mainly
obtained from the second scan except SPAEEN-100, for which Tg was determined
in
the first run. DSC curves of SPAEEN in potassium form are plotted in Figure 6.
From
our previous studies and from the literatureg'1719, it is known that glass
transition is
less pronounced in the DSC curves of poly(aryl ether ketone)s or poly(aryl
ether
sulfone)s containing sulfonate groups, compared to their parent polymers.
However,
SPAEEN copolymers, as can be seen from Figure 6, exhibited distinctive glass
transitions in their DSC curves. All samples were amorphous and only a single
Tg is
discernable on each curve. The obtained Tgs are listed in Table 2. It is seen
that the
Tgs of copolymers in potassium form increase with SC values varying from 308
to
371 C for SPAEEN-Qs and from 339 to 371 for SPAEEN-Bs. Compared with the
unsulfonated nitrile copolymer (PEEN), which shows a Tg value at 144 C, all
the
prepared SPAEEN copolymers in potassium form show more than a two-fold
increase
in Tg. It was discussed previously18 that the increase in Tgs is mainly the
result of
introducing of sulfonate groups, which increases the intermolecular
interactions by
pendant ions, enhances molecular bulkiness, and raises the rotation activation
energy
of polymer molecular segment. This can be expressed as TgEcq/a, where c is the
concentration of ionic repeat unit in backbone, q is the cation charge, and a
is the
distance of closest approach between the centers of charge of the anion and
cation42.
Furthermore, cluster formation due to the separation of hydrophilic and
hydrophobic
domains also contributed to the increase in Tg21. For SPAEEN Q copolymers Tgs
continually goes down when SC decreases. However, for SPAEEN-B copolymers, the
Tg dependence on the SC is not the same. From SPAEEN-100 to SPAEEN-B50, Tgs
decrease with decreasing SC values initially, and then begin to increase at a
certain
CA 02527445 2005-11-18
SC value. To explain this, it should be taken into consideration that a
decrease in the
content of sulfonate groups is achieved by decrease in the feed ratio of SHQ
to
biphenol during synthesis. However, the increase in polymer rigidity caused by
replacement of short benzene ring by long rigid biphenyl between two ether
linkages,
which results in an increase in Tgs of SPAEEN-B copolymers. Therefore, when
the
content of biphenyl structures in polymer chain reaches a high enough value,
the
change in Tgs depends not only on SC values, but is also influenced by
backbone
structure. The weak transitions occurring between 100 to 200 C in the DSC
curves of
SPAEEN with high SC values were caused by evaporation of residual water,
strongly
bound by the copolymers, which is difficult to remove completely. Glass
transitions
for SPAEEN in acid form were not observed.
Water uptake and swelling ratio
[0025] The proton conductivity and mechanical stability of PEMs are strongly
related
to the presence of water. In sulfonated poly(aryl ether ketone) or sulfonated
poly(aryl
ether sulfone)8' 43, hydrophilic sulfonic acid clusters are distributed in
continuous
hydrophobic domains. These domains swell with imbibed water and are inter-
connected to form continuous ionic pathways. Water uptake and swelling ratio
of
SPAEEN as determined by measuring the changes in weight and length are listed
Table 3.
16
CA 02527445 2005-11-18
Table 3 Water uptake and swelling ratio of SPAEEN copolymers
Room temperature 80 C
Polymer Water uptake (%) Swelling ratio (%) Water uptake (%)
Salt form Acid form Salt form Acid form Salt form Acid form
SPAEEN-100 190 Swelled 51 Swelled D D
SPAEEN-B90 61 250 14 61 PD D
SPAEEN-B80 32 81 11 28 PD D
SPAEEN-B70 19 51 3.9 17 190 SW
SPAEEN-B60 16 22 0.61 9.1 29 58
SPAEEN-B50 7.7 19 0.50 6.9 13 36
SPAEEN-Q90 76 520 23 110 PD D
SPAEEN-Q80 42 160 16 52 880 SW
SPAEEN-Q70 25 90 10 30 690 S
SPAEEN-Q60 22 50 2.8 16 140 410
SPAEEN-Q50 15 31 2.2 9.4 34 96
*D: dissolved; PD: partially dissolved; SW: swelled
17
CA 02527445 2005-11-18
[0026] The water uptake and swelling ratio increase with SC or IEC values and
temperature. At room temperature, membrane films, both in salt and acid forms,
show
gradual increases in water uptake up to a certain SC value then increase
sharply. At
80 C, SPAEEN copolymers with SC values lower than 0.7 for SPAEEN-Bs and 0.6
for SPAEEN-Qs show moderate water uptake and swelling. SPAEENH copolymers
with higher SC values swelled too much or completely dissolved in hot water.
Membrane films in acid form have higher water uptake and swelling ratio values
than
in salt form due to hydrogen bond interactions between H2O and sulfonic acid
groups.
Compared with Nafion 117 (IEC = 0.91 mmol/g) membrane, which shows a water
uptake of 35% at room temperature44, SPAEENH copolymers absorbing similar
proportion of water have much higher IEC values. In other words, SPAEENH with
the same sulfonic acid content have lower water uptake values than Nafiono
117. The
copolymers imbibe less water than Nafion 117 at room temperature since the
aromatic chain of SPAEENHs is more rigid than that of Nafion 117 and the
sulfonic
acid groups have lower acidity. In addition, the strong ionic interaction
between
sulfonic acid groups increases rigidity of network structure. A combination of
these
two effects results in the restriction of free volume for water adsorption and
a
decrease in the water uptake of SPAEENH copolymers. At elevated temperatures,
the
polymer chain mobility and the free volume for water adsorption increase. As a
result,
the rigid network structure of the membrane is weakened or even destroyed for
copolymers with high SC, since in hot water the ionic interactions between
macromolecules is gradually replaced by hydrogen-bonding between H2O and
sulfonic acid groups.
[0027] It should be also mentioned that unlike sulfonated poly(phthalazinone
ether
sulfone) previously prepared in our group'8 or other sulfonated poly(aryl
ether)s,
which showed some brittleness in the dry state at high SC, all SPAEEN
copolymers
even up to SC1.0 (IEC 3.46) showed good film-forming properties and yielded
membranes that were tough and flexible. This could be the result of the good
proportion of the flexible ether linkage in polymer backbone and the rigid
polar
chemical groups as a side substitute instead of in the polymer backbone.
18
CA 02527445 2005-11-18
Tensile properties
[0028] Tensile properties of SPAEENH copolymers were tested at room
temperature
during two days and the results are tabulated in Table 4.
Table 4 Tensile properties of SPAEENH copolymers
Polymer Tensile Elongation at break Tensile strength at break
strength (%) (MPa)
(MPa)
SPAEEN-B90 67 63 61
SPAEEN-B80 64 65 71
SPAEEN-B70 77 70 81
SPAEEN-B60 81 45 74
SPAEEN-B50 82 16 72
SPAEEN-Q90 51 39 71
SPAEEN-Q80 66 37 71
SPAEEN-Q70 73 24 67
SPAEEN-Q60 80 21 73
SPAEEN-Q50 75 25 71
Nafion 117 10 623 15
[0029] In general, all SPAEENH copolymers exhibited good tensile strengths
ranging
from 51 MPa to 82 MPa, which decreased with increasing SC values. The
elongations
at break ranged from 16% to 70%, increasing with the SC values. Since SPAEENH-
100 swelled excessively during the process for conversion from salt to acid
form and
then wrinkled when dry, the film dimension was difficult to measure
accurately. Thus,
tensile properties of SPAEENH-100 were not reported. However, its tensile
curve
also exhibited the same trend. For comparison, Nafionl17 was also tested for
tensile
properties under the same conditions and the results are also listed in Table
4. All
19
CA 02527445 2005-11-18
SPAEENH copolymers exhibited tensile strength values several-fold higher and
less
elongation at break compared with Nafionl17. In addition, all SPAEENH
copolymers
showed yield behavior, while Nafion117 exhibited a continuous increase in
tensile
strength before break. In other words, Nafion117 exhibited tensile behavior
between
elastomer and thermoplastic whereas the SPAEENH copolymer exhibited
thermoplastic behavior.
Proton Conductivity
[0030] In our previous study23, X-ray fluorescence spectroscopy confirmed that
all
the sodium sites were effectively converted into sulfonic acid use the method
described in experimental part. Proton conductivities of the acid form of
SPAEEN-Q
and SPAEEN-B copolymers were measured at 100% relative humidity as a function
of SC and temperature and are shown in Figures 7 and 8 respectively. For
comparison
the proton conductivity of Nafionl17 measured under the same experimental
conditions is also shown in the Figures.
[0031] All SPAEEN membrane films showed room temperature proton conductivities
higher than 10-2 S/cm, which makes them placed among the promising PEMs as
representing a practical interest for use in fuel cells. As expected, membrane
proton
conductivity increases with sulfonic acid groups' content. The membranes with
lower
IEC values, SPAEEN-B50, SPAEEN-B60, SPAEEN-Q50, and SPAEEN-Q60 showed
room temperature proton conductivities comparable to Nafion117. The films with
higher IEC values, SPAEEN-B70, SPAEEN-B80, SPAEEN-B90, SPAEEN-Q70 and
SPAEEN-Q80 showed room temperature proton conductivities higher than
Nafion117, in some cases even exceeding 0.1 S/cm. SPAEEN-Q90 showed
unexpectedly lower proton conductivity than SPEEN-Q70 and SPAEEN-Q80,
although its proton conductivities was still higher than 10-2 S/cm. This
apparent
inconsistency is attributed to its higher water uptake and swelling ratio.
From Table 3,
it can be seen that SPAEEN-Q90 exhibited high water absorption and a more
substantial dimensional change than other samples. This large dimensional
change
resulted in a large decrease in the SC content per unit of volume of wet
membrane. In
other words, although it has a high SC value, the sulfonic acid groups in the
excessively swollen membrane are highly diluted, resulting in a decrease in
its proton
conductivity. Figure 7 also illustrates the fact that a higher proton carrier
CA 02527445 2005-11-18
concentration in dry membranes does not necessarily ensure a higher proton
conductivity in humidified material. Thus, high proton conductivities of PEMs
cannot
be pursued solely by increasing the IEC values. SPAEEN-100 swelled excessively
even at room temperature; its proton conductivity could not be measured.
However,
since the entire SPAEEN copolymer series up to SPAEEN-100 showed good
membrane-forming properties and could be cast into tough and flexible membrane
films in the dry state, there may be application for the materials having high
SC in a
non-aqueous environment which would not cause excessive swelling.
[0032] In general, proton conductivities increased with temperature and the
SPAEEN-
Q series showed higher proton conductivities than the SPAEEN-B series at the
same
SC values, which may be explained by the difference in their equivalent
molecular
weights. Compared with Nafion117, SPAEEN copolymers exhibited a more sensitive
change in proton conductivities with temperature which suggests that SPAEEN
copolymers have higher activation energy for proton conductivity that
Nafion117.
Summary
[0033] Wholly aromatic sulfonated poly(aryl ether ether nitrile)s (SPAEEN)s
were
prepared via K2CO3 mediated direct polymerizations of commercially available
monomers: 2,6-difluorobenzonitrile (2,6-DFBN), potassium 2,5-
dihydroxybenenesulfonate (SHQ), or 2,8-DHNS-6 (see below) and a third monomer
4,4'-biphenol or hydroquinone, in NMP. The sulfonic acid group content (SC) in
the
copolymers was controlled by varying the ratio of the sulfonated diol monomer
to
either biphenol or hydroquinone diol monomers.
[0034] The sulfonic acid group content (SC), expressed as a number per repeat
unit of
polymer, ranged from 0.5 to 1.0 and was obtained by changing the feed ratio of
SHQ
to the unsulfonated bisphenol. Membrane films in potassium salt and acid forms
were
obtained by casting N,N-dimethylacetamide (DMAc) solution of SPAEENs, followed
by immersing in 2 N sulfuric acid at room temperature. FT-IR confirmed the
structure
of polymer in both salt and acid forms. NMR was used to determine the obtained
SC
values of SPAEENs. Decomposition temperatures (T~ls) of SPAEENs were around
21
CA 02527445 2005-11-18
300 C for acid form and over 400 C for potassium form. Water uptake and
swelling
ratio values increased with SC and temperature. All SPAEENH copolymers were
mechanically stronger than Nafionl 17 and exhibited a reasonable flexibility.
The
proton conductivities of acid form membrane at different SC values were close
to or
higher than that of Nafionl 17, and reached 10-1 S/cm. The best compromise on
PEM
mechanical strength, water swelling and proton conductivity, was achieved at
SC
ranged from 0.5 to 0.7.
