New Anion Conducting Membranes Based on Functionalized Styrene

ButadieneStyrene Triblock Copolymer for Fuel Cells Applications

MARCO FARAJ,1 ELISABETTA ELIA,1 MASSIMILIANO BOCCIA,2 ANTONIO FILPI,1 ANDREA PUCCI,1,3,4 FRANCESCO CIARDELLI1,5
1

Dipartimento di Chimica e Chimica Industriale, Universita` di Pisa, Via Risorgimento 35, I-56126 Pisa, Italy

Acta S.p.A., via di Lavoria 56/G, I-56042 Crespina (PI), Italy

CNR NANO, Istituto Nanoscienze-CNR, piazza San Silvestro 12, I-56127 Pisa, Italy

INSTM, Unita` di Ricerca di Pisa, Via Risorgimento 35, 56126 Pisa, Italy

Spin-Pet s.r.l., Via Risorgimento 35, 56126 Pisa, Italy

Received 14 March 2011; accepted 13 May 2011

DOI: 10.1002/pola.24781
Published online 7 June 2011 in Wiley Online Library (wileyonlinelibrary.com).

ABSTRACT: In this work, the functionalization of polystyrene-b-

poly(butadiene)-b-polystyrene triblock copolymer (SBS) with

vinylbenzyl chloride and benzoyl peroxide (BPO) or a,a0 -azobis-isobutyronitrile (AIBN) as free radical initiators was
reported. The functionalization degree (FD), calculated by 1H
NMR spectroscopy and confirmed by elemental analysis, was
highly tunable (from 4 to 10 mol %) and positively correlated
to the starting percentage of radical initiator. More specifically,
at the same initiator molar percentage grafting efficiency is
higher using BPO rather than AIBN. Quaternization reaction of
the grafted benzyl chloride groups with the bifunctional tertiary
amine 1,4-diazabicyclo[2.2.2]octane (Dabco) led to a chemically
and thermally stable homogeneous anion-exchange membrane. Electrochemical parameters were evaluated for Dabco-

quaternized grafted copolymers having different FDs, and compared with a commercial Tokuyama benchmark membrane. Experimental data showed a positive correlation between FD and
both water swelling and ionic conductivity. Best trade-off
between ionic conductivity and water swelling was found for
membrane having FD 9.1 mol %, which conductivity is comparable with the Tokuyama benchmark one and water uptake is
only slightly higher. The results are discussed based on the
molecular parameters with particular reference to ionic content
C 2011 Wiley Periodicals, Inc. J Polym Sci
and distribution. V
Part A: Polym Chem 49: 34373447, 2011
KEYWORDS: block copolymers; fuel cell; functionalization of

polymers; ionomer; membranes

INTRODUCTION Proton exchange membrane fuel cells

(PEMFCs) have been recognized as promising energy source
systems due to their high energy-conversion efficiency, high
power density, and low pollutant emission. However, the crucial drawback of PEMFCs, that is, the slow kinetics of the oxygen reduction reaction, still requires high loading of precious metal catalysts on the cathode side, thus strongly
restricting their commercialization.14

face polymer (ionomer) that allows the preparation of a

membrane electrode assemblies (MEAs, which is an
assembled stack of ion-exchange membrane, catalyst, and flat
plate electrode)3,18,19 that do not contain any cations.
Actually, using ionomer-binded MEAs,1921 the carbonate
precipitation issue is overcome as the cationic sites (typically
quaternary ammonium sites) are grafted and immobilized on
the polymer skeleton.

Recently, increasing attention has focused on anion-exchange

membrane fuel cells,513 because it enables the use of firstrow transition metal-based catalysts which are intrinsically
stable and show similar activity to platinum in alkaline
media.11,1416

In the last few years, a number of new anion-exchange polymers11,15,2226 have been developed, along with non-noble
metal catalysts for the oxygen reduction reaction,27,28 which
together offer the possibility of assembling MEAs to develop
anion membrane fuel cells. The current technologies of
anion-exchange membranes (AEMs) for AMFC application
show several limitations related to the possibility of obtaining a membrane having: high ionic conductivity, chemical
stability in strong alkaline media, low permeability to fuel
crossover, low swelling, and good mechanical properties.

A major issue related to traditional aqueous electrolyte alkaline fuel cells was the presence of mobile cations (e.g., K,
Na) which could precipitate as carbonate, thus, hindering
the activity of the catalyst layers.17 The breakthrough was
the successful development of a crosslinked alkaline inter-

Additional Supporting Information may be found in the online version of this article. Correspondence to: A. Pucci (E-mail: apucci@ns.dcci.unipi.it)
C 2011 Wiley Periodicals, Inc.
Journal of Polymer Science Part A: Polymer Chemistry, Vol. 49, 34373447 (2011) V

FUNCTIONALIZATION OF SBS TRIBLOCK COPOLYMER, FARAJ ET AL.

