Highly stable anion exchange membranes based on quaternized polypropylene

Article abstract of DOI:10.1039/c5ta01420d

A series of novel quaternized polypropylene (PP) membranes with 'side-chain-type' architecture was prepared by heterogeneous Ziegler-Natta catalyst mediated polymerization and subsequent quaternization. Tough and flexible anion exchange membranes were prepared by melt-pressing of bromoalkyl-functionalized PP (PP-CH₂Br) at 160°C, followed by post-functionalization with trimethylamine (TMA) or N,N-dimethyl-1-hexadecylamine (DMHDA) and ion exchange. By simple incorporation of a thermally crosslinkable styrenic diene monomer during polymerization, crosslinkable PP-AEMs were also prepared at 220°C. PP-AEM properties such as ion exchange capacity, thermal stability, water and methanol uptake, methanol permeability, hydroxide conductivity and alkaline stability of uncrosslinked and crosslinked membranes were investigated. Hydroxide conductivities of above 14 mS cm^-1 were achieved at room temperature. The crosslinked membranes maintained their high hydroxide conductivities in spite of their extremely low water uptake (up to 56.5 mS cm^-1 at 80°C, water uptake = 21.1 wt%). The unusually low water uptake and good hydroxide conductivity may be attributed to the "side-chain-type" structures of pendent cation groups, which probably facilitate ion transport. The membranes retained more than 85% of their high hydroxide conductivity in 5 M or 10 M NaOH aqueous solution at 80°C for 700 h, suggesting their excellent alkaline stability. It is assumed that the long alkyl spacer in the 'side-chain-type' of 9 carbon atoms between the polymer backbone and cation groups reduces the nucleophilic attack of water or hydroxide at the cationic centre. Thus, PP-based AEMs with long "side-chain-type" cations appear to be very promising candidates with good stability for use in anion exchange membrane fuel cells (AEMFCs).

Full text of DOI:10.1039/c5ta01420d

View Article Online
View Journal
Journal of
Materials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Zhang, J. Liu,
Y. Wang, L. An, M. D. Guiver and N. Li, J. Mater. Chem. A, 2015, DOI: 10.1039/C5TA01420D.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/materialsA

Page 1 of 14
Journal of Materials Chemistry A
Dynamic Article Links ►
Journal Name
View Article Online
DOI: 10.1039/C5TA01420D
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Highly Stable Anion Exchange Membranes Based on Quaternized
Polypropylene
Min Zhanga, Jinling Liua, Yiguang Wang,*a Linan Anc Michael D. Guiver,d,e and Nanwen Li*b
Received (in XXX, XXX) Xth XXXXXXXXX 2015, Accepted Xth XXXXXXXXX 20XX
5
DOI: 10.1039/b000000x
A series of novel quaternized polypropylene (PP) with ‘sideꢀchainꢀtype’ architecture was prepared by
heterogeneous ZieglerꢀNatta catalyst mediated polymerization and subsequently quaternization. Tough
and flexible anion exchange membranes were prepared by meltꢀpressing of bromoalkylꢀfunctionalized PP
(PPꢀCH2Br) at 160 °C, followed by postꢀfunctionalization with trimethylamine (TMA) or N,Nꢀdimethylꢀ
10 1ꢀhexadecylamine (DMHDA) and ion exchange. By the simple incorporation of a thermallyꢀcrosslinkable
styrenic diene monomer during polymerization, crosslinkable PPꢀAEMs were also prepared at 220 °C.
PPꢀAEM properties such as ion exchange capacity, thermal stability, water and methanol uptake,
methanol permeability, hydroxide conductivity and alkaline stability of uncrosslinked and crosslinked
membranes were investigated. Hydroxide conductivities of above 14 mS cm–1 were achieved at room
15 temperature. The crosslinked membranes maintained their high hydroxide conductivities in spite of their
extremely low water uptake (up to 56.5 mS cm–1 at 80 °C, water uptake = 21.1 wt %). The unusually low
water uptake and good hydroxide conductivity may be attributed to the “sideꢀchainꢀtype” structures of
pendent cation groups, which probably facilitate ion transport. The membranes retained more than 85 %
of their high hydroxide conductivity in 5 M or 10 M of NaOH aqueous solution at 80 °C for 700 h,
20 suggesting their excellent alkaline stability. It is assumed that the long alkyl spacer in ‘sideꢀchainꢀtype’ of
9 carbon atoms between polymer backbone and cation groups reduce the nucleophilic attack of water or
hydroxide at the cationic centre. Thus, the PPꢀbased AEMs with long “sideꢀchainꢀtype” cations appear to
be very promising candidates with good stability for use in anion exchange membrane fuel cells
(AEMFC)s.
thermal crosslinking.6,11,12 For example, AEMs with long alkyl
25 Introduction
chains of 6–16 carbon atoms grafted onto QA groups had
improved alkaline stability. The steric hindrance effect of long
alkyl chains was believed to effectively reduce nucleophilic
50 attack of water or hydroxide at the QA centres, thereby
improving alkaline stability. Recently, Hibbs and coꢀworkers13
reported similarly enhanced alkali stability for a poly(phenylene)
with a ‘sideꢀchainꢀtype’ QA group attachment by hexamethylene
spacers. Furthermore, polymer structures containing flexible
55 pendent side chains linking the hydrophobic polymer with
hydrophilic QA groups typically exhibit nanophase separation
between hydrophilic and hydrophobic domains, enhancing
hydroxide conductivity.
Additionally, the structure of the polymer backbone was
60 observed to have a significant influence on the AEM properties
and performance. For example, Ramani and coworkers14
demonstrated the degradation of aromatic polysulfone backbone
which arose from both the quaternary carbon and ether
hydrolysis, using 2D NMR spectroscopy (correlation
65 spectroscopy (COSY) and heteronuclear multiple quantum
correlation spectroscopy (HMQC)). Subsequently, Hickner15 et
al. reported that quaternary ammonium groups displayed different
Anion exchange membrane fuel cells (AEMFCs) have been
gaining significant interest over the past several years owing to
their significant advantages of improved oxygen reduction and
fuel oxidation kinetics, which can lead to higher efficiencies,
30 while the alkaline environment allows utilization of nonꢀprecious
metal catalysts, greatly reducing the device cost.1–5 Although
AEM technology has improved immensely in the past decade,
AEM fuel cells continue to perform unfavorably compared to
their PEM counterparts because of a significant performance gap
35 of lower ionic conductivities and ineffective alkaline stability.
Given the high performance of PEM, it is evident that their
structure and chemistry have significantly influenced AEM
design. In the pursuit of hydrocarbonꢀbased AEMs with improved
properties, we and others have shown several promising
40 approaches to enhance quaternary ammonium (QA)ꢀbased AEM
properties and performance, which are (1) to position the QA
group on side chains grafted onto the polymer main chain (sideꢀ
chainꢀtype), (2) to form combꢀshaped quaternized PPO AEMs
using a long alkyl chain of 6–16 carbon atoms pendant to the QA
45 groups6–10 and (3) to crosslink the AEM matrix by catalyzed or
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

Journal of Materials Chemistry A
Page 2 of 14
View Article Online
DOI: 10.1039/C5TA01420D
alkaline stabilities when attached to different polymer backbones,
subsequent thermal crosslinking. The properties of the resulting
by comparison of the alkaline stabilities of quaternized 60 membranes, such as the alkaline stability, mechanical strength,
poly(arylene ether sulfone) (PAES), poly(phenylene oxide)
(PPO) and polystyrene (PS) using the quantitative 1H NMR
spectroscopy. It was observed that quaternized PS offered a
significant increase in stability over quaternized PAES.
Therefore, to render the cations and backbone less susceptible to
degradation in alkaline environment, the design of both the cation
and the selection of polymer backbones are crucially important.
Polyolefins, such as high density polyethylene (HDPE) or
isotactic polypropylene (iPP), are known to be highly stable in
alkaline38 and in electrochemical environments, such as
rechargeable lithium ion batteries.16–18 Several polyolefinꢀbased
ion exchange membranes have been reported.19–21 However,
water uptake behaviour, methanol permeability, and hydroxide
conductivity were investigated in detail.
5
Experimental Section
Materials
65
All O2 and moisture sensitive manipulations were carried out
inside an argon filled Vacuum Atmospheres dryꢀbox.
