The preparation and catalytic performance of graphene-reinforced ion-exchange resins

Article abstract of DOI:10.1039/c4ra08972c

Graphene-based polymer nanocomposites (GPNCs) were prepared by in situ suspension polymerization and the prepared polymers were chloromethylated with chloromethyl ethyl ether and then aminated with trimethylamine to obtain graphene-based nanocomposite ion-exchange resins (GPNC-IERs). Raman spectroscopy revealed the existence of a chemically bonded interaction between graphene and the polymer networks via the increased intensity and shift in the vibration bands of graphene in the nanocomposites. The reinforced thermal and structural properties of the nanocomposites were investigated with graphene contents from 0 to 1.0 wt%. The thermal stability was evaluated by thermogravimetric-differential thermal analysis (TG-DTA). It showed that the addition of graphene to the polymer matrix greatly increased the onset degradation temperature by 20°C. The structural properties of the nanocomposites such as swelling capacity and total exchange capacity were improved with 0.4 wt% graphene, which confirmed that the good dispersion of graphene throughout the polymer matrix with strong interaction within the polymer networks played an important role in the properties enhancement effect. With the excellent anti-swelling property and thermal stability, the GPNC-IER catalysts presented a better performance in hydration of ethylene oxide (EO) to monoethylene glycol (MEG) including reaction activity, selectivity and stability in 500 h tests.

Full text of DOI:10.1039/c4ra08972c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
The preparation and catalytic performance of
graphene-reinforced ion-exchange resins
Cite this: RSC Adv., 2015, 5, 2550
*
Yanan Li, Fengping Yu, Wenjun He and Weimin Yang
Graphene-based polymer nanocomposites (GPNCs) were prepared by in situ suspension polymerization
and the prepared polymers were chloromethylated with chloromethyl ethyl ether and then aminated
with trimethylamine to obtain graphene-based nanocomposite ion-exchange resins (GPNC-IERs).
Raman spectroscopy revealed the existence of a chemically bonded interaction between graphene and
the polymer networks via the increased intensity and shift in the vibration bands of graphene in the
nanocomposites. The reinforced thermal and structural properties of the nanocomposites were
investigated with graphene contents from 0 to 1.0 wt%. The thermal stability was evaluated by
thermogravimetric-differential thermal analysis (TG-DTA). It showed that the addition of graphene to the
polymer matrix greatly increased the onset degradation temperature by 20 ꢀC. The structural properties
of the nanocomposites such as swelling capacity and total exchange capacity were improved with 0.4
wt% graphene, which confirmed that the good dispersion of graphene throughout the polymer matrix
with strong interaction within the polymer networks played an important role in the properties
enhancement effect. With the excellent anti-swelling property and thermal stability, the GPNC-IER
catalysts presented a better performance in hydration of ethylene oxide (EO) to monoethylene glycol
(MEG) including reaction activity, selectivity and stability in 500 h tests.
Received 20th August 2014
Accepted 30th September 2014
DOI: 10.1039/c4ra08972c
www.rsc.org/advances
polymers, such as swelling and deactivation of IERs under
reaction condition, restrict their further industrial application.
Introduction
Higher crosslinking density helped with reducing the swelling
at a cost of ion-exchange capacity and reactive sites of the IER
catalysts.4,13–16 Xie et al.17 have reported the use of carbon
nanotubes (CNTs) as a reinforcing ller in polymer composites
to deal with the swelling problem of the IERs. It was found that
the anti-swelling capacities and thermal stability of the poly-
mer/CNTs composites had been improved, and the catalysts
performed well in the EO hydration reaction test.
Recently, graphene has played an important role in nano-
science owning to its exceptional mechanical, electrical,
chemical and thermal properties.18–21 Its mechanical strength is
comparable to that of CNTs and the two-dimensional honey-
comb layer of sp2-bonded structure has potential applications
in many elds as composite materials, energy storage and
conversion, sensors and nanoscale electronic components, etc.
Recent studies have reported that the scalable production cost
of graphene sheets in large quantities could be much lower than
CNTs, which might rival the CNTs in many technological
applications.18,19 Graphene-based polymer nanocomposites
(GPNCs) have attracted considerable interest in order to
enhance multifunctional properties of polymer matrix. As a
reinforcing phase, graphene sheets should be incorporated and
distributed homogenously into various polymer matrices.
However, graphene could form aggregation due to the strong
van der Waals forces and p–p interactions between the layers.21
Monoethylene glycol (MEG) is an important intermediate for
the production of polyethylene terephthalate (PET). Generally,
MEG is produced predominantly by non-catalytic liquid-phase
hydration of ethylene oxide (EO) with signicant amounts of by-
products diethylene glycol (DEG) and triethylene glycol (TEG).
In order to obtain high selectivity of MEG, a large excess of water
(about 20–25 molar ratio of water/EO) must be used industrially,
which increases production costs due to energy consumption
and distillation unit operation investment.1,2 Therefore, many
academic and industrial researchers have focused on the cata-
lysts for EO hydration to MEG at low water/EO molar ratio, for
instance, anions of salts (such as acetate, formate, carbonate
and bicarbonate) immobilized on ion-exchange resins (IERs),3–7
quaternary phosphonium halides,8 polymeric organosilane
ammonium salts,8,9 macrocyclic chelating compounds and
supported metal oxides.10–12 In addition, immobilized IER is
considered to be one of the best catalysts because of its excellent
reaction performance and easy separation progress. Shvets
et al.5 reported that under a reaction temperature of 95 ꢀC and a
water/EO molar ratio of 18, the EO conversion and MEG selec-
tivity on the IER catalysts exhibited 95.6% and 98.1%, respec-
tively. However, the instinct drawbacks of the ion-exchange
SINOPEC Shanghai Research Institute of Petrochemical Technology, Shanghai 201208,
PR China. E-mail: yangwm.sshy@sinopec.com
2550 | RSC Adv., 2015, 5, 2550–2561
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
To obtain the well dispersed graphene in the polymer matrix, interface with whole-process stirring. Aer 30 min pre-blending
many methods have been developed to reduce the aggregation, at 40 ꢀC, raised temperature gradually to 70 ꢀC and maintained
including solvent-based exfoliation, ultrasonication with for 1 h. The polymerization occurred and proceeded at 80–85 ꢀC
ꢀ
surfactant and chemical functionalization. Chemical converted for 6 h and then at 95–97 C for 4 h. The polymer composite
graphene sheets or graphene oxide could readily form stable precipitated out as beads, which was then separated and
dispersions in the organic solvents. However, it was found that washed by hot deionized water. The obtained GPNC was dried at
chemical modication couldn't provide a good control of 90 ꢀC and designated as X-GEP (X dened as the weight content
polymer architecture,22 and additional reduction procedure is of graphene from 0.1 up to 1.0 wt%). For instance, sample
required using hydrazine as a chemical reducing agent, which is 0.1-GEP indicates the graphene content in the nanocomposite
not environmentally friendly.23,24
is 0.1 wt%.