3437

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA

The purpose of this study was to develop and characterize

an AEM based on polystyrene-b-poly(butadiene)-b-polystyrene triblock copolymer (SBS), a thermoplastic elastomer
with a block structure largely investigated as starting polymer matrix for the preparation of functional materials2937
and widely used in the rubber industry.
The block structure was selected, because it was recently
reported for proton exchange membranes that, in contrast to
the random design, the block or graft chain architectures significantly improve ionic conductivity due to the formation of
hydrophilic and hydrophobic domains.3841
The polymer matrix was modified by radical grafting using
vinylbenzyl chloride (VBC) as functional monomer and benzoyl peroxide (BPO) as radical initiator.27,29,42,43 The resulting
grafted chlorobenzylic moieties were then converted into the
anion exchange sites by quaternization with aliphatic diamines
such as the 1,4-diazabicyclo[2.2.2]octane (Dabco)10,4446 thus
providing an anion exchange thermoplastic membrane in the
form of a thin sheet (50100 microns), in which the Dabco
group acts also as a dicationic crosslinker. The final properties
of this AEM were determined in terms of water uptake (WU),
ion-exchange capacity (IEC), and ionic conductivity (r).
EXPERIMENTAL

Materials
All the solvents used were used as received or purified by
standard procedures. The SBS linear triblock copolymer (Calprene 501) was provided by Softer S.p.A. (Forl`, Italy). The
polymer is characterized by a butadiene/styrene ratio of 69/
31 by weight, an average molecular weight (MW) of blocks
of 930041,4009300, a content of the vinyl units in the butadiene block of 10 wt % and a density of 0.94 g cm
3. The
polymer was used after purification; about 4 g of SBS were
dissolved in 60 mL of chloroform and precipitated into 600
mL of methanol. The procedure was repeated twice, and the
material was collected by filtration and dried under vacuum.
4-Vinylbenzyl chloride [VBC; 4-(chloromethyl)styrene,
Aldrich, 90%] was purified by washing twice with 25%
aqueous sodium hydroxide, and then thoroughly washed
with distilled water until neutral. Finally, the monomer was
distilled at reduced pressure and stored under nitrogen over
molecular sieves at
20  C.

reduced pressure (conv. 30 wt %. The low value of conversion is merely due to the experimental apparatus used for
the polymerization. The polymerization was stopped when
the viscosity of the medium reached a value at which the
rotation of stirring bar was blocked).
Preparation of the Anion Exchange Membranes
SBS Functionalization Reaction
In a glass cylindrical reactor equipped with a mechanical
stirrer, 2 g of SBS was introduced under nitrogen atmosphere and dissolved in 9 mL of VBC (9.747 g, 6.39  10
2
mol). The desired amount of the radical initiator (from 0.25
to 1.2 mol % with respect to SBS, AIBN, or BPO) was then
dissolved into 1 mL of VBC at 0  C and added under nitrogen to the polymer solution, and the resulting mixture was
mixed at 80  C for 2 h. The reaction was stopped after 2 h,
that is, when the viscosity of the reacting medium aimed the
polymer to stick on the mechanical stirrer surface, thus
strongly limiting its mixing. After cooling down to room temperature, the crude product was diluted with 80 mL of chloroform containing dissolved about 5 g of the free radical
scavenger BHT and recovered by precipitation in methanol
as a white solid powder. The material was then soaked in acetone to remove from the functionalized polymer the amount
of PVBC oligomer obtained during the reaction. The remaining amount of the latter was completely removed by extraction with acetone in a Kumagawa-type extractor for 8 h. The
two phases [a residue (r-Fi) and an extract (e-Fi)] were then
collected and accurately characterized.
Amination of the Functionalized SBS
The purified functionalized SBS with VBC units (r-Fi) was
dissolved in chloroform and poured in a TeflonV Petri dish.
The film obtained after slow solvent evaporation (carried
out under saturated chloroform air atmosphere at a temperature of 20  C; the films were recovered after 12 h with a
thickness of about 6080 lm) was then immersed into a
Dabco or a TMA 1 M methanol solution at 60  C for 72 h. As
an alternative, a weighed amount of the r-Fi film was
immersed into 20 mL of methanol at 60  C and a stochiometric amount of Dabco, calculated with respect to the
amount of VBC grafted on SBS, was dissolved into 120 mL of
methanol, and slowly added (0.03 mL per min) over a period
of 72 h.
R

a,a0 -Azobisisobutyronitrile (AIBN; Carlo Erba) was purified

by recrystallization from acetone. BPO (dibenzoyl peroxide,
Luperox_A98, Aldrich, 97%) was purified by recrystallization
from methanol.

In both cases, the unreacted amine present on the membrane surface was removed by washing with excess of water
and methanol and then dried in a desiccator for 48 h until
constant weight.

Dabco (Aldrich, 98%), triethylamine (TMA; Sigma-Aldrich,

99%), and 2,6-di-tert-butyl-4-methylphenol (BHT) were
used as received.

Instrumentations and Methods

FT Infrared Spectroscopy Analysis
Infrared spectra were performed with a Fourier transform
(FTIR) spectrometer PerkinElmerTM Spectrum One on films
cast from chloroform solution on a KBr window.

