A novel poly(p-styrenesulfonic acid) grafted carbon nanotube/graphene oxide architecture with enhanced catalytic performance for the synthesis of benzoate esters and fatty acid alkyl esters

Article abstract of DOI:10.1039/c5ra14813h

Considering the issue of low yield in the synthesis of benzoate esters and fatty acid alkyl esters, designing a high catalytic activity composite catalyst is very significant and attractive. In this study, the rational design strategy was used to develop a novel poly(p-styrenesulfonate acid, namely PSSF) grafted multi-walled carbon nanotube composite with graphene oxide nanomaterial (PSSF-mCNTs-GO) using a simple two-step method. FT-IR and Raman spectroscopy, XRD, SEM, TEM, and NH₃-TPD were used to characterize the inorganic-organic hybrid material. In particular, the addition of GO remarkably enhanced its catalytic performance in the production of fatty acid alkyl esters (92.16%) and benzoate esters (90.27%), in which the conversion was more than doubled as a result of its strong π-π interaction with the substrate. In addition, PSSF-mCNTs-GO can be separated from the substrate conveniently and still maintained a relatively high catalytic activity even after 6 times recycling, which indicates its rather good reusability. This novel catalyst is promising in the synthesis of biodiesel and benzoate esters.

Full text of DOI:10.1039/c5ra14813h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
A novel poly(p-styrenesulfonic acid) grafted carbon
nanotube/graphene oxide architecture with
enhanced catalytic performance for the synthesis
of benzoate esters and fatty acid alkyl esters†
Cite this: RSC Adv., 2015, 5, 90757
a
a
a
b
a
c
Gang Bian, Pingping Jiang,* Weijie Zhang, Kelei Jiang, Ling Hu, Zhang Jian,
a
a
Yirui Shen and Pingbo Zhang
Considering the issue of low yield in the synthesis of benzoate esters and fatty acid alkyl esters, designing
a high catalytic activity composite catalyst is very significant and attractive. In this study, the rational design
strategy was used to develop a novel poly(p-styrenesulfonate acid, namely PSSF) grafted multi-walled
carbon nanotube composite with graphene oxide nanomaterial (PSSF-mCNTs-GO) using a simple two-
3
step method. FT-IR and Raman spectroscopy, XRD, SEM, TEM, and NH -TPD were used to characterize
the inorganic–organic hybrid material. In particular, the addition of GO remarkably enhanced its catalytic
performance in the production of fatty acid alkyl esters (92.16%) and benzoate esters (90.27%), in which
the conversion was more than doubled as a result of its strong p–p interaction with the substrate. In
addition, PSSF-mCNTs-GO can be separated from the substrate conveniently and still maintained
a relatively high catalytic activity even after 6 times recycling, which indicates its rather good reusability.
This novel catalyst is promising in the synthesis of biodiesel and benzoate esters.
Received 26th July 2015
Accepted 30th September 2015
DOI: 10.1039/c5ra14813h
www.rsc.org/advances
systems, low emission of chemical waste liquid and low toxicity.
Numerous chemistry researchers have focused their attention
Among the countless esterication products, fatty acid alkyl on heteropolyacids, sulfonated ion exchange resins, organic–
1
Introduction
9
10
1
1
12
esters (FAAE) and benzoate esters are very important. Fatty acid inorganic hybrid solid acids, solid superacids and SO H
3
1
,2
13
alkyl esters, called biodiesel, are well-known as clean green functionalized ionic liquids.
14
energy and renewable liquid fuel compared to mineral fuel.
In 2014, Mario et al. successfully applied Sn(IV)-based
Moreover, benzoate esters can be applied in the synthesis of organometallics to the production of fatty acid alkyl esters.
a number of drug substances and our study is mainly focused When the reaction temperature rose to 160 C, the conversion of
on these two aspects.
fatty acid with MeOH reached 94% in 3 h. In addition, with
In general, the esterication reaction is catalyzed by liquid respect to benzoate esters, Wu et al. used SO
Bronsted acids such as concentrated H SO , H PO , HF, HCl, ceramic as acid catalysts for the esterication of benzoic acid.
3–5
ꢀ
15
2ꢁ
4
3 2
/Ti AlC
2
4
3
4
6
and p-toluenesulfonic acid. Although these liquid catalysts Aer 34 h, conversion increased to 80.4% while the catalyst
16
exert rather good activity and have a moderate price, the spent amount was more than 10 wt%. Ladero's group also reported
acid and waste water produced in esterication reactions causes that an industrial free Candida antarctica lipase B was rather
severe environmental pollution. Furthermore, most liquid efficient in the esterication of benzoic acid with glycerol.
7
Bronsted acids belong to homogeneous catalysts, which are Conversion was obtained higher than 80% when the reaction
difficult to separate from substrates. Recently, solid acid was completed.
