Porous solid acid with high Surface area derived from emulsion templating and hypercrosslinking for efficient one-pot conversion of cellulose to 5-hydroxymethylfurfural

Article abstract of DOI:10.1039/c4ra10383a

This work addressed efficient one-pot conversion of cellulose to 5-hydroxymethylfurfural by porous solid acid in the presence of 1-ethyl-3-methyl-imidazolium chloride ([Emim]Cl). On the basis of stable water-in-oil pickering high internal phase emulsions (HIPEs), porous solid acid (HC-PDVB-SS-SO₃H) was prepared by polymerizing divinylbenzene (DVB) and sodium p-styrenesulfonate (SS), succedent sulfonation in H₂SO₄ and hypercrosslinking process. HC-PDVB-SS-SO₃H with a stable network exhibited mesopores within a macropore structure, high BET specific surface area and more super-strong acid sites. To test catalytic activity, the reaction conditions were optimized via response surface methodology, and the sequence of the three variables affecting HMF yield followed the order, temperature > reaction time > amount of catalyst. By comparing with the other as-prepared porous solid acid, it can also be concluded that BET specific surface area and super-strong acid site both played key roles in cellulose conversion. Moreover, HC-PDVB-SS-SO₃H was very easily reused at least four times without significant loss of activity. This journal is

Full text of DOI:10.1039/c4ra10383a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Porous solid acid with high Surface area derived
from emulsion templating and hypercrosslinking
for efficient one-pot conversion of cellulose to 5-
hydroxymethylfurfural
Cite this: RSC Adv., 2014, 4, 59175
a
a
Jianming Pan, Heping Gao,ab Yunlei Zhang,a Jun Zeng,ab Weidong Shi,
*
*
Changhua Song,a Yongsheng Yan,a Longbao Yua and Dandan Changa
This work addressed efficient one-pot conversion of cellulose to 5-hydroxymethylfurfural by porous solid
acid in the presence of 1-ethyl-3-methyl-imidazolium chloride ([Emim]Cl). On the basis of stable water-in-
oil pickering high internal phase emulsions (HIPEs), porous solid acid (HC-PDVB-SS-SO3H) was prepared by
polymerizing divinylbenzene (DVB) and sodium p-styrenesulfonate (SS), succedent sulfonation in H2SO4
and hypercrosslinking process. HC-PDVB-SS-SO3H with a stable network exhibited mesopores within a
macropore structure, high BET specific surface area and more super-strong acid sites. To test catalytic
activity, the reaction conditions were optimized via response surface methodology, and the sequence of
the three variables affecting HMF yield followed the order, temperature > reaction time > amount of
catalyst. By comparing with the other as-prepared porous solid acid, it can also be concluded that BET
specific surface area and super-strong acid site both played key roles in cellulose conversion. Moreover,
HC-PDVB-SS-SO3H was very easily reused at least four times without significant loss of activity.
Received 13th September 2014
Accepted 21st October 2014
DOI: 10.1039/c4ra10383a
www.rsc.org/advances
wood-based biomass (e.g., cellulose).5,6 Unfortunately,
improving the efficiency in cellulose convert into HMF is a
1. Introduction
challenge owing to the complex structure and difficulty of dis-
solving in general solvents.