Poly(vinylbenzyl chloride) (PVBC) was prepared for comparison purposes by radical bulk polymerization of VBC in the
presence of 1% by mol of AIBN. The solution was stirred at
80  C for 2 h under nitrogen atmosphere, and the crude
product was purified by double dissolution in chloroform
and precipitation in methanol and recovered after drying at

3438

H NMR Spectroscopy Analysis and Evaluation of the

Functionalization Degree
Proton nuclear resonance spectra were recorded by using a
Varian Gemini 300 MHz on 510% CDCl3 (Aldrich, 99.8 atom
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE

% D) solutions. NMR spectra were registered at 20  C, and

the chemical shifts were assigned in ppm using the solvent
signal as reference.
The functionalization degree (FD) of SBS was calculated as
the mol of grafted VBC per 100 mol of repeating units of the
polymer from the 1H NMR spectra as analogously reported
in literature for different functionalized polymers.4749 Being:

d 7.1 and 6.5 ppm attributed to 5H of styrene and 4H

of VBC ! Integral A;

d 5.4 ppm attributed to 2H of 1,4 butadiene and 1H of

1,2 butadiene ! Integral B;

d 4.9 ppm attributed to 2H 1,2 butadiene ! Integral C;

d 4.5 ppm attributed to 2H ACH2Cl of VBC ! Integral D.
The relative contributions of each repeating unit were calculated as follows:
VBC

integral D
2

integral C
8 2
9
integral C > 1
:integral B

;
1; 4 butadiene >
2
2
8
9
integral
D
1
>integral A

>
styrene :
 4;
2
5
1; 2 butadiene

FD was then calculated according to eq 1:

FD VBC mol % P

integral D  12
 100
all repeating units

(1)

Size Exclusion Chromatography Analysis

The SBS MW was obtained by using the instrument as follows: Jasco PU-1580 equipped with Jasco 830-RI as refractometer index and two columns PL gel mixed C 5 m (Polymer Laboratories). The analysis were performed onto
chloroform-diluted solutions (0.1 wt %). The calibration
curve was realized with polystyrene standard samples.
Differential Scanning Calorimetry
The differential scanning calorimetry (DSC) analyses were
performed under nitrogen flux (80 mL min
1) with a Mettler
Toledo/DSC 822e equipped with a cooling system. The calibration was performed with zinc and indium. Heating and
cooling thermograms were carried out at a standard rate of
10  C min
1.
Thermogravimetric Analysis
Thermogravimetric analysis (TGA) scans were obtained with
a Mettler Toledo Starc System moduloTGA/SDTA851e under
nitrogen flux, at a scan rate of 20  C min
1.
Elemental Analyses
Elemental analyses (EA) were accomplished by the microanalysis laboratory at the faculty of pharmacy, University of
Pisa.
Scanning Electron Microscopy
The scanning electron microscopy (SEM) analysis was performed with a Jeol 5600-LV microscope, equipped with

FUNCTIONALIZATION OF SBS TRIBLOCK COPOLYMER, FARAJ ET AL.

Oxford X-rays EDS microprobe, instrument at the Chemical

Engineering Department of Pisa University.
Characterization of the Anion Exchange Membranes
Deionized (DI) water (q  10 MX cm) was used in all
experiments. A multichannel potentiostat/galvanostat, having
an impedance channel and a 4 current booster (VMP3
from Bio-Logic SA), was used to perform electrochemical impedance spectroscopy measurements.
Determination of the Ion-Exchange Capacity
The IEC of the polymer was determined by acid-base back titration. A piece of the polymer film was activated in 1 M
KOH at 50  C for 30 min to convert it to the OH-form. The
piece was then washed using N2-degassed water and soaked
for 30 min in a measured volume of a standardized HCl solution, where OH
from the polymer is neutralized. Excess of
HCl was then titrated using standardized KOH. Endpoint was
detected by using visual or potentiometric methods.
Determination of the Water Uptake
Dry weight (Wdry) of the polymer film was measured after 12
h drying at 50  C using a four-digit analytical balance. The
film was activated in 1 M KOH at 50  C for 30 min to
exchange its counteranions into the OH-form. Temperature
control was ensured using a thermostatic oil bath. Wet weight
(Wwet) was measured after conditioning in decarbonated DI
water (q  10 MX cm) at the envisaged temperature for 30
min. Before the weigh, excess of water was removed from the
membrane surface using a humidified filter paper. WU, as
mass percentage, was calculated using eq 2:
WU%

Wwet
Wdry
 100
Wdry

(2)

where wet- and dry-film weights were averaged over three

measurements.
Determination of the In-Plane (rIP) Conductivity
In-Plane (IP) conductivity was measured in a Bekktech PTFE
4-probe flow-cell, assembled with a 5 cm2 cell hardware supplied by Fuel Cell Technologies allowing in situ AEM
exchange to the OH
form (HCO

3 form) flowing a 1 M KOH
(or 1 M NaHCO3) solution. This was followed by removal of
the electrolyte excess via a continuous flow of N2-degassed
water. A multichannel potentiostat/galvanostat (VMP3 from
Bio-Logic, SA) was used to record electrochemical impedance
spectra.
Determination of the Through-Plane (rTP) Conductivity
Through-plane conductivity was measured using a 5-cm2 cell
hardware supplied by Fuel Cell Technology allowing in situ
ion exchange, analogous to the IP tests, again carefully avoiding air exposure of the AEM materials. Platinum-based
anion-exchange ionomer-based MEAs were used.
RESULTS AND DISCUSSION

SBS Functionalization
The thermoplastic elastomer SBS49 was introduced in a glass
cylindrical reactor under nitrogen atmosphere and dissolved

3439

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA

TABLE 1 Characteristics of the Grafting Reactions of VBC on SBS Performed Using Different Type and Amount of Radical Initiators
Mol % Radical
Initiator

