Sulfo-Phenylated Polyphenylenes Containing Sterically Hindered Pyridines

Article abstract of DOI:10.1021/acs.macromol.8b02289

We systematically investigated the effect of incorporating a sterically hindered pyridyl group into a sulfo-phenylated polyphenylene to control the polymer's physicochemical properties through acid-base interactions. Homopolymers with similar molecular weights and comparable structures that vary by only one atom (N- vs C-) per repeat unit along the polymer chain were prepared. Compared to a non-pyridyl reference membrane, incorporation of a pyridyl group improves the oxidative stability against free radicals, increases the elongation at break to 55% (from 37%), and enhances the thermal stability to 326 °C (from 246 °C). In an accelerated fuel cell degradation test, polymeric membranes containing the sterically encumbered pyridyl unit exhibited exceptional stability (0.16 mV h^-1 degradation rate over 1000 h) and retained ～80% of their peak power density over this time.

Full text of DOI:10.1021/acs.macromol.8b02289

Article
pubs.acs.org/Macromolecules
Cite This: Macromolecules XXXX, XXX, XXX−XXX
Sulfo-Phenylated Polyphenylenes Containing Sterically Hindered
Pyridines
Shaoyi Xu,† Michael Adamski,† Miho Killer,† Eric M. Schibli,‡ Barbara J. Frisken,‡
and Steven Holdcroft*,†
‡
†Department of Chemistry and Department of Physics, Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6,
Canada
S
* Supporting Information
ABSTRACT: We systematically investigated the effect of incorporating a sterically hindered pyridyl group into a sulfo-
phenylated polyphenylene to control the polymer’s physicochemical properties through acid−base interactions. Homopolymers
with similar molecular weights and comparable structures that vary by only one atom (N− vs C−) per repeat unit along the
polymer chain were prepared. Compared to a non-pyridyl reference membrane, incorporation of a pyridyl group improves the
oxidative stability against free radicals, increases the elongation at break to 55% (from 37%), and enhances the thermal stability
to 326 °C (from 246 °C). In an accelerated fuel cell degradation test, polymeric membranes containing the sterically
encumbered pyridyl unit exhibited exceptional stability (0.16 mV h−1 degradation rate over 1000 h) and retained ∼80% of their
peak power density over this time.
example, Guiver and co-workers prepared poly(benzothiazole-
co-benzimidazole)-based PEMs that possessed low dimensional
swelling and high mechanical strength.12 Kallitsis et al.
reported an aromatic polyether bearing pyridine group that
retained its mechanical integrity after exposure to Fenton’s
Reagent for 72 h.10 A series of basic polybenzimidazole (PBI)
blend polymers with low water uptake and enhanced oxidation
resistivity were investigated for high-temperature PEMs by
Kerres et al.3,13
Despite the aforementioned, sulfonated hydrocarbon-based
polymers for PEMs incorporating N-heterocycle moieties
usually suffer from reduced proton conductivity due to the
Brønsted−Lowry acid−base neutralization of dissociated
protons. The N-heterocycle groups (e.g., benzimidazole,
pyridine, pyrazole, triazole, and imidazole) commonly
employed in these acid−base polymer systems can be readily
neutralized by available protons from nearby sulfonic acid
groups to form the conjugate acid.4,9,14−17 The stronger the
1. INTRODUCTION
A wide variety of hydrocarbon polymers functionalized with
sulfonic acid have been investigated for use in proton exchange
membrane fuel cells (PEMFCs), redox-flow batteries, and
PEM electrolyzers.1−4 Among these, sulfonated polypheny-
lenes are attracting increased attention due to their low gas
permeability, high thermal and chemical stability, and
mechanical robustness.5−8 However, this class of polymer
requires relatively high sulfonic acid content to achieve a
percolation network for proton and water transport and often
swells to much greater extents compared to incumbent PFSA
ionomer membranes. For instance, the volumetric expansion of
reported sulfonated phenylated polyphenylene (sPPP-H+)
homopolymers in water can be as high as 300 vol % at
ambient temperature, which limits their useful integration into
electrochemical devices.5,6
It has been well-documented that the incorporation of
monomer units containing basic N-heterocycle groups in a
sulfonated polymer can serve to reduce swelling due to the
creation of intermolecular acid−base interactions.9,10 The
incorporation of basic N-heterocycle groups is also reported to
enhance oxidative durability and thermal stability.9−12 For
Received: October 24, 2018
Revised: March 6, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Synthesis of Sulfo-Triphenylated Pyridylpolyphenylenes (sTPPyPP-H+) and Sulfo-Triphenylated
Phenylpolyphenylenes (sTPPPP-H+)
base, the stronger the proton is bound to the resulting
conjugate acid, which reduces the extent of dissociate of the
proton and reduces proton conduction.18−20 As a result, their
utility in solid polymer electrolyte electrochemical devices is
compromised. Thus, the addition of basic N-heterocycle
groups to a sulfonated aromatic polymer creates a paradox:
the resulting acid−base interactions serve to reduce swelling
and enhance the mechanical properties of the solid polymer
electrolyte, but proton conductivity and their performance in
electrochemical devices are reduced.
To address this dilemma, physical blending and copoly-
merization have been employed to control the ratio of N-
heterocycle moieties to sulfonic acid groups in poly-
mers.4,13,15,16 In other examples, the number of acidic groups
attached to a polymer has been increased to excessive values to
compensate for the presence of acid-neutralizing basic groups
and to obtain a desired ion exchange capacity (IEC) based on
the concentration of “free” protons.15,19−25 Often, the
molecular structures of such polymers are not easily defined
as they possess block or random structures or consist of
physical mixtures of acidic and basic polymers. Because the
acid−base interactions require molecular proximity of the
corollary groups, a lack of control of the polymer structures
translates into a lack of control of the acid−base interactions
within polymer chains. Given this, it would be useful to
synthesize hydrocarbon polymers that bear both acidic and N-
heterocycle groups with more precise molecular control and
spatial resolution so that the effect of incorporating basic
groups into acidic hydrocarbon polymers could be clearly
distinguished.
triphenylated pyridine-containing polyphenylene (sTPPyPP)
and the corollary all-phenylene analogue, triphenylated
phenylpolyphenylene (sTPPPP). Notably, the pyridine in the
TPPy molecule is congested by three attached phenyl groups
to provide steric hindrance around the amine. This design was
incorporated to reduce the pKa for the conjugate acid of the
pyridine moiety and facilitate proton transport, the details of
which are discussed later. It is known that incorporating
hydrophobic phenyl groups into proton conducting polymers
decreases their ion-exchange capacity (IEC) and consequently
reduces the water uptake and swelling characteristics.26−29
Therefore, to compare the pyridine-containing polymer with a
non-pyridine-containing polymers, the linker TPP was
synthesized, and a the novel polymer, sTPPPP, was prepared.
Thus, sTPPyPP and sTPPPP are structurally identical except
for one N atom per repeat unit, as shown in Scheme 1. For
comparison, a first-generation sulfonated phenylated poly-
phenylene homopolymer prepared by our group in this series,
sPPP-H+, is also shown.6
2. EXPERIMENTAL SECTION
2.1. Synthesis. The details of materials, synthetic procedures, and
reaction yields can be found in the Supporting Information.
2.2. Membrane Preparation. A 6 wt % polymeric solution was
prepared by dissolving the polymer in DMSO. After filtering through
a sintered glass funnel, the solution was poured into a glass Petri dish
and placed for 12 h into an oven set to 86 °C. Once the solvent had
evaporated, water was added to the Petri dish to help desorb the
polymer membrane. The membrane was then soaked in 1 M H2SO4
for 24 h and subsequently in water for 3 × 12 h to remove residual
acid. Membranes were dried under vacuum at 80 °C overnight.
2.3. Water Uptake and Swelling. Membrane samples were
predried in a vacuum oven at 80 °C for 24 h. The weight of dry
To this end, we designed and synthesized two novel
sulfonated homopolymers (Scheme 1): a sterically hindered
B
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
samples Wdry was recorded. After immersion in Milli-Q water for 24 h,
fully hydrated membrane samples were blotted dry with an absorbent
paper and their weights Wwet were recorded. The experiment was
typically repeated five times to obtain average values. Water uptake
was then calculated by using eq 1:
2.6. Density of Polymers. The density of polymers ρ was
calculated via the equation
m
V
ρ =
(10)
where m is the mass of a membrane sample under the desired
condition and V is the volume of a membrane sample under the same
conditions.