Although today coal, natural gas, petroleum, and other non-
renewable fossil resources supply most energy and chem-
icals,1 in the future decades these fossil resources are unlikely to
meet the growing needs of humanity. Moreover, consumers and
governments pay more and more attention to environmental
issues and environment protection and are demanding renew-
able energy and products. With the development of conversion
technology, abundant biomass resources have the potential to
occupy the main position as feedstocks, particularly as renew-
able carbon sources for the production of fuels and chemicals.2
One of those, the platform chemical 5-hydroxymethylfurfural
(HMF) plays an important role in biobased energy, because
HMF can be converted by effective methods into several useful
acids, such as levulinic acid, 2,5-furandicarboxylic acid (FDA),
and 5-hydroxy-4-keto-2-pentenoic acid,3 as well as the promising
fuel 2,5-dimethylfuran (DMF).4 Several efforts toward HMF
production have used edible carbohydrate-based biomass such
as fructose and glucose, which are easy to convert into valuable
products. However, their consumption for the manufacture of
biofuels would cause a food deciency, contrary to the case of
In recent years, several studies have adopted various types of
solid acid catalysts for the conversion of cellulose to HMF in
ionic liquid because solid acid and ionic liquid possesses ultra
strong acid strength and unique solubility for cellulose. For
example, we fabricated two acid-chromic chloride bi-
functionalized solid acid catalysts by graing the –SO3H and
Cr(III) onto the surface of treated attapulgite (ATP) and halloysite
nanotubes (HNTs), and their catalytic activities were evaluated
for the one-pot conversion of cellulose to 5-hydroxy-
methylfurfural (HMF) in an ionic liquid 1-ethyl-3-methyl-
imidazolium chloride ([EMIM]-Cl).7 Very recently, we applied
precipitation polymerization and Pickering emulsion polymer-
ization combined with a sulfonation process for synthesizing
two types of polymeric solid acid for cellulose conversion in
[EMIM]-Cl, which showed a satisfying yield of HMF.8 Never-
theless, solid acids, which have a high yield of HMF, as well as a
simple synthesis process in combination with easy recovery are
also practically unavailable.
Recently, mesoporous solid acids9–13 have been widely used
in HMF production, because of their high specic surface area
and large pore volume. For example, Clayton et al.14 success-
fully obtained 29% HMF yields using a xed bed porous metal
oxide-based catalytic process in two liquid phases. Peng et al.15
aSchool of Chemistry and Chemical Engineering, Jiangsu University, Zhenjiang
212013, China. E-mail: pjm@ujs.edu.cn; shiwd999@163.com; Fax: +86 88791800;
Tel: +86 88791708
bSchool of Environmental and Chemical Engineering, Jiangsu University of science and
technology, Zhenjiang 212003, China
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 59175–59184 | 59175

View Article Online
RSC Advances
Paper
have prepared acidic large-pored mesoporous silica nano-
particles for a one-pot catalysis of cellulose to HMF with a low
yield of 18%. Xiao's group successfully designed and synthe-
sized a stable mesoporous polymeric solid acid for cellulose
conversion.16 The excellent catalytic activity was determined
for these mesoporous materials, but extra isolation steps for
catalyst recycling mass and slow mass transfer to the active
sites on the internal surface limited their further application.17
The mass transport limitations of a mesoporous solid acid can
be overcome using macroporous polymers based on high
internal phase emulsions (HIPEs), which possess very large
pores (1–100 mm).18 HIPEs, as emulsions with a volume frac-
tion of the droplet phase higher than 74%, provide a very
convenient route to synthesize macroporous polymer supports
(polyHIPEs) by polymerizing the monomers in the continuous
phase.19 However, conventional polyHIPEs synthesized from
surfactant stabilized HIPEs have poor mechanical properties
and low surface area.20 In the literature, it has been shown that
the poor mechanical properties usually observed for poly-
HIPEs can be improved by particle reinforcements, and one of
the typical method was utilizing particle-stabilized HIPEs
templates (Pickering HIPEs) to produce macroporous poly-
mers. Bismarck's group21 made use of pickering HIPEs incor-
porated with small amounts of non-ionic polymeric surfactant
to form a macroporous polyHIPEs with an open porous
structure. To address the bottleneck of low surface area, an
alternative technology is the development of extensive post
polymerisation crosslinking of swollen or dissolved polymers,
namely hypercrosslinking (HC).22 Krajnc and co-workers23 had
successfully created micropores within macroporous mono-
liths by post polymerisation hypercrosslinking of styrene/
divinylbenzene polyHIPEs. Their work demonstrated that the
hypercrosslinking process could signicantly increase BET
surface area, while retained the macroporous morphology,
which is a typical of polyHIPEs materials.