Mn
(g mol
1,
Peak II)c

Mw
(g mol
1,
Peak II)c

AIBN

Wt %
Residue
(r-Fi)*

Wt %
Extract
(e-Fi)*

FD
(mol %)a

0.25

40

60

0.30

38

62

3.7

1.8

179,600

382,800

0.45

37

63

4.4

2.1

191,900

399,100

0.60

44

56

5.3

2.6

228,700

462,100

0.70

43

57

6.7

3.2

230,100

459,000

0.80

48

52

9.1

4.0

250,300

465,500

0.85

55

45

9.6

4.4

0.90

56

44

10.2

4.7

182,900

425,900

8b

1.20

9.3

4.2

212,700

438,500

Entries
(Fi)

BPO

1
2

Chlorine
(wt %)b

0.40

39

61

1.0

0.5

190,400

376,000

10

0.60

38

62

2.5

1.2

181,800

384,900

11

0.80

38

62

3.1

1.6

203,100

393,800

12

1.20

37

63

4.5

2.3

182,900

425,900

Weight percentage of residual (SBS-g-VBC) and extracted (PVBC)

phases using acetone as extracting solvent; residual phases (SBS-gVBC) composition calculated from 1H NMR spectroscopy and elemental
analysis; and molecular weight (MW) of the residue r-Fi products evaluated by size exclusion chromatography after deconvolution of the
chromatographic traces.
* Average of two or three entries.

a
Evaluated by means of 1H NMR spectroscopy investigations (see eq 1
experimental part).
b
Evaluated from elemental analysis.
c
For all entries the Mn and Mw values corresponding to the I peak coincided with those of pure SBS (216,300 and 220,800 g mol
1, respectively) with variations less than 5%.

in the minimum amount of VBC. The desired amount of the

radical initiator (AIBN or BPO, from 0.25 to 1.1 mol % with
respect to SBS, see Table 1) was then added under nitrogen
to the polymer solution (1617 wt % of SBS), and the
resulting mixture was mixed at 80  C for 2 h. After cooling
down to room temperature, the crude product was dissolved
in chloroform containing a large amount of the free radical
scavenger BHT. The purification protocol of the functionalized polymer isolated two distinct phases: a residue phase
(r-Fi, consisted in the functionalized polymer SBS-g-VBC and
the unreacted SBS unsoluble in acetone (SBS and SBS-g-VBC
showed similar solubility making difficult their complete separation) and an extract phase (e-Fi, composed by the PVBC
oligomer soluble in acetone) that were then fully
characterized.

In particular, (a) peaks in the range 7.16.5 ppm were attributed to the styrene units of SBS and to the aromatic protons
of the grafted chlorobenzyl moieties; (b) peaks in the range
5.25.0 ppm were attributed to the double bonds of the poly
(butadiene) block of SBS; (c) peak at 4.5 ppm was attributed
to the two chloromethylenic protons of the VBC units.
It is worth noting for both investigation techniques, the progressive emersion of the signals associated to the grafted
functional units of VBC by increasing the amount of BPO
from 0.45 to 0.9 mol %. On the contrary, no significant

Differently from AIBN, the amount of the residue SBS-g-VBC

(r-Fi) increased with increasing BPO concentration and
became larger than the extract (e-Fi) when the amount of
BPO used as radical initiator was higher than 0.8 mol %.
FTIR spectroscopy analyses performed on the residue products showed the typical signals attributed to the stretching
of the C
C units at 1640 cm
1 and the bending of the
C
CAH units at 965 and 910 cm
1 of the SBS polymer matrix and to the bending of the ACH2Cl at 1265 cm
1 and the
stretching CACl at 675 cm
1 of the grafted chlorobenzylic
moieties (Fig. 1).
1

H NMR spectra of the r-Fi samples showed resonance characteristic of the SBS-g-VBC functionalized polymer (Fig. 2).

3440

FIGURE 1 FTIR spectra and attribution of the main signals of

SBS and residual phases (r-Fi, SBS-g-VBC) at different % of
BPO initiator.

WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE

FIGURE 2 1H NMR spectra in CDCl3 and attribution of the main signals of SBS and residual phases (r-Fi, SBS-g-VBC) at different %
of BPO initiator.

modification of the absorption or resonance intensities

attributed to the SBS matrix was recorded.

only evaluated by using the highest amount of AIBN (i.e., 1.2

mol %).

Moreover, FTIR spectroscopy and 1H NMR investigations carried out on the extract e-Fi phases showed signals totally
similar to that characteristic of a PVBC matrix (Supporting
Information Figs. S1 and S2).

Interestingly, when BPO was used in the concentration range

of 0.60.9 mol % of initiator seemed to strongly promote the
grafting efficiency of the VBC functional monomer. These
data suggest that BPO promotes VBC grafting on SBS matrix
over its homopolymerization, the opposite trend was instead
observed for AIBN.

The functionalization degree (FD) of the SBS-g-VBC samples

was calculated, as the mol of grafted VBC per 100 mol of
repeating units of the functionalized polymer, by means of
1
H NMR spectroscopy according to the eq 1 reported in the
experimental part. The chlorine content (wt %) of the functionalized polymers was evaluated by elemental analysis.
The data reported in Table 1 showed that BPO appeared to
be more efficient than AIBN in the grafting reaction of VBC
into the SBS backbone. FD values comprised from about 4 to
about 10 mol % were actually calculated for SBS-g-VBC samples by using BPO concentration in the range from 0.3 to
1.20 mol %. Conversely, FD values as high as 4.5 mol % was

FUNCTIONALIZATION OF SBS TRIBLOCK COPOLYMER, FARAJ ET AL.