Wwet − W
dry
water uptake (%) =
× 100%
W
dry
(1)
2.7. Mechanical Stress Tests. The membranes were cut into
barbell shapes using a standard ASTM D638-4 cutter. The mechanical
tests were performed under ambient conditions on a single column
system (Instron 3344 Series) using a crosshead speed of 5 mm min−1.
Five specimens were tested for each polymer’s mechanic properties.
The determined tensile strength, Young’s moduli, and elongation at
break were averaged over five samples. The error reported is the
standard deviation for five measurements.
The volume expansion was calculated via the equation
Vwet − V
dry
volume expansion (%) =
× 100%
V
dry
(2)
where Vwet and Vdry are the volume of wet membrane and dry
membrane, respectively, and V = length × width × thickness.
2.4. Ion-Exchange Capacity (IEC) Measurements. A small
piece of membrane was predried (vacuum oven at 80 °C for 24 h,
final weight Wdry) was immersed in a 25 mL solution of 2 M NaCl
solution for 24 h. The membrane was removed, and the displaced H+
ions were titrated to neutral pH using a 0.01 M NaOH standard
solution with phenolphthalein indicator. The IEC of the membrane
was calculated using eq 3:
2.8. TGA Measurements. Six milligrams of the polymer sample
was heated under Ar at 10 °C min−1 from 23 °C to reach 800 °C. The
weight loss was recorded by a PerkinElmer STA6000.
2.9. Oxidative Stability Test (Fenton’s Test). The oxidative
stability of polymers sTPPPP-H+ and sTPPyPP-H+ was studied by
immersing dry, preweighed (average 20 mg, three samples for each
hour) membrane samples in Fenton’s reagent (3% H2O2, 3 ppm Fe2+)
at 80 °C. Every 1 h, a membrane sample was retrieved, washed with
water (24 h × 3 times), and soaked in 1 M HCl (24 h × 3 times) and
again in water (24 h × 3 times). After the membrane was dried in a
vacuum oven at 80 °C for 24 h, the weight loss was calculated
according to the following equation:
0.01 × VNaOH
IEC =
W
dry
(3)
wherein VNaOH is the volume of the 0.01 M NaOH solution used for
titrating the proton concentration in the NaCl solution.
2.5. Hydration Number (λ), Analytical Acid Concentration
M0 − M1
+
weight loss (%) =
× 100%
[−SO3H], Proton Mobility μH , and Water Content. The
M0
(11)
hydration number (λ) was calculated via the equation
where M0 is the initial dry mass before Fenton’s test and M1 is the dry
mass after the degradation test.
1000 × water uptake
18.01 g/mol × IEC
where the molecular weight of H2O is 18.01 g mol−1.
The effective acid concentration in the whole membrane and the
water-rich have [−SO3H]mem and [−SO3H]water, respectively, were
calculated via the following equations:
hydration number (λ) =
(4)
2.10. Proton Conductivity Measurements. A small piece of the
membrane (1 cm × 0.5 cm) was assembled into a conductivity cell in
a two-probe configuration, as previously described.5 The cell was
placed in an Espec model SH-241 humidity chamber, which allowed
for temperature and humidity modulation. The ionic conductivity (σ)
was calculated from the resistance (R) and the membrane dimensions
using the equation
m
dryIECtitr
[−SO3H]mem
[−SO3H]water
=
=
Vwet membrane
(5)
L
σ =
(12)
R × A
m
dryIECtitr
mdryIECtitr
=
× 1 g/mL
where A is the cross-sectional area of membrane for the resistance
measurement and L is the distance between the two electrodes. R was
measured using an impedance analyzer (Solartron 1260 frequency
response analyzer (FRA) using ac frequencies ranging from 10 MHz
to 100 Hz).
Vwater absorbed
mwet − mdry
(6)
where mdry and mwet are the mass of dry and hydrated membrane,
respectively, IECtitr is the experimentally determined IEC obtained
from titration experiments, and Vwet is the volume of a wet membrane
sample.
2.11. MEA Fabrication. The catalyst inks for fuel cell tests were
prepared via the following method: Water was used to wet Pt/C
powder (TKK TEC-10E50E: 46.4 wt % Pt on graphitized carbon),
after which methanol was added to achieve the desired mixture
(MeOH:H2O = 3:1). Nafion D520 ionomer was added dropwise to
the stirring mixture to obtain 1% solids containing 30 wt % ionomer
and 70 wt % Pt/C. The resulting catalyst ink was then deposited via a
spray coater (Sono-Tek ExactaCoat SC) onto the membrane surface
at 60 °C for an electrode area of 5 cm2, yielding a catalyst coated
membrane (CCM): 0.4 mg cm−2 Pt/C was deposited on the cathode,
and 0.2 mg cm−2 Pt/C was deposited on the anode. The CCMs were
then sandwiched between conventional GDLs (Sigracet 24BC) and
laid between Teflon gaskets with thicknesses allowing for 20−30%
GDL compression by a torque wrench set to 50 in.-lb.
+
+
The proton mobility μH ,mem and μH ,water were calculated
according to the equations
+
σH
+
μH ,mem
=
=
F[−SO3H]mem
(7)
(8)
+
σH
+
μH ,water
F[−SO3H]water
+
where F is Faraday’s constant, σH is the proton conductivity at 30 °C
and 95% RH, and [−SO3H] is the effective acid concentration.
The water contentwet and water contentexpanded were calculated via
the following equation:
2.12. Fuel Cell Tests and Chemical Accelerated Stress Test
(AST). The fuel cell performances for membranes were evaluated
using a single cell fuel cell test station at 80 °C, 100%, 95%, 90%, and
50% RH, zero back-pressure, and O2/H2 or air/H2. After the above
evaluation for sTPPyPP-H+, the AST experiment was followed. The
conditions were: OCV was held was 90 °C, 30% RH at anode and
cathode, zero back-pressure under H2/Air for 1000 h. Then the cell
mwet − mdry
Vwet × 1 g/mL
water contentwet
=
(9)
where the mdry and mwet are the mass of dry and hydrated membrane
sample, respectively, and Vdry and Vwet are the volume of dry and
hydrated membrane sample, respectively.
C
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
Scheme 2. Synthesis of Diene Monomer BTCS-TEA
a
Reagents and conditions: (a) Pd(PPh3)2Cl2, CuI, TEA, 50 °C, overnight; (b) I2, DMSO, 155 °C, overnight; (c) dibenzyl ketone, KOH, reflux, 45
min; (d) TMS-OSO2-Cl, DCE, overnight; (e) n-BuOH, triethylamine, THF.
a
Scheme 3. Synthesis of Dienophiles TPPy and TPP
a
Reagents and conditions: (a) trimethylsilylacetylene, Pd(PPh3)2Cl2, CuI, TEA, 50 °C, overnight; (b) benzaldehyde, KOH, NH4OH, EtOH, RT, 3
days; (c) trimethylsilylacetylene, Pd(PPh3)2Cl2, CuI, TEA, RT, overnight; (d) 2-isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, n-BuLi, THF,
−78 °C, 1 h; (e) phenylboronic acid, Pd(PPh3)4, Na2CO3, THF/H2O, 90 °C, overnight; (f) 13, Pd(PPh3)4, Na2CO3, THF/H2O, 90 °C, overnight;
(g) K2CO3, Et2O/MeOH, RT, 1 h.
was reconditioned, and a polarization was performed to examine the
fuel cell recoverability after AST. The details of fuel cell experiments
can be found in the Supporting Information.
2.13. X-ray Scattering Experiment. We utilized a SAXSLab
Ganesha 300XL instrument for X-ray scattering measurements. This
instrument operates a copper anode source at 50 kV and 0.6 mA,
selecting Kα radiation with a wavelength of 1.54 Å−1. Measurements
were performed in four standard configurationsextremely small
angle, small angle, medium angle, and wide angleproviding a usable
q-range of 0.003−2.7 Å−1. All samples were allowed to equilibrate
with ambient conditions overnight, and hydrated samples were
immersed in deionized water for 1 h and then patted dry with a tissue.
Samples were measured in sealed paste cells. A transmission-corrected
D
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
+
background measurement of an empty sample cell was subtracted
pixel-by-pixel from each measurement.