8
heterogeneous catalysts have been drawing researchers'
Graphene oxide (GO), since rst prepared by the classic
17
attention because of their high selectivity, easy separation from Hummers method, has gained considerable interest and
tremendous attention due to its unique physical, chemical and
18
electrical properties. During its preparation, GO undergoes
exhaustive oxidation, thus leading to rich oxygen-containing
a
The Key Laboratory of Food Colloids and Biotechnology, School of Chemical and
Material Engineering, Jiangnan University, Wuxi 214122, P. R. China. E-mail:
ppjiang@jiangnan.edu.cn
3
functionalities such as –OH, –SO H, –COOH, and epoxide
b
groups. These abundant decient sites endow GO with
Wuxi WeiFu Lida Catalytic Converter Co., Ltd, Wuxi 214177, P. R. China
19
c
School of Chemistry and Environmental Science, Lanzhou City University, Lanzhou moderate acidic and oxidizing properties. In addition, due to
7
†
1
30000, P. R. China
its high surface area, GO is always used as a support to immo-
bilize active species in catalytic reactions. As an efficient carbon-
Electronic supplementary information (ESI) available. See DOI:
0.1039/c5ra14813h
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 90757–90765 | 90757

View Article Online
RSC Advances
Paper
catalyst, GO has been employed in many organic acid-catalytic the benzene ring on catalytic activity over PSSF-mCNTs-GO is
20
reactions such as the aza-Michael addition, Friedel–Cras discussed. The synthetic route is displayed in Scheme 1. Aer
21
22
23
addition, hydration and oxidation. Recently, Zhu et al.
examining a series of different alcohols and acids, PSSF-
successfully employed GO to catalyze the esterication of levu- mCNTs-GO was found to have higher catalytic activity on
linic acid with different alcohols, which nally generated alkyl aliphatic and aromatic acids with alcohols. When GO was
levulinates. The yield of methyl levulinate reached as high as introduced, due to the p–p stacking interaction between GO
24
9
9.4% in only 6 h. Simultaneously, Sanny Verma et al. pointed and the benzene ring, the conversion of benzoate esters was
out that GO can act as a catalyst carrier in the synthesis of esters. more than doubled. Furthermore, PSSF-mCNTs-GO showed
According to their report, a magnetic nanoparticle composite good reusability, and there was no signicant decrease in
2
+
was prepared by loading Pd on the surface of GO with the catalytic activity even aer recycling six times.
addition of Fe/FeO. 90% of phenylcarbinol can be catalyzed to
methyl benzoate in methanol aer 6 h. In summary, all the
2
Experimental
research mentioned above indicates that GO is a promising
25
prospect in esterication reactions. However, Maryam et al.
2.1 Synthesis of graphene oxide
presented that, in spite of the favourable catalytic activity of GO
as solid acid catalyst in esterication, its reusability is unsatis-
factory. The yield decreased to 14% aer recycling twice. The
active sites disappear, which indicates the poor stability of GO.
Furthermore, some researchers have paid attention to the
Graphene oxide was synthesized from ake graphite (44 mm
average particle diameter, Qingdao Jinrilai Graphite Co. Ltd.,
Qingdao, China) via the Hummers method described by Jiang
29
Du et al. with some modications. Briey, aky graphite
(
1.25 g) and NaNO
necked ask. Then, concentrated H
added with stirring in an ice-water bath. KMnO
3
(0.94 g) were added to a 500 mL three-
SO (93.75 mL) was slowly
(5.63 g) was
26
effect of the p–p interaction caused by GO in reactions. Li et al.
demonstrated that the p–p stacking interaction between dopa-
2
4
4
mine and the graphene surface may accelerate electron transfer.
gradually added over 1 h under vigorous stirring and the
mixture was stirred for another 2 h. Aer keeping vigorous
stirring at room temperature for 4 days, 5 wt% H SO aqueous
27
Jin's group prepared a reduced graphene oxide/TiO
2
photo-
catalyst and proved that the p–p interaction can combine the
2
4
28
substrate with GO more tightly. Yin et al. also found that an
iron phthalocyanine can be uniformly dispersed and anchored
on graphene's surface with the assistance of p–p stacking. Such
strong interactions between substrates and graphene oxide may
be benecial for the synthesis of benzoate esters.
solution (175 mL) was used to dilute the mixture over 1 h, and
ꢀ
then stirring was continued for 2 h at 98 C. Aer the system
ꢀ
cooled to 60 C, 30 wt% H
2
O
2
(7.5 mL) was added, and the
mixture was stirred for another 2 h. When the reaction was
completed, the mixture obtained was centrifuged and washed
According to the characteristics of GO and problems
mentioned above, a novel catalyst, poly(p-styrenesulfonate acid)
graed multi-walled carbon nanotubes composited with gra-
phene oxide (PSSF-mCNTs-GO), was prepared using a two-step
method. To the best of our knowledge, this is the rst time
that PSSF-mCNTs has been combined with GO and applied as
a catalyst in esterication reactions for the synthesis of fatty
acid alkyl esters and benzoate esters. The inuence of chain
branching degree, chain length and the presence or absence of
2 4 2 2
with a 3 wt% H SO /0.5 wt% H O aqueous solution (500 mL)
for 15 times. The solids at the bottom were washed with a 3 wt%
HCl aqueous (500 mL) and one additional wash with deionized
H O solution (500 mL). Aerwards, the solid was dispersed
2
again in H O (500 mL) and ultrasonicated for 30 min. The water
2
solution was treated with a weak basic ion-exchange resin to
remove the remaining HCl acid and was centrifuged once again.
Finally, the mixture (0.8 g) was freeze dried for use in the
following steps.