In agreement with these results, it is reported in literature

that different radical initiators have different effects on grafting. More specifically, the grafting of monomers onto the styrene repeating units by radical initiators is rather limited,50
whereas AIBN and BPO can react through different mechanisms with the butadiene blocks (predominantly with 1,2
units) which are the free radical reactive units of SBS.30,33 It
is actually reported in literature that the grafting efficiencies
of styrene-based monomers are generally higher for vinyl-1,2
butadiene backbone polymers than for cis-1,4-polybutadiene.51 This is attributed to the fact that the allylic

3441

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA

SCHEME 1 Proposed mechanism of the radical functionalization of SBS copolymer with VBC functional monomer in the presence
of BPO (a) or AIBN (b) as radical initiators.

hydrogen bond of the vinyl-1,2 butadiene unit results weaker

than that of the cis-1,4 butadiene moiety due to its contemporary tertiary position. As a consequence, the chain transfer
mechanism is more likely to occur at the 1,2-double bonds
of butadiene units.
Primary radicals generated from BPO initiator most likely
reacts by a chain transfer mechanism creating allylic macroradicals,30,5153 whereas AIBN produces carbon located free
radicals, which preferentially react with double bonds. Then,
with BPO, the process involves the formation of free radicals
on 1,2 butadiene units that can successively start VBC grafting followed by its homopolymerization to short chains
[Scheme 1(a)].
With AIBN, the formation of a free radical on SBS could
occur by the attack of the primary radical on vinyl double

3442

bond, which then could bring to short VBC grafted chains as

well [Scheme 1(b)]. This process is reported to occur with
great difficulty; AIBN prefers to react with styrene-based
monomers and only to a low extent with the butadiene
blocks.30,5355 Clearly, competition with VBC homopolymerization occurs in both the cases,54,56 but it seems more important for the latter initiator thus providing functionalized
polymers with lower FD.
The MW of the residue r-Fi products evaluated by size exclusion chromatography revealed for all the samples similar
trends to neat SBS consisting of a main peak (I) characterized by a MW of about 200  103 g mol
1 and a minor contribution at lower MW (70  103 g mol
1) attributed to
SBS impurities (1%). Moreover, the chromatograms of the
residue materials showed also a well-pronounced shoulder

WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE

SEM coupled with energy dispersive X-ray spectroscopy

(EDS) of the films obtained by solution casting of the residual phase showed an homogeneous structure with no microscopic defects [Fig. 4(a)] and a chlorine percentage comparable with that obtained from 1H NMR investigation and
elemental analysis [Fig. 4(b)].

FIGURE 3 GPC chromatogram and its deconvoluted curves of a

SBS-g-VBC residual phase consisted in the functionalized polymer SBS-g-VBC and the unreacted SBS.

SBS-g-VBC Quaternization
Benzyl chloride groups grafted onto the SBS backbone were
converted into quaternary ammonium groups with a tertiary
amine such as the Dabco. Even though there are several bhydrogens in its bicyclic structure, it can produce crosslinked
structure during the quaternization reaction (Scheme 2),
thus preventing interconversion between conformers and
limiting Hofmann degradation.44,5761

at higher MW (II, 400300  103 g mol
1) attributed to

highly functionalized SBS-g-VBC samples or to partially
branched, but still soluble, SBS macromolecules. SBS crosslinking appeared limited for all the experiments likely due to the
high concentration of VBC monomers composing the reacting
mixture that makes unfavorable SBS macroradicals coupling.
The size exclusion chromatography trace was then deconvolved by a curve-fitting procedure into three main contributions, and all the derived peaks were analyzed in terms of
Mn and Mw (Fig. 3).
As the average MW values evaluated for the main I peak
were approximately the same for all r-Fi products and similar to neat SBS (216,300 and 220,800 g mol
1 for Mn and
Mw, respectively) with variations less than 57%, only the
data acquired for peak II were reported in Table 1.
For samples obtained by using BPO as radical initiator, the
MW associated to the peak II increased progressively with
BPO concentration up to the amount of 0.80 mol %, above it
MW values showed a little decreasing.
It is worth noting that the increasing of MW associated to
peak II appears to be related to the grafting efficiency of the
functionalization reaction (Table 1). In other words, the peak
II might be likely attributed to functionalized SBS-g-VBC
macromolecules that are composed of r-Fi products, and the
MW trend as a function of initiator concentration confirms
the existence of a precise range of BPO concentration that is
able to provide the highest grafting efficiency.
On the contrary, more restricted variations were observed in
FD and MW varying initiator percentage when AIBN was
used.
Thermogravimetric analysis carried out on SBS matrix and
residual phases (r-F5 and r-F6) showed decomposition temperatures (Td) located at 430, 445, and 450  C, respectively.
The positive correlation between Td and the calculated FD
(i.e., 6.7 mol % for r-F5 and 9.1 mol % for r-F6) can be
explained in terms of thermal stabilizing effect provided by
VBC-grafted fragments, which can delocalize free radicals
generated on polymer matrix during the heating.

FUNCTIONALIZATION OF SBS TRIBLOCK COPOLYMER, FARAJ ET AL.

FIGURE 4 (a) Scanning electron microscopy (SEM) coupled

with energy spectroscopy (EDS) of r-F7 residual phase; (b)
scanning line for EDS chlorine analysis across the film fracture.

3443

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA

SCHEME 2 Quaternization reactions between grafted chlorobenzylic groups and Dabco.