The mid-q feature in the data from the hydrated samples was fit to
the broad peak model (BPM) as implemented in the NIST small-
angle scattering analysis package30
As outlined in Scheme 1, polymer sTPPPP-HNEt3 was
prepared by Diels−Alder polymerization of BTCS-HNEt3+ and
1
TPP at 180 °C for 72 h. H NMR analysis was employed to
1
confirm the successful polymerization. H NMR (DMSO-d6,
400 MHz) δ: 9.00 (s, 4H), 8.25−6.0 (m, 48H), 3.01 (q, 24H),
1.25 (t, 36H). The characteristic 1H resonance signals of
methyl, methylene, and aromatic protons of the polymers are
observed at δ 1.10, δ 3.05, and δ 6.00−8.35 ppm, respectively
(Figure S27). The integration ratios are consistent with the
A
C
I(q) =
+
+ B
qn
1 + (|q − q0|ξ)m
(13)
where q0, is the peak position, ξ is the correlation length, m and n are
Porod exponents, A and C are scale parameters, and B is the q-
independent background. q0 is inversely proportional to the
characteristic spacing D, which is commonly approximated according
to the Bragg equation:
1
experimental H NMR spectra (−CH3 36H: −CH2− 24H:
+
aromatic protons 48H). The HNEt3 counterions were
converted to the Na+ form via a methanolic NaOH wash,
after which they were exchanged to the acidic form (H+) by
acidification with aqueous H2SO4 to yield polymer sTPPPP-
H+. Gel permeation chromatography (GPC) analysis of
D = 2π/q0
(14)
sTPPPP-H+ revealed an estimated molecular weight (Mw
=
However, this may underestimate length scales in polymer systems by
up to 22%.31
271 kDa; Đ = 3.0). In a similar fashion, Diels−Alder
polymerization of the presulfonated monomer BTCS-HNEt3
yielded polymer sTPPyPP-HNEt3 . The structure of polymer
+
+
3. RESULTS AND DISCUSSION
1
1
was verified by H NMR spectroscopic analysis . H NMR
(DMSO-d6, 400 MHz) δ: 8.9 (s, 4H), 8.25−6.46 (m, 47H),
3.01 (q, 24H), 1.25 (t, 36H). The chemical shifts associated
with the methyl, methylene, and aromatic protons occur at δ
1.10, δ 3.05, and δ 6.00−8.35, and the integration ratios are
36H, 24H, and 47H, respectively (Figure S25). After the
3.1. Synthesis. As shown in Scheme 1, the polymers
sTPPyPP-H+ and sTPPPP-H+ were synthesized from the diene
+
BTCS-HNEt3 and dienophiles TPPy and TPP, respectively.
+
The presulfonated monomer BTCS-HNEt3 (Scheme 2) was
synthesized according to our previous report.6 The detailed
procedures and reaction yields can be found in the Supporting
Information.
conversion from HNEt3 to H+ forms, polymer sTPPyPP-H+
+
was obtained with a Mw of 295 kDa (Đ = 2.56). Polymer
membrane FT-IR data showed minimal differences (Figure 1).
The N-heterocycle dienophile TPPy was obtained via a two-
step synthesis according to literature procedures (Scheme
3).32,33 The procedures commenced with Sonogashira
coupling between trimethylsilylacetylene and bromo-
acetophenone (8) to yield the acetophenone-bearing alkynyl
moieties (9). Subsequently, 9 was reacted with benzaldehyde
to yield the target molecule TPPy (10). During the reaction,
cyclization to yield pyridine rings and the cleavage of
trimethylsilyl groups were achieved via a one-pot reaction
under basic conditions. The structure of TPPy (10) was
confirmed by 1H NMR, 13C NMR, and high-resolution
MS−1H NMR (CDCl3, 400 MHz) δ: 8.20 (d, 4H), 7.92 (s,
2H), 7.76 (d, 4H), 7.67 (d, 4H), 7.58−7.50 (m, 3H), 3.12 (s,
2H). 13C NMR (CDCl3, 100 MHz) δ: 156. HRMS (EI) m/z
calculated for C27H18N+ (M + H+) 356.1439; found 356.1449.
The route to the fully phenylene analogue, TPP, is also
depicted in Scheme 3. A Sonogashira reaction was employed to
couple bromoiodobenzene (11) with trimethylsilylacetylene,
affording the alkyne-functionated bromobenzene (12). A
lithium-halide exchange was performed with Br of 12, followed
by addition of triisopropyl borate to yield the boronate ester
(13). After Suzuki cross-coupling between 14 and phenyl-
boronic acid, the resulting product 15 was subject to another
Suzuki coupling with 13, yielding trimethylsilyl-protected TPP
(16). During the reaction, although an excess of Na2CO3 was
used as the base, the trimethylsilyl groups were not completely
removed. Thus, deprotection using K2CO3 was employed in
the last step was to ensure complete removal of trimethylsilyl
groups, giving the product (TPP, 17). The structures of 16 and
17 were confirmed by 1H NMR, 13C NMR, and high-
resolution MS (see the Supporting Information for 16). TPP
(17)−1H NMR (CDCl3, 400 MHz) δ: 8.20 (d, 4H), 7.92 (s,
2H), 7.77 (d, 2H), 7.67 (d, 4H), 7.57−7.53 (m, 6H), 3.18 (s,
2H). 13C NMR (CDCl3, 100 MHz) δ: 142.70, 141.54, 141.32,
140.83, 132.64, 128.88, 127.71, 127.32, 127.18, 125.43, 124.84,
Figure 1. FT-IR spectra of sTPPPP-H+ and sTPPyPP-H+ membranes.
The spectrum obtained for sTPPPP-H+ was very similar to that
previously reported for sPPP-H+.34 The spectrum obtained for
sTPPyPP-H+ showed a new signal at 1617 cm−1 (vs), and
greater absorbance at 1223 cm−1 (m), which is characteristic of
pyridine mixed ring vibrations typically observed between
1581−1590 cm−1 and 1217−1218 cm−1, respectively.35 Shifts
to higher frequencies of the skeletal vibrations of pyridine have
previously been attributed to hydrogen bonding due to the
presence of hydrogen bond donors such as water or
methanol.36 In the case of sTPPyPP-H+, the observed shift
may be due to intermolecular hydrogen bonding with
neighboring −SO3H moieties.
3.2. Physical Properties. As shown in Table 1, the
experimentally derived IEC of the fully phenylated poly-
phenylene sTPPPP-H+ (IECexp = 2.90
0.04 mmol g−1) is
+
121.48, 83.45, 77.92. HRMS (EI) m/z calculated for C28H19
(M + H+) 355.1481; found 355.1472.
close to its theoretical value (IECth = 3.06 mmol g−1).
E
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 1. IEC, Water Uptake, and Swelling of sTPPyPP-H+, TPPPP-H+, and sPPP-H+
a
b
c
d
theor IEC (mmol g−1
)
theor IEC (mmol g−1
)
exptl IEC (mmol g−1
)
water uptake (wt %) vol expansion (%)
hydration no.
sTPPyPP-H+
sTPPPP-H+
sPPP-H+
3.05
3.06
3.70
2.29
3.06
3.70
1.98 0.08
2.90 0.04
3.47
54.4 0.7
79.8 2.6
319.3
62.4 1.9
81.8 2.5
364.3
15.2
15.3
50.7
a
b
Based on sulfonic acid content. Based on “free” H+ calculated by taking account of acid−base neutralization (one sulfonic acid group per pyridine
c
d
group) into consideration. Volume increase ratio (length × width × thickness). See the Supporting Information. Based on experimental IEC.
Compared to the previously reported homopolymer sPPP-H+
(IECexp = 3.47 mmol g−1), the reduced IEC of sTPPPP-H+ is
attributed to the introduction of enhanced molecular mass of
the triphenylphenylene compared to phenylene. Because of its
structural similarity, the theoretical IEC of sTPPyPP-H+ (IECth
= 3.05 mmol g−1) is similar to that of sTPPPP-H+ when based
simply on the number of sulfonic acid groups present. The
titration-derived IECexp of sTPPyPP-H+ is found to be 1.98
0.08 mmol g−1. This can be explained by assuming that each
basic pyridine group neutralizes a sulfonic acid, such that only
three acidic protons are available per polymer repeat unit. In
this case, the theoretical IEC value (i.e., the “free” H+) for
sTPPyPP-H+ is reduced to 2.29 mmol g−1, closer to the
experimental value. This indicates that indeed the proton
bound to the conjugate acid of pyridine moiety is not
exchanged with Na+ ions in the Na+/H+ exchange process used
in the titration.
forming an acid−base salt. The third stage of decomposition,
observed at temperatures >550 °C, is attributed to the
decomposition of the polyphenylene backbone.5,7 The TGA
data show that sTPPyPP-H+ possesses a greater thermal
stability than sTPPPP-H+, which in turn is consistent with
previously reported polymers based completely on phenyl-
enes,5,7 indicating that incorporation of pyridine moieties into
acid-bearing polymers is an effective approach to enhancing
their thermal stability, which may aid their thermal processing.