30
2.2 Preparation of PSSF-mCNTs
5.0 g of sodium p-styrenesulfonate was dissolved in deionized
water (200 mL), and 100 mg of mCNTs were immediately added.
The mixture was then kept under vigorous stirring for 3 h aer 1
h of high-power sonication. Aerwards, the mixture was trans-
ferred into a three-neck ask (1 L), and 1.0 g of KPS was added
as a radical initiator. To remove oxygen in the reaction, the
mixture was bubbled with dry nitrogen for 1 h. Aer this the
three-neck ask was placed in an oil bath, and the solution was
ꢀ
ꢀ
le stirred uniformly at 80 C for 3 h and 90 C for 1 h
subsequently.
A mixing broid membrane (0.45 mm) was used to lter the
reaction solution when the heating process was over, and the

ltered black paste was washed with deionized water 5 times
using a supercentrifuge to remove residual sodium p-styr-
enesulfonate. Aer centrifugation, the black paste was reuxed
ꢀ
Scheme 1 Synthetic route of PSSF-mCNTs-GO.
with concentrated hydrochloric acid at 120 C for 12 h, and the
90758 | RSC Adv., 2015, 5, 90757–90765
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
reaction product was ltered and washed again with deionized alcohols (N-butyl alcohol, sec-butyl alcohol, tert-butyl alcohol, 1-
water 4 times using a supercentrifuge to remove residual octyl alcohol, and benzyl alcohol) were carried out in a two-neck
concentrated hydrochloric acid and sodium.
.3 Synthesis of the PSSF-mCNTs-GO nanocomposite
ask (25 mL) with a water separator under magnetic stirring.
Cyclohexane was used as the water-carrying agent in the ester-
ication system. Typically, acids and alcohols (1 : 4 mol ratio)
with PSSF-mCNTs-GO catalyst (2 wt%) were added to the reactor
and the solution was heated to the desired temperature for
a specied time. The obtained product samples (400 mL drawn
from the solution at 0 h, 0.5 h, 1 h, 2 h, 4 h, 6 h and 10 h) were
separated by centrifugation and analyzed by GC. Aer each run,
the ‘used’ catalyst was recycled from the reaction mixture by
centrifugation and reused in sequential runs aer washing with
ethanol twice. Moreover, the corrected area normalization
method was used to quantify the products. Reaction tempera-
ture, time, catalyst species, and substrate species were changed
to investigate the dependence of the main variables for
comparison using the same conditions. Blank experiments
2
To form a homogeneous suspension, the as-prepared PSSF-
mCNTs (100 mg) and GO (50 mg) were dispersed in ethanol
(100 mL) in a 250 mL three-neck ask, followed by 2 h sonica-
tion. Subsequently, the homogeneous suspension was ther-
ꢀ
mally reuxed at 90 C for 2 h. Aer the reaction was completed,
the solution was ltered and washed with deionized water 3
times using a supercentrifuge (9000 rpm) to remove ungraed
GO. Finally, the black paste was dried in a vacuum oven at 60 C
overnight to obtain the PSSF-mCNTs-GO nanocomposite as
a black solid.
ꢀ
2.4 Catalyst characterization
(without catalyst) were also conducted as references.
Powder X-ray diffraction (XRD) patterns were obtained on
a BRUKER D8-Focus Bragg-Brentano X-ray Powder Diffractom-
3
Results and discussion
˚
eter equipped with a Cu sealed tube (l ¼ 1.54178 A) at 40 kV and
ꢀ ꢀ
0 mA. The scanning region 2q was from 6 to 80 with a scan- 3.1 Catalyst characterization
4
ꢀ
ꢁ1
ning rate of 4 min . Scanning electron microscope (SEM)
investigations were carried out with a S-4800 at an accelerating
voltage of 3 kV. Bright eld transmission electron microscopy
The synthesized GO was characterized using XRD, Raman
spectroscopy, SEM and TEM, which are shown in the ESI.† A
ꢀ
sharp and strong peak at 10.0 in the XRD patterns (Fig. S1A†)
(TEM) images were acquired using a JEOL JEM 2100 electron
and D band and G band as shown in the Raman spectra
microscope under an accelerating voltage of 200 kV. High-
resolution transmission electron microscopy (HR-TEM)
images were acquired using the same equipment at 200 kV.
Samples for TEM (HR-TEM) were prepared by dispersing the
powder sample in ethanol with ultrasonic cleaning over 1 h, and
then the mixed solution was dropped onto a carbon coated
copper grid. Raman spectra were obtained on a confocal
microscopic Raman spectrometer (Renishow In-Via, USA) with
31
(
Fig. S1B†) all provide evidence for the formation of GO. GO
presented a crumpled and layered structure, as observed from
its SEM (Fig. S2A†) image, and TEM (Fig. S2†) images revealed
that the prepared GO was well-shaped. All these characters
displayed in the spectra indicate that GO would be an excellent
catalyst carrier.