Aer polymerization, the Friedel–Cra reaction was carried
To the best of the authors' knowledge, there are limited
reports describing the synthesis and properties of well-dened out to introduce chloromethyl groups to the cross-linked
graphene-based nanocomposite ion-exchange resins (GPNC- networks of the GPNCs. The composite (25.0 g) and chlor-
IERs) by in situ suspension polymerization. This work employed omethyl ethyl ether (75 mL) were mixed to swell the cross-linked
the modied in situ suspension polymerization approach to polymer for 2 h at room temperature. Then, ZnCl2 (9.6 g) was
prepare several strong basic GPNC-IERs without any surfactant. added to the mixture in two batches. The reaction was allowed
Then, the as-prepared GPNC-IER composites were characterized to proceed at 38–40 ꢀC for 10 h. The chloromethylation product
by scanning electron microscopy (SEM), Raman spectroscopy, (designated as X-GEP-Cl) was washed by acetone and dried until
Fourier transform infrared (FTIR) spectroscopy, X-ray photo- a constant weight was reached. Then trimethylamine hydro-
electron spectroscopy (XPS), thermogravimetric-differential chloride was added to the products in three batches and the
ꢀ
thermal analysis (TG-DTA), differential scanning calorimetry amination reaction ran for 6 h at 30 C. Aer the reaction, the
(DSC) and the measurement of anti-swelling properties. The product was washed by deionized water, NaOH and NaHCO3
effects of graphene acting as a reinforcing phase in the resins aqueous solution to obtain the strong basic IERs in hydroxylic
and those of exchange capacity, thermal stability and catalytic and bicarbonate form (designated as X-GEP-OH and X-GEP-HC).
performance are conducted with reference to the incorporation Schematic diagram of preparation steps of the GPNC-IERs is
and interaction of graphene within the polymer networks.
shown in Scheme 1.
As a comparison, styrene-divinylbenzene cross-linked poly-
mer IERs were also synthesized using the same procedure in the
absence of graphene. The products were designated as PS-DVB,
PS-DVB-Cl, PS-DVB-OH and PS-DVB-HC, respectively. All the
polymer IERs with a crosslinking density (CD) of 6% were
controlled and presented according to the following eqn (1).25
Experimental section
Materials
Styrene (St, Sinopharm Chemical Reagent Co., Ltd., 99%) and
divinylbenzene (DVB, Sigma-Aldrich Chemicals, technical
grade, 80%) were washed by NaOH solution (5 wt%) thrice to
remove inhibitors, then washed with deionized water, stirred
over magnesium sulfate overnight and then over calcium
hydride, distilled, and stored under a nitrogen atmosphere at 4
ꢀC. The purication of radical initiator benzoyl peroxide (BPO,
Sinopharm Chemical Reagent Co., Ltd., 98%) was prepared by
recrystallization from methanol solution. Chloromethyl ethyl
ether (96%) was purchased from Tokyo Chemical Industry Co.,
Ltd. Other chemicals including gelatin (CP), zinc chloride
(ZnCl2, AR), tetrahydrofuran (THF, AR), trimethylamine hydro-
chloride (98%) and 1,2-dichloroethane (99%) were purchased
from Sinopharm Chemical Reagent Co., Ltd., and used as
received. Pristine graphene (Nanjing JCNANO Technology Co.,
Ltd., JCGNP-15-10, 99.5%) was calcined to remove amorphous
À
Á
mðDVBÞ
CD % ¼
ꢁ 100%
(1)
mðStÞ þ mðDVBÞ
Characterization and measurement
Scanning electron microscopy (SEM, Philips XL30E) was used to
observe the general size and surface morphology of the polymer
beads. The samples were coated with a thin layer of gold on the
surface. The pressure during scanning was 5.2 ꢁ 10ꢂ9 MPa and
the accelerated voltage was 5 kV. Fracture surface of the nano-
composites were observed by a FEI Nova NanoSEM 450 instru-
ment with accelerating voltage of 5 kV. The Raman spectra were
collected with LabRam-1B micro-Raman spectrometer (Jobin
Yvon Instrument), using the He–Ne laser excitation line at 632.8
nm. Acquisition time was 30 s with a nal laser power of about 3
mW at the surface. Fourier transform infrared (FTIR) spectra
were obtained in dried KBr pellets (1 : 49 sample-to-KBr mass
ꢀ
carbon at 110 C for 1 h in air.
Synthesis of the GPNCs and GPNC-IERs
Graphene was suspended in 10 mL styrene monomer, and then ratio) on Nicolet 5700 spectrometer (ThermoElectron, Madison,
ultrasonicated for 10 min to obtain the required dispersions. 2.5 WI, US). All the spectra were recorded at a range of 4000–400
g gelatin was dissolved in 200 mL deionized water at 40 ꢀC in a cmꢂ1 with 32 scans and a resolution of 4 cmꢂ1 by subtracting
500 mL round-bottom ask. The monomer mixture consisted of the background spectra, and such difference spectra are
45.3 g St, 4.4 g DVB, 0.5 g BPO and the prepared graphene reported herein. The X-ray photoelectron spectroscopy (XPS)
suspension. The suspension polymerization was carried out in experiments were carried out on a Kratos AXIS Ultra DLD (Shi-
the presence of the monomer mixture at the organic–aqueous madzu Corporation, Kyoto, Japan) spectrometer using
a
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 2550–2561 | 2551

View Article Online
RSC Advances
Paper
Scheme 1 Schematic diagram of reaction steps involved in the preparation of GPNC-IERs.
monochromatic Al Ka (hn ¼ 1486.6 eV) irradiation source at 75.0 2000, which describes the determination of the exchange
W. All XPS spectra were recorded with pass energy of 80 eV for capacity of anion-exchange resins in hydroxylic form.13 The
the survey spectra, and 40 eV for high resolution spectra. The methodology was as follows: the sample was rst pre-treated
samples were directly pressed to a self-supported disk (10 ꢁ 10 according to National Standard of People's Republic of China
mm2), mounted onto a sample holder, and then degassed GB/T 5476. Then it was washed with HCl aqueous solution
overnight in the vacuum chamber. The binding energies were (2 mol Lꢂ1), deionized water, and NaOH aqueous solution
calibrated with the containment carbon (C1s ¼ 284.8 eV). (2 mol Lꢂ1) in sequence. The sample was measured and soaked
ꢀ
Thermogravimetry-differential thermal analysis (TG-DTA) was in 100 mL standard HCl solution at 40 C for 2 h, then 25 mL
performed on a TA Instruments SDT Q600 TGA thermogravi- soaked solution was titrated with standard NaOH solution
ꢀ
metric analyzer from room temperature to 900 C at a heating using phenolphthalein as the indicator. The total exchange
rate of 10 ꢀC minꢂ1 under a dynamic (100 mL minꢂ1) air or N2 capacity (Q) was calculated according to eqn (3).
atmosphere. The differential scanning calorimetry (DSC)
100c1 ꢂ 4c2Vt
Q ¼
(3)
cycling curves were measured with DSC Q2000 (TA Instruments-
Waꢀters LLC, USA) in 50 mL minꢂ1 N2 as the purge gas at a rate of
10 C minꢂ1. The procedures of each DSC run were as follows:
heat from room temperature to 200 ꢀC, then hold isothermal at
200 ꢀC for 5 min, cool equilibrated to 50 ꢀC, and then heat from
50 ꢀC to 160 ꢀC. Samples of approximately 7–9 mg in these
examinations were weighed with a precision of 0.01 mg
aluminium pans prior to analysis. All measurements were
repeated at least twice.