This phenomenon is consistent with previous reports, which
also have pointed out that improved thermal stability can be
ascribed to acid−base interactions.12,38
As shown in Table 2 and Figure S29, the sTPPyPP-H+
membranes exhibit a tensile strength of 43.3
1.1 MPa,
a
Table 2. Mechanical Properties of sTPPyPP-H+, TPPPP-
H+, sPPP-H+, and N-211
The results of dimensional swelling measurements are
summarized in Table 1. At ambient temperature, sTPPPP-H+
membranes exhibit a water uptake of 79.8 2.6% and volume
expansion of 81.8 2.5%. The decrease in swelling compared
to previously published data on sPPP-H+ is attributed to the
increased molecular mass of the TPP moiety and the resulting
reduced IEC compared to sPPP-H+. In comparison, the water
uptake and volume expansion of the protonated triphenylpyr-
tensile strength
(MPa)
elongation at break
(%)
Young’s modulus
ionomers
(MPa)
sTPPyPP-
43.3 1.1
55.5 3.1
401.8 23.8
H+
sTPPPP-H+
sPPP-H+5
N-211
54.8 1.9
43.5 1.9
17.3 0.4
36.8 1.7
29.1 1.4
584.1 81.5
1059 34
270 17
148
4
a
The mechanical analyses were performed under ambient conditions
(23 °C, 50% RH).
idyl polymer membrane, sTPPyPP-H+, are 54.4
62.4 1.9%, respectively, a decrease of 32% and 24%,
0.7% and
respectively, compared to sTPPPP-H+. In both cases, the
values are significantly lower than those measured for sPPP-H+
(Table 1). Direct comparison between sTPPPP-H+ and
sTPPyPP-H+ indicates that the acid−base interaction between
pyridine and sulfonic acid groups reducing swelling. Both
polymers sTPPPP-H+ (λ = 15.2) and sTPPyPP-H+ (λ = 15.3)
display a lower hydration number λ than sPPP-H+ (λ = 50.7).
Thermogravimetric analyses (TGA) of sTPPPP-H+ and
sTPPyPP-H+ exhibited similar mass loss patterns with three
steps consistent with previous reports on sulfonated poly-
phenylenes.5,7 As shown in Figure S30, the first stage of mass
loss occurs between 100 and 200 °C due to loss of residual
water. Degradation of sulfonic acid groups occurs during the
second stage. TGA data indicate that the −SO3H groups of
pyridyl polymer sTPPyPP-H+ decompose at 326 °C, much
higher in temperature than sTPPPP-H+, which occurs at 246
°C. This is the highest thermal stability among reported
sulfonated polyphenylenes.5−8,37 During this second stage,
sTPPPP-H+ loses 23.6% of its mass, close to the theoretical
value for loss of four −SO3H groups (24.4%). Two
decomposition steps in the desulfonation stage are observed
for sTPPyPP-H+: an 18.8% mass loss at 326 °C, corresponding
to three −SO3H groups (theoretical = 18.3%), and a 6.1%
mass loss at 511 °C, corresponding to the fourth −SO3H
group (theoretical = 6.4%). This two-step behavior suggests
that each pyridine group may function as a protecting group
for a sulfonic acid moiety during thermolysis, by virtue of it
Young’s modulus of 401.8
23.8 MPa, and elongation at
break of 55.5
sTPPPP-H+ yields comparable values: 54.8 1.9 MPa, 584.1
81.5 MPa, and 36.8 1.7%, respectively. Both polymers
3.1% under ambient conditions. Polymer
show a higher tensile strength and Young’s modulus but lower
elongation at break than N-211. The N-heterocycle-containing
polymer sTPPyPP-H+ was less strong but tougher than
sTPPPP-H+. Membrane toughness is an important consid-
eration for material handling, electrochemical cell construction,
and durability in long-term, dynamic operation of electro-
chemical devices. It is commonly known that sulfonated,
phenylated polyphenylenes possess a relatively high tensile
strength and Young’s modulus but a low elongation at break
compared to that of PFSA ionomer membranes. For example,
the postsulfonated polyphenylenes reported by Fujimoto et al.
possessed a tensile strength of 74.0 MPa, Young’s modulus of
1200 MPa, and elongation at break of 12%.7 The presulfonated
sPPP-H+ reported previously listed values of 43.5 1.9 MPa,
1059 34 MPa, and 29.1 1.4%, respectively.5
Both polymers sTPPPP-H+ and sTPPyPP-H+ showed tensile
strength (54.8
1.9 and 43.3
1.1 MPa, respectively)
comparable to that of previously reported materials.5,7 Despite
the lower Young’s moduli (584 81.5 and 401.8 23.8 MPa,
respectively) as compared to sPPP-H+, both polymers showed
improved elongation at break (36.8 1.7% and 55.5 3.1%,
respectively) and generally improved membrane flexibility and
robustness. The higher elongation at break observed in
F
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 2. (a) In-plane proton conductivities of sTPPyPP-H+, sTPPPP-H+, and N-211 at 30 °C (open) and 80 °C (solid) under different RH and
(b) respective log plot.
+
+
Table 3. Analytical Acid Concentration ([−SO3H]), Effective Proton Mobility (μH ), and Water Content of sTPPPP-H and
sTPPyPP-H+
a
a
a
(mS cm−1
)
water contentmem
[−SO3H]mem (M)
[−SO3H]water (M)
μH ,mem (cm2 V−1 S−1
)
μH ,water (cm2 V−1 S−1
)
+
+
membranes
σ
sTPPPP-H+
sTPPyPP-H+
115
85
0.46
0.41
1.66
1.48
3.6
3.6
0.72 × 10−3
0.60 × 10−3
0.33 × 10−3
0.24 × 10−3
a
At 30 °C and 95% RH.
sTPPPP-H+ vs sPPP-H+ (+7.7%) may be due to the slightly
more kinked molecular structure of the repeating unit induced
by the TPP moiety, allowing for a more flexible polymer
backbone. The further enhancement in elongation at break
observed for sTPPyPP-H+ (+26.4% vs sPPP-H+) may also be
ascribed to the kinked structure of the repeating unit due to
the TPPy moiety present, together with the acid−base
interactions. These data suggest that such acid−base
interactions exert a role in improving membrane strain, as
sTPPyPP-H+ exhibits the highest elongation at break among
previously reported sulfonated polyphenylenes.5−8,37 It is
noted that the densities of dry sTPPyPP-H+ and dry
sTPPPP-H+ are 1.18 and 1.04 g cm−3, respectively, and the
densities of hydrated sTPPyPP-H+ and hydrated sTPPPP-H+
are 1.15 and 1.03 g cm−3, respectively. That is, sTPPyPP-H is
denser than sTPPPP-H+ by virtue of the acid−base
interactions present.
3.3. Proton Conductivity. The RH-dependent proton
conductivities for polymer sTPPPP-H+ and sTPPyPP-H+ were
determined at 30 and 80 °C (Figures 2). At 95% RH, sTPPPP-
H+ (155 mS cm−1 at 80 °C, 115 mS cm−1 at 30 °C) exhibited
greater proton conductivities than sTPPyPP-H+ (119 mS cm−1
at 80 °C, 85 mS cm−1 at 30 °C) due to its higher IEC (2.90
mmol g−1). Both polymers exhibited proton conductivities
higher than a PFSA reference, Nafion N-211 (110 mS cm−1 at
80 °C, 78 mS cm−1 at 30 °C).
polymers prepared by Huang’s group39 possessed a high
proton conductivity (264 mS cm−1) for high IEC (2.73 mmol
g−1) membranes, which also decreased disproportionately
(70%, to 80 mS cm−1) at lower IEC (31%, 1.89 mmol g−1)
when measured at 95% RH and 80 °C. In contrast, while
sTPPPP-H+ and sTPPyPP-H+ display a similar discrepancy in
IEC (32% difference from 2.90 to 1.98 mmol g−1), the
associated decrease in proton conductivities is only 23% (155
to 119 mS cm−1) when measured at 95% RH and 80 °C.