Polymerizationꢀgrade propylene was used as received. TiCl3·AA,
AlEt2Cl (10 wt% in toluene), trimethylamine (TMA) solution
(~45 wt% in H2O), N, Nꢀdimethylꢀ1ꢀhexadecyamine (DMHDA),
70 11ꢀbromoꢀ1ꢀundecene, calcium hydride and heptanes were
purchased from SigmaꢀAldrich. Other chemicals, including
methanol, sodium hydroxide, allylmagnesium bromide (1.0 M
solution in diethyl ether), diethyl ether (anhydrous), 4ꢀ
vinylbenzyl chloride and hydrochloric acid were analytical grade
75 and were also from SigmaꢀAldrich and used as received. Heptane
was distilled over sodium/potassium alloy with benzophenone as
indicator under argon gas protection. 11ꢀBromoꢀ1ꢀundecene was
distilled over CaH2 under reduced pressure before use.
10
15 because of synthetic difficulties – catalyst poisoning22,23 by
functional groups in the direct ZieglerꢀNatta mediated αꢀolefin
polymerization process, most functionalized polyolefins have
been prepared by postꢀfunctionalization, such as irradiationꢀ
mediated free radical graft polymerization with functionalized
20 monomer. Unfortunately, postꢀfunctionalization of free radical
graft polymers usually results in very complicated molecular
structures which have rarely been characterized.24 Recently,
Chung and coꢀworkers25,26 reported the preparation of aminoꢀ
functionalized polyolefins from silaneꢀprotected α,ωꢀaminoolefin
25 via metallocene catalyst mediated copolymerization. The
quaternized polyethyleneꢀbased AEM having ‘sideꢀchainꢀtype’
quaternary ammonium groups was obtained by interconverting
the silaneꢀprotected amino groups. High ionic conductivities of
119.6 mS/cm in 2 M HCl were observed. In addition, Coats et
30 al.27 used ringꢀopen metathesis polymerization (ROMP) with
functionalized monomer and subsequent hydrogenation reaction
to prepare solutionꢀprocessable uncrosslinked PEꢀbased AEMs,
which exhibited a high hydroxide conductivity of 65 mS·cm–1 at
50 °C. They further28 developed this synthetic approach to
35 prepare crossꢀlinked AEMs via ringꢀopening metathesis
copolymerization of cycloꢀolefins and their quaternaryꢀ
ammonium derivatives. However, the crosslinked unsaturated
ROMP polymer, in which the unsaturated double bonds might not
be alkaline stable, was not susceptible to hydrogenation. Isotactic
40 polypropylene (PP) structures usually possess higher mechanical
properties, higher melting temperatures and better filmꢀforming
capabilities than those of PE, which makes PP a viable candidate
for AEMs application. To the best of our knowledge, there are
few reports related to PP based AEMs, which is probably because
45 a smaller selection of comonomers that can be incorporated into
the PP backbone in high molar ratio in comparison with relatively
less reactive ethylene monomer.22,23
Synthesis of crosslinker p-(3-butenyl)styrene (BSt)
80
To a dry 500 mL threeꢀnecked roundꢀbottom flask equipped
with addition funnel, condenser, and magnetic stir bar was
transferred 200 mL (0.2 mol) of allylmagnesium bromide
solution. A 20 mL (0.14 mol) aliquot of 4ꢀvinylbenzyl chloride
diluted with 50 mL of diethyl ether was added dropwise to
85 allylmagnesium bromide at ice bath temperature. After complete
addition of 4ꢀvinylbenzyl chloride, the mixture was warmed to
room temperature and refluxed for 12 h. Then, 200 mL of
distilled water was slowly added to the mixture. The aqueous
layer was separated and washed three times with diethyl ether.
90 The organic phases were combined and dried with anhydrous
sodium sulfate. After evaporation of the solvent the residue was
further dried with calcium hydride and distilled under vacuum
before use. Yield: 20.1 g, 94%. 1H NMR spectrum: δ 7.1–7.5 (m,
4 H, phenyl−H), 6.7 (m, 1 H, phenyl−CH=CH2), 5.9 (m, 1 H,
95 −CH=CH2), 5.7 (d, 1 H, phenyl−CH=CH2), 5.3 (d, 1 H, phenyl
−CH=CH2), 5.1 (m, 2 H, −CH=CH2), 2.7 (t, 2 H, phenyl−CH2),
2.4 (m, 2 H, −CH2−CH=CH2).
Synthesis
of
functionalized
polypropylene
without
crosslinkable groups
100
In a typical copolymerization reaction, 100 mL of heptanes
was charged into a Parr 450 mL stainless autoclave equipped with
a mechanical stirrer in a dryꢀbox. After removal from the box, the
reactor was injected with a certain amount of 11ꢀbromoꢀ1ꢀ
undecene as comonomer and then charged with 20 psi (~1.4 atm)
The objective of the present research is the preparation and
systematic property evaluation of AEMs based on the highly
50 stable polypropylene backbone containing "sideꢀchainꢀtype" QA
groups. PP copolymer with a bromoalkyl side chain was first
synthesized via ZieglerꢀNatta catalyzed copolymerization of
propylene and 11ꢀbromoꢀ1ꢀundecene. Subsequent conversion of
the bromoalkyl to quaternary ammonium by treatment with
55 trimethylamine (TMA) or N,Nꢀdimethylꢀ1ꢀhexadecylamine
(DMHDA) and ion exchange reaction provided a PPꢀAEM with
hydroxideꢀconductive properties. Crosslinked AEM was also
prepared by introduction of crosslinkable diene monomer and
105 of propylene to saturate the heptane solution at 70 °C. About 0.4
g of TiCl3·AA and 5.0 mL of AlEt2Cl (10 wt% in toluene) was
mixed together and stirred for 30 minutes in the dryꢀbox at room
temperature to activate the catalyst and then injected into the
reactor under rapid stirring to initiate the copolymerization.
110 Additional propylene was fed continuously into the reactor to
maintain a constant pressure (20 psi) during the entire course of
the polymerization. After a certain reaction time elapsed at 70 °C,
the reaction solution was quenched by methanol, followed by
pouring the polymer solution into acidic methanol solution. The
115 polymer, denoted as PPꢀBr, was filtered and washed with THF
2
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 3 of 14
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA01420D
and methanol thoroughly and dried under vacuum at 60 °C for 12
Membrane characterization
h.
All highꢀtemperature 1H NMR spectra were recorded on a
Bruker AMꢀ300 instrument in 1,1,2,2ꢀtetrachloroethaneꢀd2 at 110
°C. The polymer molecular weights were also analyzed on a PLꢀ
60 220 series high temperature gel permeation chromatography
(GPC) unit equipped with four PLgel MixedꢀA (20µm) columns
(Polymer Laboratory Inc.). The oven temperature was at 150 °C
and the temperatures of the autosampler’s hot and the warm
zones were at 135 °C and 130 °C respectively. The solvent 1,2,4ꢀ
65 trichlorobenzene (TCB) containing ~200 ppm tris(2,4ꢀdiꢀtertꢀ
butylphenyl) phosphite (Irgafos 168) was nitrogen purged. The
flow rate was 1.0 mL/min and the injection volume was 200 µL.
A 2 mg/mL sample concentration was prepared by dissolving the
sample in N2 purged and preheated TCB (containing 200 ppm
70 Irgafos 168) for 2.5 h at 160 °C with gentle agitation. The melting
temperatures of the polymers were measured by differential
Synthesis of functionalized polypropylene with crosslinkable
groups
5
Terpolymerization of propylene with 11ꢀbromoꢀ1ꢀundecene
and pꢀ(3ꢀbutenyl)styrene was carried out under similar
procedures. 100 mL of heptanes was charged into a Parr 450 mL
stainless autoclave equipped with a mechanical stirrer in a dryꢀ
box. After removal from the box, a certain amount of 11ꢀbromoꢀ
10 1ꢀundecene and pꢀ(3ꢀbutenyl)styrene were sequentially
introduced into the autoclave under propylene atmosphere (20
psi). About 0.4 g of TiCl3·AA and 5.0 mL of AlEt2Cl (10 wt% in
toluene) was mixed together and stirred for 30 minutes in the dryꢀ
box and then injected into the flask under rapid stirring to initiate
15 the polymerization at 70 °C. After 60 minutes at 70 °C, the
reaction solution was quenched by methanol, followed by
pouring the polymer solution into acidic methanol solution. The
polymer, denoted as nxꢀPPꢀBr, was filtered and washed with THF
and methanol thoroughly and dried under vacuum at room
20 temperature for 12 h.
scanning calorimetry (DSC) using
a PerkinꢀElmer DSCꢀ7
o
instrument controller with a heating rate of 20 C/min. Thermal
stability of the samples were investigated by using a TA
75 instrument thermogravimetric analyzer (TGA) instrument model
TA SDT Q600. Preheating of the polymer samples was
performed at 100 °C for 40 min under argon atmosphere to
remove moisture. Small amounts of polymer were placed in
ceramic cells and the temperature was increased from 30 to 700
80 °C under nitrogen atmosphere with a heating rate of 20 °C/min.