Wð1 ꢂ AÞ
where c1 is the molar concentration of standard HCl solution, c2
is the molar concentration of standard NaOH solution. The
volume of the titrating solution of NaOH was recorded as Vt,
while W is the weight of the IER sample. A is the water retention
capacity of the sample measured by National Standard of Peo-
ple's Republic of China GB/T 5759-2000.
The as-prepared resins were characterized by various struc-
tural parameters such as swelling capacity (SC), water retention
capacity (A), mean mesh size (Dv), average mesh density (r) and
total exchange capacity (Q). The swelling capacity was deter-
mined through the procedures given below. The resins were
immersed in deionized water for 24 h and dried at 60 ꢀC for 72
h. The volumes of the resins were recorded aer the drying
process as Va. As mentioned above, the anti-swelling property of
the IER catalysts are crucial to the catalytic hydration of EO. The
swollen volumes in various solvent systems were measured as
Vs. The swelling capacity in the solvents were calculated
according to eqn (2).26
Catalytic application in hydration of EO
The prepared GPNC-IERs and PS-DVB IER were applied in
hydration reaction of EO. Typically, the IER catalysts were
placed in a xed bed reactor with a dimension of 10 mm (i.d.) ꢁ
300 mm (length). The experiments were carried out with a
liquid hourly space velocity of 3 hꢂ1 (water/EO molar ratio of 10)
under pressure of 1.2 MPa in the temperature range of 90–102
ꢀC. The composition of effluent product was analyzed by a gas
chromatograph (Agilent 7890A GC) equipped with a 30 m
(length) ꢁ 0.25 mm (i.d.) ꢁ 0.25 mm (lm) HP-INNOWAX
capillary column and a ame ionization detector (FID). Both EO
conversion and MEG selectivity were taken into account as the
criterion for comparison of IER catalysts' performance in this
study.
À
Á
Vs ꢂ Va
SC % ¼
ꢁ 100%
(2)
Va
À
Á
For evaluating the catalytic performance, total exchange
capacity of the prepared resins were measured according to
National Standard of People's Republic of China GB/T 5760-
nðMEGÞ þ 2nðDEGÞ þ 3nðTEGÞ
n0ðEOÞ
XEO % ¼
ꢁ 100%
(4)
2552 | RSC Adv., 2015, 5, 2550–2561
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
À
Á
nðMEGÞ
SMEG % ¼
ꢁ 100% (5)
nðMEGÞ þ 2nðDEGÞ þ 3nðTEGÞ
where n(MEG), n(DEG) and n(TEG) are the molar compositions
in the effluent product, and n0(EO) is the molar composition of
EO in the feed.
Fig. 2 SEM images of PS-DVB and GPNC-IERs in bicarbonate form: (a)
PS-DVB; (b) 0.1-GEP; (c) 0.4-GEP; (d) 0.7-GEP; (e) 1.0-GEP; (f)
PS-DVB-HC; (g) 0.1-GEP-HC; (h) 0.4-GEP-HC; (i) 0.7-GEP-HC; (j)
1.0-GEP-HC.
Results and discussion
Morphology of the GPNCs and GPNC-IERs
Various graphene-based polymer composites were prepared via
the modied suspension polymerization. The quality of gra-
phene dispersion in the monomer during polymerization not
only affects the further dispersion in the polymer matrix, but
also directly correlates with their effectiveness for improving
thermal, anti-swelling and other properties. The dispersion
stability against van der Waals aggregation of graphene in
styrene for 2 h was compared in Fig. 1. Aer the mixture was
ultrasonicated for 10 min, it appeared that graphene could
disperse well in styrene. Although some particulates were
settled to the bottom of the vials over 2 h, graphene stayed
homogenously for at least 30 min. Moreover, graphene shows
better dispersion behavior at low concentration ([graphene] #
0.7 mg mLꢂ1) even aer several hours. For example, a graphene
suspension with 1 h ultrasonication at a low concentration of
0.1 mg mLꢂ1 was stable for more than 12 h. At higher concen-
tration, the suspension was found to have lower stability, and
some graphene particulates were settled to the bottom of the
vial. This indicates that the equilibrium aggregation of gra-
phene could be controlled and the concentration of graphene in
the suspension is critical. The highly stable graphene suspen-
sion in monomer is possibly related to strong interactions
between graphene and monomer, thus favoring the homoge-
neity of the beads during the suspension polymerization.
Fig. 2 presented the SEM micrographs of the polymer beads
and IERs in bicarbonate form. By suspension polymerization,
graphene was proximately homogeneous dispersing in the
monomer and bound in or to polymer cross-linked structure
gradually with the elevated temperature without causing
aggregation. Therefore, the GPNCs resulted in very ne beads
with spherical shapes as well as PS-DVB, and the size of the
beads was in the same range of 400–600 mm. Fig. 2 also showed
the surface morphologies of the GPNC-IERs. The structure of
GPNCs obviously changed from smooth sphere with homoge-
neously dispersed graphene to the beads with some crumpled
cracks and isolated patches on the surface when the graphene
content reached 1.0 wt%. These results indicate that graphene
had an optimum content in the polymer networks.
High-resolution SEM images of the fracture surface of
GPNCs were shown in Fig. 3. In the case of 0.4-GEP, graphene
was well dispersed throughout the matrix, without agglomer-
ates formed during the polymerization. More importantly, as
seen in the Fig. 3(a), graphene was completely separated in
pieces and tightly embedded in the matrix, suggesting the
excellent dispersion and compatibility of graphene within the
polymer matrix. For the SEM image of 1.0-GEP, some graphene
sheets exhibited a stack on the fracture surface, showing the
formation of defects and higher roughness on the fracture
surface. These SEM results revealed that lower graphene
contents in GPNCs could effectively prohibit the restacking of
graphene, providing the strengthened interfacial interaction
and a graphene–polymer 3D-network.
TEM results also conrmed the well dispersed graphene in
the polymer in Fig. 4. Pristine graphene exhibited a wrinkled
morphology with lamellar structure. Fig. 4(b) showed the 0.4-
GEP consisted of single-to-few layer graphene closely associated
with the polymer, forming the strong interfacial action within
the polymer matrix.