To explore the mechanism explaining the unusually high
conductivity of sTPPyPP-H+ having an IEC < 2 mmol g−1, we
examined the water content, analytical acid concentration
+
([−SO3H]), and effective proton mobility (μH ) for the two
polymers. Besides IEC, water content has an impact on proton
conduction in a PEM.41 As shown in Table 3, the water
contents for sTPPyPP-H+ and sTPPPP-H+ are close, 0.41 vs
0.46, respectively. For hydrocarbon-based PEMs, a higher IEC
often leads to a higher water content.41 However, in our case,
the slightly decreased water content of sTPPyPP-H+ in
comparison to sTPPPP-H+ suggested that the water content
is not significantly affected by the large IEC difference, which
may be caused by the lower volume swelling of the former
membrane. Because sTPPyPP-H+ displays a unique behavior in
swelling, IEC, and proton conductivity, we calculated the
+
volume of the absorbed water, [−SO3H], and μH with respect
to the volume of both the hydrated membrane and the water
phase (see the Experimental Section for details). In this
analysis, [−SO3H]mem refers to the analytical acid concen-
tration per the volume of wet membrane, while [−SO3H]water
refers to the analytical acid concentration per the volume of
absorbed water. The latter recognizes that H+ resides in the
water phase only. The [−SO3H] value of sTPPyPP-H+ in
either case is slightly lower or identical compared to that of
sTPPPP-H+ (1.48 M vs 1.66 and 3.6 M vs 3.6 M, respectively).
According to eq 5, [−SO3H]mem is dependent on both IECexp
The ion exchange capacity of sulfonated polyphenylene-
based polymers plays an important role in their proton
conductivitya small difference in IEC can lead to a large
change in conductivity.7,39,40 For example, in the case of the
sulfo-phenylated polyphenylenes reported by Cornelius and
co-workers, an IEC reduction from 1.8 to 1.4 mmol g−1 (a 22%
decrease) resulted in a disproportionate reduction in proton
conductivity in water at 80 °C, from 87 to 49 mS cm−1 (44%
decrease).7 In another report, sulfonated polyphenylene
G
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
and the volume of the wet membranes. IECexp of sTPPyPP-H+
is smaller than that of sTPPPP-H+ (1.98 mmol g−1 vs 2.90
mmol g−1), which can decrease the [−SO3H]mem value for the
former. However, the volume swelling of the sTPPyPP-H+ in
the hydrated state is smaller (62.4% vs 81.8%), which
compensates for the decrease in [−SO3H]mem caused by the
difference in IEC. Similarly, in regards to the acid
concentration in the water phase in the membrane
([−SO3H]water), the volume of absorbed water for sTPPyPP-
H+ is also lower compared to that of sTPPPP-H+ (eq 6), which
Scheme 4. Scheme Demonstrating How Steric Hindrance
Leads to a Lower pKa of the Protonated Form of TPPy
Relative to Py
makes up the above-mentioned decrease in [−SO3H]water
.
Thus, it explains why the sTPPyPP-H+ and sTPPPP-H+
membranes possessed similar values in [−SO3H]mem or
[−SO3H]water. Given that the proton conductivities at 30 °C
and 95 RH (Table 3) and using eqs 7 and 8, we calculated the
effective proton mobilities for the two membranes. The
sTPPyPP-H+ membrane exhibited μH ,mem (based on the
+
volume of the wet membrane) of 0.60 × 10−3 cm2 V−1 S−1 and
+
μH ,water (based on the volume of absorbed water) of 0.24×
of the sTPPPP-H+ polymer, which possesses a 50% higher IEC
(2.9 mequiv g−1).
10−3 cm2 V−1 S−1, which are smaller than those of sTPPPP-H+
(0.72 × 10−3 cm2 V−1 S−1 and 0.33 × 10−3 cm2 V−1 S−1,
respectively). Considering the large IEC difference between
the two polymers, it is surprising that the proton mobility
values of sTPPyPP-H+ are as high as they are based on
previously reported correlations of IEC versus proton mobility.
Because the effective proton mobility is related to acid
dissociation and tortuosity in the PEM,42 we examined
possible reasons for the high mobility in the context of pKa
of the acidic functionality, which controls the fraction of free
H+ carrier, and morphology, which controls the tortuosity.
A possible reason for this unusually high proton conductivity
behavior of sTPPyPP-H+ with respect to its low IEC is due to
the presence of the sterically hindered pyridine (TPPy). As
previously mentioned, in the pyridine-containing dienophile
monomer, the basic N atom is hindered by two flanking phenyl
rings, resulting in a sterically hindered pyridine, which was
reported to possess a lower basicity than pyridine itself.43 It has
been well-studied that the steric effect introduced by the
pyridine limits the solvation of the resulting protonated
pyridine and lowers the pKa for sterically hindered pyridinium
(H+).43 It has been experimentally verified that the pKa of
protonated pyridine in 70% aqueous ethanol is 3.69, while the
more sterically hindered protonated 2,4,6-triphenylpyridine
(TPPy) is 1.84.44 According to the extrapolated equation
(pKa,water − pKa,70% ethanol = 1.56),43 the pKa of the conjugate
acid of 2,4,6-triphenylpyridine is 3.40 in water at 25 °C, which
is lower than that of pyridine (5.25).45 On the basis of this, 1
M of the conjugated acid 2,4,6-triphenylpyridinium (H+) in
water dissociates 2.0%. Under the same conditions, 1 M of
pyridine-H+ will dissociate 10 times less (∼0.2%). Because of
the reduced pKa of TPPy, as shown in Scheme 4, the proton on
TPPy may dissociate to a greater extent than a proton bound
to pyridine, which will participate in proton conduction. Liu et
al. demonstrated this concept by incorporating 1H-1,2,4-
triazole (pKa of its conjugate acid = 2.39) in place of imidazole
(pKa of its conjugate acid = 7.18) to enhance proton
conductivity of a hydrocarbon-based PEM.46 Miyatake and
co-workers also reported a triazole-based polyimide possessing
a high proton conductivity.47 Thus, the reduced pKa of TPPy-
H+ caused by the sterically hindrance around the nitrogen may
explain why the sTPPyPP-H+ polymer, with an IEC of <2.0
mequiv g−1, exhibited a proton conductivity comparable to that
Another reason for the higher than expected proton
conductivity of sTPPyPP-H+ could lie in differences between
its morphology and that of sTPPPP-H+, given that differences
in conductivity can be tied to the nanoscale organization of the
polymer membrane. We performed X-ray scattering experi-
ments to investigate the morphology of the materials. The
results are shown in Figure 3. Under ambient conditions, the
Figure 3. Small-angle X-ray scattering spectra of dry and hydrated
sTPPyPP-H+ and sTPPPP-H+.
X-ray scattering profiles for sTPPyPP-H+ (black) and sTPPPP-
H+ (blue) are indistinguishable, suggesting that the morphol-
ogies of dry membranes are similar. However, when the two
membranes are hydrated, the sTPPPP-H+ membrane (red)
reveals a much more distinct feature in the mid-q region than
sTPPyPP-H+ (magenta), indicating a sharper interface
between the water- and polymer-rich regions of the material.