Fourier transform infrared spectroscopy (FTIR) was recorded on
a PEꢀ1710 spectrometer from 4000 to 400 cm–1 with a 4 cm–1
resolution in 64 scans using polymer thin films. All mechanical
properties (tensile strength, tensile modulus, elongation at break,
85 etc.) were measured according to the ASTM Dꢀ1708 method. The
dogꢀbone specimens (38 mm × 15 mm overall size and 5 mm ×
22 mm in the gauge area) were die cut and performed using an
Instron 5866 universal tester, with a load cell of 100 N and a
constant cross head speed of 1 mm/min. At least 6 samples were
90 tested in order to minimize possible errors. To assess the
chemical stability, anion exchange membrane samples of 5 cm
diameter were immersed in 5 M and 10 M NaOH solution at 80
°C for time period ranging 0~700 h. After contact, the
membranes samples were separated, and washed with DI water
95 until the absence of the alkalinity in the effluent, followed by IEC
determination and conductivity measurement.
Evaluation of crosslinking efficiency
The resulting crosslinkable terpolymer nxꢀPPꢀBr was dissolved
in xylene at 130 °C, and the heated polymer solution was directly
cast into films (thickness 10–20 ꢁm) at 125 °C. The resulting
25 films were then heated in a vacuum oven at 220 °C for 15
minutes. The resulting crossꢀlinked films were subjected to
vigorous Soxhlet extraction with refluxing xylene for 24 h under
argon to ensure removal of the soluble polymer fraction before
drying the films and determining the gel content. The gel fraction
30 of the crosslinked membranes was calculated by measuring the
residual mass of the sample and then computed by the equation:
gel fraction = (Wd / Wi) × 100%, where Wi is the initial weight of
dried membranes and Wd is the weight of the dried insoluble
fraction of membranes after extraction.
35 Membrane Preparation
Membranes were prepared by hotꢀpress processing at 160 °C
under a pressure of 24000 psi to remove defects, providing an
average thickness in the range of 40–60 ꢁm. In the nxꢀPPꢀBr
terpolymer case, the membranes were further treated at 220 °C
40 for 15 minutes under vacuum to complete the crosslinking
reaction.
Smallꢀangle Xꢀray scattering curves of unstained dry
membranes were obtained using a Rigaku (formerly Molecular
Metrology) instrument equipped with a pinhole camera with an
100 Osmic microfocus Cu Ka source and a parallel beam optic.
Typical counting times for integration over a multiwire area
detector were 1 h with typical membrane thicknesses on the order
of 100 m. Measurements were taken under vacuum at ambient
temperature on dry samples. Scattering intensities were
105 normalized for background scattering and beam transmission.
Quaternization and Ion Exchange
PPꢀBr membranes in vials were stirred in about 30 mL of 45%
aqueous solution of TMA or DMHDA at about 35–40 °C for 36
45 h. After completion of the reactions, the membranes were washed
with deionized water to remove excess TMA or DMHDA and
then dried at 60 °C overnight, followed by vacuum at 70 °C for 8
h. The obtained membranes were treated with 1 M NaOH
solution at room temperature for 48 h to obtain PPꢀRꢀy (R =
50 TMA or DMHDA, y = Br content incorporated). The membranes
were washed thoroughly to remove residual NaOH and stored in
DI water prior to analysis. The nxꢀPPꢀBr terpolymer film samples
after thermal crosslinking at 220 °C were quaternized and ion
exchanged using similar procedure; the samples were denoted as
55 xꢀPPꢀRꢀy (R = TMA or DMHDA, y = Br content incorporated).
Ion-Exchange Capacity (IEC)
The ionꢀexchange capacities (IECs) of the membranes were
measured in triplicate using a typical titration method. Accurately
weighed samples in OH– form were immersed into 25 mL of 0.05
110 M HCl solution and equilibrated for 48 h under argon atmosphere,
after which the HCl solution was back titrated by 0.05 M NaOH
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

Journal of Materials Chemistry A
Page 4 of 14
View Article Online
DOI: 10.1039/C5TA01420D
solution using phenolphthalein as indicator. IEC values of the 45 where CA and CB are methanol concentrations of the membrane
samples were calculated using the following equation:
feed side and permeate side, respectively. A, L and VB are the
effective membrane area, thickness and the liquid volume of
permeate compartment, respectively. DK is defined as the
methanol permeability and t0 is the time lag.
n1
•
HCl − n2 HCl
•
IEC =
(meq
g−1)
•
Mdry
50 Hydroxide Conductivity
Where n1·HCl and n2·HCl are the amounts (mmol) of
hydrochloric acid required before and after equilibrium
respectively, and Mdry is the mass (g) of the dried sample. The
average value of the three samples calculated from the above
equation is the IEC value of the measured membrane.
5
AC impedance spectroscopy was used to measure the
membrane conductivity on
a
Solartron SI 1260A
Impedance/GainꢀPhase Analyzer and an electrochemical interface
(Solartron 1287, Farnborough, Hampshire, UK). The electrode
55 system was installed in an electrically shielded thermoꢀ and
hygroꢀcontrolled chamber. The impedance measurement was
carried out at a given temperature and relative humidity (RH).
The OH– conductivity (σ) was obtained from the following
equation:
Water and Methanol Uptake
10
The water and methanol uptake of the previously vacuum dried
membranes were measured for 24 h in DI water and methanol,
respectively. The reported measurements are based on the
average of two readings of each sample of dimension about 30
mm × 30 mm. Water uptake was measured at 20, 30, 40, 50, 60,
15 70 and 80 °C while methanol uptake was measured at 20 °C. The
temperature was controlled by means of a thermostatic water bath.
The weight percentage uptake W(%) was determined by the
following equation:
L
σOH – (S⋅cm−1) =
60
RA
where σ is the hydroxide conductivity (S·cm–1, R is the ohmic
resistance of the membrane sample (ꢂ). L is the distance between
the electrodes to measure the voltage drop (cm), and A is the
crossꢀsectional area of the membrane samples (cm2). In order to
65 avoid the effect of carbonation, the membranes were converted to
OH– form and washed just before the conductivity measurement.
Moreover, the tests were also conducted in water which
minimizes the exposure to air. The impedance of each sample
was measured repeatedly to ensure reproducibility for the
70 measured data.
(Ww −Wd )
W(%) =
×100%
Wd
20
where Wd and Ww are the weight of the dry and fully hydrated
membrane, respectively.
The swelling ratio was characterized by the linear expansion ratio,
which was determined by the difference between the wet and dry
dimensions of membrane samples after immersion in water at 20,
25 30, 40, 50, 60, 70 and 80 °C for 24 h (4 cm in length and 1 cm in
width). The calculation was based on the following equation:
Fuel Cell Performance Evaluation
Xwet − X
dry ×100%
A wellꢀdispersed catalyst ink was prepared by mixing 46.4
wt% Pt/C catalysts (TKK, Japan) with deꢀionized water, 1ꢀ
propanol and ASꢀ4 (Tokuyama Corporation, Japan) ionomer
75 solution (5 wt% in 1ꢀpropanol) using magnetic stirring and
ultrasonication. To obtain a catalystꢀcoated substrate (CCS) for
the electrodes, the asꢀprepared ink was coated onto the surface of
SGL 25BC carbon paper (SGL group, Germany) using an air
spray gun (Iwata, Japan). The Pt loading and ionomer content in
80 the catalyst layer were ~0.60 mg/cm2 and ~20 wt.%, respectively.
Anode CCS, PPꢀTMAꢀ20 membrane (Tokuyama Corporation,
Japan) and cathode CCS were assembled together in a cell fixture
to form a membrane/electrode assembly (MEA). The electrode
Swelling(%)=
Xwet
With the obtained weight percentage uptake W(%) and IEC
value, the hydration number (λ) was calculated by following
30 equation:
W(%)×10
IEC×18
λ
=
Methanol Permeability
Methanol permeability (P, cm2·s–1) was measured at 30 °C
using a twoꢀchamber diffusion cell method. Prior to the
35 measurement, each membrane coupon (diameter = 2.5 cm) was
soaked in water to the fully hydrated state for at least 1 day. One
chamber (80 mL) contained 10 M (34 wt %) methanol solution
and the other chamber (80 mL) was filled with DI water. The
diffused methanol was periodically measured using a Shimadzu
40 GCꢀ1020A series gas chromatography machine. Peak areas were
converted into methanol concentration with a calibration curve.