Structural properties of the GPNCs and GPNC-IERs
The Raman spectra of pristine graphene, PS-DVB and GPNCs
were plotted in Fig. 5. Typically, Raman spectroscopy is effective
in investigating the structure defects in graphene by monitoring
the D- and G-band.27 The D-band peak, which occurs at ꢃ1330
cmꢂ1, is the nger print of defects since it is associated with
hexagonal framework defects or sp3 carbon, while the G-band
Fig. 1 Photographs of graphene dispersions in styrene: (a) [graphene]
¼ 0.09 mg mLꢂ1; (b) [graphene] ¼ 0.42 mg mLꢂ1; (c) [graphene] ¼ 0.66 Fig. 3 SEM images of fracture surface of GPNCs: (a) 0.4-GEP; (b) 0.7-
mg mLꢂ1; (d) [graphene] ¼ 0.98 mg mLꢂ1
.
GEP; (c) 1.0-GEP.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 2550–2561 | 2553

View Article Online
RSC Advances
Paper
prominent peaks at ꢃ1330 cmꢂ1, ꢃ1580 cmꢂ1, ꢃ2670 cmꢂ1 and
some characteristic peaks of polymer structure. Since the ID/IG
ratio increased dramatically from 0.78 of pristine graphene to
1.65 of 0.4-GEP, it could be assumed a portion of sp2-domain
carbon of graphene transformed to sp3-domain to form covalent
bonds with polymer networks during polymerization, which
gave rise to the appearance of D-band peak on graphene and
developed strong interactions between the polymer chains and
graphene.30,31 The peak positions of the G-band were not shied
appreciably and the positions of D-band and 2D-band showed
shis by ꢃ6 cmꢂ1 compared to the pristine graphene, which
was probably because the polymer chains anchored on the
graphene. This strong chemical interaction increased the
energy necessary for vibrations to occur, which is reected in
the higher frequency of the Raman peaks.32 Fig. 5(b) showed
Raman spectra comparison of the composites with different
contents of graphene. The ID/IG ratio was calculated to be 1.65,
1.13 and 1.05 for 0.4-GEP, 0.7-GEP and 1.0-GEP, respectively.
Graphene of 1.0 wt% content could chemically interact with
polymer networks to some extent as 0.4-GEP and 0.7-GEP
samples. However, it also dispersed in form of agglomeration
state in the polymer matrix, as depicted in Fig. 2 and 3.
Therefore, the Raman spectra indicated an optimum content of
graphene in a polymer matrix when graphene was introduced in
the suspension polymerization. From the viewpoint of Raman
spectra and SEM images, it can be assumed that during the
suspension polymerization of 0.4-GEP, graphene could effec-
tively tether with the polymer chain radicals or initiator frag-
ments and penetrate into the polymer networks,13,33 which leads
to well dispersion of graphene in the polymer matrix and
enhanced interaction between the graphene and the polymer
networks.
Fig. 4 TEM images of pristine graphene and 0.4-GEP nanocomposites.
The FTIR spectra were shown in Fig. 6. Aer chlor-
omethylation, the samples showed the absorption peaks at 1265
cmꢂ1 and 677 cmꢂ1, which were attributed to the in-plane
bending vibration of aromatic C–H in Ph-CH2Cl groups and C–
Cl stretching vibration.34 Concomitantly, the weak shoulder
peaks at ꢃ1421 cmꢂ1 were assigned to the bending vibration of
C–H in chloromethyl groups (–CH2Cl). Aer amination, all the
chlorine-containing groups disappeared and new peaks at ꢃ976
cmꢂ1 attributed to C–N stretching were clearly observed.
Besides, the peaks at ꢃ1635 cmꢂ1 corresponded to O–H in-
plane bending vibration. These results conrmed the existence
of –N+(CH3)3OHꢂ groups graed on the polymer chains in
successful chloromethylation and amination step.13,35,36 Aer
ion exchange process, the strong absorption bands were
observed at ꢃ1280 cmꢂ1 in PS-DVB-HC and 0.4-GEP-HC, which
were attributed to the stretching vibration of C–O in bicar-
bonate groups.
Fig. 5 Raman spectra of pristine graphene, PS-DVB and GPNCs.
peak at ꢃ1580 cmꢂ1 is due to the in-plane bond stretching
To further understand the carbonaceous species in the
motion of sp2 carbon. Besides, the spectra also exhibited polymer, the tting XPS curves of the core-level C1s spectra were
another characteristic 2D-band peak appeared at ꢃ2670 cmꢂ1
.
plotted in Fig. 7. Table 1 summarized the detailed atomic
The interaction between graphene and polymer networks could composition of the surface C and O. Some trace amounts of
be estimated by the intensity ratio of the D- and G-bands (ID/IG), other elements were not shown in the table. Since the XPS
where ID and IG were the integrated intensities calculated from experiments were carried out with degassing overnight, the
tted spectra curves using Lorentzian lineshapes parame- presence of surface O is a reference variable in the results which
ters.28,29 As shown in Fig. 5(a), the composites displayed the may be derived from the functional group on the IERs surface
2554 | RSC Adv., 2015, 5, 2550–2561
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
contained different contents of graphene in the polymer
networks, the fundamental structures of the IERs were nearly
identical. However, the structural properties, such as total
exchange capacity and swelling capacity, were highly inuenced
by the graphene contents. In good solvent such as THF, the
swelling capacity dropped rapidly with the increasing graphene
content. Besides, it was found that small amounts of graphene
dramatically improved the total exchange capacity behavior.
Among these GPNC-IERs, 0.4-GEP-HC and 0.7-GEP-HC showed
the higher value at 1.88 and 1.91 mmol gꢂ1. This result could be
explained by the fact that the high level electron cloud density in
graphene might produce p–p conjugative effect between gra-
phene and adjacent aromatic structure of styrene, and favor the
electrophilic substitution reaction of chloromethylation
process, providing more reactive sites for functional groups, as
shown in Scheme 1. From the morphology of 1.0-GEP-HC
shown in Fig. 2, some graphene aggregated in the nano-
composite without chemically bonded with polymer chains.
This might be the reason that 1.0-GEP-HC showed slight
decrease in total exchange capacity compared to 0.7-GEP-HC. It
should be noted that the exchange capacity was related to the
total basic alkaline content of the HCO3ꢂ group. Therefore, the
nanocomposites with 0.4 wt% and 0.7 wt% graphene had the
higher level of basic content, which was consistent with the XPS
results.