The broad peak model (BPM, eq 13) was fit to the mid-q
regions of the SAXS data for the hydrated membranes to
determine the distribution and extent of the water-rich regions
(Figure S35); the BPM has been used to study solvation
distributions in polymers.48 Results produced similar values for
the separation D of water-rich regions for both films: 36 1 Å
for sTPPPP and 35 1 Å for sTPPyPP. However, values for
H
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
the correlation length ξ, which represents the extent of the
water-rich regions, were substantially different. sTPPPP-H+
measured 13.8 0.3 Å, whereas sTPPyPP-H+ measured 6.2
0.6 Å, reflecting the broader feature in the latter. The more
intense peak with larger ξ in sTPPPP-H+ indicates that the
water-rich regions are more distinct and are of greater extent
than in sTPPyPP-H+, while the water content by volume is
similar in the two materials (Table 1).
sTPPPP-H+ lost 57.5 7.8 wt %. The sTPPPP-H+ membrane
completely dissolved in the solution after 5.3 h while 10 wt %
of sTPPyPP-H+ remained after 6 h. Enhanced oxidative
stability of N-heterocycles is recognized in the literature.9,10,21
For example, to examine the antioxidation property of
pyridine-containing polymers, Wang and co-workers50 refluxed
a pyridine-based compound in a mixture of H2O2 and glacial
acetic acid. Mass spectrometric analysis indicated that pyridine
had reacted with peroxide to form pyridine-N-oxide,
scavenging reactive species in solution. In the case of
sTPPyPP-H+, the pyridine moiety is believed to act as a free-
radical scavenger, preventing critical chain degradation. It
should be pointed out that although the hydrocarbon polymer
sTPPPP-H+ displays a lower chemical stability than its pyridyl
analogue, it is still much more resistive to oxidative
degradation than sulfonated polyaromatic ethers (20% weight
loss after 1 h in Fenton’s reagent, 3% H2O2, 2 ppm Fe2+ at 80
°C).39
Both hydrated samples showed a large peak at ∼2 Å−1,
which is consistent with bulk water.49 As this feature could
correspond to tighter intermolecular packing inside the
membranes or bulk water on the surface of the membranes
or sample holders, we avoided further interpretations of the
WAXS region.
3.4. Stability against Fenton’s Reagent. The oxidative
stability of the polymers was examined by immersing
polymeric membranes in Fenton’s reagent. As shown in Figure
4, in the first 3 h, sTPPyPP-H+ and sTPPPP-H+ showed a
3.5. Investigation in Fuel Cells. Single cell performance
of membrane−electrode assemblies incorporating sTPPPP-H+
and sTPPyPP-H+ membranes (thickness in dry state: 28 and
28 μm, respectively) were evaluated in a fuel cell testing station
operated under zero back-pressure, at 80 °C, and under fully
humidified conditions. Catalyst ink containing PFSA ionomers
and Pt/C was spray coated onto the membranes to form
catalyst-coated membranes (CCMs) to achieve 0.4 mg cm−2
loading on the anode and 0.2 mg cm−2 on the cathode. MEAs
were subjected to fuel cell conditioning and RH cycling, the
duration of which was 15 h, after which polarization
experiments were conducted. As shown in Figure 5a, using
H2/O2, the peak power densities of sTPPPP-H+ and sTPPyPP-
H+ membranes are 1080 and 931 mW cm−2, respectively.
Using H2/air, the peak power density of sTPPyPP-H+ is similar
to that of sTPPPP-H+ (532 vs 521 mW cm−2, Figure 5b).
MEAs prepared from N-212 tested under the same conditions
yield peak power densities of 793 mW cm−2 using H2/O2 and
455 mW cm−2 using H2/air.5 For MEAs prepared from 25 μm
N-211 the maximum power densities in H2/O2 and H2/air
under identical conditions are 1337 and 537 mW cm−2,
respectively (Figure 5), which reflects the reduced thickness of
the N-211 membranes. An electrochemical surface area
(ECSA) measurement was performed via cyclic voltammetry.
The ECSA values of N212 and N211 are 68.6 and 49.2 m2 g−1
Figure 4. Accelerated oxidative stability test of membranes sTPPPP-
H+ and sTPPyPP-H+ in Fenton’s reagent (3% H2O2, 3 ppm Fe2+) at
80 °C (the thickness of dry membranes used in the Fenton’s test is 28
5 μm; sTPPPP-H+ membrane dissolved completely at 5.3 h).
similar rate of degradation. However, after 4 h of exposure to
the Fenton’s reagent, sTPPyPP-H+ degraded more slowly. The
mass loss of sTPPyPP-H+ was 37.5 9.1 wt % after 4 h, while
Figure 5. In situ polarization (left axis, solid) and power density (right axis, open) under (a) H2/O2 and (b) H2/air. Conditions in both cases were
80 °C, 100% RH, 0.5/1.0 slpm anode/cathode gas flows, and zero back-pressure.
I
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Pt, respectively, while sTPPPP-H+ displayed an ECSA value of
34.5 m2 g−1 Pt and sTPPyPP-H+ showed that of 55.1 m2 g−1
Pt. The ohmic resistances (iR-drop method) of sTPPPP-H+
and sTPPyPP-H+ polymers are 63 and 65 mohm·cm2,
respectively, indicating that the two polymers exhibit high in
situ proton conductivities.
The sTPPPP-H+ and sTPPyPP-H+ polymers displayed open
circuit voltages (OCVs) of 0.966 and 0.936 V, respectively. In
addition, after 15 h of fuel cell conditioning and RH cycling,
fuel crossover of sTPPPP-H+ (0.52 mA cm−2) and sTPPyPP-
H+ (0.46 mA cm−2) membranes determined by chronoamper-
ometry (CA) measurements were exceptionally low, further
verified by comparison to a N-211 standard (1.36 mA cm−2)
under the same conditions. Both OCV and CA data show that
both membranes possess good gas barrier properties with
similar gas crossover rates, further supporting that the addition
of the pyridine moiety and resulting acid−base interactions
does not lead to membranes with reduced quality or
performance.
Figure 6. Accelerated stress test (AST) of sTPPyPP-H+ (the
thickness in the dry state: 28 μm, plus conditioning and polarization
experiments before and after the AST) and N-211 (AST only). The
open circuit voltage (OCV) was held at 90 °C, 30% RH, and zero
back-pressure using H2/air. During the AST of N-211, several
polarizations and gas crossover measurements were conducted, the
interruption of which caused the observable spikes.
Polarization studies under different RH were performed on
sTPPyPP-H+ and sTPPPP-H+. The peak power densities
reached 931/878/846/297 mW cm−2 and 1083/1075/1042/
470 mW cm−2, respectively, under 100/95/90/50% RH using
H2/O2 (Figures S32 and S33). After performing the fuel cell
test at 50% RH, the polarization data at 100% RH were
reobtained to examine the recoverability for two polymers.
While sTPPPP-H+ experienced a decrease of 17% in peak
power density (1083 to 896 mW cm−2) at 100% RH following
polarization at 50% RH, the two IV curves of sTPPyPP-H+
were identical (Figure S34). Full recoverability was observed
for the pyridine-containing polymeric membrane, possibly due
to its tougher nature (cf. mechanical properties). Following the
above polarization experiments, an in situ chemical accelerated
stress test (AST) was performed to evaluate the chemical
stability of sTPPyPP-H+ within a fuel cell device. During this
test, the open circuit voltage (OCV) was held at 90 °C and
30% RH using H2/air. It has been reported that such harsh
conditions can generate hydrogen and peroxide radicals, which
are highly reactive and detrimental to membrane integrity. For
example, N-211 fails after 153 h under these condition.5 In
contrast, after 1000 h of AST, an OCV of 0.733 V was
observed for sTPPyPP-H+a decrease of only 0.16 mV h−1
was observed for the first hour, after which the degradation
rate decreased to 0.11 mV h−1. By comparison, N-211
degraded by 2.14 mV h−1 under the same conditions (Figure
6). After the AST, the sTPPyPP-H+ MEA was reconditioned
using the procedure described above, and a second set of
polarization experiments were performed. The peak power
density after 1000 h of AST was found to be 728 mW cm−2 at
80 °C, 100% RH using H2/O2. (Figure 7). That is, after 1000 h
of accelerated in situ oxidative chemical degradation, sTPPyPP-
H+ experienced a decrease in performance of only 21%, and
continued to serve its function as a PEM, providing useful
polarization data with no signs of cell failure. This exceptional
in situ performance is attributed to a combination of increased
membrane toughness, low gas crossover, and high oxidative
stability.