Comparing FTIR spectra before and after quaternization

reaction with Dabco for r-F2 residual phase film peak at
1265 cm
1, associated to CH2-Cl wagging, disappears
whereas new peaks at 1313 and 1058 cm
1 show up, assignable to amine absorptions (Supporting Information Fig.
S3). The same behavior is observed for all the residual
phases.
The amination procedure did not introduce evident microscopic defects in the polymeric membrane as revealed by
scanning electron micrographs (Supporting Information Fig.
S4), whereas thermogravimetric investigations showed thermal stability of the membrane up to 270  C, temperature on
which quaternary ammonium groups start to decompose.
DSC (Supporting Information Fig. S5 for a Dabco-quaternized
r-F5 film) carried out on SBS-g-VBC-aminated films both in
chloride and hydroxyl forms showed two glass transition
temperatures at
85  C (butadienic block) and at about 75

C (styrenic blocks), suggesting that VBC grafting and successive Dabco quaternization did not stiffen the polymer chain.
As crosslinking tends to increase the glass transition temperature, this result also suggests that quaternized films had a
very low crosslinking ratio after amination.

Electrochemical Characterization
Electrochemical parameters were evaluated as described in
the experimental section for membranes prepared using rF5_Dabco, r-F6_Dabco, and r-F8_Dabco films after being quaternized with a 1 M solution of Dabco in methanol. Electrochemical parameters obtained for a commercial membrane
(Tokuyama A-201) were used as benchmark values to optimize FD value (Table 2).

In Figure 5, an histogram of WU data for the hydroxyl form

of the considered SBS-based films is reported: increasing the
FD, water swelling increases as well.
This trend was predictable: an higher FD (i.e., higher IEC)
would result in a comb-type polymer having more side
chains or longer side chains. In either case, charge density of
the butadienic block would be higher, increasing thus its
hydrophilicity and the water swollen at the equilibrium.
These data suggest that increasing the ion-exchange groups
concentration newly bonded ionic groups are reachable by
the solvent; hence, a higher ionic conductivity is expected
increasing the FD. Nevertheless, IECs of the SBS-based membrane are lower compared to the benchmark, and their swelling is slightly higher but still promising.
In Figure 6, IP ionic conductivities, measured at 30  C, are
reported for both bicarbonate and hydroxyl forms of the
three SBS-based and benchmark membranes: as suggested
by WU data analysis, increasing the FD, ionic conductivity
increases as well.
Nevertheless, the IEC measured for r-F8_Dabco-based membrane was lower than the benchmark membrane one,
whereas its ionic conductivity at 30  C is higher in both bicarbonate and hydroxyl forms. Moreover, ionic conductivities
for bicarbonate forms are approximately fourfold lower than
that of the hydroxyl form, consistent with the approximately
fourfold lower equivalent ionic conductance in water of bicarbonate ions compared to hydroxide, suggesting an ion
transport mechanism similar to the one that occurs in
water.
Best trade-off between ionic conductivity and water swelling
was found for r-F6_Dabco-based membrane, having FD 9.1

TABLE 2 Comparison Between the Functionalization Degree (FD), the Ion-Exchange Capacities
of SBS-Based Membranes Measured by Back Titration (IEC) and the In-Plane (rIP) and
Through-Plane (rTP) Conductivities at 30 8C for the Bicarbonate Form of Three DabcoQuaternized SBS-Based Membranes

Samples

FD
(mol %)

IEC
(meq/gpolymer)

HCO3-Form,
rIP (mS cm
1)

HCO3-Form,
rTP (mS cm
1)

Tokuyama A-201

1.84 6 0.04

7.3

7.1

r-F5_Dabco

6.7

0.52 6 0.05

3.2

3.3

r-F6_Dabco

9.1

0.82 6 0.05

5.3

5.5

r-F8_Dabco

10.2

1.21 6 0.04

9.2

9.1

Tokuyama A-201 membrane IEC is reported as benchmark.

3444

WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE

SCHEME 3 Random and clustered distribution schematics of

ion-exchange groups.
FIGURE 5 Water uptake (WU) comparison, expressed as wt %,
at different temperatures of three Dabco-quaternized SBSbased membranes (OH-form) having different FD. Tokuyama A201 swelling (OH-form) is reported as benchmark.

mol %, whose conductivity is comparable with the benchmark

material, and WU is slightly higher but still acceptable. More
specifically, simply tuning the FD of these SBS-based membranes, ionic conductivities higher than the commercial benchmark membrane was reached (r-F8_Dabco), though WU has
yet to be reduced. This phenomenon could be probably
addressed to the way in which VBC grafting on the SBS matrix
occurs. In fact, assuming constant overall IEC, the distribution
of the ion-exchange groups is not fixed. If VBC homopropagation is preferred over its uniform grafting on the SBS matrix,
the resulting SBS-based membrane will have a clustered distribution of the ion-exchange groups over the butadienic blocks
(Scheme 3), that is, the charge will be unevenly distributed
with local peaks of its density thus increasing the WU.
CONCLUSIONS

A new anion exchange polymer based on SBS were prepared

through the bulk radical grafting of VBC monomers, initiated

either by BPO or a,a0 -azo-bis-isobutyronitrile (AIBN) and successive amine treatment.