The successful graing of PSSF onto mCNTs was conrmed
by FT-IR and TEM. As shown in Fig. 1A, aer washing the
catalyst several times, the intensity of the main characteristic
ꢁ
1
5
32 nm laser light irradiation from 400 to 3000 cm
a duration time of 10 s. The samples were pressed into slices
before analysis. NH temperature-programmed desorption
NH -TPD) measurements were conducted on an AUTO 2920
at
peaks barley decreased. –SO
3
H exhibited a relatively weak band
3
ꢁ1
at 1191 cm , which is associated with S]O stretching vibra-
(
3
32
ꢁ1
ꢁ1
tions. The characteristic peaks at 1069 cm and 620 cm
instrument (Micromeritics, USA), and a thermal conductivity
detector was used for the continuous monitoring of desorbed
ammonia. Prior to TPD analysis, samples were pretreated at 350
ꢁ
both belong to –SO3 . In addition, due to –SO –H, –O–H also
3
ꢁ1
exhibited a strong broad band at 3435 cm . All these charac-
teristic peaks in the FT-IR spectrum provide strong evidence of
a tight gra. To demonstrate the improved dispersity of PSSF-
mCNTs compared with pristine mCNTs, a dispersity experi-
ment in methanol was carried out (when the dispersity of
catalyst improves, the active sites are dispersed better, which
ꢀ
C for 1 h in a ow of ultra-pure helium gas, and the sample was
ꢀ
cooled to 50 C. The pretreated sample was then saturated with
8

% anhydrous ammonia gas for 45 min and subsequently
ushed with helium gas for 1 h to remove physisorbed
ammonia. The heating rate of the TPD measurements, which
33
leads to a promotion of catalytic performance ). Pristine
ꢀ
ꢀ
ꢀ
ꢁ1
ranged from 50 C to 800 C, was 10 C min . Thermogravi-
metric analysis (TGA) was carried out using a Mettler TGA/SDTA
mCNTs could not be well dispersed in methanol, and it was
totally precipitated in a few minutes (Fig. 1b). Aer graing with
PSSF, PSSF-mCNTs could be well dispersed in methanol, which
formed a black solution uniformly, due to the existence of
ꢀ ꢀ
851 E analyzer in the temperature range of 50 C to 700 C at
ꢀ
ꢁ1
a heating rate of 20 C min . The concentration of C, O, S and
H elements in the composite catalysts was experimentally
measured by ICP emission spectrometry (Inductively Coupled
Plasma) (Shimadzu ICP-7500).
3
copious –SO H groups (Fig. 1b). This solution was rather stable,
and just a little deposition was detected even aer one month.
Transmission electron microscopy (TEM) is routinely
employed as an auxiliary means of characterization to realize
the state of dispersity. As can be observed in Fig. 1C, due to van
2.5 Catalytic tests
34
Esterication catalytic tests for different acids (acetic acid, der Waals attractions between the carbon nanotubes, the
lauric acid, stearic acid, and benzoic acid) with different pristine mCNTs tangled together and appeared relatively
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 90757–90765 | 90759

View Article Online
RSC Advances
Paper
Fig. 1 (A) FT-IR spectra of PSSF-mCNTs (before and after washing); (B) dispersity experiment of PSSF-mCNTs and mCNTs in methanol; TEM
images of (C) pristine mCNTs and (D) PSSF-mCNTs.
smoother. When PSSF was graed, PSSF-mCNTs (Fig. 1D) was totally precipitated 3 months later. Simultaneously, PSSF-
appeared to have better dispersity in methanol than the pristine mCNTs (Fig. 3C) deposited the most in 3 months. However,
mCNTs because of the insertion of sulfonic groups. The organic the solution containing PSSF-mCNTs-GO (Fig. 3C) was quite
groups can afford adequate electrostatic repulsions to eliminate stable, which formed a homogeneous black solution, and not
35
van der Waals attractions. Furthermore, around the surface of a little deposition was detected even aer 3 months. In general,
PSSF-mCNTs (Fig. 1D), many small black particles, which are
poly-p-styrenesulfonate groups, can be distinguished. This gave
corroborative evidence to prove the existence of –SO H groups,
3
which made the mCNTs much coarser.
FT-IR, SEM and TEM were used to characterize the success-
fully synthesized PSSF-mCNTs-GO. As shown in Fig. 2,
comparing PSSF-mCNTs-GO with PSSF-mCNTs, the broad
stretching vibration peak belonging to –OH between 3000 and
ꢁ
1
3
–
700 cm was enhanced aer the doping of GO. Moreover, the
ꢁ1
C–O stretching vibration peak at 1285 cm , which is due to
the carboxylic groups in GO, was enhanced, and this gave strong
36
evidence for the doping of GO. Furthermore, the small sharp
ꢁ
1
peak at 1345 cm , which is associated with the –O]S]O
ꢁ
stretching vibrations, and the weak band of the –SO
symmetric stretching vibration at 1035 cm
3
ꢁ
1
were both
strengthened compared with the non-graed GO. The infor-
mation shown in the FT-IR spectrum indicated a successful
gra of PSSF and a tight combination between GO and PSSF-
mCNTs.
Dispersity experiments (Fig. 3C) were conducted again to
conrm the excellent dispersity of PSSF-mCNTs-GO in meth-
Fig. 2 FTIR spectra of PSSF-mCNTs-GO (a), mCNTs (b), PSSF-mCNTs
anol, using SEM and TEM analysis. Pristine mCNTs (Fig. 3C) (c) and GO (d).
90760 | RSC Adv., 2015, 5, 90757–90765
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
3
7,38
with literature reports.