Figure 7. In situ polarization (left axis, solid) and power density (right
axis, open) of sTPPyPP-H+ before and after AST. Conditions were 80
°C, 100% RH, 0.5/1.0 slpm anode/cathode gas flows, and zero back-
pressure under H2/O2.
sTPPPP-H+ and sTPPyPP-H+, were designed and synthesized
for comparison. Both polymers possess well-defined structures
with a nonrandom distribution of −SO3H groups because of
the prefunctionalization approach, and their compositions of
repeating units differ by only one atom. These minimal
structural variations enabled precise analysis of the effect of the
pyridine moiety (N atom) on physicochemical properties such
as the dimensional stability, elongation at break, and thermal
and oxidative properties, which were significantly enhanced. In-
situ and ex-situ electrochemical experiments revealed a minimal
reduction in proton conductivity and fuel cell performance of
sTPPyPP-H+ versus that of sTPPPP-H+. Notably, to the best
of our knowledge, many properties of sTPPyPP-H+, such as
the elongation at break at ambient conditions (55.5 3.1%),
the initial thermal degradation temperature (326 °C via
thermolysis), and chemical durability in fuel cells (0.16 mV h−1
degradation rate in 1000 h of AST), are exceptional among
reported sulfonated aromatic polymers. The results indicate
that incorporation of a sterically hindered pyridine moiety into
sulfonated polyphenylenes is an efficient means of exploiting
4. CONCLUSIONS
In this work, we systematically investigated the effect of a
sterically hindered pyridine moiety on the properties of
proton-conducting sulfonated polyphenylenes. Two novel
sulfonated, phenylated polyphenylene homopolymers,
J
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
acid−base interactions and improving the physical and
chemical attributes of polyphenylenes without compromising
proton conductivity. We propose two possible explanations for
this: the morphological differences caused between the
pyridine and non-pyridine form and the reduced pKa of
pyridinium (H+) caused by the sterically hindered amine
structure; however, more possibilities remain to be explored.
We expect this simple pyridine moiety to be of broad utility in
the design and synthesis of sulfonated aromatic-based
polymers for next-generation proton-conducting membranes.
REFERENCES
■
(1) Schwenzer, B.; Zhang, J.; Kim, S.; Li, L.; Liu, J.; Yang, Z.
Membrane Development for Vanadium Redox Flow Batteries.
ChemSusChem 2011, 4, 1388−1406.
(2) Goni-Urtiaga, A.; Presvytes, D.; Scott, K. Solid Acids as
̃
Electrolyte Materials for Proton Exchange Membrane (Pem)
Electrolysis: Review. Int. J. Hydrogen Energy 2012, 37, 3358−3372.
(3) Kerres, J. Development of Ionomer Membranes for Fuel Cells. J.
Membr. Sci. 2001, 185, 3−27.
(4) Kreuer, K. D. On the Development of Proton Conducting
Polymer Membranes for Hydrogen and Methanol Fuel Cells. J.
Membr. Sci. 2001, 185, 29−39.
(5) Adamski, M.; Skalski, J. G. T.; Britton, B.; Peckham, J. T.;
Metzler, L.; Holdcroft, S. Highly Stable, Low Gas Crossover, Proton-
Conducting Phenylated Polyphenylenes. Angew. Chem., Int. Ed. 2017,
56, 9058−9061.
(6) Skalski, T. J. G.; Britton, B.; Peckham, T. J.; Holdcroft, S.
Structurally-Defined, Sulfo-Phenylated, Oligophenylenes and Poly-
phenylenes. J. Am. Chem. Soc. 2015, 137, 12223−12226.
(7) Fujimoto, C. H.; Hickner, M. A.; Cornelius, C. J.; Loy, D. A.
Ionomeric Poly(phenylene) Prepared by Diels-Alder Polymerization:
Synthesis and Physical Properties of a Novel Polyelectrolyte.
Macromolecules 2005, 38, 5010−5016.
(8) Miyake, J.; Taki, R.; Mochizuki, T.; Shimizu, R.; Akiyama, R.;
Uchida, M.; Miyatake, K. Design of flexible polyphenylene proton-
conducting membrane for next-generation fuel cells. Sci. Adv. 2017, 3,
eaao0476.
(9) Geormezi, M.; Deimede, V.; Gourdoupi, N.; Triantafyllopoulos,
N.; Neophytides, S.; Kallitsis, J. K. Novel Pyridine-Based Poly(ether
sulfones) and their Study in High Temperature PEM Fuel Cells.
Macromolecules 2008, 41, 9051−9056.
(10) Kallitsis, J. K.; Geormezi, M.; Neophytides, S. G. Polymer
electrolyte membranes for high-temperature fuel cells based on
aromatic polyethers bearing pyridine units. Polym. Int. 2009, 58,
1226−1233.
(11) Li, Q.; Rudbeck, H. C.; Chromik, A.; Jensen, J. O.; Pan, C.;
Steenberg, T.; Calverley, M.; Bjerrum, N. J.; Kerres, J. Development
of ionomer membranes for fuel cells. J. Membr. Sci. 2010, 347, 260−
270.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.8b02289.
Materials and instruments information, full synthetic
schemes for each monomer and polymer starting from
commercially available reagents, the synthetic details and
NMR spectra for all compounds and polymers
described, stress vs strain curves of polymers at ambient
conditions, TGA profiles of polymers under nitrogen
flow, additional experimental information about fuel cell
setup and characterization, in situ polarization and power
density curves of both polymers described compared to
Nafion 211 and 212 membranes, in situ polarization and
power density curves of both polymers described at
different RHs using H2/O2, in situ polarization of both
polymers described before and after 50% RH polar-
ization using H2/O2, fit small-angle X-ray scattering
spectra of both polymers described in their hydrated
state (PDF)
AUTHOR INFORMATION
Corresponding Author
*(S.H.) E-mail holdcrof@sfu.ca.
■
(12) Wang, G.; Lee, K. H.; Lee, W. H.; Shin, D. W.; Kang, N. R.;
Cho, D. H.; Hwang, D. S.; Zhuang, Y.; Lee, Y. M.; Guiver, M. D.
Durable Sulfonated Poly(benzothiazole-co-benzimidazole) Proton
Exchange Membranes. Macromolecules 2014, 47, 6355−6364.
(13) Giffin, G. A.; Galbiati, S.; Walter, M.; Aniol, K.; Ellwein, C.;
Kerres, J.; Zeis, R. Interplay between Structure and Properties in Acid-
Base Blend PBI-Based Membranes for HT-PEM Fuel Cells. J. Membr.
Sci. 2017, 535, 122−131.
ORCID
Barbara J. Frisken: 0000-0001-7234-891X
Steven Holdcroft: 0000-0002-1653-1047
Funding
Financial support for this study was provided by Natural
Sciences and Engineering Research Council of Canada
(NSERC). Research described in this work made use of the
4D LABORATORIES shared facilities at SFU supported by
the Canada Foundation for Innovation (CFI), British
Columbia Knowledge Development Fund (BCKDF), Western
Economic Diversification Canada (WD), and Simon Fraser
University (SFU).
(14) Kreuer, K. D.; Fuchs, A.; Ise, M.; Spaeth, M.; Maier, J.
Imidazole and pyrazole-based proton conducting polymers and
liquids. Electrochim. Acta 1998, 43, 1281−1288.
̀
(15) Jones, D. J.; Roziere, J. Recent Advances in the Functionalisa-
tion of Polybenzimidazole and Polyetherketone for Fuel Cell
Applications. J. Membr. Sci. 2001, 185, 41−58.
(16) Kerres, J.; Ullrich, A.; Meier, F.; Haring, T. Synthesis and
Characterization of Novel Acid−Base Polymer Blends for Application
in Membrane Fuel Cells. Solid State Ionics 1999, 125, 243−249.
(17) Mehta, V.; Cooper, J. S. Review and Analysis of PEM Fuel Cell
Design and Manufacturing. J. Power Sources 2003, 114, 32−53.
(18) Kopitzke, R. W.; Linkous, C. A.; Anderson, R. H.; Nelson, G. L.
onductivity and Water Uptake of Aromatic-Based Proton Exchange
Membrane Electrolytes. J. Electrochem. Soc. 2000, 147, 1677−1681.
(19) Kulkarni, M. P.; Peckham, T. J.; Thomas, O. D.; Holdcroft, S.
Synthesis of Highly Sulfonated Polybenzimidazoles by Direct
Copolymerization and Grafting. J. Polym. Sci., Part A: Polym. Chem.
2013, 51, 3654−3666.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors acknowledge Dr. Alexandra Furtos-Matei and Dr.
Marie-Christine Tang from the Centre Regionale de
■
́
́
́
́
Spectrometrie de Masse de l’Universite de Montreal for the
high resolution mass spectrum analysis. We thank Thomas
Holmes for assistance with syntheses; Johnathan Ward, Wei Li,
Thomas Skalski, and Benjamin Britton for assistance with
TGA, GPC, MS, and fuel cell analyses, respectively; and Drs.
Jiantao Fan, Chi Zhang, Timothy Peckham, and David
Novitski for useful discussions.