BPO was more efficient than AIBN to give grafting reaction
of VBC on SBS more than VBC homopolymerization with FD
(defined as the mol of grafted VBC per 100 mol of repeating
units of the functionalized polymer) ranging from 4 to 10
mol % depending on the amount of initiator. The functionalization reaction did not induce remarkable crosslinking phenomena of the polymer matrix thus preserving its thermal
properties and solubility.
The grafted benzyl chloride groups of the SBS-g-VBC were
quantitatively converted of the into quaternary ammonium
units by treatment with the tertiary bis-amine Dabco. A positive correlation between the FD, tunable during the synthesis, of the synthesized Dabco-quaternized SBS-g-VBCbased
membranes and both water swelling and ionic conductivity,
was found.
Moreover, simply tuning the FD of these SBS-based membranes ionic conductivities higher than the commercial
benchmark membrane was obtained, but still large WU was
observed likely due to a clustered distribution of the ionexchange groups over the butadienic blocks. Future development will move in this direction, by tuning the crosslinking
degree during quaternization and VBC-grafted unit distribution to promote large water swelling decrease with small
reduction in ionic conductivity.
M. Boccia thanks Tokuyama Corp. for providing commercial
anion-exchange materials. H. Gasteiger is kindly acknowledged
for helpful discussions.

REFERENCES AND NOTES

FIGURE 6 In-plane conductivities (rIP) comparison at 30 C,

both in hydroxyl (light gray bars) and bicarbonate (dark gray
bars) forms, of three Dabco-quaternized SBS-based membranes having different FD. Tokuyama A-201 in-plane conductivities (both hydroxyl and bicarbonate forms) at 30  C are
reported as benchmark.

FUNCTIONALIZATION OF SBS TRIBLOCK COPOLYMER, FARAJ ET AL.

1 Gasteiger, H. A.; Kocha, S. S.; Sompalli, B.; Wagner, F. T.

Appl Catal B 2005, 56, 935.
2 Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.;
Garland, N.; Myers, D.; Wilson, M.; Garzon, F.; Wood, D.; Zelenay, P.; More, K.; Stroh, K.; Zawodzinski, T.; Boncella, J.;
McGrath, J. E.; Inaba, M.; Miyatake, K.; Hori, M.; Ota, K.;
Ogumi, Z.; Miyata, S.; Nishikata, A.; Siroma, Z.; Uchimoto, Y.;

3445

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA

Yasuda, K.; Kimijima, K.-i.; Iwashita, N. Chem Rev 2007, 107,

39043951.

26 Tamain, C.; Poynton, S. D.; Slade, R. C. T.; Carroll, B.; Varcoe, J. R. J Phys Chem C 2007, 111, 1842318430.

3 Guha, A.; Zawodzinski, T. A.; Schiraldi, D. A. J Power Sources 2007, 172, 530541.

27 Piana, M.; Catanorchi, S.; Gasteiger, H. A. ECS Trans 2008,

16, 20452055.

4 Zhou, C.; Guerra, M. A.; Qiu, Z.-M.; Zawodzinski, T. A., Jr.;

Schiraldi, D. A. Macromolecules 2007, 40, 86958707.

28 Varcoe, J. R.; Slade, R. C. T.; Wright, G. L.; Chen, Y. J Phys

Chem B 2006, 110, 2104121049.

5 Zhang, X.; Tay, S. W.; Liu, Z.; Hong, L. J Power Sources

2011, 196, 54945498.

29 Cordella, C. D.; Cardozo, N. S. M.; Baumhardt Neto, R.;

Mauler, R. S. J Appl Polym Sci 2003, 87, 20742079.

6 Wang, J.; Li, S.; Zhang, S. Macromolecules 2010, 43,

38903896.

30 Jiang, D. D.; Wilkie, C. A. J Polym Sci Part A: Polym Chem

1997, 35, 965973.

7 Robertson, N. J.; Kostalik, H. A. I. V.; Clark, T. J.; Mutolo, P.

F.; Abruna, H. D.; Coates, G. W. J Am Chem Soc 2010, 132,
34003404.

31 Jiang, D. D.; Wilkie, C. A. Eur Polym J 1998, 34,

9971006.

8 Piana, M.; Boccia, M.; Filpi, A.; Flammia, E.; Miller, H. A.;
Orsini, M.; Salusti, F.; Santiccioli, S.; Ciardelli, F.; Pucci, A. J
Power Sources 2010, 195, 58755881.
9 Guo, M.; Fang, J.; Xu, H.; Li, W.; Lu, X.; Lan, C.; Li, K. J
Membr Sci 2010, 362, 97104.
10 Sollogoub, C.; Guinault, A.; Bonnebat, C.; Bennjima, M.;
Akrour, L.; Fauvarque, J. F.; Ogier, L. J Membr Sci 2009, 335,
3742.
11 Varcoe, J. R.; Slade, R. C. T. Fuel Cells 2005, 5,
187200.
12 Varcoe, J. R.; Poynton, S. D.; Slade, R. C. T. Handb Fuel
Cells 2009, 5, 322333.
13 Jagur-Grodzinski,
785799.

J.

Polym

Adv

Technol

2007,

18,

14 Matsumoto, K.; Fujigaya, T.; Yanagi, H.; Nakashima, N. Adv

Funct Mater 2011, 21, 10891094.
15 Varcoe, J. R.; Slade, R. C. T.; Yee, E. L. H. Chem Commun
2006, 14281429.