As the calcination temperature
ꢀ
increased to 250 C, the S]O bonds decomposed and total
acidity rapidly decreased. In addition, it was obvious that the
peak position of PSSF-mCNTs-GO moved to the high tempera-
ture direction and a stronger acidity was revealed aer the
addition of GO, which contributed to the abundant acidic
3
functionalities on the surface of GO such as –SO H and –COOH
and the tight van der Waals attractions between the mCNTs and
GO. GO in the composite catalyst not only provided a suitable
substrate, but also improved the acid strength, which led to the
enhancement of catalytic performance. These results indicated
ꢀ
that the optimum heat treatment temperature was 250 C and
PSSF-mCNTs-GO had more acid sites, which is very benecial to
esterication reactions.
To further investigate the thermostability of the composite
catalysts, thermogravimetric analysis (TGA) was performed on
Fig. 3 (A) SEM image of PSSF-mCNTs-GO (fresh); (B) TEM image of PSSF-mCNTs and PSSF-mCNTs-GO, which are displayed in the
PSSF-mCNTs-GO (fresh); (C) images from the dispersity experiment of
PSSF-mCNTs-GO, PSSF-mCNTs and mCNTs in methanol (3 months).
ESI.† As shown in Fig. S4,† no detectable weight loss step for
ꢀ
PSSF-mCNTs appeared before 180 C, and the approximate 5%
ꢀ
weight loss before 180 C of PSSF-mCNTs-GO is attributed to
physically adsorbed water. Both the PSSF-mCNTs and PSSF-
mCNTs-GO had a slight weight loss (2.61% and 9.78%,
the outstanding dispersity of PSSF-mCNTs-GO provided
a foundation for efficient catalysis in esterication. A similar
situation was observed in the SEM and TEM images. No obvious
aggregation was found in the SEM image (Fig. 3A) between the
nanoparticles, and the carbon nanotubes were uniformly laying
on the graphene oxide sheets, which matched well with the
results in the dispersity experiment. Carbon nanotubes were
interspersed randomly between sheets of graphene oxide, thus
preventing further aggregation of the graphene oxide
nanosheets.
ꢀ
ꢀ
respectively) in the range of 180 C to 270 C, which was due to
the –SO H graed onto the composite catalysts and abundant
3
oxygen-containing functional groups, such as –OH and –COOH,
on the surface of GO. In general, PSSF-mCNTs and PSSF-
mCNTs-GO presented considerable stability below the temper-
ꢀ
ꢀ
ature of 180 C. As the reaction temperature was set at 120 C in
the catalytic test, it was predictable that the PSSF-mCNTs and
PSSF-mCNTs-GO would exert a comparative catalytic stability in
the esterication reaction.
NH -TPD proles were obtained for PSSF-mCNTs, PSSF-
3
mCNTs-GO and GO at various sintering temperatures (Fig. 4).
It can be observed that the acid sites of PSSF-mCNTs and PSSF- 3.2 Catalytic activity
ꢀ
ꢀ
mCNTs-GO were both located in the range of 100 C to 400 C,
which can be considered as the presence of acid sites, such as
–
Reaction time and temperature have a signicant impact on
catalytic activity, therefore before esterication, the optimum
reaction time and temperature were investigated. As shown in
Fig. 5A, n-butanol and stearic acid were used as the probe
substrates to conrm the optimum reaction time. Before time
reached 10 h, conversion increased steadily over time due to the
sufficient contact between reactants and catalyst. However, the
conversion seldom increased aer 10 h because of the equilib-
3
SO H, with moderate strength. These results are in agreement
39
rium in esterication. Thus, we set the optimum reaction time
as 10 h for economical reasons. Comparing the inuence of
time, temperature also had a major effect on esterication
conversion (Fig. 5B). Before the temperature increased to
ꢀ
1
25 C, conversion increased along with the improvement of
temperature, due to the increase of thermal energy. However,
ꢀ
when the temperature was higher than 130 C, conversion
began to decrease because of the volatilization of acids and
alcohol. Considering the low boiling point of N-butyl alcohol,
sec-butyl alcohol and tertiary butyl alcohol, we nally set the
ꢀ
optimum reaction temperature at 120 C for esterication.
Before the as-prepared catalyst was used in esterication, the
catalytic activity of PSSF-mCNTs-GO was compared with PSSF-
2 4
mCNTs, concentrated H SO , and a commercial catalyst
Fig. 4 Temperature-programmed desorption profiles of ammonia on
PSSF-mCNTs, PSSF-mCNTs-GO and GO at various temperatures.