2. Material and methods
2.1 Materials
1-Ethyl-3-methyl-imidazolium chloride ([EMIM]-Cl), cellulose
(powder, ca. 50 micron), 5-hydroxymethylfurfural (HMF)
(>99%), tetraethoxysilane (TEOS), divinylbenzene (DVB), 3-
methacryloxypropyltrimethoxysilane (KH-570) were supplied by
Aladdin reagent Co., Ltd (Shanghai, China). 2,20-Azo-bis-(2-
methyl-propionitrile)
(AIBN,
99%), toluene, ethanol,
methanol, n-hexane, Span 80, K2SO4, FeCl3 (anhydrous),
potassium persulfate (KPS), NH3$H2O (25 wt%), sodium
p-styrene sulfonate (SS), 1,2-dichloroethane and acetone were
purchased from Sinopharm Chemical Reagent Co., Ltd.
(Shanghai, China). All the other chemicals were supplied by
local suppliers and used without further purication.
2.2 Instruments and characterization
FTIR spectra were measured with a Nicolet NEXUS-470 FTIR
apparatus (U.S.A.). The morphology of the silica particles and
the as-prepared solid acid catalysts were characterized by eld
emission scanning electron microscopy (SEM, JSM-7100F).
Energy Dispersive Spectra (EDS) were obtained on a SEM
instrument. XPS spectra were performed on a Thermo ESCALAB
250 with Al Ka radition at y ¼ 901 for the X-ray source, the
binding energies were calibrated using the C1s peak at 284.9 eV.
Elemental composition was recorded by the vario EL III
elemental analyzer (Elementar, Hanau, Germany). The acidic
features of the samples were all measured according to NH3
temperature-programmed desorption (NH3-TPD) curves, which
were recorded with a thermal conductivity detector (TCD, DAS-
7000, Beijing, China). TGA of the samples were performed for
powder (near 10 mg) using a Diamond TG/DTA instrument
(Perkin-Elmer, U.S.A.) under a nitrogen atmosphere up to
800 ꢀC at a heating rate of 5.0 ꢀC minꢁ1. The water contact angle
of silica particles were measured with the Optical Contact Angle
Measuring Device (KSV CM200). A detailed test procedure is
described as follows: 2.0 mL of deionized water was injected on
the sample surface by a syringe pump, and then the images of
the water droplet were obtained using the camera of the
measuring device 30 s aer the water droplet was formed on the
sample surface. Finally, the these images were analyzed using
the supplied soware to determine the contact angle of the
sample.
Herein, a series of porous polymerized solid acids were
successfully designed and produced by W/O Pickering HIPEs
template method and a succedent hypercrosslinking process.
First, the as-prepared hydrophobic silica particles and span 80
were both used to stabilize W/O Pickering HIPEs with an
internal phase volume ratio of 84.8%. Then, Pickering HIPEs
had an external phase containing divinylbenzene (DVB)
monomer, and an internal phase containing DVB when the
external phase containing sodium p-styrene sulfonate (SS)
monomer were applied to fabricated porous materials PDVB
and PDVB-SS, respectively. Aer sulfonation in 98% H2SO4, two
porous polymerized solid acids PDVB-SO3H and PDVB-SS-SO3H
2.3 Synthesis of hydrophobic silica particles as stabilizers
were obtained. Moreover, the third solid acid HC-PDVB-SS-SO3H Hydrophobic silica particles, which were used to stabilize W/O
was prepared by further hypercrosslinking reaction and Pickering HIPEs, were synthesized via two steps: First, hydro-
sulfonation of PDVB-SS. Then, three polymerized solid acids philic silica particles were prepared using the well-known
were all characterized, and the catalytic activities were hydrolysis and condensation reaction according to a modied
24
¨
discussed in detail by optimizing the reaction time, tempera- Stober's method. 6.0 mL of TEOS was slowly injected into a
ture and catalyst loading amounts in catalyzing cellulose to uniform solution containing 90 mL ethanol, 10 mL H2O and
HMF. HC-PDVB-SS-SO3H with a large surface area, ultra 3.14 mL 25 wt% NH3$H2O. Then, the mixtures were reacted for
ꢀ
strong acid strength and stable network possessed good 1.0 h at 30 C with mechanical stirring, and subsequently the
recyclability and catalytic activity compared with PDVB-SO3H reaction products were separated and washed several times with
and PDVB-SS-SO3H.