[0035] The resulting copolymers had high Tgs ranging from 308 to 371 C in
potassium salt form and were thermally stable up to at least 289 C in acid
form.
SPAEEN copolymers were organic soluble and could be cast into membrane films,
that were tough and flexible in the dry state, even at high IEC values. The
SPAEEN
membranes with high IEC are superior to sulfonated (aryl ether ketone)s and
sulfonated (aryl ether sulfone)s, which are often brittle at high IEC values
when dry.
SPAEENH copolymers were readily prepared from inexpensive commercially
available chemicals via one pot reactions. All SPAEENH copolymers exhibited
reasonable flexibility and high tensile strength. Both potassium and acid form
sulfonated membrane films show continuous increases in water uptake and
swelling
ratio with SC and temperature, and the acid form membrane films show higher
and
more rapid increases than those in the potassium form. SPAEEN copolymers
obtained
from high sulfonated diol monomer ratios swelled excessively or dissolved in
water.
The copolymers showed a similar or slightly higher room temperature proton
conductivities compared with Nafion 117. Nitrile groups are also anticipated
to
promote adhesion of the polymers to catalyst or to carbon black in the
membrane
electrolyte assemblies (MEA) and might be beneficial for preparation of the
composite membrane, increasing bonding to embedded second phase37. This
combination of desirable properties makes SPAEENH copolymers potentially good
candidate for proton conducting membrane materials for fuel cells
applications.
[0036] Some examples of other sulfonated diphenol monomers that can be used in
Scheme I. in the preparation of novel SPAEEN co-polymers instead of just the
sulfonated hydroquinone Na, and K substituents are interchangeable, are shown
below
22
CA 02527445 2005-11-18
Q OH
HO OH NaSO3 OH
SO3K
SO3Na
NaSO So3Na NaSO SO3Na
/ I \
HO OH
OH OH
CH3
HO OH HO OH
CH3
NaSO3 SO3Na NaSO3 SO3Na
CF3
HO OH HO O
CF3 N- N
NaSO3 SO3Na NaSO3 H
OH
OH HO OH
/ \ -
NaSO3 \ I / \ /
SO3Na
NaSO3 SO3Na
HO OH
HO OH NaSO3 SO3Na
[0037) The following is additional basic information regarding data for PEMs
prepared using yet another monomer in the preparation of novel SPAEEN co-
23
CA 02527445 2005-11-18
polymers according to Scheme 1. The data for the -50 and -60 polymer show very
good conductivity and low swelling.
OH CN
F,11 F -
H O H
NaO3S
K2C03, DMSO
0 CN CN
NaO3S
Experimental Part
[0038] SPAEEN's containing naphthalene structure with sulfuric acid groups
meta to
ether linkage.
Materials.
[0039] 2,8-dihydroxynaphthalene-6-sulfonate sodium salt (2,8-DHNS-6) was
purchased from Rintech, Inc. and recrystallized from deionized water. DMSO and
4,4'-biphenol were purchased from Aldrich and was vacuum distilled and
purified by
sublimation respectively before usage. All other chemicals (obtained from
Aldrich)
were reagent grade and used as received.
Copolymerization.
[0040] Synthesis of the polymers by nucleophilic substitution reactions was
based on
the procedure reported by McGrath25. In a typical reaction, 10 mmol 2,6-DFBN,
5
mmol 2,8-DHNS-6(2,8-dihydroxynaphthalene-sulfonate sodium salt), 5 mmol 4,4'-
biphenol, and 15 mmol K2CO3 were added into a three-neck equipped flask with a
magnetic stirrer, a Dean-Stark trap, and an argon gas inlet. Then, 10 mL DMSO
and
mL chlorobenzene were charged into the reaction flask under an argon
atmosphere.
The reaction mixture was heated to 130 C. After dehydration and removal of
chlorobenzene, the reaction temperature was increased to about 160-170 C. When
the
24
CA 02527445 2005-11-18
solution viscosity had apparently increased, the mixture was cooled to 100 C
and
coagulated into a large excess of ethanol or water with vigorous stirring. The
resulting polymer was designated SPAEEN-50, where 50 refers to the 2,8-DHNS-6
content of aromatic phenol monomers. SPAEENs were washed with water to remove
salt. The results for water uptake, swelling and proton conductivity of the
various
SPAEENs are illustrated in Figures 14-16.
Copolymer Analysis and Measurement.
[0041] 1H-NMR spectra were obtained on a Varian Unity Inova NMR spectrometer
operating at a proton frequency of 399.95 MHz. Deuterated dimethylsulfoxide
(DMSO-d6) was the NMR solvent and tetramethylsilane (TMS) was used as the
chemical shift reference (0 ppm). IR spectra were measured on a Nicolet 520
Fourier
transform spectrometer with membrane film samples in a diamond cell.
[0042] A TA Instruments thermogravimetric analyzer (TGA) instrument model 2950
was used for measuring Td. Polymer samples for TGA analysis were preheated to
150 C at 10 C/min in either nitrogen or air and held isothermally for 40 min
for
moisture removal. Samples were then heated from 90 C to 750 C at 10 C/min for
Td
measurement. A TA Instruments differential scanning calorimeter (DSC) model
2920
calibrated with Tin at 231.93 C and Zinc at 419.53 C was used for measuring
Tg.
Samples in sodium form for DSC analysis were initially heated rapidly at a
rate of
C/min under nitrogen atmosphere to 380 C. When the DSC cell had cooled to
around 50 C, the samples were reheated at a rate of 10 C/min to 430 C. The
procedure for samples in acid form was similar except that the end point of
the initial
heating was 250 C.
[0043] Intrinsic viscosities were determined using an Ubbelohde viscometer for
N,N-
dimethylacetamide (DMAc) solutions of copolymer at 25 C.
Preparation of Membrane Films.
[0044] An amount of 1 g copolymer in the sodium salt form was dissolved in 20
mL
of DMAc and filtered. The filtered solution was poured onto a leveled glass
plate
having a circular glass retaining wall and dried at about 40 C under a
constant purge
of nitrogen for about one day. The acid form (m-SPAEENH) membrane films were
CA 02527445 2005-11-18
obtained by immersing corresponding sodium form m-SPAEEN membrane films in 2
N H2SO4 for 24 h at room temperature, and then in deionized water for another
24 h
during which time the water was changed several times. The thickness of all
membrane films was in the range of 40 to 70 m.
Water Uptake Content Measurement and Swelling Ratio
[0045] The membrane films were dried at 100 C overnight prior to the
measurements.
After measuring the lengths and weights of dry membranes, the sample films
were
soaked in deionized water for 24 h at predetermined temperatures. Before
measuring
the lengths and weights of hydrated membranes, the water was removed from the
membrane surface by blotting with a paper towel. The water uptake content was
calculated by
Uptake content (%) = eL -Wiry X100%
(Wry
Where o)drp and U~vet are the masses of dried and wet samples respectively.
The
swelling ratio was calculated from films 5-10 cm long by:
Swelling ratio (%) = Let - Id,,. x100%
ldrv
Where Id, and lrovet are the lengths of dry and wet samples respectively.
Tensile test
[0046] Tensile tests were performed on an Instron tensile tester (model 1123)
at a
strain speed of 50 mm/min at room temperature. Membrane films with typical
size of
40 mm x 4mm x 0.05 mm were used for testing.
Proton Conductivity
[0047] The proton conductivity measurements were performed on m-SPAEENH
membrane films by AC impedance spectroscopy over a frequency range of 1-107 Hz
with oscillating voltage 50-500 mV, using a system based on a Solartron 1260
gain
phase analyzer. Proton conductivities in the longitudinal direction were
measured.
Prior to the proton conductivity measurements, membranes were immersed in 98 C
water for 36 h to attain hydration equilibrium, and then cool to room
temperature. A
20 x 10 mm membrane sample was lightly clamped between two electrodes and
placed in a temperature controlled cell open to the air by a pinhole where the
sample
26
CA 02527445 2005-11-18
was equilibrated at 100%RH at ambient atmospheric pressure. The proton
conductivity (a) of the samples in the longitudinal direction was calculated
from the
impedance data, using the relationship a = l / Rdw, where 1 is the distance
between the
electrodes, d and w are the thickness and width of the films, respectively and
R was
derived from the low intersect of the high frequency semi-circle on a complex
impedance plane with the Re (Z) axis, where Re refers to "Real" in the complex
impedance plane. The impedance data were corrected for the contribution from
empty
and short circuited cell.
Results and Discussion
Synthesis and Characterization of m-SPAEEN copolymers
[0048] m-SPAEEN copolymers were synthesized by copolymerization of 2,6-DFBN,
2,8-DHNS-6 and 4,4'-biphenol as shown in Scheme 1. DMSO was used as
polymerization solvent instead of N-methyl-2-pyrrolidone (NMP), which was used
in
most of our previous polycondensation reactions since complete precipitation
of
reactants or oligomers in NMP took place at the bisphenoxide formation stage.
Water
generated during this stage was removed as an azeotrope with chlorobenzene. It
should be noted that replacement of chlorobenzene with toluene for dehydration
also
caused precipitation during the bisphenoxide formation. Polymerization
solutions
appeared cloudy yellow color. Polymerization compositions, details of the
resulting
polymers such as equivalent weight per sulfonate group (EW), ion exchange
capacity
(IEC) of m-SPAEEN-X copolymers are summarized in Table 10.
Table 10 Syntheses of m-SPAEEN copolymers
4,4'-BP 2,6-DFBN 2,8-DHNS-6 Na EW (IEC) expected SC SC from 'H-
expected NMR data
Polymer mmol mmol mmol dL/g g/mol SO3 (Meq g 1)
rn-PAEEN-0 5 5 0 - - 0 0
m-SPAEEN-20 8 10 2 0.85 1480 (0.68) 0.2 0.19
m-SPAEEN-30 7 10 3 1.0 1005 (1.0) 0.3 0.31
in-SPAEEN-40 6 10 4 0.62 767 (1.3) 0.4 0.38
m-SPAEEN-50 5 10 5 0.9 625 (1.6) 0.5 0.50
m-SPAEEN-60 4 10 6 0.82 530(l.9) 0.6 0.57
a Measured at 25 C in DMAc.
27
CA 02527445 2005-11-18
[0049] Intrinsic viscosity values of 0.62 to 1.0 dL/g in DMAc at 25 C indicate
the
success of polymerization in producing high molecular rn-SPAEEN-20 to m-
SPAEEN-60. However, due to the angled structure of 2,8-DHNS-6, m-SPAEEN
copolymers with SC values >0.7 were limited not obtained with high molecular
weights. This was probably due to excessive entanglement in the polymer chains
that
contained less linear biphenol segments. Indeed, there is no need to further
increase
the sulfonic acid content in m-SPAEEN. From Table 10, it can be seen that the
m-
SPAEEN-60 already has a high IEC value. It can also be seen in Fig. 20 that m-
SPAEENH-60 (the acid form of m-SPAEEN-60) has a swelling of 24% at 100 C,
which indicates that SC > 0.70 in m-SPAEENH copolymers will result in
unacceptable swelling in membrane at high temperatures, and cause the decrease
in
the membrane morphologic stability and mechanical stability. m-SPAEEN-20 to m-
SPAEEN-60 have good solubility in N,N-dimethylacetamide and were cast into
strong
transparent and flexible membrane films. Since all the reactants used in the
preparation of m-SPAEEN copolymers are commercially available and inexpensive,
the present sulfonated nitrile copolymers are much cheaper to manufacture than
Nafion.
[0050] The chemical structures of in-SPAEEN copolymers were initially
confirmed
by FT-IR. As seen in Fig. 17, characteristic bands of the aromatic sulfonate
salt are
observed at 1045, 1084, and 1108 cm' for m-SPAEEN copolymers compared with
unsulfonated m-PAEEN and the intensity of these characteristic absorption
bands
increase with SC. The characteristic symmetric stretching band of nitrile
groups was
observed at 2242 cm-1. The absorption bands at 1211 and 1246 cm'are assigned
to
phenoxy groups. The absorption bands at 1463 and 1495 cm' were assigned to
phenyl
ring and the bands at 1587 and 1606 cm' are attributed to C=C stretching.