The methanol concentration in the receiving compartment as a
function of time is given by the following equation:
size was 2.25 cm
×
2.25 cm (~5 cm2). Fuel cell testing was
85 conducted with a commercial fuel cell (Teledyne, USA) testing
system at 50 °C and 100 RH %. Fully humidified hydrogen was
supplied to the anode at 500 SCCM as indicated in the data, while
fully humidified oxygen was supplied to the cathode at 500
SCCM. After full activation of the MEA for 1 h, the fuel cell
90 polarization curve was measured under galvanostatic mode, i.e.
holding the fuel cell at serial constant currents (for example, 50,
100 and 250 mA/cm2, etc.) for 3 min. The cell voltage as a
function of current density was recorded using fuel cell testing
software.
A
DK
L
CB (t) =
×
×CA ×(t− t0 )
VB
95
4
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 5 of 14
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA01420D
and typical PDI (5.7–7.0) resulting from commercially available
ZieglerꢀNatta catalyst, namely TiCl3·AA. Moreover, a higher
concentration of bromoalkyl functionalized propylene in the PPꢀ
30 Br copolymer was also achieved (7.4–20.8 mol %). As shown in
Fig. 1, the signal at 3.49 ppm could be attributed to the chemical
shift of methylene next to bromine. Three major signals at 0.95,
1.35 and 1.65 ppm, correspond to methene, methylene and
methyl groups respectively in the PP chain. The concentration of
35 bromoalkyl in resulting copolymer PPꢀBr was calculated to be as
Results and Discussion
Synthesis and Characterization of Bromoalkyl-Functionalized
Polypropylene
Anion exchange membranes based on quaternized
polypropylene derived from bromoalkyl functionalized
polypropylene precursor (PPꢀBr) were synthesized using Zieglerꢀ
Natta catalyst mediated copolymerizations of propylene and 11ꢀ
bromoꢀ1ꢀundecene, as shown in Scheme 1. While it is general
known that polar monomers deactivate the ZieglerꢀNatta catalysts
5
1
high as 20.8 mol % according to the H NMR results, when the
monomer concentration was 1.2 M and reaction time increased to
60 min. By introduction of the crosslinkable monomer pꢀ(3ꢀ
butenyl)styrene (Bst) during the copolymerization, crosslinkable
40 nxꢀPPꢀBr polymer with 20.1 mol % of bromoalkyl was also
synthesized. As shown in Fig. 1b, there are several minor
chemical shifts that are associated with the incorporated BSt
comonomer units, including three distinct olefinic proton
chemical shifts at 6.7, 5.7, and 5.2 ppm and two equal intensity
45 aromatic proton signals at 7.1–7.2 and 7.3–7.4 ppm. With their
relative chemical shift intensities, it is clear that most of the
incorporated BSt units in the copolymer contain pendent styrene
moieties. The integrated intensity ratio of the chemical shifts
between 0.9 and 1.7 ppm and the chemical shifts between 7.1–7.4
50 ppm, and the number of protons that both chemical shifts
represent, determines the concentration of BSt in the copolymers
to be 0.7 mol %.
10 commonly used for the polymerization of olefinic
hydrocarbons.29 there are examples reported in the literature
where compounds containing atoms with unshared electron pairs
do not interfere with the polymerization. Bacskai and coworkers30
reported that 8ꢀbromoꢀ1ꢀoctene could be incorporated into
15 polypropylene using Ziegler catalyst catalyzed copolymerization,
but only in around 7.0 mol % content. Herein, in comparison with
8ꢀbromoꢀ1ꢀoctene possessing 6 carbon atoms as spacer, 11ꢀ
bromoꢀ1ꢀundecene with a longer spacer of 9 carbon atoms was
selected because of its higher flexibility, which should allow it to
20 be able to compete with less bulky and reactive propylene during
copolymerization and therefore incorporate a higher molar ratio
of bromoalkyl moieties. Table
1 summarizes the reaction
conditions, polymer yield and properties. Although the catalyst
activity systematically decreased with the increase of the feed
25 content of 11ꢀbromoꢀ1ꢀundecene, the PPꢀBr showed a favorably
high molecular weight (Mw > 129 kg/mol) (Table 1 and Fig. S1)
Scheme 1 The synthesis of PPꢀRꢀy and xꢀPPꢀRꢀy membranes (R = TMA or DMHDA; y = Br content incorporated; nx: nonꢀcrosslink; x:
55 crosslink).
Table 1 Summary of ZieglerꢀNatta catalyst mediated copolymerization reaction between propylene, 11ꢀbromoꢀ1ꢀundecene (M1) and
crosslinker pꢀ(3ꢀbutenyl)styrene (M2)
Polymerization Conditions
Copolymerization results
Samples a
b
b
d
e
Propylene M1
(psi)
M2
Time
Yield
(g)
[M1]c
[M2]c
Mw
Tm
(oC)
∆He
(J/g)
PDId
(M) (M) (min)
(mol%) (mol%) (Kg/mol)
PP
PPꢀBrꢀ7
PPꢀBrꢀ20
nxꢀPPꢀBrꢀ20
20
20
20
20
0
0
0
0
30
30
60
60
10.5
6.4
2.6
2.3
0
7.4
20.8
20.1
0
0
0
590
386
148
129
5.7
6.1
6.6
7.0
162.9 78.9
150.4 53.8
115.3 31.6
114.1 27.9
0.2
1.2
1.2
0.1
0.7
a General conditions: 0.4 g of TiCl3•AA, 5 ml of Et2AlCl, 100 ml of heptanes, reaction temperature: 70 oC; b M1: 11ꢀbromoꢀundecene, M2: pꢀ(3ꢀ
butenyl)styrene (BSt); c Comonomer content (mol%) in the coꢀ and terꢀpolymers determined by 1H NMR under 110 °C in 1,1,2,2ꢀtetrachloroethaneꢀd2; d
60 Weightꢀaverage molecular weights and polydispersity index were determined by GPC at 150 °C in 1,2,4ꢀtrichlorobenzene vs narrow polystyrene
standards; e Melting temperature and heat of fusion determined by DSC.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

Journal of Materials Chemistry A
Page 6 of 14
Dynamic Article Links ►
Journal Name
View Article Online
DOI: 10.1039/C5TA01420D
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Membrane Preparation and Ion Exchange Capacity (IEC)
Although the molecular weights of PPꢀBr and nxꢀPPꢀBr were
much lower than homopolymer PP (Table 1), tough and flexible
membranes were obtained by melt hotꢀpress processing at 160 °C
under vacuum. Further thermal treatment of nxꢀPPꢀBr film at 220
°C under vacuum for 15 min led to crosslinked xꢀPPꢀBr
membrane by cycloaddition between two pendant styrene units.31
The crosslinking was further confirmed by the high gelꢀfraction
5
o
(> 91.7 % under xylene extraction at 130 C for 24 h). As shown
10 in Fig. S5 and S6, the thermal regiospecific [2+4] cycloaddition
reaction between two styrene units starts at a relatively low
o
temperature (
∼
80 C) and achieves a high degree of crosslinking
o
at 220 C after about 20 min. Subsequently, the PPꢀBr and xꢀPPꢀ
Br membranes were quaternized with TMA or DMHDA aqueous
15 solution by the Menshutkin reaction to get anion conductive
membranes. The obtained membranes in the bromide salt form
were then treated with 1 M aqueous NaOH at room temperature
for 24 h to produce anion exchange membranes. There was no
evidence of polymer chain degradation occurring under these
20 conditions, as indicated by the mechanical properties of resulting
1
polymer membranes. Fig. 2 shows the H NMR spectrum of PPꢀ
DMHDAꢀ7 in the hydroxide form. Comparison of PPꢀDMHDAꢀ7
with its parent copolymer PPꢀBrꢀ7 revealed that the signals
assigned to the nonꢀquaternized methylene groups at 3.49 ppm
25 changed to much smaller in PPꢀDMHDAꢀ7, while the polymer
mainꢀchain protons remained after the quaternized reaction. The
new signals assigned to the methylene and methyl groups in
quaternary ammonium appeared at 3.35 ppm in PPꢀDMHMAꢀ7.
However, the signal at 3.49 ppm did not disappear completely,
30 which indicates the incomplete conversion of bromoalkyl to
quaternary ammonium groups. Thus, the IEC values of PPꢀ
DMHDAꢀ7 membranes were readily calculated to be 0.91 meq g–
Fig. 1 1H NMR spectra of (a) PPꢀBrꢀ20 and (b) nxꢀPPꢀBrꢀ20 in
1,1,2,2ꢀtetrachloroethaneꢀd2.