The TGA-DTA of the GPNC and GPNC-IERs were measured
and depicted in Fig. 8. As observed in the TGA curves, pristine
graphene exhibited a high thermal stability with the initial
decomposition temperature at 637.8 ꢀC and the maximum
weight loss rate occurred around 722.0 ꢀC. PS-DVB-HC exhibited
three degradation steps occurring at the temperature ranges of
100–200 ꢀC, 200–450 ꢀC and 450–600 ꢀC, respectively. The initial
10% weight loss was due to the elimination of the moisture in
the samples at the rst stage. Then PS-DVB-HC showed a major
weight loss stage about 70%, which was due to the degradation
of the branching chains and crosslinking networks to short
chains or volatile compounds induced by oxidative atmo-
sphere.20,40,41 The third stage weight loss of 20% was assigned to
the combustion of char residue of carbonaceous to volatile
fragments (such as CO2 aꢀnd water), and all the samples
completely burnt off at 600 C. In the case of GPNC-IERs, the
last stage showed higher amount of char residue compared to
the PS-DVB-HC, which was related to the decomposition/
combustion of graphene and the residue of carbonaceous. The
extrapolated degradation onset temperatures (Td) and the
Fig. 6 FTIR spectra of PS-DVB and GPNC-IERs.
introduced by amination step and absorbed moisture. 0.7-GEP-
OH showed the highest level of 12.49% surface O atomic
composition, indicating the highest value of –OH group in the
amination product. The C1s core-level spectra were deconvo-
luted into three peaks at the binding energy of ꢃ284.8 eV,
ꢃ286.2 eV and 288.0 eV, which could be assigned to the sp2 C]
C/sp3 C–C, –C–O and –C]O components.37–39 It could be seen
that the addition of graphene largely suppressed the binding
energy intensity of sp2 C]C/sp3 C–C, while that associated with
–C–O species increased accordingly. The increased surface –C–
O/–C–C ratios of GPNC-IERs with 0.1–0.7 wt% conrmed the
successful functionalization process of well dispersed nano-
composites. 1.0-GEP-OH sample exhibited a decrease in O/C
and –C–O/–C–C ratios on the surface, which was highly
consistent with SEM and Raman results.
temperature of the maximum weight loss rate (Tmax1 and Tmax2
)
in Table 3 showed the effect of graphene content on the
degradation behavior and thermal stability of the GPNC-IERs.
The increase in Td and Tmax indicated the enhancement in the
thermal stability of GPNC-IERs. PS-DVB-HC reached the
ꢀ
maximum weight loss velocity at 417.7 C. The Td and Tmax of
The physical properties of the GPNC-IERs were listed in
Table 2. To determine the values of the water retention capacity,
total exchange capacity and swelling capacity, all the samples
were measured at least twice. The structural properties of
GPNC-IERs resembled the PS-DVB-HC, including average mesh
density and mean mesh size. Although the composites
the GPNC-IERs all shied to higher temperature, especially for
0.4-GEP-HC and 0.7-GEP-HC, it increased by ꢃ20 ꢀC compared
to PS-DVB-HC.
A thermal transition of a segmental motion of the polymer
chains was examined and the glass transition temperature (Tg)
obtained by DSC measurements were also summarized in Table
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 2550–2561 | 2555

View Article Online
RSC Advances
Paper
Fig. 7 Core-level C1s XPS spectra of PS-DVB and GPNC-IERs: (a) PS-DVB-OH; (b) 0.1-GEP-OH; (c) 0.4-GEP-OH; (d) 0.7-GEP-OH; (e) 1.0-GEP-OH.
3. In the case of pure PS-DVB-HC, Tg was at 89.6 ꢀC and slightly with 0.4 wt% and 0.7 wt% graphene. Normally, the resin catalyst
ꢀ
increased by 3 C for 0.4-GEP-HC.
stability is the critical factor to restrict the application because
the resins swell and deactivate under the reaction conditions.
These improved GPNC-IER resin catalysts were applied in
catalytic hydration of EO.
As seen in Fig. 9, the EO conversion of all the IER catalysts
increased with the temperature up to 102 ꢀC, whereas the MEG
selectivity slightly decreased at higher temperature. This was
quite similar to the result reported by Yu et al.13 Although high
temperature could accelerate hydration reaction rate with high
EO conversion, it was likely that the decrease of MEG selectivity
Catalytic hydration of ethylene oxide on the GPNC-IER
catalysts
The characterization results conrmed the enhanced anti-
swelling capacity and thermal stability of GPNCs and GPNC-
IERs and the amounts of surface quaternary ammonium group
increased as well. In particular, the structural and thermal
stability were improved signicantly in the nanocomposites
Table 1 Surface atomic composition of the GPNC-IERs and PS-DVB IER samples in hydroxylic form
Surface atomic composition and ratio (%)
Surface covalent bonds ratioa (%)
Samples
O1s
C1s
O/C ratio
–C–O/–C–C ratio
Pristine graphene
PS-DVB-OH
0.1-GEP-OH
0.4-GEP-OH
0.7-GEP-OH
1.0-GEP-OH
1.83
6.61
6.38
11.96
12.49
7.12
97.87
89.28
89.74
83.17
81.69
89.88
1.9
7.4
7.1
14.4
15.3
7.9
2.1
21.2
20.5
47.6
50.4
34.2
a
Calculated by the area of tting curves of the core-level C1s spectra.
2556 | RSC Adv., 2015, 5, 2550–2561
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 2 Physical properties of the GPNC-IERs and PS-DVB IER samples
Swelling capacity
(SC, %)
Mesh densitya
Mean mesh size
(Dv, nm)
Water retention
capacity (A, %)
Total exchange
Samples
(r, g cmꢂ3
)
capacity (Q, mmol gꢂ1
)
Water
THF
PS-DVB-HC
0.1-GEP-HC
0.4-GEP-HC
0.7-GEP-HC
1.0-GEP-HC
0.53
0.53
0.54
0.56
0.57
0.44
0.43
0.54
0.57
0.54
21.1 ꢄ 0.1
19.9 ꢄ 0.2
39.8 ꢄ 0.2
40.2 ꢄ 0.1
28.7 ꢄ 0.2
0.28
0.33
1.88
1.91
1.47
29.2
24.9
18.4
18.3
19.1
43.8
31.7
28.0
26.6
28.9
a
Measured by volume-weighted mean mesh density.
was due to the side reactions in the circumstance of higher MEG catalysts were inevitable to be affected and undergo thermal
concentration with elevated temperature.1,2,42 On the other degradation and swelling even at a low water/EO molar ratio of
hand, the MEG selectivity was also inuenced by the undesir- 10, thus resulting in low MEG selectivity. According to the
able swelling and thermal deactivation of the IER catalysts aforementioned results, the higher values of [HCO3ꢂ] and
during reaction. The quaternary ammonium groups on the IER [OHꢂ] in the GPNC-IERs facilitated the hydration reaction with
catalysts surface were the active sites for the hydration of EO via higher EO conversion. Meanwhile, good graphene dispersion
the ring-opening reaction, and high water/EO molar ratio was and graphene–polymer network facilitated heat transfer and
normally used for obtaining high MEG selectivity. The IER allowed the heat of the reaction to travel efficiently through the
polymer matrix and graphene. Therefore, the GPNC-IERs
maintained stable under the reaction conditions because of the
improved anti-swelling capacity and thermal stability.