[0051] The structural properties of the synthesized polymers were also studied
by
liquid phase 'H NMR spectroscopy with DMSO-d6 as the solvent. As an example,
the
'H-NMR spectrum of the aromatic region of m-SPAEEN-30 is shown in Fig. 18. The
benzonitrile proton signals for H-f,h (2H) appeared at the same low
frequencies (6.6-
6.9 ppm) as observed before in the SPAEEN polymers22. The intensity of the
distinct
signals at high frequencies (8.15-8.40 ppm) for H-b,c of the DHNS monomers
were
28
CA 02527445 2005-11-18
used to estimate and compare the experimental SCs with the expected SCs from
the
feed ratios. Using m-SPAEEN-30 as an example in Fig. 18, the intensity of H-
b,c is
0.62 therefore the ratio of DHNS : biphenol is 0.31 : 0.69 for 1.0 DFBN hence
an
experimental SC of 0.31. Table 10 shows the experimental SC values obtained
from
NMR are in close agreement with the expected SCs from the feed ratios.
Thermal properties of m-SPAEEN
[0052] Thermal properties of m-SPAEEN copolymers were evaluated by their Td
and
Tg data. Tjs were determined in both nitrogen and air in order to detect their
inherent
thermal stabilities and thermal stabilities in air. Fig. 19 shows that the TGA
curves of
m-SPAEEN copolymers in both nitrogen and air are very similar to those of
other
sulfonated high performance polymers reported beforel7-19, 22 Each copolymer
showed only one weight loss steps for sodium form polymers at around 460-500 C
attributed to the degradation of polymer chain, and two distinct weight loss
steps for
acid form polymers, of which the earlier weight loss at around 260-300 C is
caused
by cleavage of -SO3H. Fig. 19 also shows that air didn't cause earlier
cleavage of -
S03H than in nitrogen and m-SPAEEN copolymers are both thermally stable and
thermo-oxidatively stable enough for PEM usage. Table 11 summarizes the
observed
Td5% (5% weight loss) and onset weight loss temperatures (Ti!) data of m-
SPAEEN.
29
CA 02527445 2005-11-18
Table 11
Thermal properties of polymers
Polymer Tg ( C) Td5% ( C)
extrapolated on
In nitrogen In air In nitrogen
Na form Acid form Na form Acid form Na form Acid form Na form Acid foi
m-SPAEEN-20 233 230 498 373 477 390 491 268
m-SPAEEN-30 261 244 493 353 468 361 484 299
m-SPAEEN-40 310 247 485 345 449 352 482 300
m-SPAEEN-50 313 260a 468 345 467 341 488 288
m-SPAEEN-60 336 - 473 339 463 344 460 307
a approximate value
CA 02527445 2005-11-18
[0053] Table 11 also summarizes Tgs of m-SPAEEN copolymers in both sodium and
acid forms. The Tgs of copolymers increase with SC values varying from 233 to
336 C in sodium form and from 230 to 260 C in acid form. The Tgs of sodium
form
copolymers were all observed before the onset of thermal decomposition.
However,
the acid form copolymer m-SPAEENH-60 had a decomposition onset lower than its
Tg and the T. of m-SPAEENH-50 was observed at a temperature that decomposition
had already started, so the reported value is only an approximation. The
combination
of Tgs and Td values leads us to conclude that all m-SPAEENH copolymers have
good
thermal stabilities.
Water uptake, swelling ratio and proton conductivity and mechanical properties
[0054] There is a considerable body of scientific evidence to suggest that the
electrostatic interactions in sulfonic acid-containing polymers results in
microphase
separation to hydrophobic and hydrophilic regions in their membrane films. The
majority of ion exchange sites and counter ions aggregate to form hydrophilic
phase,
which form clusters in a continuous hydrophobic phase. Upon hydration, the
hydrophilic regions imbibe water and increase the clusters sizes into
interconnecting
channels for protons. Hydrophobic blocks of polymers are tightly packed to
provide
the membrane films with dimensional and mechanical stabilities. Consequently,
morphology concerning the distribution of hydrophilic and hydrophobic phases
is
greatly affected by absorbed water, which in turn further affects the proton
conductivity and mechanical properties of PEMs.
[0055] The water uptake and swelling ratio of m-SPAEENH membranes are plotted
as functions of SC values and temperatures in Fig. 20. PEMFCs are normally
operated
at temperatures from RT to 80 C based on the properties of state-of-art
polymer
electrolyte Nafion. However, since elevated operation temperatures will raise
the
tolerance ability of catalysts to CO, PEMs that can endure temperatures higher
than
100 C are preferred. Considering these experimental conditions, the water
uptake and
swelling ratios were tested at room temperature, 80 C, and 100 C respectively.
Fig. 20
shows that the m-SPAEENH copolymers absorbed water in the range of 5.7% to 69%
with increases in the 2,8-DHNS-6 content and temperature from RT to 100 C
after 24
h immersion in water and increased their linear dimensional sizes by 1.7% to
24%.
Normally, swelling of 25% may be considered as an acceptable value for
adequate
31
CA 02527445 2005-11-18
dimensional stability of PEM under humidified conditions. Using these
criteria, all the
present m-SPAEENH copolymers had low or adequate dimensional swelling when
fully hydrated. Fig. 20 also indicates that an SC value of -0.6 is the highest
practical
one for FC application. A further increase in 2,8-DHNS-6 content in the
copolymer
will result in an over-uptake of water, which will weaken the interactions of
hydrophobic phase and cause excessive dimensional swelling. The present
nitrile
copolymers show much lower water uptakes and swelling ratios, when compared
with
our previously prepared sulfonic acid-containing poly(aryl ether)s (Table 12)
of
similar proton conductivity values, including both poly(aryl ether ketone) and
poly(aryl ether nitrile) from flexible hydroquinone monomer.
32
CA 02527445 2005-11-18
Table 12
Comparisons of swelling and proton conductivities of different polymers
Polymer EW expected Room temperature 80 C
g/mo1SO3 Swelling ratioa Conductivity Swelling ratio Conductivity
(%) (S/cm) (%o) (S/cm)
m-SPAEEN-50 625 6.2 3.5x10 10 6.4x10
m-SPAEEN-60 530 7.5 8.3x10-2 15 1.4x10-1
SPAEEKK-100 575 18 - Excessively swollen -
[21]
SPAEEN-B50 575 6.9 1.2x10-2 11 3.5x10-2
[431
SPAEEN-B60 479 9.1 4.5x10-2 17 1.0x10-1
[43]
SPAEEN-Q50 498 9.4 3.0x10-2 26 1.0x10-1
[431
Nafion 117 1100 13 7.5x10-2 20 9.6x10-2
a % length gain of 5-10 cm strips of films after 24 h.
33
CA 02527445 2005-11-18
[0056] We attribute the lower swelling ratios to the combination of polar
nitrile
groups and hydrophobic naphthalene structures to network the film structure.
In
nano-phase separated hydrated film, the hydrophobic domains in nitrile-
containing
polymers are more intensively packed than other polymer films via their strong
polar
intermolecular actions and enhance the hydrophobic phases, which consequently
improve the dimensional stability of membrane films.
[0057] All the membranes maintained good shape and were mechanically strong
after
the hydration pretreatment of immersion in 98 C water for 36 h. This
pretreatment
differs from previous ones we employed in past studies, where the films were
simply
soaked in water at room temperature. This is because the nitrile copolymers
were
apparently more difficult to hydrate initially, as observed by conductivity
profiles.
The proton conductivity measurements of m-SPAEENH copolymers were run at
100% relative humidity as a function of SC and temperature in the longitudinal
direction by AC impedance spectroscopy and the results are shown in Fig. 21.
It
shows that the proton conductivities of m-SPAEENH copolymers increase with
both
SC and temperature. However, their SC dependent tendency seems quite different
from the temperature dependence. All m-SPAEENH copolymers display temperature-
dependant proton conductivity curves parallel to that of Nafion 117, i.e.,
their
logarithmic conductivities are linearly dependant on the reciprocal of the
temperature
from RT to 100 C, indicating their similar proton transfer mechanism and
activation
energy to Nafion, involving hydronium ions.
[0058] An increase in SC from 20% to 30% resulted in a two orders of magnitude
increase in proton conductivity. At SC > 30%, the rate of increase in proton
conductivity with SC slows down gradually. When the proton conductivities were
plotted against SC (Fig. 22), it is noticeable that the proton conductivity
initially
increases exponentially with SC and then the rate of increase diminishes. This
phenomenon can be explained by channel formation upon hydration. As described
previously, sulfonic acid groups form clusters in the continuous hydrophobic
phase,
which increase in size into interconnecting channels for protons upon
hydration. At
low SC values, hydrated sulfonic acid groups formed mainly distributed
clusters and
less connected channels, which resulted in low proton conductivities. An
increase in
the SC will considerably improve the connection and promote the proton
conductivity
34
CA 02527445 2005-11-18
greatly. At high SC values, however, obvious dimensional swelling in hydrated
membranes will dilute the volume concentration of sulfonic acid groups in the
membrane, which negatively affects the increase of proton conductivity with
increasing SC. This is because the swelling increases at an accelerate rate
with SC and
retards the increase in conductivity. m-SPAEENH-50 and m-SPAEENH-60 show high
proton conductivities comparable with Nafion117 from room temperature to 100
C,
ranging from 6.2x10-2 to 1.5x10-I S/cm. The comparisons on swelling and
conductivity of in-SPAEENH copolymers and Nafioni 17 in addition to our
selected
previously prepared sulfonic acid-containing poly(aryl ether)s are summarized
in
Table 12. SPAEEKK-100 prepared from 1,3-bis(4-fluorobenzoyl)benzene and sodium
6,7-dihydroxy-2-naphthalenesulfonate was reported with proton conductivities
values
which were lower than that of Nafion117 under the same measurement conditions
[22]. It should be noted here that the proton conductivities were originally
measured
transversely (through the membrane), which give values significantly lower
than
those measured longitudinally (along the membrane). As Table 12 shows,
previously
reported sulfonic acid-containing poly(aryl ether)s have either Nafion-
comparable
proton conductivities but excessive swelling or reasonable swelling but lower
proton
conductivities than Nafion. However, m-SPAEENH copolymers, especially m-
SPAEENH-60, have both low dimensional swellings and high proton
conductivities,
comparable to Nafion117. The reason for low water uptake and swelling has been
discussed before. Here, the high proton conductivity is explained. Normally,
post-
sulfonation of poly(aryl ether)s results in the sulfonic acid group being
located ortho
to the ether linkage, which deactivates the acidity of sulfonic acid and
lowers the
proton conductivity. The sulfonated poly(aryl ether sulfone) or poly(aryl
ether ketone)
copolymers prepared from biphenol and SDCDPS or 3,3'-disulfonate-4,4'-
difluorobenzophenone (S-DFB) also have sulfonic acid groups ortho to ether
linkage,
which are both deactivated by the ether linkage and activated by the SDCDS or
S-
DFB units at the same time. Unlike those polymers, m-SPAEENH has sulfonic acid
groups meta to the ether linkage, which is a less deactivating position,
resulting in a
higher acidity of sulfonic acid groups. The effect of sulfonic acid located on
different
sites will be further discussed in another paper.
[0059] Kim [49] reported that high-temperature acidification of solvent-cast
films
gave high proton conductivities due to different microstructure formation
occurring
during various acidification process. In the present work, the films were all
acidified
CA 02527445 2005-11-18
at room temperature. However, they were soaked in deionized water at different
temperatures before conductivity test. The results are plotted in Fig. 23.
Unlike m-
SPAEENH copolymers, which were soaked in hot water for 36 h prior to
conductivity
measurement, m-SPAEENH-50-untreated and m-SPAEENH-60-untreated were only
immersed in deionized water at room temperature for 24 h. Fig. 23 shows that
proton
conductivities of hot water-soaked m-SPAEENH-50 and m-SPAEENH-60 are less
temperature-dependant than those of m-SPAEENH-50-untreated and m-SPAEENH-
60-untreated and at high temperatures they to converge to approximately the
same
values for the membranes treated differently. This is explained by taking into
consideration the synergic effect of a water confinement effect of 2,8-DHNS-6
structure, the nitrile group polarity as well as the microstructure effect
discussed by
Kim [50]. m-SPAEENH-50-untreated and m-SPAEENH-60-untreated were not
completely hydrated at low temperatures, consequently inadequately formed
proton
channels were improved with temperature and the proton conductivities
increased
more sensitively with temperature. The angled structure of 2,8-DHNS-6
increases the
interchain spacing and creates permanent pores lined with -SO3H groups. Once
H2O
enters the pore at high temperature, it is held very strongly. In addition,
polar nitriles
in the hydrophilic domains of nano phased separated film may also interact
with the
water molecules confined in the pore, which would assist in enhancing the
water
confinement. As a result, hot water-soaked m-SPAEENH-50 and m-SPAEENH-60
show higher proton conductivities even after having been cooled down to room
temperature. At higher temperatures, water uptake and proton channels tend to
be the
same no matter how the membranes were treated initially; their proton
conductivities
tend to be the same.