ethanol by centrifugation. Finally, the uniform sizes of
59176 | RSC Adv., 2014, 4, 59175–59184
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
hydrophilic silica particles were obtained by drying for 8.0 h at PDVB-SS (1.0 g) was swollen in 1,2-dichloroethane (40 mL)
ꢀ
40 C in a vacuum oven.
under N2 for 3.0 h. Then, FeCl3 (1.6 g) in 20 mL of 1,2-
Second, the as-prepared hydrophilic silica particles were dichloroethane was added into the mixture of PDVB-SS and
further modied with KH-570. In a typical run, hydrophilic 1,2-dichloroethane, and the reaction mixture was stirred for
silica particles (1.0 g), toluene (150 mL), H2O (15 mL) and KH- another 18 h at 80 ꢀC. The resulting hypercrosslinked products
ꢀ
570 (3.5 mL) were mixed and stirred for 24 h at 40 C in a 250 were ltered and washed three times with water and methanol,
mL three-neck ask. Subsequently, the reaction products were and then washed with methanol in a Soxhlet extraction for 24
separated and washed with ethanol. Finally, hydrophobic silica h. Subsequently, the products (HC-PDVB-SS) were dried
particles were obtained by drying for 12 h at 40 ꢀC in a vacuum under vacuum at 40 ꢀC for 24 h. Next, the sulfonation process
drying oven.
of HC-PDVB-SS was followed by the step described above.
Consequently, the obtained products were identied as
HC-PDVB-SS-SO3H.
2.4 Preparation of PDVB-SO3H and PDVB-SS-SO3H
PDVB-SO3H was prepared by emulsion polymerization and a
subsequent sulfonation process. The rst step was the synthesis
of porous PDVB polymers via a Pickering HIPEs template
method, and the HIPEs recipes are listed in Table 1. A contin-
uous phase was formed by stirring the hydrophobic silica
particles (0.3 g), AIBN (0.04 g), DVB (3.0 mL) and span 80 (0.4
mL) in a 100 mL three-neck ask. Next, 16.8 mL of the internal
aqueous phase comprised of K2SO4 (0.08 g) was gradually added
with continuous stirring, and the stable W/O Pickering HIPEs
were obtained. Then, the obtained Pickering HIPEs were
transferred into a plastic centrifuge tube and polymerized in a
2.6 The conversion of cellulose to HMF and analysis of HMF
A typical procedure for the conversion of cellulose to HMF
according to modied Wei's method25 is detailed as follows.
First, cellulose (0.1 g) and [EMIM]Cl (2.0 g) were heated with
stirring at 125 ꢀC for 30 min until a clear solution was formed,
in order to destroy the degree of crystallization and regularity
of the cellulose. Then, the as-prepared HC-PDVB-SS-SO3H (45
mg) was added into the mixed solution, and was allowed to
continue reacting for 2.0 h at 125 ꢀC under continuous stir-
ring. Next, the reaction was immediately quenched with cold
water, and diluted 5000 times with deionized water. Finally,
liquid samples were analyzed by a high performance liquid
chromatography (HPLC) with an Agilent 1200 instrument.