1
by comparing the integration ratios of the individual signals,
which correlated well with the titrated value 0.87 meq g–1 (Table
35 3), but apparently lower than its theoretical value of 1.26 meq g–1.
40 Fig. 2 1H NMR spectra of (a) PPꢀBrꢀ7 and (b) PPꢀDMHDAꢀ7
membrane in the hydroxide form in 1,1,2,2ꢀtetrachloroethaneꢀd2.
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 6

Page 7 of 14
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA01420D
1
The H NMR spectra of PPꢀTMAꢀ20 and PPꢀDMHDAꢀ20 were
not able to be obtained due to their poor solubility in organic
solvent. However, the chemical structures of PPꢀAEMs were
further analyzed by FTꢀIR spectra, as shown in Fig. 3 and Fig.
S2. The absorption bands in the ranges of 2825–2870 cm–1 and
2870–2965 cm–1 are characteristic of symmetric and asymmetric
stretching of aliphatic –CH2 and –CH3 groups, while C–H
stretching and bending absorption bands for PPꢀbased polymer
appeared at 1472 cm–1 and 1463 cm–1, respectively. The C–N
Thermal and Mechanical Properties
5
10 stretching bands at 1100 cm–1 and 1126 cm–1 in xꢀPPꢀTMAꢀ20
and xꢀPPꢀDMHDAꢀ20 supported successful quaternization of the
membrane. Furthermore, the broad absorption band in the range
of 3400–3200 cm–1 in xꢀPPꢀDMHDAꢀ20 and xꢀPPꢀTMAꢀ20
membranes is attributed to O–H stretching in H2O. These spectral
15 features further supported that high concentration of –Br group
was incorporated into the PPꢀbased membrane and subsequently
quaternized with TMA or DMHDA to form anion conductive
membranes. The experimentally obtained IEC values of the
membranes were titrated to be 1.48 and 1.42 meq. g–1 for PPꢀ
20 DMHDAꢀ20 and xꢀPPꢀDMHDAꢀ20 respectively, which was
lower than the calculated values (1.66 and 1.65 meq. g–1) from
the copolymer composition and the degree of bromination further
suggesting that incomplete quaternization of the bromoalkyl
groups occurred. Moreover, the titrated IEC values of PPꢀTMAꢀ
25 20 and xꢀPPꢀTMAꢀ20 membranes based on TMA (1.56 and 1.34
meq g–1 respectively) are also apparently lower than the
theoretical values 2.56 and 2.33 meq g–1 (based on the degree of
45
Fig. 4 DSC thermograms of set
I (top): a) PPꢀBrꢀ20 copolymer
(Tm=115.3 °C, ∆H=31.6 J/g), b) PPꢀTMAꢀ20 (Tm=115.1 °C,
∆H=27.1 J/g) and c) PPꢀDMHDAꢀ20 membranes (Tm=115.0 °C,
∆H=25.4 J/g); set II (bottom): d) nxꢀPPꢀBrꢀ20 (Tm=114.1 °C,
50 ∆H=27.9 J/g), e) xꢀPPꢀTMAꢀ20 (Tm=111.2 °C, ∆H=21.2 J/g) and
f) xꢀPPꢀDMHDAꢀ20 (Tm=111.5 °C, ∆H=20.3 J/g) membranes.
The DSC curves of PPꢀBrꢀ20 copolymer and two corresponding
membranes PPꢀTMAꢀ20, PPꢀDMHDAꢀ20 in the hydroxide form
are shown in set
I of Fig. 4 respectively ((a)–(c) from bottom to
55 top). The endothermic peak at 115.3 °C represents the Tm of PPꢀ
Brꢀ20 copolymer. Negligible decreases in Tm transition were
observed after quaternization, indicating the original crystallites
in the modified membranes remained intact, since Tm is a function
of changes in the crystalline structure. Meanwhile, quaternization
60 and subsequent ion exchange results in a dilution effect on the
crystallinity of the copolymer, which leads to significant
decreases in the areas (equal to ∆H values) under the curve. The
DSC curves in set II ((d)–(f) from bottom to top) exhibit the nxꢀ
PPꢀBrꢀ20, xꢀPPꢀTMAꢀ20 and xꢀPPꢀDMHDAꢀ20 after thermal
65 crossꢀlinking reaction, quaternization and ion exchange. As
shown in Fig. 4 set II, the crossꢀlinking slightly reduced Tm and
crystallinity. For instance, the Tm of nxꢀPPꢀBrꢀ20 was 114.1 °C,
which was slightly higher than that of crosslinked xꢀPPꢀTMAꢀ20
(Tm=111.2 °C) and xꢀPPꢀDMHDAꢀ20 (Tm=111.5 °C). Significant
70 decreases of ∆H values were also observed after crossꢀlinking or
quaternization, which could be contributed to both the crossꢀ
linking induced crystallinity reduction effect and dilution effect
from quaternization.
functionality assuming 100% amination efficiency, 100
%
conversion to hydroxide form, and 100 % water removal when
30 drying). The values determined by titration were between 58 %
and 61 % of the calculated values, which could be attributed to
incomplete conversion of bromoalkyl groups to quaternary
ammonium groups. The similarly incomplete amination has been
observed previously in chloromethylated poly(arylene ether
35 sulfone).32
The thermal stability of the membranes in their hydroxide form
75 was evaluated by TGA under argon atmosphere. As shown in Fig.
5, there was no weight loss up to 165 °C, because all the samples
were preꢀheated at 120 oC for 12 h to remove absorbed water. As
a control experiment, the thermal stability of PP and nxꢀPPꢀBrꢀ20
was also compared by TGA. The PP and nxꢀPPꢀBrꢀ20 showed
80 excellent thermal stability. The main chain initial degradation
temperature was up to 310 and 360 °C respectively. A twoꢀstep
degradation profile was observed for quaternized PP copolymers
in their hydroxide form (Fig. 5). The first weight loss occurred
above 165 oC, which is associated with the degradation of the
85 quaternary ammonium groups. This degradation temperature is
higher than typically encountered quaternized aromatic polymers,
such as quaternized PPO (~145 oC).7 The high decomposition
Fig. 3 FTꢀIR spectra of (a) PP, (b) nxꢀPPꢀBrꢀ20, (c) xꢀPPꢀ
DMHDAꢀ20 and (d) xꢀPPꢀTMAꢀ20 membranes in the hydroxide
form.
40
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

Journal of Materials Chemistry A
Page 8 of 14
View Article Online
DOI: 10.1039/C5TA01420D
Table 2 Summary of mechanical properties of the polymers
membranes and quaternized polypropylene membranes in the
hydroxide form at room temperature, 30 % RH
temperature suggests that quaternary ammonium groups attached
to pendent alkyl have better thermal stability. The main weight
loss at around 310–450 oC is related to the degradation of the
polymer chain. The decomposition temperature of the polymer
backbone was similar with its parent polymer, suggesting that the
decomposition of the quaternary ammonium groups did not
trigger degradation of the polymer backbone, which has been
observed in some aromatic AEMs.33
Young’s
modulus
(MPa)
Elongation
at break
(%)
Tensile strength
(MPa)
sample
5
PP
PPꢀBrꢀ20
35±5
21±4
17±5
16±4
19±4
62±5
64±5
1285±125
785±65
635±55
588±45
751±60
458±82
346±55
328±48
312±45
315±65
233±47
216±42
PPꢀTMAꢀ20
PPꢀDMHDAꢀ20
nxꢀPPꢀBrꢀ20
xꢀPPꢀTMAꢀ20
xꢀPPꢀDMHDAꢀ20
1588±155
1627±162
45 Water Uptake Behaviors and Dimensional Swelling
Water uptake (WU) of AEMs is an important parameter for IEC,
ionic conductivity, dimensional stability, mechanical strength and
membrane electrode compatibility. Table 3 compares the IEC,
water uptake, swelling ratio in the inꢀplane direction of the PPꢀ
50 based membranes. Generally, the water uptake of membranes
increased with IEC value, due to increased hydrophilicity. The
o
highest water uptake was 36.5 wt % at 20 C for the highest IEC
membrane PPꢀDMHDAꢀ20 (1.48 meq g–1). The WU value was
much lower than those of quaternized polyethylene (IEC=1.29
55 meq g–1, WU = 97 wt %)34 and aromatic copolymers.35 The
number of absorbed water molecules per quaternary ammonium
(QA) group, the hydration number λ, was calculated to be lower
than 14, which is also much lower than that of the quaternized
polyethylene (λ=41.8).36 The lower water uptakes are considered
60 to be related to the hydrophobicity of the long alkyl chain and the
‘sideꢀchainꢀtype’ architecture, which could contribute to
suppression of excessive water uptake and swelling. Lower water
uptake of PPꢀbased membranes leads to lower dimensional
swelling in water, as shown in Table 3. The inꢀplane water
65 swelling ratio at 20 oC was less than 10 %. Moreover, the thermal
crosslinking drastically lowered the water uptake and thus the
dimensional swelling ratio of the materials (Table 3).