32 Lasalle, V. L.; Failla, M. D.; Valles, E. M.; Cepeda-jimenez, C.

M.; Torregrosa-macia, R.; Martin-martinez, J. M. J Adhes Sci
Technol 2003, 17, 17131726.
33 Lassalle, V. L.; Failla, M. D.; Valles, E. M.; Martin-Martinez,
J. M. J Adhes Sci Technol 2004, 18, 18491860.
34 Maurano, C. H. F.; Portal, L. L.; Baumhardt Neto, R.; Mauler,
R. S. Polym Bull 2001, 46, 491498.
35 Moad, G. Prog Polym Sci 1999, 24, 81142.
36 Passaglia, E.; Ghetti, S.; Picchioni, F.; Ruggeri, G. Polymer
2000, 41, 43894400.
37 Pracella, M.; Chionna, D. Macromol Symp 2004, 218,
173182.
38 Gross, M.; Maier, G.; Fuller, T.; MacKinnon, S.; Gittleman, C.
Handb Fuel Cells 2009, 5, 283299.
39 Elabd, Y. A.; Hickner, M. A. Macromolecules 2011, 44,
111.
40 Tsang, E. M. W.; Zhang, Z.; Shi, Z.; Soboleva, T.; Holdcroft,
S. J Am Chem Soc 2007, 129, 1510615107.

16 Antolini, E.; Gonzalez, E. R. J Power Sources 2010, 195,

34313450.

41 Tsang, E. M. W.; Zhang, Z.; Yang, A. C. C.; Shi, Z.; Peckham,

T. J.; Narimani, R.; Frisken, B. J.; Holdcroft, S. Macromolecules
2009, 42, 94679480.

17 Gulzow, E.; Schulze, M. J Power Sources 2004, 127,

243251.

42 Wilhelm, H. M.; Felisberti, M. I. J Appl Polym Sci 2002, 86,

366371.

18 Eikerling, M.; Ioselevich, A. S.; Kornyshev, A. A. Fuel Cells

2004, 4, 131140.

43 Wilhelm, H. M.; Felisberti, M. I. J Appl Polym Sci 2003, 87,

516522.

19 Kim, Y. S.; Einsla, M.; McGrath, J. E.; Pivovar, B. S. J Electrochem Soc 2010, 157, B1602B1607.

44 Bauer, B.; Strathmann, H.; Effenberger, F. Desalination 1990,

79, 125144.

20 Bender, G.; Zawodzinski, T. A.; Saab, A. P. J Power Sources

2003, 124, 114117.

45 Stoica, D.; Ogier, L.; Akrour, L.; Alloin, F.; Fauvarque, J. F.

Electrochim Acta 2007, 53, 15961603.

21 Valade, D.; Boschet, F.; Roualdes, S.; Ameduri, B. J Polym

Sci Part A: Polym Chem 2009, 47, 20432058.

46 Agel, L.; Bouet, J.; Fauvarque, J. F. J Power Sources 2001,

101, 267274.

22 Li, Y.; Xu, T.; Gong, M. J Membr Sci 2006, 279,

200208.

47 Ciardelli, F.; Aglietto, M.; Ruggeri, G.; Passaglia, E.; Castelvetro, V. Macromol Symp 1997, 118, 311.

23 Danks, T. N.; Slade, R. C. T.; Varcoe, J. R. J Mater Chem

2002, 12, 33713373.

48 Passaglia, E.; Ghetti, S.; Picchioni, F.; Ruggeri, G. Polymer

2000, 41, 43894400.

24 Danks, T. N.; Slade, R. C. T.; Varcoe, J. R. J Mater Chem

2003, 13, 712721.

49 Pucci, A.; Barsocchi, C.; Rausa, R.; DElia, L.; Ciardelli, F.

Polymer 2005, 46, 14971505.

25 Komkova, E. N.; Stamatialis, D. F.; Strathmann, H.; Wessling, M. J Membr Sci 2004, 244, 2534.

50 Passaglia, E.; Coiai, S.; Ricci, L.; Ciardelli, F. Macromol

Symp 2004, 218, 6170.

3446

WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
51 Huang, N.-J.; Sundberg, D. C. J Polym Sci Part A: Polym
Chem 1995, 33, 25872603.
52 Zhang, A.; Chao, L. Eur Polym J 2003, 39, 12911295.
53 Cameron, G. G.; Qureshi, M. Y. J Polym Sci Polym Chem Ed
1980, 18, 21432153.
54 Huang, N.-J.; Sundberg, D. C. J Polym Sci Part A: Polym
Chem 1995, 33, 25712586.
55 Cameron, G. G.; Qureshi, M. Y. J Polym Sci Polym Chem Ed
1980, 18, 3149.
56 Huang, N.-J.; Sundberg, D. C. J Polym Sci Part A: Polym
Chem 1995, 33, 25512570.

FUNCTIONALIZATION OF SBS TRIBLOCK COPOLYMER, FARAJ ET AL.

57 House, H. O. Modern Synthetic Reactions (The Organic

Chemistry Monograph Series), 2nd ed; Benjamin, W. A., Ed.;
Menlo Park, 1972.
58 Chempath, S.; Einsla, B. R.; Pratt, L. R.; Macomber, C. S.;
Boncella, J. M.; Rau, J. A.; Pivovar, B. S. J Phys Chem C 2008,
112, 31793182.
59 Doering, W. von E.; Meislich, H. J Am Chem Soc 1952, 74,
20992100.
60 Cope, A. C.; Trumbull, E. R. Org React 1960, 11, 317 493.
61 Hofmann, A. W. Liebigs Ann 1851, 78, 253286.

3447