HMF was detected with a UV-vis detector at a wavelength of
283 nm. The column used was an Agilent TC-C18 Column (4.6
ꢂ 250 mm, 5.0 mm). The mobile phase was mixed water and
methanol with the volume ratio of 30 : 70, and the ow rate
and column oven were set to 0.7 mL minꢁ1 and 25 ꢀC,
respectively. Moreover, the HMF content was obtained
according to the standard curve method. In this work, the
HMF yield was dened as the ratio of total mole number of
carbon atoms of HMF (n1) and total moles number of carbon
atoms in initial cellulose loaded in the reactor (n0), which was
expressed in eqn (1).
ꢀ
circulating air oven at 60 C. Aer 12 h, the as-prepared PDVB
was washed with deionized water and acetone in turn, and then
macroporous PDVB monoliths were dried at 80 ꢀC for 12 h. The
second step was graing the above PDVB with –SO3H by a
sulfonation process. In brief, PDVB monoliths (1.0 g) and 98%
H2SO4 (30 mL) were mixed and stirred at 70 ꢀC for 12 h.
Then, the reaction products were ltered and washed to remove
the excess of H2SO4 using volumes of deionized water.
Finally, PDVB-SO3H was obtained aer drying in vacuum at
ꢀ
80 C for 3.0 h.
The preparation of PDVB-SS-SO3H also had two steps, such
as the fabrication of macroporous PDVB-SS monoliths and a
subsequent sulfonation process. For the preparation of PDVB-
SS, the SS (0.2 g) and KPS (0.04 g) were added into the
internal aqueous phase (Table 1), and the other processes were
the same as the formation of macroporous PDVB monoliths.
Moreover, the sulfonation process was followed by the step
described above. Finally, the obtained products were identied
as PDVB-SS-SO3H.
YHMF (%) ¼ 100 ꢂ (n1/n0)
(1)
3. Results and discussion
2.5 Preparation of HC-PDVB-SS-SO3H
3.1 Fabrication of hydrophobic silica particles
The preparation of HC-PDVB-SS-SO3H included the post The water contact angle images of hydrophilic and hydro-
polymerisation hypercrosslinking of PDVB-SS and a subse- phobic silica particles were shown in Fig. 1a and b, and it
quent sulfonation process. In a typical run, the as-prepared could be clearly observed that the water contact angles were
Table 1 Recipes of W/O Pickering HIPEs
External organic phase
Internal aqueous phase
Samples
DVB (mL)
AIBN (g)
Silica (g)
Span80 (mL)
Water (g)
K2SO4 (g)
KPS (g)
SS (g)
PDVB-SO3H
PDVB-SS-SO3H
3.0
3.0
0.04
0.04
0.3
0.3
0.4
0.4
16.8
16.8
0.08
0.08
0
0.04
0
0.2
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 59175–59184 | 59177

View Article Online
RSC Advances
Paper
3.2 Characterizations of porous polymerized solid acid
3.2.1 Morphological features and porous structure of
catalysts. Fig. 2 also shows the SEM images of PDVB-SO3H (c),
PDVB-SS-SO3H (d) and HC-PDVB-SS-SO3H (e). Clearly,
PDVB-SO3H (Fig. 2c) displayed
a well-dened open-cell
structure (void size of 12 ꢃ 2 mm) and interconnecting pore
throat sizes in the range 0.5–3.0 mm. However, PDVB-SS-SO3H
(Fig. 2d) had a close-cell structure with few interconnecting pore
throats, and a polymer lm covered region between two voids.
The mechanism for the formation of pore throat in polyHIPEs
indicates that the lms for the separation of the droplets in
HIPEs must be sufficiently thin to break and form pore
throats.27 Moreover, the polymerized SS monomer layer in the
internal phase of PDVB-SS-SO3H increased the thickness of the
contact domains between the neighboring droplets during
polymerization, and the function against coalescence resulted
in the close-cell structure. It was also determined that the void
size of PDVB-SS-SO3H (30 to 50 mm) was larger than that of
PDVB-SO3H. It was possibly because dissolved SS monomers in
the internal phase of PDVB-SS-SO3H brought about high
interfacial energy and low electric potential in the oil–water
interface, which was of benecial for increasing the sizes of
HIPEs droplets. Compared with PDVB-SS-SO3H, no signicant
change in the morphological features was observed for
HC-PDVB-SS-SO3H (Fig. 2e). This result was also investigated
and proved by Krajnc's group. They suggested that post
polymerisation crosslinking involved reactions between
pre-formed polymer chains, and the morphological feature of
the material was slightly affected.17
Fig. 1 Water contact angle images of hydrophilic (a) and hydrophobic
(b) silica particles.