Specifically, the crosslinked xꢀPPꢀTMAꢀ20 membrane had a
water uptake of 20.2 wt % and swelling ratio of 3.5 % at room
70 temperature, which is significantly improved compared to the
uncrosslinked PPꢀTMAꢀ20 that absorbed 34 wt % water and
swelled 8.1 % in water. These values are also lower than that of
QAꢀPAES membrane.35 At higher temperature (80 oC), the
crosslinked membranes retained excellent dimensional stability
75 with equal to or less than 25 wt % water uptake and 6 % swelling
ratio, as shown in Fig. 6. Considering the remarkable agreement
between the high solvent resistance and low swelling ratio, a
strong covalent crosslinking network was likely formed by
thermal crosslinking, which was further confirmed by the high
80 gel fraction as mentioned above. The low inꢀplane swelling is
beneficial in reducing mechanical failures in the membrane
electrode assembly (MEA), since high inꢀplane swelling deforms
the membrane electrode interface. Unlike several other previously
reported AEMs in which higher temperature induced excessive
85 water uptake and dimensional swelling, the water uptake and
swelling ration of PP AEMs shown very low dependence on
temperature, as shown in Fig. 6. These results indicated that the
'sideꢀchainꢀtype' from hydrophobicity of spacer alkyl chain
restricts water absorption of membranes more efficiently at
90 higher temperatures.
10 Fig. 5 TGA curves of (a) PP and (b) nxꢀPPꢀBrꢀ20, (c) xꢀPPꢀ
TMAꢀ20 and (d) xꢀPPꢀDMHDAꢀ20 membranes in their
hydroxide form.
The mechanical properties of the membranes are shown in Table
2. It is well known that PP film has good mechanical properties.
15 Although the introduction of bromoalkyl during copolymerization
decreased the molecular weight of PP to some extent (as
discussed above, Table 1), the PPꢀBrꢀ20 and nxꢀPPꢀBrꢀ20 still
had sufficiently high molecular weight and exhibited excellent
mechanical properties. The membranes displayed values of
20 tensile strength at maximum load of 16–21 MPa, Young's moduli
of 0.58–0.78 GPa, and values of elongation at break of 310–350
%. These results indicate that PPꢀBr precursor membranes were
tough and flexible for quaternization as AEMs. Furthermore, the
slightly lower tensile strength, Young's modulus and elongation
25 at break for quaternized PP AEMs was observed compared to
their corresponding precursor polymers. Like previously reported
ionꢀcontaining aromatic copolymers,33 in which the ionic groups
induced lower tensile strength and Young's modulus due to
plasticization from absorbed water, the quaternization tended to
30 decrease the mechanical properties, as shown in Table 2.
However, the crosslinked quaternized xꢀPPꢀTMAꢀ20 and xꢀPPꢀ
DMHDAꢀ20 membranes had tensile strengths of ~ 62 MPa and
Young's moduli of more than 1.5 GPa (Table 2). These values are
much higher than that of precursor nxꢀPPꢀBrꢀ20 or even PP due
35 to the more compact chain architecture results from the
crosslinking.41 Similar behavior was also observed in crosslinked
quaternized aromatic copolymers.42,43
40
8
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 9 of 14
Journal of Materials Chemistry A
Dynamic Article Links ►
Journal Name
View Article Online
DOI: 10.1039/C5TA01420D
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Table 3 Physicoꢀchemical properties of PP based AEMs at room temperature (20 oC).
WU
(wt %)
Swelling
(%)
MUc
(wt %)
PMeOH
σ
Samples
IECa
IECb
λ
(×10ꢀ7 cm2 sꢀ1)
(mS cmꢀ1)
PPꢀDMHDAꢀ7
PPꢀTMAꢀ20
0.87
1.56
1.26
2.57
8.3
34.0
2.4
8.1
8.9
12.1
7.1
31.2
0.17
2.19
6.1
17.4
PPꢀDMHDAꢀ20
xꢀPPꢀTMAꢀ20
1.48
1.34
1.42
1.66
2.33
1.65
36.5
20.2
21.1
8.4
3.5
3.8
13.7
8.4
32.8
16.2
17.2
2.48
0.39
0.45
19.2
14.6
16.3
xꢀPPꢀDMHDAꢀ20
8.3
QAꢀPE55
QAꢀPAES56
QAꢀPPOs28
1.2
1.82
1.39
97.0
20.0
25.9
ꢀꢀ
11
7
41.8
5.7
10.4
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
40.0
22.0
4.0
a measured by titration (meq. gꢀ1); b theoretical IEC values (meq. gꢀ1); c methanol uptake (MU) at 20 oC.
25 normalizedꢀhydroxide conductivity than that of traditional QAꢀ
PPO and PEꢀbased AEMs (Fig. 7b) if the water uptake of AEMs
is taken into account. Particularly, the crosslinked xꢀPPꢀ
DMHDAꢀ20 membranes exhibited extremely high normalizedꢀ
hydroxide conductivity of 1.96 mS·cm–1. This value is much
30 higher than that of QAꢀPPO8 and QAꢀPE34 (0.38 and 0.96 mS
cm–1, respectively). Thus, it is believed that the sideꢀchainꢀtype
architecture in PPꢀbased AEMs contributes strongly to the high
hydroxide conducting. The sideꢀchainꢀtype structure induces the
PPꢀbased AEMs to utilize water molecules more efficiently for
35 hydroxide transport. As shown in Fig. S7, the polymers showed
obvious characteristic ionomer peaks in the SAXS, indicating the
formation of nanophase separation with ionic domains. These
results suggest that our concept to apply the 'sideꢀchainꢀtype'
structure is effective for the AEMs for mitigating water swelling
40 and improving conductivity. However, both the hydroxide
conductivities and water uptake normalizedꢀhydroxide
conductivities are much lower than those of QAꢀPAES AEM35
with few exception37 in spite of their similar IEC value. This is
because the PPꢀbased AEMs were fabricated by a hotꢀpress
45 process of nonꢀionic bromoalkylated PP membrane and
subsequent amination reaction. Wellꢀdefined microphase
separation is hard to form for hydroxide transport during the hotꢀ
press of nonꢀionic bromoalkylated PP membranes. The lack of a
second order scattering peak for all membranes in SAXS results
50 (Fig. S7) suggests that the arrangement of the phase separated
domains is only locally correlated, and no longꢀrange ordered
structures formed in these materials. The correlation between
hydroxide conductivity and temperature for PPꢀbased AEMs is
shown in Fig. 8. The conductivity steadily increased with
55 temperature due to enhanced water mobility, which allows
transport of hydroxide ions by the vehicle mechanism. The
hydroxide conductivity values of 51.8–58.2 mS cm–1 for PPꢀ
AEMs are comparable to the reported values for polysulfoneꢀ
based AEMs having high IEC values (IEC>2 meq g–1).35
5
Fig. 6 The dependence of water uptake and swelling ration on
temperature.
Hydroxide Conductivity
Hydroxide conductivity is a critical performance metric for
AEMs. There is a requirement for the hydroxide conductivity to
10 be above 10 mS·cm–1 and for water uptake to be less than 30
wt %. The hydroxide conductivity of the PPꢀbased AEMs was
measured in water and compared with that of QAꢀPPO and PEꢀ
based AEMs, and reported in Table 3 are expressed in Fig. 7a as
a function of IEC. These values were above 10 mS cm–1 at 20 oC.
15 The PPꢀDMHDAꢀ20 membrane having a water uptake of 36.5
wt % achieved the highest hydroxide conductivity of 19.2 mS
o
cm–1 at 20 C. The important point to note is that the crosslinked
membranes showed a much lower water uptake but a comparable
conductivity value to the uncrosslinked PPꢀbased AEMs (Table 3
20 and Fig. 7a). For example, the crosslinked xꢀPPꢀDMHDAꢀ20
exhibits a hydroxide conductivity of 16.3 mS cm–1, although its
water uptake is only 21.1 wt %. While lower hydroxide
conductivities was observed for PPꢀAEMs than that of QAꢀPE at
a similar IEC value, the PPꢀAEMs displayed higher water uptake
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 9

Journal of Materials Chemistry A
Page 10 of 14
View Article Online
DOI: 10.1039/C5TA01420D
morphology and high proton mobility, but its high methanol
25 permeability is a shortcoming in its application in DMFC. Thus,
many strategies have been developed to decrease methanol
uptake and permeability in polymer electrolyte membranes,
including using hydrocarbon polymer backbone, crosslinking etc.