(20) Thomas, O. D.; Peckham, T. J.; Thanganathan, U.; Yang, Y.;
Holdcroft, S. Sulfonated Polybenzimidazoles: Proton Conduction and
Acid−Base Crosslinking. J. Polym. Sci., Part A: Polym. Chem. 2010, 48,
3640−3650.
K
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(21) Kalamaras, I.; Daletou, M. K.; Gregoriou, V. G.; Kallitsis, J. K.
Sulfonated Aromatic Polyethers Containing Pyridine Units as
Electrolytes for High Temperature Fuel Cells. Fuel Cells 2011, 11,
921−931.
(41) Peckham, T. J.; Holdcroft, S. Structure-morphology-property
relationships of non-perfluorinated proton-conducting membranes.
Adv. Mater. 2010, 22, 4667−90.
(42) Peckham, T. J.; Schmeisser, J.; Holdcroft, S. Relationships of
Acid and Water Content to Proton Transport in Statistically
Sulfonated Proton Exchange Membranes: Variation of Water Content
Via Control of Relative Humidity. J. Phys. Chem. B 2008, 112, 2848−
2858.
(22) Ye, Y.-S.; Chen, W.-Y.; Huang, Y.-J.; Cheng, M.-Y.; Yen, Y.-C.;
Cheng, C.-C.; Chang, F.-C. Preparation and Characterization of High-
Durability Zwitterionic Crosslinked Proton Exchange Membranes. J.
Membr. Sci. 2010, 362, 29−37.
(43) ALDER, R. W. Strain Effects on Amine Basicities. Chem. Rev.
1989, 89, 1215−1223.
̀
(23) Peron, J.; Ruiz, E.; Jones, D. J.; Roziere, J. Solution sulfonation
of a novel polybenzimidazole. J. Membr. Sci. 2008, 314, 247−256.
(24) Jouanneau, J.; Mercier, R.; Gonon, L.; Gebel, G. Synthesis of
Sulfonated Polybenzimidazoles from Functionalized Monomers:
Preparation of Ionic Conducting Membranes. Macromolecules 2007,
40, 983−990.
(44) Hopkins, H. P. J.; Jahagirdar, D. V.; Moulik, P. S.; Aue, D. H.;
Webb, H. M.; Davidson, W. R.; Pedley, M. D. Basicities of the 2-, 4-,
2,4-Di-, and 2,6-Disubstituted tert-Butyl Pyridines in the Gas Phase
and Aqueous Phase: Steric Effects in the Solvation of tert-Butyl-
Substituted Pyridines and Pyridinium Cations. J. Am. Chem. Soc.
1984, 106, 4341−4348.
̀
(25) Roziere, J.; Jones, D. J.; Marrony, M.; Glipa, X.; Mula, B. On
the doping of sulfonated polybenzimidazole with strong bases. Solid
State Ionics 2001, 145, 61−68.
(45) Katritzky, A. R.; Leahy, D. E.; et al. Solution and Gas-phase
Basicities of Pyridines with Bulky a-Substituents. J. Chem. Soc., Perkin
Trans. 2 1983, 2, 45−48.
(26) Tan, N.; Xiao, G.; Yan, D. Sulfonated Polybenzothiazoles: A
Novel Candidate for Proton Exchange Membranes†. Chem. Mater.
2010, 22, 1022−1031.
(46) Li, S.; Zhou, Z.; Zhang, Y.; Liu, M.; Li, W. 1H-1,2,4-Triazole:
An Effective Solvent for Proton-Conducting Electrolytes. Chem.
Mater. 2005, 17, 5884−5886.
(27) Wang, G.; Xiao, G.; Yan, D. Soluble sulfonated polybenzo-
thiazoles derived from 3,3′-disulfonate-4,4′-dicarboxylbiphenyl for
proton exchange membranes. Int. J. Hydrogen Energy 2012, 37, 5170−
5179.
(47) Saito, J.; Miyatake, K.; Watanabe, M. Synthesis and Properties
of Polyimide Ionomers Containing 1H-1,2,4-Triazole Groups. Macro-
molecules 2008, 41, 2415−2420.
́
(28) Alvarez-Gallego, Y.; Ruffmann, B.; Silva, V.; Silva, H.; Lozano,
(48) Horkay, F.; Hammouda, B. Small-angle Neutron Scattering
from Typical Synthetic and Biopolymer Solutions. Colloid Polym. Sci.
2008, 286, 611−620.
A. E.; de la Campa, J. G.; Nunes, S. P.; de Abajo, J. Sulfonated
polynaphthalimides with benzimidazole pendant groups. Polymer
2008, 49, 3875−3883.
(29) Li, N.; Cui, Z.; Zhang, S.; Xing, W. Sulfonated polyimides
bearing benzimidazole groups for proton exchange membranes.
Polymer 2007, 48, 7255−7263.
(30) Kline, S. R. Reduction and analysis of SANS and USANS data
using IGOR Pro. J. Appl. Crystallogr. 2006, 39, 895−900.
(31) Klug, H. P.; Alexander, L. E. X-ray Diffraction Procedures for
Polycrystalline and Amorphous Materials; John Wiley & Sons: New
York, 1954; p 631.
(49) Hura, G.; Sorenson, J. M.; Glaeser, R. M.; Head-Gordon, T. A
high-Quality X-Ray Scattering Experiment on Liquid Water at
Ambient Conditions. J. Chem. Phys. 2000, 113, 9140−9147.
(50) Sun, G.; Han, K.; Yu, J.; Zhu, H.; Wang, Z. Non-planar
Backbone Structure Polybenzimidazole Membranes with Excellent
Solubility, High Proton Conductivity, and Better Anti-oxidative for
HT-PEMFCs. RSC Adv. 2016, 6, 91068−91076.
(32) Ran, Q.; Ma, J.; Wang, T.; Fan, S.; Yang, Y.; Qi, S.; Cheng, Y.;
Song, F. Synthesis and fluorescence study of conjugated polymers
based on 2,4,6-triphenylpyridine moieties. New J. Chem. 2016, 40,
6281−6288.
(33) Huo, J.; Arulsamy, N.; Hoberg, J. O. Facile Synthesis and
Platinum Complexes of 4’,5,5′’-Trisubstituted-2,2’:6’,2’’-Terpyridines.
Dalton transactions 2011, 40, 7534−40.
(34) Skalski, T. J. G.; Adamski, M.; Britton, B.; Schibli, E. M.;
Peckham, T. J.; Weissbach, T.; Moshisuki, T.; Lyonnard, S.; Frisken,
B. J.; Holdcroft, S. Sulfophenylated Terphenylene Copolymer
Membranes and Ionomers. ChemSusChem 2018, 11, 4033−4043.
́ ́
(35) Urena, F. P.; Gomez, M. F.; Gonzalez, J. J. L.; Torres, E. M. n.
̃
A new insight into the vibrational analysis of pyridine. Spectrochim.
Acta, Part A 2003, 59, 2815−2839.
(36) Zinth, W.; Kaiser, W. In Molecular Aggregates in Liquids Resolved
by a Novel Raman Spectroscopy; Reineker, P., Haken, H., Wolf, H. C.,
Eds.; Springer-Verlag: 1983.
(37) Lim, Y.; Lee, H.; Lee, S.; Jang, H.; Hossain, M. A.; Cho, Y.;
Kim, T.; Hong, Y.; Kim, W. Synthesis and properties of sulfonated
poly(phenylene sulfone)s without ether linkage by Diels−Alder
reaction for PEMFC application. Electrochim. Acta 2014, 119, 16−23.
(38) Tan, N.; Chen, Y.; Xiao, G.; Yan, D. Synthesis and properties of
sulfonated polybenzothiazoles with benzimidazole moieties as proton
exchange membranes. J. Membr. Sci. 2010, 356, 70−77.
(39) Lee, H.-F.; Huang, Y.-C.; Wang, P.-H.; Lee, C. C.; Hung, Y.-S.;
Gopal, R.; Holdcroft, S.; Huang, W.-Y. Synthesis of highly sulfonated
polyarylene ethers containing alternating aromatic units. Materials
Today Communications 2015, 3, 114−121.
(40) Higashihara, T.; Matsumoto, K.; Ueda, M. Sulfonated aromatic
hydrocarbon polymers as proton exchange membranes for fuel cells.
Polymer 2009, 50, 5341−5357.
L
DOI: 10.1021/acs.macromol.8b02289
Macromolecules XXXX, XXX, XXX−XXX