[0060] The tensile properties were measured at room temperature and are
summarized
in Table 13.
36
CA 02527445 2005-11-18
Table 13
Tensile properties of m-SPAEENH copolymers
Polymer Tensile strength Elongation at break
(MPa) (%)
m-SPAEEN-20 78 4.5
m-SPAEEN-30 73 10
m-SPAEEN-40 39 3.5
m-SPAEEN-50 56 4.6
m-SPAEEN-60 54 4.6
Nafion 117 10 623
[0061] All membranes show small strain with elongation at break of 3.5% to 10%
compared to 623% of Nafionll7. Tensile strengths are from 39 MPa to 78 MPa,
several-fold higher that the 10 MPa value of Nafionl 17.
Conclusions
[0062] A series of aromatic poly(aryl ether ether nitrile)s containing
sulfonic acid
groups meta to ether linkage (m-SPAEEN) having 0-60 mol % of 2,8-
dihydroxynaphthalene-6-sulfonate (2,8-DHNS-6) segment, have been successfully
prepared from commercially available inexpensive monomers via one-step
polycondensation reactions. m-SPAEENH copolymers have good thermal stabilities
with decomposition temperatures higher than 250 C and Tgs higher than 230 C. m-
SPAEENH membranes have tensile strength from 39 to 78 MPa, several times
higher
than 10 MPa of Nafion117 and elongation at break from 3.5-10%, several hundred
times smaller than 623% of Nafion117. Pendant nitrile groups increase the
dipole
interactions between polymer chains and decrease the membrane swelling, even
up to
100 C. The angled structure of 2,8-DHNS-6 increases the interchain spacing and
confines he water molecules, which improves the proton conductivities of
membranes
at lower temperatures. The location of the sulfonic acid groups meta to the
ether
linkage results in the copolymer sulfonic acid groups being less deactivated,
giving
membranes with high proton conductivity due to the increased acidity.
Furthermore,
37
CA 02527445 2005-11-18
the meta position is expected to reduce hydrolytic instability. The
combination of
inexpensive monomers, high thermal stability, low dimensional swelling, good
mechanical properties and high proton conductivity makes m-SPAEENH-50 and m-
SPAEEN-60 attractive as PEM materials for fuel cells applications.
EXPERIMENTAL (SPPEKN)s
Materials
[0063] DHPZ was synthesized according to the procedure reported previously45-
48
SDFB-Na was prepared according to the procedures described early and in
literature
articles 16,18 N-methyl-2-pyrrolidone (NMP) was vacuum distilled before use.
All
other chemicals were obtained from Aldrich and were reagent grade and used as
received.
MEMBRA-CELTM dialysis tubing (MWCO 3500) was obtained from Serva
Electrophoresis (Germany)
Copolymerization Reaction
[0064] As depicted in scheme 1A, the SPPEKN copolymers were synthesized via
nucleophilic polycondensation reaction.
38
CA 02527445 2005-11-18
Scheme 1A Synthesis of SPPEKN copolymers
CN 0
OH /
F F :0o3I + /N N/ H
F ~ F
\ O
~.
0
O
CN 11
O 303X N ",N \ vv n
N O
~ \N
SO3NX
O
Wherein X=Na or K or H
[0065] The synthesis of SPPEKN-40 is used as a typical example, where n (40)
denotes to the SDFB-Na monomer feed mole ratio of difluoro monomers. To a
three-
neck flask with a magnetic stirrer, a Dean-Stark trap and condenser, and an
argon inlet,
0.8605 g 2,6-DFBN (4.04 mmol), 1.706 g SDFB-Na (6 mmol), 2.383 g DHPZ (10
mmol), and 1.8 g potassium carbonate (13 mmol) were added. Then 12 mL of NMP
and 20 mL chlorobenzene were charged into the reaction flask under an argon
atmosphere. The reaction mixture was heated to 140 C. After dehydration and
removal of chlorobenzene (--3-4 h), the reaction temperature was increased to
around
165 C. After a period of 5-7 h, when the solution viscosity had obviously
increased,
several mL of NMP was added to dilute the solution and the reaction was
continued
for a further 3-5 h. Then, the mixture was cooled to 100 C and coagulated into
a large
excess of ethanol with vigorous stirring. After recovering the product, SPPEKN-
40
was washed with deionized water to remove residual solvent and salt.
Copolymer Analysis and Measurement
[0066] 1H NMR spectra were obtained on a Varian Unity Inova NMR spectrometer
operating at a proton frequency of 399.95 MHz. Deuterated dimethylsulfoxide
39
CA 02527445 2005-11-18
(DMSO-d6) was the most convenient NMR solvent for SPPEK; the TMS signal at 0
ppm was used as the chemical shift reference. IR spectra were measured on a
Nicolet
520 Fourier transform spectrometer with membrane film samples in air.
[0067] A TA Instruments thermogravimetric analyzer (TGA) instrument model 2950
was used for measuring the degradation (weight loss) temperatures (Td) and a
TA
Instruments differential scanning calorimeter (DSC) model 2920 calibrated with
Tin
at 231.93 C was used for measuring the Tgs. Copolymer samples for TGA analysis
were preheated to 120 C at 10 C/min under air atmosphere and held isothermally
for
60 min for moisture removal. Samples were then heated from 90 C to 750 C at
C/min for Tel measurement. Samples in potassium form for DSC analysis were
initially heated rapidly at a rate of 20 C/min under nitrogen atmosphere to
450 C,
followed by quenching in liquid nitrogen. When the DSC cell had cooled to
around
50 C, the samples were replaced in the cell and heated at a rate of 10 C/min
to 400 C.
The procedure for samples in acid form was similar except that the initial
heating rate
was 10 C/min and the end point was below the polymer 7(l point.
[0068] Intrinsic viscosities were determined using an Ubbelohde viscometer for
N,N-
dimethylacetamide (DMAc) solutions of polymer at 25 C.
Preparation of Membrane Films
[0069] An amount of 1 g sulfonated polymer in sodium form was dissolved in 20
mL
of DMAc and filtered. The filtered solution was poured onto a glass plate and
dried at
40 C for about two days. The acid form (SPPEKNH) membrane films were obtained
by immersing sodium form membrane films in 2 N I12SO4 for 24 h at room
temperature, followed by deionized water for 24 h during which deionized water
was
change several times in order to remove excess acid, then dry in the air for
several
hours.
Water Uptake Content Measurement and Swelling Ratio
[0070] The sample films were soaked in deionized water for 24 h. The membrane
films were then dried at 80 C for 24 h. Weights of dry and hydrated membranes
were
measured. The water uptake content was calculated by
CA 02527445 2005-11-18
C1).ret - LOdrv
Uptake content (%) _ "' x100%
Where 0 "y and O'et are the masses of dried and wet samples respectively. The
swelling ratio was calculated from films 5-10 cm long by:
Let - ldry
Swelling ratio (%) = ldrv x100%
Where ldry and lwet are the lengths of dry and wet samples respectively.
Tensile test
[0071] Tensile tests were performed on an Instron tensile tester (model 1123)
at a
strain speed of 50 mm/min at room temperature. Membrane films with typical
size of
40 mm x 4mmx 0.05 mm were used for testing.
Proton Conductivity
[0072] The proton conductivity measurements were performed on SPPEKNH
membrane films by AC impedance spectroscopy over a frequency range of 1-1049
IIz
with oscillating voltage 50-500 mV, using a system based on a Solatron 1260
gain
phase analyzer. A 20 x 10 mm membrane sample was placed in a temperature
controlled cell open to the air by a pinhole where the sample was equilibrated
at
100%RH at ambient atmospheric pressure and clamped between stainless steel
electrodes. Specimens were soaked in deionized water for 24 to 48 h prior to
the test.
The conductivity (a) of the samples in the longitudinal direction was
calculated from
the impedance data, using the relationship r = d/RS where d and S are the
thickness
and face area of the sample respectively and R was derived from the low
intersect of
the high frequency semi-circle on a complex impedance plane with the Re (Z)
axis.
The impedance data were corrected for the contribution from empty and short
circuited cell.
41
CA 02527445 2005-11-18
RESULTS AND DISCUSSION
Copolymerization
[0073] In order to obtain polymers with target proton conductivities of >10-2
S/cm, a
specific range of monomer compositions were designed for SPPEKN copolymers.
The series of SPPEKN copolymers were prepared by nucleophilic substitution
reactions with DHPZ as the diphenol, SDFB-Na and 2,6-DFBN as activated
dihalides
and K2CO3 as a weak base. All SPPEKN copolymers were obtained under the same
polymerization conditions used for the synthesis of SPPEKs and SPPESs'8: under
inert argon atmosphere, NMP was used as the reaction solvent, chlorobenzene
was
used to remove the water generated from the reaction of diphenol and K2CO3 at
130
to 140 C. When an obvious increase in the viscosity of reaction solutions was
observed at a reaction temperature of 170 C, which indicated a high molecular
weight
had been reached, the reaction solutions were precipitated in ethanol or
water. As
shown in Scheme IA, the sulfonate content and the nitrile group content in
SPPEKN
copolymers were balanced by varying the feed ratio of SDFB-Na to 2,6-DFBN. The
sulfonate content (SC) is used to define the SPPEKN copolymers and is
expressed as
the molar ratio of SDFB-Na units (bearing the -SO3Na group) to 1.0 molar DHPZ
unit. For example, the average repeat unit of SPPEKN-30 copolymer is composed
of
0.3 unit of SDFB-Na, 0.7 unit of 2,6-DFBN and 1.0 unit of DHPZ. Table 5 lists
the
polymerization conditions and details of the resulting polymers.
42
CA 02527445 2005-11-18
Table 5 Viscosity data for SPPEKNs
SDFB-Na 2,6-DFBN DHPZ [ry]a Meq (IEC) expected SC expected
Polymer mmol mmol mmol dLIg g/molSO3 (Meq g-1)
SPPEKN-30 3.0 7.07 10 2.32 702 (1.42) 0.6
SPPEKN-35 3.5 6.56 10 1.71 631 (1.58) 0.7
SPPEKN-40 4.0 6.06 10 1.39 562 (1.78) 0.8
SPPEKN-45 4.5 5.56 10 1.20 515 (1.94) 0.9
SPPEKN-50 5.0 5.05 10 1.45 478 (2.09) 1.0
SPPEKN-55 5.5 4.55 10 0.684 447 (2.24) 1.1
SPPEKN-60 6.0 4.04 10 1.02 422 (2.37) 1.2
a Measured at 25 C in DMAc.
43
CA 02527445 2005-11-18
[0074] For the purpose of attaining polymers with proton conductivities higher
than
10-2 S/cm, only SPPEKN copolymers with SC values ranged from 0.3 to 0.6 were
synthesized. All SPPEKN copolymers had intrinsic viscosities higher than 0.6
in
DMAc at 25 C indicating the polymerizations proceeded to high molecular
weights.
All SPPEKN copolymers were cast into free-standing films, also confirming the
high
molecular weight of the resulting copolymers. Equivalent molecular weight
(Meq)
and ionic exchange capability (IEC) were also calculated theoretically and
listed in
Table 5 for comparison.
FT-IR
[0075] FT-IR is a convenient method to confirm the sulfonate or nitrile groups
in new
polymers. Figure 9 illustrated the FT-IR spectra of SPPEKN copolymers, and the
spectrum of SPPEK-50 (composed of 0.5 unit of SDFB-Na, 0.5 unit of 4,4-
difluorobenzophenone and 1.0 unit of DHPZ) was also illustrated for
comparison. In
all the spectra of SPPEKN copolymers, characteristic bands of the aromatic
sulfonate
salt symmetric and asymmetric stretching vibrations were observed at 1027 and
1096
cm 1 and the characteristic symmetric stretching band of nitrile groups was
observed
at 2247 cm 1, which was not observed in the spectrum of SPPEK-50. It was
observed
that the intensity of two characteristic sulfonate absorption bands increases
with SC
values, meanwhile the intensity of the characteristic nitrile absorption
decrease with
SC values . This confirms that sulfonate groups and nitrile groups were
introduced
into the copolymers varying with the monomer feed ratio. Some other
characteristic
absorption in SPPEKN copolymers was also assigned. The absorption bands around
1600 cm' is attributed to C=C stretching in phenyl ring. The absorption bands
around
1668 cm -1 is attributed to carbonyl group. The characteristic absorption
bands of
bands of 1,4- aromatic ring substitution at 1510 cm' also decrease with SC
values.