As shown in Table 3, the highest methanol uptake of PPꢀbased
30 AEMs was ~32.8 %, which is much lower than that of Nafion
212 (103 %) at 20 °C. The methanol permeabilities of PPꢀbased
o
AEMs were also measured at 20 C in 2 M aqueous methanol.
The PPꢀbased AEMs exhibited comparatively lower methanol
permeability to Nafion, as shown in Fig. 9. The permeability
35 values for 2 M methanol at room temperature were in the range of
0.17 × 10–7 to 2.48 × 10–7 cm2 s–1, which is almost one order of
magnitude lower than that of Nafion (1.67 × 10–6 cm2 s–1).
Although the PPꢀbased AEMs have lower conductivities than
Nafion, the hydroxide conductivities are sufficiently high to
40 achieve much better selectivity, which is the ratio of ion
conductivity to methanol permeability (Fig. 9). Moreover, the
crosslinked PPꢀbased AEMs xꢀPPꢀTMA and xꢀPPꢀDMHDA
showed decreased methanol permeability and increased
selectivity when compared with the uncrosslinked counterpart, as
45 shown in Fig. 9. These results indicate that the crosslinked AEMs
help reduce methanol permeability while retaining effective
hydroxide transport.
Fig. 7 (a) Conductivity of PPꢀbased AEMs, QAꢀPE55, QAꢀ
PAES56 and QAꢀPPO28 membranes at 20 °C as a function of IEC
values; (b) the number of absorbed water molecules per
quaternary ammonium (QA) group (λ) normalized hydroxide
conductivity of AEMs as a function of the hydration number (λ).
5
Fig. 9 Relationship of methanol permeability with hydration
50 number (λ) and IEC values of membranes.
Alkaline stability
The longꢀterm stability of AEMs is of concern due to wellꢀknown
degradation pathways for tetraalkylammonium ions under
alkaline conditions including betaꢀhydrogen Hofmann
55 elimination, direct nucleophilic substitution at an alphaꢀcarbon, or
nitrogen ylide formation.4, 38–40 The alkaline stability of the PPꢀ
based AEMs was determined by observing changes in
conductivities and IEC values with time in 5 M and 10 M NaOH
solution at 80 °C for 700 h. All of the membranes remained tough
60 and flexible throughout the test due to the exceptional stability of
the polyproplyene backbone. As shown in Fig. 10, the decreases
in conductivity are all accompanied by similar decreases in IEC,
confirming that the conductivity losses are due to cation
degradation. After an initial period of rapid conductivity decline
65 over the first 100 h, the IEC values and hydroxide conductivities
of all membranes showed a slow, steady decline, as seen in Fig.
10. The enhanced stability behavior, which is similar with our
previous AEMs based on PPO, might be due to steric shielding of
cationic domains with alkyl chains. This finding is supported by
70 the alkaline stability of PPꢀbased AEMs containing an extended
Fig. 8 Hydroxide conductivity of PPꢀbased AEMs as a function
10 of temperature.
Methanol Uptake and Permeability
Methanol can be used as fuel in solid alkaline fuel cells, thus
methanol permeability is one of the key parameters to evaluate
AEMs for alkaline fuel cells application. Methanol permeability
15 is the product of the diffusion coefficient and the sorption
coefficient and is used to describe the transport of methanol
through membranes. Diffusion or leakage of the fuel across the
membrane from anode to cathode leads not only to power loss
from mixed potentials, but also other undesirable consequences,
20 such as complicated water and thermal management. Thus, high
methanol uptake or crossover is a serious obstacle for membranes
in direct methanol fuel cell (DMFC) applications. Nafion has
good conductivity due to strongly interconnected ionic domain
10
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 11 of 14
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA01420D
alkyl chain (DMHDA). For example, the PPꢀDMHDAꢀ20
membrane in 5 M aqueous NaOH showed a lower hydroxide
conductivity drop of ~ 10 % in the initial 100 h testing period
than that of PPꢀTMAꢀ20 membrane (~18 %). Moreover,
improved alkaline stability was observed for crosslinked xꢀPPꢀ
DMHDAꢀ20 compared to nonꢀcrosslinked PPꢀDMHDAꢀ20 (Fig.
10a). While the PPꢀDMHDAꢀ20 showed a gradual decrease in
hydroxide conductivity of about ~13 % over the 700 h test,
crosslinked xꢀPPꢀDMHDAꢀ20 showed only a ~ 5 % drop, as
5
10 shown in Fig. 10a. The IEC values of the PPꢀbased AEMs
changed little during the stability testing period (Fig. 10b), which
further confirms their improved alkaline stability. It is interesting
to note that the PPꢀbased AEMs retained more than 85 % of their
initial hydroxide conductivity even under very severe conditions
o
15 (80 C, 10 M aqueous NaOH), as shown in Fig. 10c, despite the
fact that the cations are susceptible to Hofmann elimination
reactions. The comparison of FTꢀIR results for xꢀPPꢀDMHDAꢀ20
membrane before and after alkaline stability testing in 10 M
NaOH aqueous provides further evidence supporting its excellent
20 alkaline stability. As shown in Fig. S3, no obviously variation in
intensity of absorption bands of C–N and –OH etc was observed
after 700 h testing. This finding is supported by a recent
computational study using model compounds, which revealed
that the attachment of quaternary ammonium groups to the
25 polymer backbone having an alkyl spacer of > 3 carbon atoms
can lead to improved alkaline stability of the cation.36 This
stability is the result of both steric shielding of the βꢀhydrogens
when n > 3 and an increased susceptibility to SN2 attack at the
nonꢀmethyl substituent on the nitrogen atom in BTMA. A similar
30 stability result for a poly(phenylene) backbone with “sideꢀchainꢀ
Fig. 10 The changing trend for PPꢀbased AEM membranes after
immersion at 80 °C in (a) hydroxide conductivity and (b) IEC in
45 5 M NaOH, and (c) hydroxide conductivity in 10 M NaOH.
Fuel Cell Performance
type” trimethylalkylammonium cations attached by
a
hexamethylene spacer was also reported by Hibbs.13 In addition
to the “sideꢀchainꢀtype” architecture, the highly baseꢀstable PP
backbone likely also contributes to the excellent alkaline stability
35 of PPꢀbased AEMs; previous reports by Hickner and coworkers15
shown that base stable polymer backbone would induce a better
alkaline stability of AEMs. Thus, the PPꢀbased AEMs with “sideꢀ
chainꢀtype” cations appear to be very promising candidate for use
in AEMFCs.
Figure 11 shows the initial polarization curves of an H2/O2
AEMFC with PPꢀTMAꢀ20 as electrolyte membrane. The open
circuit voltages (OCVs) are close to the theoretical value of about
50 1.1 V, indicating that the PPꢀTMAꢀ20 membrane do not affect the
catalyst function of Pt significantly. The peak power density for
the AEMFC with the PPꢀTMAꢀ20 membrane was 122 mW cmꢀ2.
The higher power density and smaller internal resistance indicate
that the sideꢀchain type PP AEM is a promising electrolyte
55 membrane that shows solubility in lowꢀboilingꢀpoint solvents for
membrane processing, but is insoluble during alkaline fuel cell
operation.
40
Fig. 11 Polarization curves (open symbols) and power density
60 curves (filled symbols) of PPꢀTMAꢀ20 membrane incorporated
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 11

Journal of Materials Chemistry A
Page 12 of 14
View Article Online
DOI: 10.1039/C5TA01420D
AEMFC. Test conditions: membraneꢀthickness of 45 µm, cell
1
F. Zhang, H. Zhang and C. Qu, J. Mater. Chem., 2011, 21, 12744–
12752.
K. D. Kreuer, Chem. Mater., 1996, 8, 610–641.
W. Li, J. Fang, M. Lv, C. Chen, X. Chi, Y. Yang and Y. Zhang, J.
Mater. Chem., 2011, 21, 11340–11346.
G. Merle, M. Wessling and K. Nijmeijer, J. Membr. Sci., 2011, 377,
1–35.
J. S. Spendelow and A. Wieckowski, Phys. Chem. Chem. Phys.,
2007, 9, 2654–2675.
J. Pan, S. Lu, Y. Li, A. Huang, L. Zhuang and J. Lu, Adv. Funct.
Mater., 2009, 19, 1–8.
N. Li, L. Wang and M. A. Hickner, Chem. Comm., 2014, 50, 4092–
4095.