about 34.5ꢀ and 125.7ꢀ, respectively, suggesting that hydro-
phobic silica particles were successfully prepared by graing
vinyl groups on the surface of hydrophilic silica particles.
Moreover, the proper water contact angle meets the require-
ment for stabilizing W/O Pickering HIPEs, which was proved
by Binks's groups using the particles with the water contact
angle of 123ꢀ to stabilize W/O pickering HIPEs.26 Further-
more, the morphology of hydrophilic and hydrophobic silica
particles was both observed by SEM, which is shown in Fig. 2a
and b, respectively. Hydrophilic and hydrophobic silica
particles were both regular sphere-shaped, and their sizes
were 200 ꢃ 20 nm and 250 ꢃ 20 nm, respectively. This
phenomenon of increased sizes further conrmed that the
vinyl groups were graed onto the surface of hydrophilic
silica particles.
Porous structure has a remarkable effect on the catalytic
effect, and the nitrogen adsorption–desorption isotherms of
HC-PDVB-SS-SO3H (a), PDVB-SS-SO3H (b) and PDVB-SO3H (c)
are listed in Fig. 3. As shown in Fig. 3, they all displayed a typical
type-IV isotherm with a step capillary condensation in the
relative pressure range of 0.1–0.9, and this type of isotherm was
in agreement with the presence of obvious mesoporosity.
Moreover, the porosity property of HC-PDVB-SS-SO3H, PDVB-SS-
Fig. 2 SEM images of hydrophilic silica particles (a), hydrophobic silica
particles (b), PDVB-SO3H (c), PDVB-SS-SO3H (d) and HC-PDVB-SS- Fig. 3 Nitrogen adsorption–desorption isotherms of HC-PDVB-SS-
SO3H (e).
SO3H (a), PDVB-SS-SO3H (b), and PDVB-SO3H (c).
59178 | RSC Adv., 2014, 4, 59175–59184
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 2 The porosity properties, acidic properties and catalytic efficiency of the catalysts
Acid amounts
Total acidity
Samples
Vta (cm3 gꢁ1
)
DBJHb (nm)
SBET (m2 gꢁ1
)
Acid strength (ꢀC)
(mmol gꢁ1
)
(mmol gꢁ1
)
HMF (%)
22.3d
c
PDVB-SO3H
0.05
14.19
30.24
250
378
557
285
369
518
604
706
50
116
69
49
95
169
104
133
235
PDVB-SS-SO3H
0.13
0.20
10.67
5.94
80.06
313
237
25.2d
28.9e
HC-PDVB-SS-SO3H
323.95
a
Vt is the total pore volume determined at the relative pressure of 0.99. b DBJH is BJH pore diameter calculated by the desorption branches of the
nitrogen sorption isotherms. c SBET is the BET specic surface area. d The reaction condition are 2.0 h, 130 ꢀC and 50 mg of catalyst. e The reaction
condition are 1.75 h, 125 ꢀC and 45 mg of catalyst.