(sulfonated GO), the blank experiment was also determined
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 90757–90765 | 90761

View Article Online
RSC Advances
Paper
a
b
a
Fig. 5 Effects of reaction time (A) and reaction temperature (B) on esterification. Reaction conditions: temperature (120 C), n-butanol : stearic
b
acid = 4 : 1, and catalyst amount (2 wt%). Reaction conditions: n-butanol : stearic acid = 4 : 1, catalyst amount (2 wt%), and reaction time (10 h).
subsequently. As shown in Fig. 6, lauric acid and isooctyl comparison. All the catalytic data displayed in Table S1† indi-
alcohol were taken as reaction substrates. The conversion was cate a promotion of conversion with the addition of GO. To
nearly 65.13% aer 2 h when PSSF-mCNTs-GO was used as the further explore the trends hidden in Table S1,† three different
catalyst, which is much higher than PSSF-mCNTs (41.25%), the aspects of catalytic performance were investigated.
commercial catalyst (40.56%) and concentrated
H
2
SO
4
First, the catalytic performance of PSSF-mCNTs-GO in
(37.62%). Reacted over 10 h, the PSSF-mCNTs-GO exhibited esterication with different chain branching degree alcohols
outstanding conversion (85.25%), compared with the relatively was investigated. It can be observed from Fig. 7(A–C), when GO
low conversion of PSSF-mCNTs (65.27%), the commercial was combined with PSSF-mCNTs, the conversion of three
catalyst (63.28%) and concentrated H SO (58.43%). This different alcohols with stearic acid were all increased. Inter-
2
4
contrast experiment proved that PSSF-mCNTs-GO is superior to estingly, tertiary butyl alcohol had a better conversion (84.10%),
similar esterication catalysts and further conrmed that the compared with N-butyl alcohol (83.15%) and sec-butyl alcohol
prepared catalyst can act as a potential acid catalyst in indus- (75.17%). However, the catalyst without GO (PSSF-mCNTs) also
trial applications.
exerted a rather good conversion (80.16%) to tertiary butyl
To test the catalytic activity of PSSF-mCNTs-GO, a series of alcohol, which was close to PSSF-mCNTs-GO. On the contrary,
esterication reactions were performed. The results of the in the esterication reaction of N-butyl alcohol with stearic acid,
esterication catalytic tests for different acids (acetic acid, lau- the addition of GO increased the conversion from 55.14% to
ric acid, stearic acid, and benzoic acid) with different alcohols 83.15%, which increased by almost 30%. The addition of GO
(N-butyl alcohol, sec-butyl alcohol, tert-butyl alcohol, 1-octyl can remarkably improve the conversion of a primary alcohol in
alcohol, and benzyl alcohol) are provided in ESI (Table S1†) as esterication. The relatively high conversion of N-butyl alcohol
40
was due to less steric hindrance, which caused –OH to be
exposed, and was more benecial for the carboxyl group to
attack. To conrm this, we carried out reactions among the
same three alcohols with lauric acid subsequently. As shown in
Fig. 7(D–F), it was found that the best catalytic activity was ob-
tained in the esterication of N-butyl alcohol with lauric acid.
The conversion was as high as 81.13%, which was higher than
sec-butyl alcohol (70.38%) and tertiary butyl alcohol (80.16%).
Likewise, the conversion of N-butyl alcohol also greatly
improved when GO was added. Therefore, the abovementioned
results conrmed the nding that PSSF-mCNTs-GO can catalyze
primary alcohols better than secondary alcohols and tertiary
alcohols, which means that PSSF-mCNTs-GO has a higher
catalytic activity for lower chain branching degree alcohols in
esterication.
To study the effect of the chain branching degree of alcohols,
the catalytic activity of PSSF-mCNTs-GO for a series of alcohols
and acids with different chain lengths was examined. As dis-
played in Fig. 8(A and B), compared with the short-chain n-butyl
a
Fig. 6 Comparison of catalytic activity : blank experiment (black);
concentrated H SO
(pink); commercial catalyst (red); PSSF-mCNTs alcohol, PSSF-mCNTs-GO had a higher conversion for the long
blue); PSSF-mCNTs-GO (green). Reaction conditions: temperature chain alcohol (1-octanol), which was 83.15% and 95.63%,
2
4
a
(
(
ꢀ
120 C), isooctyl alcohol : lauric acid ¼ 4 : 1, catalyst amount (2 wt%),
respectively. Furthermore, among the acids with different chain
and reaction time (10 h).
90762 | RSC Adv., 2015, 5, 90757–90765
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
a
Fig. 7 Conversion of stearic acid with different alcohols : (A) N-butyl alcohol; (B) sec-butyl alcohol; (C) tertiary butyl alcohol. Conversion of
a
a
lauric acid with different alcohols : (D) N-butyl alcohol; (E) sec-butyl alcohol; (F) tertiary butyl alcohol. Reaction conditions: temperature
(
ꢀ
120 C), acid : alcohol (mol ratio 1 : 4), catalyst amount (2 wt%), and reaction time (10 h).
lengths (acetic acid, lauric acid and stearic acid), PSSF-mCNTs- this is the contribution of abundant carboxyl and hydroxyl
GO had the best catalytic activity for stearic acid, which had the groups on the surface of GO, which was more benecial for –OH
longest chain length and conversion was as high as 92.16%. groups exposed in aliphatic alcohols and acids to attack, thus
Especially intriguing, in the esterication of phenylcarbinol leading to a promotion in conversion. The experimental data
with lauric acid and stearic acid (Fig. 8(D and E)), PSSF-mCNTs- displayed above all ideally demonstrate that the PSSF-mCNTs-
GO gave almost twice the conversion of PSSF-mCNTs, which is GO can catalyze fatty acids and alcohols in esterication very
strong evidence that the addition of GO could signicantly well. Moreover, it also shows potential application in prepara-
promote catalytic activity to aliphatic acids. The main reason for tion of biodiesel.
a
a
Fig. 8 Conversion of stearic acid with different alcohols : (A) N-butyl alcohol; (B) 1-octanol. Conversion of phenylcarbinol with different acids :
(
a
ꢀ
C) acetic acid; (D) lauric acid; (E) stearic acid. Reaction conditions: temperature (120 C), acid : alcohol (mol ratio 1 : 4), catalyst amount (2 wt%),
and reaction time (10 h).