N. Li, T. Yan, T. ThurnꢀAlbrecht and W. H. Binder, Energy.
Environ. Sci., 2012, 5, 7888–7892.
N. Li, Y. Leng, M. A. Hickner and C.ꢀY. Wang, J. Am. Chem. Soc.
2013, 135, 10124–10133.
o
temperature of 50 C, catalyst loadings of 0.5 mg Pt cmꢀ2 (Pt/C)
for both anode and cathode, gas flow rate of 0.5 L minꢀ1 for both
H2 and O2.
60
65
70
2
3
4
5
6
7
8
9
5
Conclusions
We have designed and synthesized “sideꢀchainꢀtype”
polypropyleneꢀbased anion exchange membrane with high base
stability using heterogeneous ZieglerꢀNatta catalyst mediated
polymerization that provide multiple structural variations to allow
10 tuning of the membrane properties. The side chain type PPꢀbased
AEMs exhibited comparable hydroxide conductivity to typical
AEMs based on benzyltrimethyl ammonium motif in spite of
their low water uptake. The PPꢀbased AEMs had unusually high
alkaline stability for up to 700 h in 5 M and 10 M NaOH at 80
15 °C. The steric effects of the long alkyl chains in ‘sideꢀchainꢀtype’
architecture surrounding the quaternary ammonium center are
likely the cause of the observed good alkaline stability.
Crosslinkable PPꢀbased AEMs were also obtained simply by
copolymerization of a thermally crosslinkable monomer. All of
20 the crosslinked membranes had better overall properties including
lower water uptake, lower water swelling ratio, better mechanical
properties, and lower methanol permeability, compared to those
of the uncrosslinked counterparts, while having comparable
hydroxide conductivities. A combination of good thermal and
25 chemical stability, excellent mechanical properties and excellent
balance between hydroxide conductivity and swelling or
methanol transport makes PPꢀbased membrane attractive as AEM
materials for alkaline fuel cells applications. We consider that
these ‘sideꢀchainꢀtype’ polypropylene synthesized by commercial
30 heterogeneous ZieglerꢀNatta catalyst mediated polymerization
could lead to new materials for the production of AEMs that meet
the demanding challenges of alkaline fuel cells and other
electrochemical processes. The general approach is versatile for
varying the length of the sideꢀchain and the cations, which may
35 be lead to further tuning and refinement of AEMs to address the
various separations and energyꢀfocused applications.
75 10 T. Xu, D. Wu and L. Wu, Prog. Polym. Sci., 2008, 33, 894–915.
11 J. Miyake, K. Fukasawa, M. Watanabe, K. Miyatake, J. Polym. Sci.,
Polym. Chem., 2014, 52, 383–389.
12 L. Wang and M. A. Hickner, Polym. Chem., 2014, 5, 2928–2935.
13 M. R. Hibbs, J. Polym. Sci., Polym. Phys., 2013, 51, 1736–1742.
80 14 C. G. Arges and V. Ramani, Proc. Natl. Acad. Sci. U.S.A. 2013, 110,
2490−2495.
15 S. A. Nuñez and M. A. Hickner, ACS Macro Lett., 2013, 2, 49–52.
16 B. Bae, B. H. Chun and D. Kim, Polymer, 2001, 42, 7879–7885.
17 G. Venugopal, J. Moore, J. Howard and S. Pendawar, J. Power
85
Sources., 1999, 77, 34–41.
18 H. Matsuyama, M. Yuasa, Y. Kitamura, M. Teramoto and D. R.
Lloyd, J. Membr. Sci., 2000, 179, 91–100.
19 G. K. Kostov and S. C. Turmanova, J. Appl. Polym. Sci., 1997, 64,
1469–1475.
90 20 W. M. Trochimczuk, J. Polym. Sci.: Polym. Chem. Ed., 1975, 13,
357–363.
21 S. M. Kolhe and A. Kumar, Radiat. Phys. Chem., 2005, 74, 384–
390.
22 T. C. Chung, Functionalization of Polyolefins; Academic Press:
95
100
105
London, 2002.
23 T. C. Chung, Prog. Polym. Sci., 2002, 27, 39–85.
24 G. Ruggeri, M. Aglietto, A. Petragnani and F. Ciardelli, Euro.
Polym. J., 1983, 19, 863–866.
25 M. Zhang, H. K. Kim, E. Chalkova, F. Mark, S. N. Lvov and T. C.
Chung, Macromolecules, 2011, 44, 5937–5946.
26 M. Zhang, X. Yuan, L. Wang, T. C. Chung, T. Huang and W.
deGroot, Macromolecules, 2014, 42, 571–581.
27 T. J. Clark, N. J. Robertson, H. A. Kostalik, E. B. Lobkovsky, P. F.
Mutolo, H. D. Abruña and G. W. Coats, J. Am. Chem. Soc., 2009,
131, 12888–12889.
Acknowledgement
This work was supported by the National Natural Science
Foundation of China under Grant 21404084 and 21474126, and
40 by the State Key Laboratory of Solidification Processing in
NWPU (Grand No. G8QT0298) and the 111 Project (B08040).
28 N. J. Robertson, H. A. Kostalik, T. J. Clark, P. F. Mutolo, H. D.
Abruña and G. W. Coats, J. Am. Chem. Soc., 2010, 132, 3400–3404.
29 C. Natta, J. Polym. Sci., 1960, 48, 219–239.
30 R. Bacskai, J. Polym. Sci: Part A., 1965, 3, 2491–2510.
110 31 W. Lin, Z. Shao, J. Dong and T. C. Chung, Macromolecules, 2009,
42, 3750–3754.
Notes and references
32 M. R. Hibbs, M. A. Hickner, T. M. Alam, S. K. Mcintyre, C.
Fujimoto, J. Cornelius, Chem. Mater., 2008, 20, 2566–2573.
33 M. Tanaka, M. Koike, K. Miyatake and M. Watanabe, Polym. Chem.
a
State Key Laboratory of Solidification Processing, Northwestern
Polytechnical
University,
Xi’an,
710072,
China.
Eꢀmail:
115
120
2011, 2, 99−106.
45 wangyiguang@nwpu.edu.cn
34 H. A. Kostalik, T. J. Clark, N. J. Robertson, P. F. Mutolo, J. M.
Longo, H. D. Abruña and G. W. Coates, Macromolecules, 2010, 43,
7147–7150.
35 J. Wang, J. Wang, S. Li and S. Zhang, J. Membr. Sci., 2011, 368,
246–253.
36 S. Chempath, J. M. Boncella, L. R. Pratt, N. Henson and B. S.
Pivovar, J. Phys. Chem. C., 2010, 114, 11977−11983.
37 N. Li, Q. Zhang, C. Wang, Y. M. Lee, M. D. Guiver,
Macromolecules, 2012, 45, 2411–2419.
b
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry,
Chinese Academy of Sciences, Taiyuan, 030001, China. Eꢀmail:
linanwen@sxicc.ac.cn
c
Department of Materials Science and Engineering, Advanced Materials
50 Processing and Analysis Center, University of Central Florida, Orlando,
FL 32816, USA
d State Key Laboratory of Engines, School of Mechanical Engineering,
Tianjin University, Tianjin 300072, China
e Collaborative Innovation Center of Chemical Science and Engineering
55 (Tianjin), Tianjin 300072, China
125 38 J. R. Varcoe, P. Atanassov, D. R. Dekel, A. M. Herring, M. A.
Hickner, P. A. Kohl, A. R. Kucernak, W. E. Mustain, K. Nijmeijer,
K. Scott, T. W. Xu and L. Zhuang, Energy. Environ. Sci., 2014, 7,
3135–3191.
† Electronic Supplementary Information (ESI) available: See DOI:
10.1039/b000000x/
12
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 13 of 14
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA01420D
39 N. Li and M. D. Guiver, Macromolecules 2014, 47, 2175–2198.
40 A. D. Mohanty and C. Bae, J. Mater. Chem. A., 2014, 2, 17314–
17320.
41 W. Chiu and S. Fang, J. App. Polym. Sci., 1985, 30, 1473ꢀ1489.
42 T. Xu, F. Zhang, J. Membr. Sci. 2002, 199, 203ꢀ210.
43 J. Qiao, J. Fu, L. Liu, Y. Liu, J. Sheng, Int. J. Hydrogen
Energy, 2012, 37, 4580ꢀ4589.
5
15
20
10
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 13

Journal of Materials Chemistry A
Page 14 of 14
View Article Online
DOI: 10.1039/C5TA01420D
“Side-chain-type” and crosslinkable quaternized polypropylene was prepared by
heterogeneous Ziegler-Natta catalyst mediated polymerization for highly stable anion
exchange membranes