Xie et al.17 have reported the enhanced performance of
polymer/CNTs composites in the EO hydration reaction test. For
comparison, the EO hydration reaction results of CNTs-based
polymer nanocomposites ion-exchange resins (CPNC-IERs) and
GPNC-IERs were shown in Fig. 10. The 0.4-CNTP-HC was
prepared by the same procedure as 0.4-GEP-HC, except the
CNTs as a reinforcing material in the polymer nanocomposites.
The EO conversion and MEG selectivity for 0.4-GEP-HC exhibi-
ꢀ
ted better performance at a range of 90–102 C. These results
indicate that the enhancement in properties and reactive
performance of graphene-based nanocomposites is markedly
superior to the effect of the CNTs in the polymer matrix.20
The long-time run of the IER catalysts in the hydration of EO
ꢀ
at 98 C was carried out for 500 h on stream, and the results
were plotted in Fig. 11. It was found that the EO conversion and
MEG yield for 0.4-GEP-HC remained at 98% and 95%, which
were much higher than the conventional PS-DVB-HC catalyzed
Table 3 Thermal properties of the GPNC-IERs and PS-DVB IER
samples
Td (ꢀC) Tmax1 (ꢀC) Tmax2 (ꢀC) Charc (wt%) Tg (ꢀC)
a
b
b
d
Samples
PS-DVB-HC 351.8
0.1-GEP-HC 360.0
0.4-GEP-HC 372.3
0.7-GEP-HC 375.8
1.0-GEP-HC 370.3
417.7
429.5
438.6
441.3
419.8
548.6
549.3
556.2
560.0
555.3
1.11
1.24
1.57
1.79
2.09
89.6
90.8
92.3
91.7
85.9
a
b
Determined by the extrapolated onset temperatures. Maximum
c
weight loss temperatures in DTA curves. Determined from the TGA
d
measurements in N2 at 700 ꢀC. Determined as the midpoint
Fig. 8 TG-DTA curves of PS-DVB and GPNC-IERs: (a) PS-DVB-HC; (b)
temperature at half the complete change of heat capacity in DSC curves.
0.4-GEP-HC; (c) 0.7-GEP-HC; (d) pristine graphene.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 2550–2561 | 2557

View Article Online
RSC Advances
Paper
gradually decreased to ꢃ89% aer 500 h reaction test. This was
likely due to the thermal deactivation of the active groups on the
catalysts resulting in lower activity and selectivity. The total
exchange capacity of the IER catalysts were measured again
aer 500 h hydration test, the data in Table 4 conrmed that the
addition of graphene in the IER catalysts was benecial to
improve the stability of the surface active sites.
Effect of graphene in the nanocomposites
In this work, graphene was applied as a reinforcing material in
the polymer nanocomposites synthesized via in situ suspension
polymerization. The pre-suspension procedure was used for
assisting the formation of uniform polymer beads. The results
showed that graphene could be well dispersed in the compos-
ites, and the strong interaction between graphene and polymer
cross-linked networks was demonstrated here to be an impor-
tant factor to the enhancement effects in thermal stability and
anti-swelling properties of the composites.
It is generally accepted that two main approaches, “graing-
from” and “graing-to”, have been developed to introduce
graphene into the polymer structures. It is worth noting that the
GPNCs are obtained by both methodologies. At the beginning of
polymerization, the initiator BPO generates phenyl radicals,
which can lead to a functionalization reaction on graphene
sheets. The resulting radical functionalized graphene could
subsequently react with the adsorbed styrene monomer mole-
cule. With covalent bonding with free radical and styrene, a
portion of the sp2 carbon on graphene is converted to sp3
carbon leading to an increase in D-band peak intensity in
Raman spectra. Mylvaganam et al.43 showed by computation
that free radical functionalized graphene could initiate the
polymerization reaction with chemically bonded polymer
chains growing on graphene. Based on this “graing-from”
methodology, the polymeric fragment is propagated from gra-
phene surface by capturing more styrene molecules.33,44 Mean-
while, based on the “graing-to” methodology, polymer chain
with free radical, like [–CH2–CH(Ph)–/–CH2–CH(Ph)]c can be
directly conjugated with graphene, which allows further
Fig. 9 Catalytic performance of PS-DVB-HC, 0.4-GEP-HC and 0.7-
GEP-HC for the hydration of EO.
system during the whole 500 h test, allowing a large window for
the industrial application in the hydration of EO in the future.
Additionally, the MEG yield on the conventional PS-DVB-HC
Fig. 10 Catalytic performance of 0.4-CNTP-HC and 0.4-GEP-HC for Fig. 11 Catalytic performance of PS-DVB-HC and 0.4-GEP-HC
the hydration of EO.
catalysts within 500 h.
2558 | RSC Adv., 2015, 5, 2550–2561
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 4 Reaction results of the GPNC-IERs and PS-DVB IER catalysts at 98 ꢀC
EO conversion
(X, %)
MEG selectivity
(Y, %)
Time space yield of MEG
Total exchange capacitya
Catalysts
(STY, g MEG hꢂ1 g catꢂ1
)
(Q, mmol gꢂ1
)
PS-DVB-HC
0.4-GEP-HC
94.70
98.05
96.13
97.03
0.96
1.32
0.07
0.76
a
Measured aer 500 h reaction tests.
conjugation of a new monomer molecule or polymer chain. This dispersed graphene can induce the barrier labyrinth effect,
chain propagation prefers to occur on the opposite side of impeding the transport of degradation products of polymer into
graphene sheet to avoid steric hindrance, which appears like the gas phase. More energy is required to break the covalent
graphene incorporated into the polymer networks. The growing linkages and transport the degradation products of polymer,
polymer chains will terminate through combining with another which tends to shi the Td and Tmax to higher temperatures. As
propagating polymer chain or radical functionalized graphene described above, graphene can chemically incorporated into the
directly.
polymer networks, which introduces the steric limitation
In the as-prepared polymer composites, graphene acts as through the resin and restricts the polymer chain mobility in
covalent cross-linkers to form the polymer networks improving the nanocomposites, resulting in higher Tg value.19,44 Therefore,
interfacial interaction with polymer chains. Moreover, the well it may be concluded that covalent bonded graphene via direct
polymerization herein effectively protects against thermal
degradation and transition for the polymer networks, providing
enhanced structural and thermal stability of GPNCs and GPNC-
IERs.