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 90757–90765 | 90763

View Article Online
RSC Advances
Paper
Furthermore, in comparison with fatty acids, the acid
substrate was extended to aromatic acids to test the catalytic
activity of PSSF-mCNTs-GO. As can be observed in Fig. 9(A and
B), although the conversion of n-butyl alcohol and benzyl
alcohol was almost the same in 10 h, the conversion of benzyl
alcohol was higher (40.26–82.93%) than n-butyl alcohol (53.60–
8
1.13%) when GO existed. The same trend was shown again in
Fig. 9(C and D), wherein not only the conversion of benzoic acid
90.27%) was higher than acetic acid (70.15%), but also the
(
conversion enhanced from 40.68% to 90.27% with the addition
of GO, which was more than doubled, therefore catalytic activity
was signicantly improved. Such strong catalytic activity for
2
benzoate esters is attributed to the conjugation between the sp
hybrid C atom of the graphene oxide surface and delocalized pi
4
1,42
bond in the benzene ring.
The p–p stacking interaction can
43
accelerate the electron transfer between catalyst and reactants,
thus leading to a substantial promotion of catalytic activity.
a
a
Fig. 10 The cycling experiment of catalytic activity . Reaction
ꢀ
conditions: temperature (120 C), 1-octyl alcohol : stearic acid ¼ 4 : 1,
The reusability of PSSF-mCNTs-GO was also determined
PSSF-mCNTs-GO amount (2 wt%), and reaction time (10 h).
(Fig. 10). The ‘used’ catalyst was recycled by centrifugation and
then reused. Even aer 6 times recycling, only a slightly
decrease of conversion was detected, and catalytic activity was
almost consistent. Furthermore, to demonstrate the high
structural stability of PSSF-mCNTs-GO in catalytic reactions and
and B),† wherein the recycled catalyst still maintained a similar
morphology aer recycling 6 times. Moreover, as shown in
Table S2,† there was no signicant change of S element content
between the fresh and recycled catalyst, which indicated that
the as-prepared catalyst was quite stable in esterication
reactions.
3
illustrate the change of –SO H content between the fresh and
recycled catalyst, SEM, TEM and ICP analysis were carried out to
conrm the structure of PSSF-mCNTs-GO and characterize the S
element content in the composite catalyst. No obvious change
was observed through a comparison of Fig. 3(A and B) and S5(A
a
Fig. 9 Conversion of lauric acid with different alcohols : (A) N-butyl alcohol; (B) phenylcarbinol. Conversion of N-butyl alcohol with different
a
a
ꢀ
acids : (C) acetic acid; (D) benzoic acid. Reaction conditions: temperature (120 C), acid : alcohol (mol ratio 1 : 4), catalyst amount (2 wt%), and
reaction time (10 h).
90764 | RSC Adv., 2015, 5, 90757–90765
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
1
4 M. R. Meneghetti and S. M. P. Meneghetti, Catal. Sci.
Technol., 2015, 5(2), 765–771.
4
Conclusion
Compared to conventional solid acid catalysts, PSSF-mCNTs-GO, 15 M. Wu, J. Guo, Y. Li, et al., Ceram. Int., 2013, 39(8), 9731–9736.
which was prepared using a two-step method, showed good 16 J. J. Tamayo, M. Ladero, V. E. Santos, et al., Process Biochem.,
catalytic stability and was proven to be a highly active acid catalyst
towards the esterication of fatty and aromatic acids with alco- 17 R. E. Offeman and W. S. Hummers, J. Am. Chem. Soc., 1958,
hols. Aer the addition of GO with the appropriate amount in the
80, 1339.
composite catalyst, catalytic activity was signicantly increased. 18 Y. Zhu, S. Murali, W. Cai, et al., Adv. Mater., 2010, 22(35),
In particular, the conversion of an aliphatic (stearic acid) and
3906–3924.
aromatic acid (benzoic acid) in esterication reactions was more 19 D. R. Dreyer, H.-P. Jia and C. W. Bielawski, Angew. Chem., Int.
than doubled, and they were as high as 92.16% and 90.27%,
Ed., 2010, 122, 6965–6968.
respectively. In addition, the p–p stacking interaction between 20 S. Verma, H. P. Mungse, N. Kumar, et al., Chem. Commun.,
the benzene ring and graphene oxide was more benecial for the
2011, 47(47), 12673–12675.
synthesis of benzoate esters. Furthermore, no signicant 21 A. V. Kumar and K. R. Rao, Tetrahedron Lett., 2011, 52(40),
decrease in catalytic activity was found even aer recycling six
5188–5191.
times. It is worth mentioning that such a robust solid acid cata- 22 D. R. Dreyer, H. P. Jia and C. W. Bielawski, Angew. Chem.,
lyst holds great potential in the synthesis of biodiesel and
2010, 122(38), 6965–6968.
aromatic esters. However, more studies are still needed to explore 23 S. Zhu, C. Chen, Y. Xue, et al., ChemCatChem, 2014, 6(11),
the mechanism of such high conversion and a deeper under-
3080–3083.
standing of graphene-supported solid acids should be acquired. 24 S. Verma, D. Verma, A. K. Sinha, et al., Appl. Catal., A, 2015,
89, 17–23.