SO3H, and PDVB-SO3H are listed in Table 2. Clearly, the BET suggesting the presence of silica particles and monomers such
surface area (323.95 m2 gꢁ1) and pore volume (0.20 cm3 gꢁ1) of as DVB or SS. In addition, the peak at 1039 cmꢁ1 attributed to
HC-PDVB-SS-SO3H were obviously higher compared to those of C–S bond, and 1170 cmꢁ1 belonging to –SO3H groups were also
PDVB-SS-SO3H (80.06 m2 gꢁ1, 0.13 cm3 gꢁ1), and PDVB-SO3H clearly observed in the samples of PDVB-SO3H, regenerated
(30.24 m2 gꢁ1, 0.05 cm3 gꢁ1). The low surface area was found for HC-PDVB-SS-SO3H, HC-PDVB-SS-SO3H and PDVB-SS-SO3H,
highly permeable macroporous PDVB-SO3H polymers, and this conrmed that the sulfonic groups were successfully graed
result demonstrated that PDVB-SO3H possessed a very less onto the benzene rings of the monomers.
amount of mesopores. Although a close-cell structure was
Energy dispersive spectrometry (EDS) analysis of PDVB,
obtained for HC-PDVB-SS-SO3H, a signicant enlargement of PDVB-SO3H, and PDVB-SS-SO3H are listed in Fig. 5a–c, respec-
the amount of mesopores in the material was determined tively. When comparing Fig. 5a with b, the peak of the element S
resulting from post polymerisation hypercrosslinking. Never- appeared in the sample of PDVB-SO3H, and the same
theless, the BJH pore diameter (5.94 nm) of HC-PDVB-SS-SO3H phenomenon was also determined in the sample of
was lower compared to PDVB-SS-SO3H (10.67 nm), and PDVB-SS-SO3H when comparing Fig. 5a with c, which suggested
PDVB-SO3H (14.19 nm), which could be attributed to the that –SO3H groups were successfully graed onto the surface of
presence of SS monomers, which might block the mesopores PDVB via the sulfonation process. Furthermore, it could be
and increase the density of the network in the hypercrosslinking determined that the peak of the element S in the sample of
process.
PDVB-SS-SO3H was stronger compared to PDVB-SO3H, and the
3.2.2 Chemical composition of catalysts. Fig. 4 shows the amount of S was also more than that of PDVB-SO3H. Thus, it
FTIR spectra of PDVB-SO3H (a), PDVB (b), regenerated could be emphasized that PDVB-SS-SO3H had a PSS layer, which
HC-PDVB-SS-SO3H (c), HC-PDVB-SS-SO3H (d) and PDVB-SS- resulted in more –SO3H groups in the sample of PDVB-SS-SO3H.
SO3H (e). The peaks at 1456 cmꢁ1 associated with benzene ring, Furthermore, the results of the elemental analysis of PDVB,
and 1107 cmꢁ1 assigned to Si–O–Si antisymmetric stretching PDVB-SO3H and PDVB-SS-SO3H are all shown in the
vibration could both be clearly observed in all the samples,28 corresponding gures. Moreover, the amount of S in the sample
of PDVB-SO3H (0.65%) was increased to 2.28% in the sample of
PDVB-SS-SO3H, and S was not observed in PDVB. These results
agree with the results of EDS analysis.
Fig. 6 shows the X-ray photoelectron spectroscopy (XPS)
measurements of PDVB-SO3H and HC-PDVB-SS-SO3H. Clearly,
PDVB-SO3H and HC-PDVB-SS-SO3H both showed the signals of
S, C, Si and O, possibly indicating the presence of –SO3H groups
and hydrophobic silica particles in these samples. Corre-
spondingly, the highly resolved XPS spectrum of C1s showed
signals at around 284.7 eV, which were associated with C–S
bond in these samples. This result further proved the successful
introduction of –SO3H into the network of these samples, and
–SO3H groups would contribute to increase the acid strength of
these solid acids.29 Moreover, the signal of S2p of PDVB-SO3H
and HC-PDVB-SS-SO3H were found at the same position with a
binding energy of 169 eV, suggesting that the hypercrosslinking
process of HC-PDVB-SS-SO3H had no obvious effect on the
Fig.