To better demonstrate the graphene–polymer synergism in
thermal stability, dynamic FTIR spectra of 0.4-GEP-OH and PS-
DVB-OH at different temperatures in N2 were recorded and
presented in Fig. 12. The peaks observed in the region of 3025–
3080 cmꢂ1 and 2920–2850 cmꢂ1 were attributed to the stretch-
ing vibrations of the aromatic C–H and the C–H stretching
vibrations of the methylene groups. The broad absorption peaks
at 3420 cmꢂ1 and 1633 cmꢂ1 could be assigned to the stretching
vibration and in-plane bending vibration of associative –OH
group.35 During heating up process, the intensity of these peaks
decreased slowly, indicating the C–H and –OH dissociated with
the elevated temperature to 200 ꢀC. The obvious diminution in
peak intensity at 1310 cmꢂ1 implied the thermal decomposition
of C–N groups.45 The gradual peak intensity decrease of 1450–
1600 cmꢂ1 from 150–400 ꢀC could infer that the aromatic C]C
of the carbon skeleton decomposed to some volatile fragments,
leading to the disappearance of the absorption band at 450 ꢀC.
As for 0.4-GEP-OH, shown in Fig. 12(b), the distinct peak
intensity decrease at 3420 cmꢂ1 and 1640 cmꢂ1 were assigned to
Fig. 12 FTIR spectra in N2 atmosphere of (a) PS-DVB-OH and (b) 0.4- Scheme 2 Schematic diagram of the initial thermal degradation steps
GEP-OH.
of (a) PS-DVB-OH and (b) 0.4-GEP-OH samples.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 2550–2561 | 2559

View Article Online
RSC Advances
Paper
the initial dissociation of the quaternary ammonium hydroxide poorer than the composites with 0.4 wt% graphene. Such a
groups (–N+(CH3)3OHꢂ) with the elevated temperature to 250 ‘volcano-type’ trend on water retention capacity, exchange
ꢀC, which showed the ꢃ50 ꢀC improvement in thermal stability capacity, anti-swelling property and thermal stability might be
of functional groups. Besides, the absorption band at 1410 attributed to the dispersion of graphene within the composite
cmꢂ1 was assigned to the C–H in-plane bending vibration of structures and the interaction between graphene–polymer
–CH2–N+ groups, which decreased and disappeared until 250 networks,51 which is consistent with the mechanism proposed
ꢀC, conrming the decomposition of the functional groups on previously. Considering this trend in the GPNC-IERs, it might
the GPNC-IERs at higher temperature.46
need further study in order to exploit the optimum graphene
Scheme 2 depicted the initial thermal degradation steps of concentration for the composites' properties.
the PS-DVB-OH and 0.4-GEP-OH samples. It is well accepted
that quaternary ammonium hydroxides (–N+(CH3)3OHꢂ)
undergo thermal degradation to produce a tertiary amine and
Conclusions
an alcohol. The two possible degradation ways result in either a In summary, we have demonstrated a generic approach to
basic group or a non-active benzyl alcohol.47 PS-DVB-OH prepare GPNCs by modied surfactant-free in situ suspension
preferred to decompose started by breaking the –CH2–N+ bond, polymerization. The direct use of graphene to react with styrene
producing a benzyl alcohol structure. The degradation of 0.4- monomer could produce the GPNCs with good dispersion of
GEP-OH occurred preferentially by –OH group dissociation and graphene and strong interaction within the polymer matrix. The
obtained a weak basic anion exchange resin, as shown in structural and thermal stability of the GPNCs and GPNC-IERs
Scheme 2(b). The present of weak basic anion of –N+(CH3)2 on were improved greatly with incorporation of graphene in the
the exchange resins could prevent the deactivation of IER polymer networks. With a low content at 0.4 wt% of graphene,
catalysts, which further conrmed the catalytic reaction results. the strong basic GPNC-IERs showed enhanced activity and
Overall, the addition of graphene to PS-DVB polymer resin selectivity performance in catalytic hydration of EO to MEG. The
performed an enhanced thermal stability and anti-swelling long-term reaction test further conrmed that graphene effec-
capacity. More attractively, Raman spectra showed that gra- tively reinforced the properties with well incorporation into the
phene tends to form a covalent bond with the polymer chains at polymer networks.
the reactive sites on the graphene sheets, and/or assembles a
3D-network stability through the polymer resin, suggesting
good interfacial compatibility between the graphene and the
Acknowledgements
entangled polymer chains. The incorporated graphene func- This work was supported by Science and Technology Commis-
tions as an effective barrier to inhibit the mass loss and retards sion of Shanghai Municipality, China (13R21421900).
the transfer of degradation products of the polymer into the gas
phase, enabling the higher thermal degradation tempera-
tures.40,44,48,49 Detailed information of the TG-DTA and dynamic
Notes and references
FTIR results demonstrated that graphene mainly affect the
degradation of functional groups and cross-linked polymer
networks of the IERs. It is noteworthy that the composites with
0.1–0.7 wt% graphene exhibited enhancements in total
exchange capacity and resistance to thermal degradation.
However, the composite containing 1.0 wt% graphene exhibited
the opposite behavior, especially Tmax1 and Tg. Therefore, the
graphene content in the composites is not the only factor for the
changes in the properties. Considering the morphology results
shown in Fig. 2 and 3, graphene was well distributed all over the
polymer surface at low concentrations, while 1.0-GEP showed
some great irregularity and rupture on the surface. The aggre-
1 Z.-J. Yang, N. Ren, Y.-H. Zhang and Y. Tang, Catal. Commun.,
2010, 11, 447–450.
2 H. Yue, Y. Zhao, X. Ma and J. Gong, Chem. Soc. Rev., 2012, 41,
4218–4244.
3 M. R. Altiokka and S. Akyalçin, Ind. Eng. Chem. Res., 2009, 48,
10840–10844.
4 V. F. Shvets, R. A. Kozlovskiy, I. A. Kozlovskiy, M. G. Makarov,
J. P. Suchkov and A. V. Koustov, Chem. Eng. J., 2005, 107, 199–
204.
5 V. F. Shvets, R. A. Kozlovskiy, I. A. Kozlovskiy, M. G. Makarov,
J. P. Suchkov and A. V. Koustov, Org. Process Res. Dev., 2005,
9, 768–773.
gation of graphene accounts for
enhancement effect because of reduction of effective interfacial
area between the graphene and polymer networks.
As elaborated above, the nanocomposites with 0.7 wt% gra-
phene showed the best properties among these prepared
samples. However, whether 0.7 wt% of graphene in the polymer
a
suppression in the
6 G. R. Strickler, V. G. Landon, G.-S. J. Lee and W. J. Rievert,
WO Patent, 9931034, 1999.
7 G. H. Grosch, B. Stein and E. Gehrer, DE Patent, 19757684,
1999.