The absorption bands at 1217 and 1260 cm -1 are due to phenoxy groups.
NMR
[0076] Figure 10 displays the aromatic region of five polymers. The top and
bottom
spectra represent the polymers poly(phthalazinone ether nitrile) (PPEN) and
SPPEK
respectively whereas the three other spectra are SPPEKN with various SC.
Assignment of the PPEN signals was done using 1H and 2D-COSY and TOCSY
44
CA 02527445 2005-11-18
N'MR. The three different hydrogen spin systems (H-6,7,8,9; H-25,26; H-
31,32,33)
were easily detected and assigned using 2D. The analysis of SPPEK 1H spectrum
was
achieved earlier 18 and is shown in Fig. 10. These two fully assigned spectra
were
incorporated into Figure 10 in order to better visualize the signal changes
taking place
into the three SPPEKN spectra. The copolymer spectra are very complex as a
result of
the asymmetry of the phthalazinone monomer. As illustrated in Figure 10, both
R
groups for the copolymers can either be nitrile or sulfonated monomers. The
different
feed ratios of sulfonated monomer also add to the complexity of the spectra
therefore
we did not try to assign the signals from the three SPPEKN copolymers.
Nevertheless,
Figure 10 clearly shows downfield growing signals for sulfonated monomer and
upfield growing signals for the nitrile monomer. We observed from previous
experience' 8 of measuring the SC from 1H NMR data that the SC is usually
consistent
with the feed ratio of monomers. The apparent changes one can see are that
sulfonated
repeat unit signals increase while nitrile decrease with SC values. The
smaller signals
at lower frequency are either from chain end groups or they could also be a
result of
one of the many possible configuration of RU due to the factors described
above.
Thermal properties
[0077] The thermal properties of SPPEKN copolymers in both salt and acid forms
are
listed in Table 6 and illustrated in Figure 11.
CA 02527445 2005-11-18
Table 6 Thermal properties of copolymers
Polymer Tg ( C) TT15% ( C)
extrapo
Sodium form Acid form Sodium form Acid form Sodi
PPEK 263 487
SPPEK-4012 367 294 475 340
SPPEK-5012 390 ND 482 322
SPPEK-6012 ND ND 484 339
SPPEKN-30 359 ND 475 327
SPPEKN-35 362 ND 480 348
SPPEKN-40 365 ND 483 342
SPPEKN-45 372 ND 483 342
SPPEKN-50 ND ND 479 317
SPPEKN-55 ND ND 479 310
SPPEKN-60 384 ND 466 327
* ND: not detected
46
CA 02527445 2005-11-18
[0078] There is only one weight loss step in the TGA curves of salt form
SPPEKN
copolymers (Figure 11), which is assigned to the degradation of polymer main
chain.
Compared with PPEK and SPPEKs, no obvious difference in the thermal stability
of
salt form SPPEKN copolymers was observed and all their 5% weight loss
temperatures and the extrapolated onset temperatures are higher than 460 C,
indicating the high thermal stabilities of sodium form SPPEKN copolymers. As
observed before17-19, 22, there are two distinct weight loss steps in the TGA
curves of
SPPEKNH copolymers, and the 5% weight loss temperatures and the extrapolated
onset temperatures for the first weight loss of SPPEKNH copolymers caused by
the
loss of sulfonic acid groups are all higher than 280 C, which decrease with
increasing
SC values. The second thermal degradation around 480 C was assigned to the
degradation of the polymer main chain. Compared with SPPEK copolymers, the
introduction of nitrile groups into the polymer chain did not have an obvious
affect on
the thermal stabilities of SPPEKN copolymers in either salt or acid forms. The
SPPEKN copolymer series appear to have sufficient thermal stability for PEM
usage.
[0079] Table 6 also lists the Tgs of SPPEKN copolymers. SPPEKNH copolymers
were not detected with Tgs before their Tds. SPPEKN copolymers in sodium salt
form
show Tgs from 359 to 384 C, increasing with SC values and all higher than that
of
PPEK. The increase in the Tgs of SPPEKN copolymers are the result of the
introduction of sodium sulfonate or sulfonic acid groups, which increase
intermolecular interaction by pendant ions or hydrogen bonding and molecular
bulkiness and hinder the internal rotation of high molecular chain segment.
The
introduction of nitrile groups didn't sacrifice the Tgs of SPPEKN copolymers.
Water uptake and swelling ratio
[0080] It was reported that some sulfonated polymers are nanophase separated
into
hydrophilic and hydrophobic domains wherein sulfonate or sulfonic acid groups
and
polymer backbones aggregate separately8 50, 40, 43 The hydrophobic part
provides the
hydrated sulfonated polymer membrane films with good morphological and
mechanical stability whereas the hydrophilic domains imbibe water and provide
good
proton conductivity. Since PEMs in fuel cells are generally operated at
temperatures
close to 80 C, water uptake and swelling ratio of SPPEKN copolymers were
47
CA 02527445 2005-11-18
measured at both room temperature and 80 C in relation to the SC values, and
the
counter ions, as shown in Table 7.
48
CA 02527445 2005-11-18
Table 7 Water uptake and swelling ratio of SPPEKN copolymer
Meq (IEC) Room temperature
expected Water uptake Swelling ratio Water uptake
(%) (%) (%)
Polymer g/molSO3 Sodium form Acid form Sodium form Acid form Sodium form Acid
(Meg g-1)
SPPEK-40'2 698 (1.43) 21 24 6.0 9.5 26 2
SPPEK-5012 568 (1.76) 33 42 11 15 101 21
SPPEK-6012 482 (2.07) 47 60 14 20 410 23
SPPEKN-30 702 (1.42) 14 16 4.5 7.4 18 3
SPPEKN-35 631 (1.58) 16 20 5.6 8.5 20 4
SPPEKN-40 562 (1.78) 26 29 7.4 13 44 9
SPPEKN-45 515 (1.94) 31 34 7.9 15 65 1`
SPPEKN-50 478 (2.09) 36 51 8.3 18 92 11
SPPEKN-55 447 (2.24) - 79 - 29 - Dissc
SPPEKN-60 422 (2.37) 220 Swelled 52 Swelled Dissolved Dissc
49
CA 02527445 2005-11-18
[0081] SPPEKN-55 in sodium form was brittle when fully dehydrated. It was
difficult
to measure the changes in length and weight; hence no accurate data could be
reported.
However, the membrane film of SPPEKN-55 in sodium form maintained its shape in
the hydrated state at room temperature. In general, the water uptake and
swelling ratio
of SPPEKN copolymers in both salt and acid forms increase with SC or IEC
values
and temperature. Membrane films in acid form have higher water uptake and
swelling
ratio values than salt form ones because of the hydrogen bond interactions
between
H2O and sulfonic acid groups. Compared with Nafion 117 (IEC=0.91mmol/g)
membrane, which shows a water uptake of 35% at room temperature44 and SPPEK
copolymers prepared from DHPZ, 4,4'-SDFB-Na and 4,4'-difluorobenzophenone
reported previously18, SPPEKN copolymers with similar water uptake have higher
IEC values. In other words, SPPEKN copolymers imbibe less water and swell less
at
equivalent IEC values to SPPEK copolymers and Nafion 117. Since the
aggregation
of hydrophilic sulfonic acid groups in SPPEKN membrane films are similar to
those
in SPPEK, the greater hydrophobic aggregation occurring as a result of the
SPPEKN
leads to a relative decrease in the water uptake and swelling ratios compared
with.
The use of 2,6-DFBN for the SPPEKN copolymerizations instead of 4,4'-
difluorobenzophenone for SPPEK polymerizations resulted in the strongly polar
nitrile groups pendant on aromatic rings of SPPEKN copolymers, instead of the
weaker ketone groups in the polymer main chains of SPPEK copolymers. The
strongly polar nitrile groups in should enhance the intermolecular interaction
of
polymers and enhance the hydrophobic network structure. As a result, free
volume for
water adsorption in SPPEKN membrane films is restricted, while reduces the
water
uptake. Consequently the dimensional stability of membrane films is improved.
Tensile properties
[0082] Besides the requirement for thermal and dimensional stability of PEMs
for
fuel cells, adequate mechanical strength is also required. The tensile
properties of
SPPEKNH copolymers were tested at room temperature in both the dry and fully
hydrated states and the results are tabulated in Table 8.
CA 02527445 2005-11-18
Table 8 Tensile properties of SPPEKNH copolymers
Dry membranes Hydrated
Polymer Tensile strength Elongation at break Tensile strength at break
(MPa) (%) (MPa)
SPPEKNH-30 39 5.4 36
SPPEKNH-35 52 4.1 48
SPPEKNH-40 43 3.8 37
SPPEKNH-45 37 3.7 36
SPPEKNH-50 42 3.9 21
SPPEKNH-55 22 3.0 25
SPPEKNH-60 35 5.8 -
Nafion 117 10 623 -
51
CA 02527445 2005-11-18
[0083] In general, all SPPEKNH copolymers exhibited good tensile strength
ranging
from 22 to 52 MPa and an elongation at break ranging from 3.7% to 5.8%.
SPPEKNH-55 exhibited lower tensile strength than other SPPEKNH copolymers,
since SPPEKNH-55 had lower molecular weight than other copolymers. For
comparison, Nafion 117 was also tested under the same conditions and the
results are
also shown in Table 8. Compared with Nafion 117, all SPPEKNH copolymers
exhibited much higher tensile strength values and less elongation at break.
After
immersion in deionized water at room temperature for 24 h, during which time
the
polymer membrane films are normally fully hydrated' 8' 40, SPPEKNH copolymers
exhibited decreases in tensile strength and increase in elongations at break
except
SPPEKNH-55, which maintained its tensile strength. SPPEKNH-60 swelled
excessively and lost its mechanical strength in the fully hydrated state. The
decreases
in tensile strengths were the result of water plasticization in membrane
films. In the
dry state, the sulfonic acid groups contribute to the tensile strength of
membrane films
via ionic interactions. However, when fully hydrated, the ionic interactions
were
partly replaced by hydrogen-bonding between H2O and sulfonic acid groups. The
film
dimensional size was increased by imbibed water and the rigid network
structure of
the membrane was weakened, resulting in a decrease in the tensile strength in
membrane films. However, all hydrated SPPEKNH membrane films still have higher
tensile strength values than Nafion 117. The comparison with Nafion 117
indicates
that SPAEENH copolymers qualify for the tensile strength requirements for PEM
fuel
cells applications.
Proton Conductivity
[0084] Proton conductivities of SPPEKNH copolymers as functions of SC and
temperature were measured in air at 100% relative humidity in the longitudinal
direction by AC impedance spectroscopy and shown in Figure 12. For comparative
purpose, the proton conductivity of Nafion" 117 was also measured under the
same
experimental conditions.
[0085] The SPPEKNH membrane film series showed room temperature proton
conductivities in the range of around 10-2 to 10-1 S/cm according to the SC
values.
SPPEKNH-60 swelled excessively even at room temperature; thus its proton
conductivity could not be measured. Generally speaking, proton conductivities
increase with SC , temperature and the values are from 8.3x10.3 to 1.7x10-'
S/cm. For
52
CA 02527445 2005-11-18
comparison, Nafion 117 showed proton conductivities from 7.5x10.2 at room
temperature to 1.6x10-l S/cm at 97 C. SPPEKN-50 and SPPEKN-55 shown proton
conductivities higher than or close to Nafion 117 at all test temperatures;
however,
they swelled at elevated temperatures or even dissolved in hot water. Thus,
SPPEKNH-50 and SPPEKNH-55 are unsuitable for PEM applications at elevated
temperatures although they may be used at room temperature. SPPEKNH-35,
SPPEKNH-40 and SPPEKNH-45 films showed good proton conductivities, from 10-2
to 10-1 S/cm at different temperatures, close to the values of Nafion117. In
addition,
since they maintained reasonable dimensional stabilities and tensile strengths
in both
the dry and hydrated states, they could be considered as promising candidates
for
PEMs applications. Compared with Nafion 117, SPPEKNH copolymers exhibited a
more rapid increase in proton conductivities with temperature, suggesting that
SPPEKNH copolymers have higher activation energy for proton conductivity.