2012, 47(2), 243–250.
4
2
5 M. Mirza-Aghayan, R. Boukherroub and M. Rahimifard,
Turk. J. Chem., 2014, 38, 859–864.
Acknowledgements
The authors gratefully acknowledge the nancial support from 26 Y. Wang, Y. Li, L. Tang, et al., Electrochem. Commun., 2009,
the National “Twelh Five-Year” Plan for Science & Technology
11(4), 889–892.
2012BAD32B03), the Fundamental Research Funds for the 27 K. L. V. Joseph, J. Lim, A. Anthonysamy, et al., J. Mater. Chem.
(
Central Universities(JUSRP51507) and the Postgraduate Inno-
vation Project of Jiangsu Province (KYLX15-1156).
A, 2015, 3(1), 232–239.
28 H. Yin, C. Zhang, F. Liu, et al., Adv. Funct. Mater., 2014,
24(20), 2930–2937.
2
3
9 J. Du, X. Lai, N. Yang, et al., ACS Nano, 2010, 5(1), 590–596.
0 X. H. Zhang, Q. Q. Tang, D. Yang, et al., Mater. Chem. Phys.,
2011, 126(1), 310–313.
Notes and references
1
2
F. Ma and M. A. Hanna, Bioresour. Technol., 1999, 70(1), 1–15.
S. K. Hoekman, A. Broch, C. Robbins, et al., Renewable 31 Y. Wang, I. Djerdj, B. Smarsly and M. Antonietti, Chem.
Sustainable Energy Rev., 2012, 16(1), 143–169.
Mater., 2009, 21, 3202–3209.
X. Li, W. Eli and G. Li, Catal. Commun., 2008, 9(13), 2264–2268. 32 F. Liu, J. Sun, L. Zhu, et al., J. Mater. Chem., 2012, 22(12),
3
4
S. L. Barbosa, G. R. Hurtado, S. I. Klein, et al., Appl. Catal., A,
008, 338(1), 9–13.
5495–5502.
2
33 G. Du, S. Lim, M. Pinault, et al., J. Catal., 2008, 253(1), 74–90.
5
6
H. Shi, W. Zhu, H. Li, et al., Microwave-accelerated 34 C. Li and T. W. Chou, Compos. Sci. Technol., 2003, 63(11),
esterication of salicylic acid using Br ¨o nsted acidic ionic
liquids as catalysts, Catal. Commun., 2010, 11(7), 588–591.
(a) A. Sakakura, Y. Koshikari and K. Ishihara, Tetrahedron
Lett., 2008, 49, 5017–5020.
1517–1524.
35 S. L. Price and A. J. Stone, J. Chem. Phys., 1987, 86(5), 2859–
2868.
36 S. Stankovich, D. A. Dikin, R. D. Piner, et al., Carbon, 2007,
45(7), 1558–1565.
6
b K. Ishihara, Tetrahedron, 2009, 65, 1085–1109;
7
E. S. Stoyanov, S. P. Hoffmann, M. Juhasz, et al., J. Am. Chem. 37 G. Xiao, J. Zhou, X. Huang, et al., RSC Adv., 2014, 4(8), 4010–
Soc., 2006, 128(10), 3160–3161.
4019.
J. A. Melero, J. Iglesias and G. Morales, Green Chem., 2009, 38 H. Zhao, P. Jiang, Y. Dong, et al., New J. Chem., 2014, 38(9),
1(9), 1285–1308. 4541–4548.
F. Su, S. An, D. Song, et al., J. Mater. Chem. A, 2014, 2(34), 39 S. Y. Chin, M. A. A. Ahmad, M. R. Kamaruzaman, et al.,
4127–14138. Chem. Eng. Sci., 2015, 129, 116–125.
0 L. Ma, Y. Han, K. Sun, et al., Energy Convers. Manage., 2015, 40 M. Taherimehr, J. P. C. C. Sert ˜a , A. W. Kleij, et al.,
8, 46–53.
ChemSusChem, 2015, 8(6), 1034–1042.
1 Z. Zhang, K. Dong and Z. K. Zhao, ChemSusChem, 2011, 4(1), 41 K. S. Novoselov, A. K. Geim, S. V. Morozov, et al., science,
12–118.
2004, 306(5696), 666–669.
2 Y. Li, X. D. Zhang, L. Sun, et al., Appl. Energy, 2010, 87(1), 42 F. L. Meng, H. H. Li, L. T. Kong, et al., Anal. Chim. Acta, 2012,
56–159. 736, 100–107.
3 A. M. Hengne, S. B. Kamble and C. V. Rode, Green Chem., 43 A. Vlcek Jr, H. Kvapilov ´a , M. Towrie, et al., Acc. Chem. Res.,
013, 15(9), 2540–2547. 2015, 48(3), 868–876.
8
1
9
1
1
1
1
1
9
1
1
2
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 90757–90765 | 90765