8 K. Kawabe, US Pat., 6080897, 2000.
9 E. M. G. A. V. Kruchten, US Pat., 5874653, 1999.
resin manifests the optimum performance in these structural 10 Y. Li, B. Yue, S. Yan, W. Yang, Z. Xie, Q. Chen and H. He,
properties is not clear yet. Nonetheless, the total exchange Catal. Lett., 2004, 95, 163–166.
capacity of the GPNC-IER is not in a simple positive correlation 11 C.-F. Lien, C.-H. Ho, C.-Y. Shieh, C.-L. Tseng and J.-L. Lin, J.
with graphene contents from 0.1–1.0 wt%, but a ‘volcano-type’ Phys. Chem. C, 2008, 112, 8365–8371.
scatter similar to the results reported by Malho et al.50 At gra- 12 Z.-J. Yang, Y.-F. Li, Q.-B. Wu, N. Ren, Y.-H. Zhang, Z.-P. Liu
phene content of 1.0 wt%, the properties are even slightly
and Y. Tang, J. Catal., 2011, 280, 247–254.
2560 | RSC Adv., 2015, 5, 2550–2561
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
13 F. Yu, H. Cai, W. He, W. Yang and Z. Xie, J. Appl. Polym. Sci., 32 R. Asif, D. D. Mark, I. Ilia, F. B. Phillip and B. G. David, Chem.
2010, 115, 2946–2954. Mater., 2006, 18, 3513–3522.
14 E. J. Simpson, W. J. Koros and R. S. Schechter, Ind. Eng. 33 J. Park and M. Yan, Acc. Chem. Res., 2013, 46, 181–189.
Chem. Res., 1996, 35, 1195–1205.
15 A. R. Vaino and K. D. Janda, J. Comb. Chem., 2000, 2, 579–
596.
16 T. Ghosh Dastidar and A. Netravali, ACS Sustainable Chem.
Eng., 2013, 1, 1537–1544.
17 Z. Xie, Z. Liu, Y. Wang, Q. Yang, L. Xu and W. Ding, Int. J.
Mol. Sci., 2010, 11, 2152–2187.
18 P. A. Denis, J. Phys. Chem. C, 2009, 113, 5612–5619.
19 J. R. Potts, D. R. Dreyer, C. W. Bielawski and R. S. Ruoff,
Polymer, 2011, 52, 5–25.
20 Y. Li, F. Yu, W. He and W. Yang, J. Appl. Polym. Sci., 2015,
132, DOI: 10.1002/app.41234.
21 X. Qian, L. Song, B. Yu, B. Wang, B. Yuan, Y. Shi, Y. Hu and
R. K. K. Yuen, J. Mater. Chem. A, 2013, 1, 6822–6830.
22 Y. Jing, H. Tang, G. Yu and P. Wu, Polym. Chem., 2013, 4,
2598–2607.
34 Y. Cao, Z. Lai, J. Feng and P. Wu, J. Mater. Chem., 2011, 21,
9271–9278.
35 Y. Zhao, M. Seredych, Q. Zhong and T. J. Bandosz, RSC Adv.,
2013, 3, 9932–9941.
36 K. Kunimatsu, B. Bae, K. Miyatake, H. Uchida and
M. Watanabe, J. Phys. Chem. B, 2011, 115, 4315–4321.
37 R. Haerle, E. Riedo, A. Pasquarello and A. Baldereschi, Phys.
Rev. B: Condens. Matter Mater. Phys., 2001, 65, 045101–
045109.
38 Z. J. Han, B. K. Tay, P. C. T. Ha, J. Y. Sze and D. H. C. Chua, J.
Appl. Phys., 2007, 101, 053301–053307.
39 W. H. Lee, S. J. Kim, W. J. Lee, J. G. Lee, R. C. Haddon and
P. J. Reucro, Appl. Surf. Sci., 2001, 181, 121–127.
40 F. Laoutid, L. Bonnaud, M. Alexandre, J. M. Lopez-Cuesta
and P. Dubois, Mater. Sci. Eng., R, 2009, 63, 100–125.
41 S. Navalon, A. Dhakshinamoorthy, M. Alvaro and H. Garcia,
Chem. Rev., 2014, 114, 6179–6212.
23 W. Hu, J. Zhan, X. Wang, N. Hong, B. Wang, L. Song,
A. A. Stec, T. R. Hull, J. Wang and Y. Hu, Ind. Eng. Chem. 42 J. W. van Hal, J. S. Ledford and X. Zhang, Catal. Today, 2007,
Res., 2014, 53, 3073–3083. 123, 310–315.
24 J.-M. Thomassin, M. Triovic, W. Alkarmo, C. Detrembleur, 43 K. Mylvaganam and L. Zhang, J. Phys. Chem. C, 2013, 117,
´ ˆ
C. Jerome and C. Macosko, Macromolecules, 2014, 47, 2149–
2817–2823.
2155.
44 A. S. Patole, S. P. Patole, S.-Y. Jung, J.-B. Yoo, J.-H. An and
T.-H. Kim, Eur. Polym. J., 2012, 48, 252–259.
45 S. Ramalingam, S. Periandy, B. Narayanan and S. Mohan,
Spectrochim. Acta, Part A, 2010, 76, 84–92.
46 A. Danon, P. C. Stair and E. Weitz, J. Phys. Chem. C, 2011,
115, 11540–11549.
47 G. C. Allen, N. R. Holmes, B. J. Lee and R. R. Harries, J. Chem.
Soc., Faraday Trans., 1988, 84, 3891–3904.
25 S. Bhuyan, S. Sundararajan, D. Andjelkovic and R. Larock,
Tribol. Int., 2010, 43, 831–837.
26 Y. Huang, M. Zeng, J. Ren, J. Wang, L. Fan and Q. Xu,
Colloids Surf., A, 2012, 401, 97–106.
27 D. Fan, Y. Liu, J. He, Y. Zhou and Y. Yang, J. Mater. Chem.,
2012, 22, 1396–1402.
28 S. Sahoo, R. Palai, S. K. Barik and R. S. Katiyar, J. Raman
Spectrosc., 2013, 44, 798–802.
48 K. Takashi, G. Eric, H. Jenny, H. Richard, A. Walid and
D. Jack, Macromol. Rapid Commun., 2002, 23, 761–765.
49 O. Hyunjoon and P. F. Green, Nat. Mater., 2009, 8, 139–143.
29 M. S. Dresselhaus, A. Jorio, A. G. Souza Filho and R. Saito,
Philos. Trans. R. Soc., A, 2010, 368, 5355–5377.
30 M. Fang, K. Wang, H. Lu, Y. Yang and S. Nutt, J. Mater. 50 J.-M. Malho, P. Laaksonen, A. Walther, O. Ikkala and
Chem., 2009, 19, 7098–7105. M. B. Linder, Biomacromolecules, 2012, 13, 1093–1099.
31 V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, 51 R. Rahman, J. T. Foster and A. Haque, J. Phys. Chem. A, 2013,
N. Kim, K. C. Kemp, P. Hobza, R. Zboril and K. S. Kim,
117, 5344–5353.
Chem. Rev., 2012, 112, 6156–6214.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 2550–2561 | 2561