[0086] Since the proton conductivities of SPPEKH copolymers (Figure 13) were
measured in their transverse direction in our previous study, their data are
not directly
comparable with those of SPPEKNH copolymers, which were obtained in their
longitudinal direction. We have observed that proton conductivity values we
measured transversely are typically 3-5 times lower than those measured
longitudinally. The comparisons of each series were made against Nafion 117
which
were measured either in the transverse and longitudinal directions. In both
Figures 12
and 13, SPPEKH and SPPEKNH respectively showed proton conductivities close to
the values of Nafion117. SPPEKNH and SPPEKH copolymers with similar SC values
showed similar differences in proton conductivities compared with Nafionl 1.7.
The
introduction of nitrile groups didn't lead to a significant decrease in the
proton
conductivities of copolymers.
CONCLUSIONS
[0087] Sulfonated poly(phthalazinone ether ketone nitrile) copolymers with
high
molecular weight were prepared by direct copolymerization reaction of disodium
3,3'-disulfonate-4,4'-difluorobenzophenone (SDFB-Na), 2,6-difluorobenzonitrile
(2,6-DFBN), and 4-(4-hydroxyphenyl)-1(2H)-phthalazinone (DHPZ) under general
polycondensation reaction conditions. The sulfonic acid group content (SC) and
nitrile content in the copolymers were balanced by varying the ratio of
sulfonated
53
CA 02527445 2005-11-18
monomer SDFB-Na to unsulfonated monomer 2,6-DFBN and characterized by FT-IR
and NMR measurements. All SPEEKNH copolymers exhibited thermal stabilities up
to 270 C, but no Ts were observed up to their degradation temperatures.
[0088] Characteristic absorptions of aryl carbonyl, sulfonate and nitrile
groups were
observed in the resulting copolymers by FT-IR characterization. 1H NMR was
also
used to characterize the polymer structures. Membrane films in both salt and
acid
forms of SPPEKN copolymers with SDFB-Na to 2,6-DFBN mole feed ratios up to
60/40 were cast from the N,N-dimethylacetamide (DMAc) polymer solutions
followed by immersing in 2 N sulfuric acid at ambient temperature. An increase
of
sulfonate groups in the copolymers resulted in increased glass transition
temperature
(Tg) and membrane hydrophilicity. The sodium form copolymers were thermally
more
stable than their acid form. The introduction of highly polar nitrile groups
were
expected to increase the intermolecular forces and make the polymers less
moisture
absorbable than previously prepared sulfonated poly(phthalazinone ether
ketone)
(SPPEK) copolymers and improve the combination of polymer and catalyst in the
possible future usage
[0089] The presence of highly polar nitrile groups in SPPEKNH renders the
copolymers less moisture absorbable and swellable than the previously prepared
SPPEKH copolymers. The SPPEKNH copolymers exhibited tensile strength stronger
than Nafion 117 and reasonable flexibility in both the dry and hydrated
states. The
proton conductivities of SPPEKNH copolymers were in the range of 10-2 to 10-
15/cm,
increasing with SC values and temperature. Direct comparisons were made with
Nafion 117 measured under the same conditions, to remove variability with
measurement techniques. Nitrile groups are also anticipated to promote
adhesion of
the polymers to catalyst, carbon black in membrane electrolyte assembly (MEA)
or
certain conducting inorganics such as heteropolyacids in composite membranes4o
Considering the combination of thermal properties, tensile strength,
dimensional
stability in the hydrated state and proton conductivity values, SPPEKN-35, -40
and -
45 appear to be potentially good candidate for proton conducting membrane
materials
for fuel cells applications.
54
CA 02527445 2005-11-18
Experimental Part other SPAEENs Containing Napthalene Structure
Materials.
[0090] 6,7-Dihydroxynaphthalene-2-sulfonate sodium salt (2,3-DHNS-6), and 2,7-
dihydroxynaphthalene-3,6-sulfonate disodium salt (2,7-DHNS-3,6) were purchased
from Rintech, Inc. and recrystallized from deionized water. Dimethylsulfoxide
(DMSO) and N-methyl-2-pyrrolidone (NMP) (Aldrich) were vacuum distilled prior
to
use. 4,4'-Biphenol (Aldrich) was sublimated before usage for purification. All
other
chemicals obtained commercially were reagent grade and used as received.
Copolymerization.
[0091] In a typical reaction, 10 mmol 2,6-BFBN, 4 mmol 2,3-DHNS-6, 6 mmol 4,4'-
biphenol, and 15 mmol K2CO3 were added into a three-neck equipped flask with a
magnetic stirrer, a Dean-Stark trap, and an argon gas inlet. Then, 10 mL DMSO
(or
NMP) and 10 mL chlorobenzene were charged into the reaction flask under an
argon
atmosphere. The reaction mixture was heated to around 130 C. After dehydration
and
removal of chlorobenzene, the reaction temperature was increased to about 160
C.
When the solution viscosity had obviously increased, the mixture was cooled to
100 C and coagulated into a large excess of water or ethanol with vigorous
stirring.
P-SPAEEN-40 was resulted, where n (40) refers to the 2,3-DHNS-6 content of
aromatic phenol monomers. The product was washed thoroughly with water or
ethanol several times.
Copolymer Analysis and Measurement.
[0092] 1H NMR spectra were obtained on a Varian Unity Inova NMR spectrometer
operating at a proton frequency of 399.95 MHz. Deuterated dimethylsulfoxide
(DMSO-d6) was the NMR solvent and tetramethylsilane (TMS) was used as the
chemical shift reference (0 ppm). IR spectra were measured on a Nicolet 520
Fourier
transform spectrometer with membrane film samples in a diamond cell.
[0093] A TA Instruments thermogravimetric analyser (TGA) instrument model 2950
was used for measuring Tits. Polymer samples for TGA analysis were preheated
to
150 C at 10 C/min under air atmosphere and held isothermally for 40 min for
moisture removal. Samples were then heated from 90 C to 750 C at 10 C/min for
Td
CA 02527445 2005-11-18
measurement. A TA Instruments differential scanning calorimeter (DSC) model
2920
calibrated with tin at 231.93 C and lead at 327.50 C was used for measuring
Tgs.
[0094] Intrinsic viscosities were determined using an Ubbelohde viscometer for
NN-
dimethylacetamide (DMAc) solutions of copolymer at 30 C.
Preparation of Membrane Films.
[0095] An amount of 0.8 to1.0 g sulfonated copolymer in the sodium salt form
was
dissolved in 20 mL of DMAc and filtered. The filtered solution was poured onto
a
leveled glass plate having a circular glass retaining wall and dried at about
40 C for
about one day. The acid form (SPAEENH-n) membrane films were obtained by
immersing corresponding sodium form SPAEEN-n membrane films in 2 N HZSO4 for
24 h at room temperature, followed by deionized water for 24 h during which
time the
water was changed several times. The thickness of the membrane films was in
the
range of 40 to 70 pm.
Water Uptake Content Measurement and Swelling Ratio
[0096] The sample films were soaked in deionized water for 24 h at determined
temperatures. The membrane films were then dried at 80 C for 24 h. Weights of
dry
and wet membranes were measured. The water uptake content was calculated as
Uptake content (%) = Mve` - CWr~ X100
oily
Where ak, and yet are the masses of dried and wet samples respectively. The
swelling ratio was calculated from films 5-10 cm long as:
Swelling ratio (%) = I.,` -ldry x100
ldry
Where ldr, and lwet are the lengths of dry and wet samples respectively.
Proton Conductivity
[0097] The proton conductivity measurements were performed on SPAEENH
membrane films by AC impedance spectroscopy over a frequency range of 1-107 Hz
with oscillating voltage 50-500 mV, using a system based on a Solartron 1260
gain
phase analyzer. Prior to the proton conductivity measurements, membranes were
immersed in 98 C water for 36 h to hydrate. To avoid excessive swelling, P-
56
CA 02527445 2005-11-18
SPAEEN-H60 and D-SPAEENH-40 were pretreated in 80 C water instead of 98 C
and other D-SPAEENH copolymers were only immersed in water at room
temperature. A 20 x 10 mm membrane sample was clamped between two stainless
steel electrodes in a temperature controlled cell open to the air by a pinhole
where the
sample was equilibrated at 100%RH at ambient atmospheric pressure. The proton
conductivity ((T) of the samples in the longitudinal direction was calculated
from the
impedance data, using the relationship a = d/RS where d and S are the
thickness and
face area of the sample respectively and R was derived from the low intersect
of the
high frequency semi-circle on a complex impedance plane with the Re (Z) axis.
The
impedance data were corrected for the contribution from the empty and short
circuited
cell.
Results and Discussion
Synthesis and Characterization of SPAEEN copolymers
[0098] Copolymers from two investigated sulfonated naphthalene-based
bisphenols:
2,3-DHNS-6 and 2,7-DHNS-3,6 were denoted as P-SPAEEN and D-SPAEEN
respectively, with the prefixes of P- and D meaning sulfonic acid groups
located,
pendant on a phenyl ring and di-substituted accordingly. The above results on
M-
SPAEENs are combined for comparison. SPAEEN copolymers were synthesized by
copolymerization of 2,6-DFBN, one of the three sulfonated naphthalene-based
bisphenols under typical polymerization reaction conditions as shown in Scheme
1.
For these syntheses, the solvent selection is a key for high molecular weight.
N-
methyl-2-pyrrolidone (NMP) was initially used as solvent for polymerization,
and all
sulfonate content D-SPAEENs completely precipitated before high polymers were
obtained. P-SPAEEN-20 and P-SPAEEN-30 were successfully prepared in NMP.
However, high sulfonate containing P-SPAEENs were not obtained with molecular
weights high enough for good quality membranes due to the precipitation of
polymers
from polymerization solutions which prevented the polymer chains from further
propagation. DMSO was used instead for the polymerizations of the other
copolymers
and high molecular weights were obtained due to the improved dissolvability
with
DMSO for copolymers. However, long reaction time was required for
polymerizations with high content of more sterically hindered 2,3-DHNS-6 as
monomer. In contrast, D-SPAEEN copolymers were most easily to be obtained,
which is attributed to its stretched structure. Nevertheless, 2,3-DHNS-6 based
57
CA 02527445 2005-11-18
SPAEEN copolymers showed decreasing viscosities with increasing SC values and
high molecular weight 2,8-DHNS-6 based SPAEEN copolymers are limited with SC
values lower than 0.6 due to their angled structures, which made their polymer
chains
more entangled and more difficult to propagate. Properties of the resulting
polymers
such as theoretical equivalent weight per sulfonate group (EW), ion exchange
capacity (IEC) of SPAEEN-X copolymers are summarized in Table 14
58
CA 02527445 2005-11-18
Table 14 Syntheses of SPAEEN copolymers
4,4'-BP 2,6-DFBN sulfonated [q]a EW (IEC) expected SC SC from
biphenol expected 'H-NMR
data
Polymer mmol mmol Mmol dL/g g/molSO3 (Meq g-)
PAEEN-0 5 5 0 - - 0 0
m-SPAEEN-20 8 10 2 0.85b 1480 (0.68) 0.2 0.19
m-SPAEEN-30 7 10 3 1.0 b 1005 (1.0) 0.3 0.29
m-SPAEEN -40 6 10 4 0.62 b 767 (1.3) 0.4 0.38
m-SPAEEN -50 5 10 5 0.9 b 625 (1.6) 0.5 0.50
m-SPAEEN -60 4 10 6 0.82 b 530(l.9) 0.6 0.57
P-SPAEEN-20 8 10 2 2.3 1480 (0.68) 0.2
P-SPAEEN-30 7 10 3 1.9 1005 (1.0) 0.3
P-SPAEEN-40 6 10 4 1.3 767 (1.3) 0.4
P-SPAEEN-50 5 10 5 1.9 625 (1.6) 0.5
P-SPAEEN-60 4 10 6 1.6 530 (1.9) 0.6 0.58
P-SPAEEN-70 3 10 7 1.5 462 (2.2) 0.7
P-SPAEEN-80 2 10 8 0.87 411 (2.4) 0.8
D-SPAEEN-10 9 10 1 - 1492 (0.67) 0.2 -
D-SPAEEN-20 8 10 2 2.5 780 (1.3) 0.4
D-SPAEEN-30 7 10 3 1.4 542 (1.8) 0.6
D-SPAEEN -40 6 10 4 1.5 423 (2.4) 0.8
D-SPAEEN -50 5 10 5 2.1 352 (2.8) 1.0
D-SPAEEN -60 4 10 6 2.0 305 (3.3) 1.2
a Measured at 30 C in DMAc.
b Measured at 25 C in DMAc.
59
CA 02527445 2005-11-18
P-SPAEEN-20 to P-SPAEEN-60 have good solubility in DMAc, NMP, DMSO and
N,N-dimethylformamide (DMF) and could readily be cast into membrane films.
[0099] The chemical structures of SPAEEN copolymers were characterized by FT-
IR
and 1H NMR. In FT-IR spectra of P-SPAEEN and D-SPAEEN copolymers (Figure
24), characteristic bands of the aromatic sulfonate salt are observed at 1036
and 1107
cm -1 for P-SPAEEN copolymers and 1080 cm -1 for D-SPAEEN copolymers and the
intensity of these characteristic absorption bands increase with SC. The
characteristic
stretching band of nitrile groups was observed at 2239 cm-' for P-SPAEEN
copolymers and 2245 cm -1 for D-SPAEEN copolymers. The absorption bands at
around 1211 and 1254 cm 1 are assigned to phenoxy groups. The absorption bands
at
1463 and 1495 cm 1 were assigned to phenyl ring and the bands at 1589 and 1604
cm
are attributed to C=C stretching and only slight differences appeared with the
bands
positions for different copolymers.
[00100] The structural properties of the synthesized polymers were also
studied
by liquid phase 1H NMR spectroscopy with DMSO-d6 as the solvent and reference
material. Take P-SPAEEN copolymers as an example. Figure 25 shows three
spectra
of the aromatic protons for the sulfonated P-SPAEEN-20, 40 and 60 in sodium
form.
As expected, the spectra have some similarities with the SPAEEN spectra
published
before [51]; for example, the chemical shift for the benzonitrile and biphenol
segments are nearly the same. The five proton signals from the naphthalene
sulfonate
group appear at high frequencies due to the deshielding ring current effect of
the
adjacent benzene rings (H-a,b,d,e) and also due to the electron withdrawing
sulfonate
group (H-c,d). Two-spin systems Hb-c, Hf-g, Hg-h and Hi-j were identified by
2D
COSY. The 1H NMR spectra were used to corroborate SCs with the expected SCs
from feed ratios by simple comparison of the intensities of the benzonitrile H-
f,b
(2.OOH) with the H-j signals of the biphenol segment. As an example, the
intensity of
H-j signal for P-SPAEEN-60 was 1.71H for the four H-j protons resulting in
0.42
biphenol unit hence 0.58 naphthalene sulfonate. 1H NMR confirmed that SC
values
are in the reasonable scale of expected values.
[00101] Thermal properties including glass transition temperatures (Tgs) and
thermal decomposition temperatures in the air (Tds) of SPAEEN copolymers were
CA 02527445 2005-11-18
investigated by TGA and DSC analyses and illustrated in Figures 26 and 27
respectively. Similar to m-SPAEEN copolymers, P-SPAEEN and D-SPAEEN
copolymers showed high thermal stabilities. Sodium form copolymers lost weight
at
around 430 C due to the degradation of polymer main chain, and acid form
copolymers showed earlier weight loss at around 264-470 C differently
depending on
the sulfonic acid content due to the cleavage of -SO3H. All SPAEEN copolymers
show sufficient thermo-oxidative stabilities since PEMs are mainly used in air
at
about 100 C. For DSC measurements, copolymers in both sodium and acid forms
were initially heated at a 10 C/min rate to about 10 C below their
decomposition
temperatures based on TGA curves. After cooling, they were reheated at a
heating
rate of 10 C/min to their decomposition temperatures, and the data from the
second
scan were reported for Tgs. Figure 27 shows that Tgs of P-SPAEEN copolymers
increase with SC values in both sodium and acid forms. Sodium form copolymers
have higher Tgs than their corresponding acid forms. In addition, the glass
transitions
become less sensitive to DSC detection with the increase in SC values and the
Tgs of
P-SPAEEN-60 were not detected in both sodium and acid forms. Tgs of D-SPAEEN
were not as sensitive to DSC detection as those of m-SPAEEN and D-SPAEEN
copolymers. Table 15 summarizes the observed TT15% , onset weight loss
temperatures
(Td) data and Tgs of SPAEEN copolymers.
61
CA 02527445 2005-11-18
Table 15 Thermal properties of SPAEEN copolymers
Polymer Tg ( C) Tds%a ( C) Td ( C)
extrapolated onset for first
weight loss
Na form Acid form Na form Acid form Na form Acid form
PAEEN-0 224 - -
m-SPAEEN-20 233 230 477 390 521 264
m-SPAEEN-30 261 244 468 361 504 290
m-SPAEEN -40 310 247 449 352 491 290
m-SPAEEN -50 313 260a 467 341 489 297
m-SPAEEN -60 336 - 463 344 512 310
P-SPAEEN-20 274 261 461 472 486 345
P-SPAEEN-30 296 285 432 363 456 327
P-SPAEEN-40 317 311 469 365 462 328
P-SPAEEN-50 349 324 471 363 459 332
P-SPAEEN-60 ND ND 468 356 452 323
D-SPAEEN-20 228 ND - 370 415
D-SPAEEN-30 224 ND - 330 - 386
D-SPAEEN -40 ND ND - 301 - 328
D-SPAEEN -50 ND ND - 320 - 305
D-SPAEEN -60 ND ND - 335 - 331
ND: not detected
Not measured.
62
CA 02527445 2005-11-18
[00102] This table shows that their Tgs vary from 224 to 349 C in sodium form
and from 224 to 385 C in acid form. Since some acid form copolymers, such as P-
SPAEENH-50, were observed with Tgs around their decomposition temperature, the
reported value are only an approximation. In addition, it is noteworthy that D-
SPAEEN was detected with T,, at 362 C for the first scan, close to the T,, of
366 C for
unsulfonated PAEEN in its first scan. The combination of Tgs and T(1 values
leads us
to conclude that all P-SPAEENH copolymers have good thermal stabilities for
PEM
usages.
[00103] Properties of membranes
[00104] In PEMFCs and DMFCs, PEMs are prepared into membrane electrode
assemblies (MEA)s and serve as separators for the reactants, catalysts support
and provide
the required ionic pathway between the anode and the cathode. Therefore, their
properties
such as mechanical properties and thermal stability, water management, proton
conductivity
and adhesive ability to catalyst and other additives are crucial for the fuel
cell performance.
The thermal stabilities have been described above. The introduction of highly
polar nitrite has
been anticipated to promote adhesion of the polymers to many substrates via
interaction
with other polar chemical groups such as those of acid fillers in composite
membranes
or catalyst layer of MEA. Our ongoing work on MEA has confirmed that catalyst
layer adheres well to the SPAEENH copolymers, supporting the claims that
nitriles
facilitate catalyst layer binding through polar interactions.
[00105] As disclosed above, m-SPAEEN based films with SC up to 0.6
maintained good shapes and mechanical strengths in both dry and fully hydrated
states. P-SPAEEN-20 and P-SPAEEN-30 were cast into flexible films. P-SPAEEN-
40 and P-SPAEEN-50 were cast into robust films. The film of P-SPAEEN-60 was
fragile and P-SPAEEN-70 was brittle when completely dehydrated. The change in
strength appearances of membrane films may be related to both the contents of
entangled ortho-biphenol and molecular weights of polymers. However, all
hydrated
films of P-SPAEENs with sulfonate contents up to 60% are tough enough. All D-
SPAEEN copolymers with SC up to 0.6 are tough and flexible at dry states and
high
sulfonic acid containing D-SPAEEN copolymers swelled too much, or even
dissolved
in water especially at elevated temperatures and lost their mechanical
properties.
[00106] Researchers in McGrath's group found 25 that bisphenols have a
marked influence on water uptake of prepared sulfonated poly(aryl ether
sulfone)s
63
CA 02527445 2005-11-18
and attributed it to the hydrophobicities of bisphenols. In this
investigation,
bifluorobenzonitrile and 4,4'-biphenol form the hydrophobic part and with
exactly the
same structure in all SPAEEN copolymers. As the hydrophilic part, three
bisphenol
monomers all have naphthalene skeleton and sulfonate group and their
derivative
SPAEEN copolymers only differ from each other in the linkage and sulfonate
positions, thus they can be considered as large isomers. Thus the difference
in water
management has to be re-explained. Figure 28 shows that after 24 h immersion
in
water, P-SPAEENH copolymers absorbed water in the range of 4.0% to 168% and
increased their lengths by 1.6% to 49% depending on the 2,3-DHNS-6 content and
temperature and D-SPAEENH copolymers absorbed water higher than 8.8% and
increased their lengths by 2% till to dissolved in hot water. Their comparison
along
with 2,8-DHNS-6 based m-SPAEENH copolymers34 on water uptake and swelling
was listed in Table 16.
64
CA 02527445 2005-11-18
Table 16 Comparisons on swelling and conductivity of different polymers
Polymer EW expected Room temperature 80 C
g/molSO3 Swelling ratio Conductivity Swelling ratio Conductivity
(%) (S/cm) (%) (S/cm)
m-SPAEENH-50 625 6.2 3.5x10 10 6.4x10
m-SPAEENH-60 530 7.5 8.3x10-2 15 1.4x10-1
P-SPAEENH-50 625 9.1 3.9x10-2 13 8.8x10-2
P-SPAEEN-H60 530 16 7.9x10-2 22 1.5x101
D-SPAEENH-30 542 4.9 1.5x10-2 6.1 2.9x10-2
D-SPAEENH-40 423 19 8.0x10-2 39 1.5x10-1
SPAEEKKH-10021 575 18 - Excessively -
swollen
Nafion 117 1100 13 7.5x10-2 20 9.6x10-2
CA 02527445 2005-11-18
[00107] From Table 16, it can be seen that P-SPAEENH copolymers show
slightly higher water uptakes and swellings than m-SPAEENH copolymers at the
same EW values, especially at high sulfonic acid content. Since D-SPAEENH
copolymers have slight higher EW values than corresponding m-SPAEENH and P-
SPAEENH copolymers with same SC values. Their comparisons are not directly.
However, it still can be seen that D-SPAEENH copolymers have the lowest water
uptake and swelling ratios among the three kinds of SPAEENHs at similar EW
values.
Compared with m-SPAEENH copolymers, comb-like branched naphthalene-sulfonic
acid segment in P-SPAEENH copolymers show a kinetic diameter of 9.1 A (Figure
29) estimated by ACD/ChemSketch calculation, bigger than the value of 6.0 A of
angled naphthalene-sulfonic acid segment in m-SPAEENH. So, P-SPAEENH
copolymers have larger interchain spaces for water molecules in hydrophilic
domains
in the membranes and absorbed more water than m-SPAEENH copolymers at the
same EW values. Among the three SPAEENH copolymers, D-SPAEENHs are most
stretching and naphthalene-sulfonic acid segment has the smallest kinetic
diameter of
. 5.1 A. Thus, D-SPAEENH copolymers have the smallest interchain spaces and
smallest free volume for water molecules and therefore show lowest water
uptakes
and swellings. The P-SPAEENH copolymers with SC values up to 0.5 swelled less
than 25% and are dimensionally stable enough up to 100 C and with SC values up
to
0.6 they are dimensionally stable up to 80 C; D-SPAEENH-30 is dimensional
stable
up to 100 C and D-SPAEENH-40 swelled too much at 80 C.
[00108] Compared with our previously prepared sulfonic acid-containing
poly(aryl ether ether ketone ketone)s (SPAEEKKH) based on 2,3-DHNS-6 (Table
16),
P-SPAEENH show much lower water uptakes and considerably improved
dimensional stabilities. SPAEEKKH (EW 575 g/molSO3) excessively swelled at 80
C
and dissolved in 100 C hot water, however, P-SPAEEN-H60 (EW 530 g/mo1SO3)
only showed water uptake of 68% and 168% and swelling of 22% and 49% at 80 and
100 C respectively. These phenomena are explained by taking the molecular
structures of the different polymers into account. Compared with ketone
structures in
SPAEEKKH polymer main chain, nitrile groups have stronger polarity, and are
pendant on aromatic rings of P-SPAEENH copolymers; they enhance the
66
CA 02527445 2005-11-18
intermolecular interaction of P-SPAEENH copolymers and enhance the hydrophobic
network structure, consequently enhancing the dimensional stability of the
membrane
films. TEM was tried to be used to characterize the percolating network
structures and
explain the behaviors of membrane films, according to reference articles25'
26, with
their electro aggregation of different chain segments and phase separation,
which
forms ionic pathways. Unfortunately, film samples under TEM were fully
dehydrated
and not ionic channels were observed.
[00109] The temperature dependence of proton conductivity of P-SPAEENH
and D-SPAEENH copolymers together with Nafion 117 for comparison is plotted in
Figure 30. It shows that the proton conductivities of P-SPAEENH copolymers
increase with both SC and temperature and P-SPAEENH copolymers with SC values
of 0.4 to 0.6 all show proton conductivities higher than 10-2 S/cm from room
temperature to 100 C, a lowest value of practical interest for use as PEMs in
fuel cells.
P-SPAEENI-I-50 and P-SPAEENH-60 show high proton conductivities comparable to
Nafion117, ranging from 3.9x10-2 to 2.0x10-1 S/cm at similar water uptake and
swellings. D-SPAEENH copolymers show increase in proton conductivities with
both
temperature and SC, and for SC of 1.2, proton conductivities decrease again
due to its
excessive swelling. D-SPAEENH-30 and with much higher SC values show proton
conductivities higher than 10-2 S/cm. Comparisons on swellings and proton
conductivities of SPAEENH copolymers are listed Table 16. It can be seen that
P-
SPAEENH films normally show somewhat higher proton conductivities than
corresponding m-SPAEENH films at the same SC values and same temperatures, in
accordance with their higher water uptake and swellings; and their increase in
proton
conductivity with SC and temperature is parallel to their swelling. As
aforementioned,
it is the result of their different kinetic diameters. However, although D-
SPAEENHs
have small kinetic diameters, the films show smaller increase in proton
conductivities
with their swellings than m-SPAEENHs and P-SPAEENHs. It may be the effect
result of ether linkage in polymer chains on sulfonic acid groups. In m-
SPAEENH and
P-SPAEENH polymer chains, sulfonic acid groups are located on meta position to
ether linkage or on a different benzene ring from ether linkage and are less
deactivated by electro-donating ether group. However, in D-SPAEENH polymer
67
CA 02527445 2005-11-18
chain, sulfonic acid are located ortho to ether linkage and deactivated.
Proton
conductivity depends on the concentration of protons in membrane, their
ability to
dissociate and their speed of diffusion. Thus, for similar sulfonic acid
concentration
membranes at a determined temperature, the different dissociation abilities of
sulfonic
acid and diffusion of protons will result in their different conductivities.
In D-
SPAEENH, although more water is absorbed, their ether-deactivated sulfonic
acid
groups are more difficult to dissociate and hold water more tenaciously and
lower the
overall transport rates of protons. As a result, the proton conductivities are
lowered.
In addition, different distribution of sulfonic acid groups in polymer chains
may also
contribute to their different conductivities
Conclusion
[00110] Three series of naphthalene-based copoly(aryl ether ether nitrile)s
containing sulfonic acid groups have been successfully prepared from
commercially
available monomers via one-step polycondensation reactions. They were solvent-
cast
into dry membranes, which are from flexible, robust to fragile. All SPAEENH
copolymers have good thermal stabilities with decomposition temperatures
higher
than 264 C and Tgs higher than 224 C. P-SPAEENH and m-SPAEENH films show
lower water uptakes and swellings but more obvious increase in proton
conductivities
with water content than D-SPAEENH films. P-SPAEENH films show somewhat
higher swelling and proton conductivities than in-SPAEENH films due to their
increased intermolecular distance. The meta linkage or iso-ring connection of
sulfonic
acid groups to ether linkage makes them less deactivated in sulfonic acids and
lead to
higher proton conductivities of P-SPAEENH and m-SPAEENH films than D-
SPAEENH films. P-SPAEENH and m-SPAEENH films show proton conductivities
close to or higher than those of Nafion 117 at similar water uptake and
swelling
values. The combination of inexpensive monomers, high thermal stability, low
dimensional swelling, and high proton conductivity makes rn-SPAEENH-50, m-
SPAEENH-60, P-SPAEENH-50 and P-SPAEEN-60 attractive as PEM materials for
fuel cells applications up to 100 C or 80 C respectively.
[00111] Example: Test of MEA containing SPAEEN-60, compared to Nafion
115 membranes using GDEs w/Nafion ionomers, at 30 C cell temperature, 100 %
68
CA 02527445 2005-11-18
humidity H2/air and 0.100 L/min flow, with anode and cathode GDEs: 0.25 mg
pt/em 2(20 wt% Pt/C), 30 wt%Nafion, 5 cm2 active area. A graph of the results
is
shown in Figure 31.
69
CA 02527445 2005-11-18
References
[1] Roziere, J.; Jones, D. J. Annu. Rev. Mater. Res. 2003, 33, 503-55.
[2] Mecerreyes, D.; Grande, H.; Miguel, 0.; Ochoteco, E.; Marcilla, R.;
Cantero, I.
Chem. Mater. 2004, 16, 604-607.
[3] Yang, Y.; Shi, Z.; Holdcroft, S. Macromolecules 2004, 37, 1678-1681.
[4] Wang, 1.; Meng, Y. Z.; Wang, S. J.; Shang, X. Y.; Li, L.; Hay, A. S.
Macromolecules 2004, 37, 3151-3158.
[5] Ueda, M.; Toyota, H.; Ouchi, T.; Sugiyama, J.; Yonetake, K.; Masuko, T.;
Teramoto, T. J. Polym. Sci., Part A: Polym. Chem. Ed. 1993, 31, 853-858.
[6] Genies, C.; Mercier, R.; Sillion, B.; Cornet, N.; Gebel, G.; Pineri, M.
Polymer
2001, 42, 359-373.
[7] Miyatake, K.; Hay, A. S.; J. Polym. Sci., Part A: Polym. Chem. Ed. 2001,
39,
3211-3217.
[8] Wang, F.; Hickner, M.; Kim, Y. S.; Zawodzinski, T. A.; McGrath, J. E. J.
Membrane Sci. 2002, 197, 231-242.
[9] Faure, S.; Cornet, N.; Gebel, G.; Mercier, R.; Pineri, M.; Sillion, B.; in
Proceedings of the Second International Symposium on New Materials for Fuel
Cell and Modern Battery Systems, (Savadogo, 0.; Roberge, P. R. eds.),
Montreal, Canada, 1997, July 6-10, p. 818.
[10] Nolte, R.; Ledjeff, K.; Bauer, M.; Mulhaupt, R. J. Membrane Sci. 1993,
83, 211-
220.
[11] Kobayashi, T.; Rikukawa, M.; Sanui, K.; Ogata, N. Solid State Ionics
1998, 106,
219- 225.
[12] Glipa, X.; Haddad, M. E.; Jones, D. J.; Roziere, J. Solid State Ionics
1997, 97,
323-331.
[13] Kerres, J.; Cui, W.; Reichle, S. J. Polym. Sci., Part A: Polym. Chem. Ed
1996,
34, 2421-2438.
[14] Soczka-Guth, T.; Baurmeister, J.; Frank, G.; Knauf, R. International
Patent WO
1999, 99/29763.
[15] Kim, Y. S.; Dong, L.; Hickner, M. A.; McGrath, J. E. Macromolecules 2003,
36,
6281.
[16] Harrison, W. L.; Wang, F.; Mecham, J. B.; Bhanu, V. A.; Hill, M.; Kim, Y.
S.;
McGrath, J. E. J. Polym. Sci. Part A: Polym. Chem. 2003, 41, 2264.
CA 02527445 2005-11-18
[17] Gao, Y.; Robertson, G. P.; Guiver, M. D.; Jian, X. J. Polym. Sci. Part A:
Polym.
Chem. 2003, 41, 497.
[18] Gao, Y.; Robertson, G. P.; Guiver, M. D.; Jian, X.; Mikhailenko, S. D.;
Wang,
K.; Kaliaguine, S. J. Polym. Sci. Part A: Polym. Chem. 2003, 41, 2731.
[19] Gao, Y.; Robertson, G. P.; Guiver, M. D.; Jian, X.; Mikhailenko, S. D.;
Wang,
K.; Kaliaguine, S. J. Membrane Sci. 2003, 22 7,39.
[20] Xiao, G.; Sun, G.; Yan, D.; Zhu, P.; Tao, P. Polymer 2002, 43, 5335.
[21] Xiao, G.; Sun, G.; Yan, D. Macromol. Rapid Commun. 2002, 23, 488.
[22] Gao, Y.; Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.; Li, X;
Kaliaguine,
S. Macromolecules 2004, 37, 6748
[23] Xing, P; Robertson, Robertson, G. P.; Guiver, M. D.; Mikhailenko, S. D.;
Kaliaguine, S. Macromolecules 2004, 37, 7960
[24] Cotter, R. J. Engineering Plastics: Handbook of Polyarylethers; Gordon
and
Breach Science Publishers S.A., Basel Switzerland, 1995.
[25] Wang, S.; McGrath, J. In Synthetic Methods in Step-Growth Polymers;
Rogers,
M. E.; Long, T.E. (Eds.); Hoboken, N. J. Wiley, 2003; Chapter 6.
[26] Kricheldorf, H. R.; Meier, J.; Schwarz, G. Macromol. Chem., Rapid Commun.
1987, 8, 529.
[27] Kricheldorf, H. R.; Berghahn, M. Macromol. Chem., Rapid Commun. 1991, 12,
529.
[28] Kricheldorf, H.R.; Garaleh, M.; Schwarz, G. J. Polym. Sci. Part A: Polym.
Chem. 2003, 41, 3838.
[29] Mohanty, D. K.; Waistrom, A. M.; Ward, T. C.; McGrath, J. E. Polym.
Prepr.1986, 27, 147.
[30] Mohanty, D. K.; Hedrick, J. L.; Gobetz, K.; Johnson, B. C.; Yilgor, I.;
Yilgor, E.;
Yang, R.; McGrath, J. E. Polym. Prepr.1982, 23, 284.
[31] Blinne, G.; Bender, H.; Neumann, P. US Patent 1986, 4567248.
[32] Hoehn, H. H.; Richter, J. W. US Patent 1975, 3899309.
[33] Heath, D. R.; Wirth, J. G. US Patent 1973, 3730946.
[34] Murakami, T. US Patent 1990, 4972016.
[35] Krizan, T. D. US Patent 1992, 5080698.
[36] Matsuo, S.; Murakami, T.; Takasawa, R. US Patent 1987, 4703104.
[37] Matsuo, S.; Murakami, T. US Patent 1987, 4663427.
[38] Matsuo, S.; Murakami, T. US Patent 1987, 4640975.
71
CA 02527445 2005-11-18
[39] Sakaguchi, Y.; Kitamura, K.; Nagahara, S.; Takase, S. Polym. Prepr. 2004,
45,
56.
[40] Sumner, M. J.; Harrison, W. L.; Weyers, R. M.; Kim, Y. S.; McGrath, J.
E.;
Riffle, J. S.; Brink, A.; Brink, M. H. J. Membrane Sci. 2004, 239, 199.
[41] Meng, Y. Z.; Tjong, S. C.; Hay, A. S.; Wang, S. J. J. Polym. Sci. Part A:
Polym.
Chem. 2001, 39, 3218.
[42] Besso, E.; Eisenberg, A. Properties and structures of ionomers and
ionomeric
membranes, in Proceedings of the Symposium on Membranes and Ionic and
Electronic Conducting Polymers, (Yeager, E. B.; Schumm, B.; Mauritz, Jr. K.;
Abbey, K.; Blankenship, D.; Akridge, J. eds.), Cleveland, United States, 1982,
May 17-19, p. 4.
[43] Gieke, T. D.; Munn, G. E.; Wilson, F. C. J Polym Sci, Polym Phys 1981, 19
1687.
[44] Zawodzinski, T. A.; Derouin, C. ; Raszinski, S. ; Sherman, R. J.; Smith,
V. T.;
Springer, T. E.; Gottesfeld, S. J. Electrochem. Soc. 1993, 140, 1041.
[45] Jian, X. G.; Meng, Y. Z.; Zheng, H.B. Chin. Pat. 93109180.2 (1993).
[46] Jian, X. G.; Meng, Y. Z.; Zheng, H.B. Chin. Pat. 93109179.9 (1993).
[47] Yoshida, S.; Hay, A. S. Macromolecules, 1995, 36, 2579-258 1.
[48] Ding, Y.; Hlil, A. R.; Hay, A. S. Polym Prepr, 1997, 38, 187-188.
[49] Kim YS, Wang F, Hickner M, Mccartney S, Hong YT. Harrison W,
Zawodzinski TA, McGrath JE. J Polym Sci Part B: Polym Phys 2003; 41: 2816.
[50] Kim, Y. S.; Dong, L.; Hiekner, M. A.; McGrath, J. E. Macromolecules 2003,
36,
6281.
[51] Gao Y, Robertson GP, Guiver MD, Mikhailenko SD, Li X, Kaliaguine S.
Macromolecules 2005; 38: 3237
72