4 FTIR spectra of PDVB-SO3H (a), PDVB (b), regenerated
HC-PDVB-SS-SO3H (c), HC-PDVB-SS-SO3H (d) and PDVB-SS-SO3H (e).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 59175–59184 | 59179

View Article Online
RSC Advances
Paper
Fig. 5 Energy dispersive spectra (EDS) and elemental analysis of PDVB (a), PDVB-SO3H (b) and PDVB-SS-SO3H (c).
Fig. 7 TGA curves of PDVB-SO3H (a), HC-PDVB-SS-SO3H (b) and
PDVB-SS-SO3H (c).
HC-PDVB-SS-SO3H (16.57%), and PDVB-SS-SO3H (17.64%)
resulting from decomposition of –SO3H groups could also be
seen.31 Interestingly, it could be found that the more –SO3H
groups the catalyst had, the higher adsorption content of water,
which suggested that –SO3H groups provided more hydrophilic
surface. When the temperature further increased to 800 ꢀC,
Fig. 6 X-ray photoelectron spectroscopy measurements of survey (a)
C1s, (b) S2p, (c) Si2p, (d) of PDVB-SO3H and HC-PDVB-SS-SO3H.
introduction of –SO3H groups, and the results also supported
the fact that the HC-PDVB-SS-SO3H catalyst possessed a rigid PDVB-SO3H, HC-PDVB-SS-SO3H and PDVB-SS-SO3H had a
network structure and good chemical stability. Furthermore, signicant weight loss with 46.75%, 36.76%, and 30.76%,
the peak position of Si2p of PDVB-SO3H and HC-PDVB-SS-SO3H respectively, due to the destruction and decomposition of
were both at 102.2 eV, and the phenomenon showed that spatial structure of catalysts. At 800 ꢀC, the remaining mass for
hydrophobic silica particles might not be involved in other catalysts may be put down to rigid silica particles and residual
carbon by calcining polymers.32
reactions except in the emulsifying process.
3.2.3 The acidic properties of catalysts. The acidic features
Fig.
7 showed the TGA curves of PDVB-SO3H (a),
of PDVB-SO3H, PDVB-SS-SO3H and HC-PDVB-SS-SO3H catalysts
HC-PDVB-SS-SO3H (b), and PDVB-SS-SO3H (c). Slight weight loss
of PDVB-SO3H (6.78%), HC-PDVB-SS-SO3H (9.09%) and were determined by use of NH3-TPD. Additionally, the type of
PDVB-SS-SO3H (11.29%) within the initial temperature range distinguishing acid sitꢀes was basedꢀon desorption temperatures
ꢀ
ꢀ
ꢀ
(<150 C) could be observed, which could be attributed to the of NH3, and 150–250 C, 250–350 C, 350–500 C and >500 C
loss of H2O in these samples.30 With the temperature increased were measured as weak, medium, strong and super-strong acid
to 450 ꢀC, the obvious weight losses of PDVB-SO3H (11.37%), sites, respectively.33 As shown in Fig. 8, it can be clearly seen that
59180 | RSC Adv., 2014, 4, 59175–59184
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
medium, strong and very strong acid sites existed in
PDVB-SO3H and PDVB-SS-SO3H, but HC-PDVB-SS-SO3H
possessed only super-strong acid sites that can also be readily
observed. Moreover, the areas of the peak represented the
amount of acidic sites were further calculated and summarized
in Table 2. The total acidic amount of 235 mmol gꢁ1 for
3.3 Conversion of cellulose to HMF
The experimental data of the correlation of three variables, such
as catalyst dosage (m, mg), temperature (T, ꢀC) and reaction
time (t, h), were tted using a low-order polynomial equation to
evaluate the effect of each independent variable on the
response. Moreover, in this work, the conversion efficiency was
evaluated by the analysis of variance (ANOVA) with the HMF
yield (Y, mol%) as the response, and the nal obtained equation
of HC-PDVB-SS-0-SO3H in terms of the actual factors is
expressed in eqn (2).
PDVB-SO3H (i.e. respective acidic amounts of 50 mmol gꢁ1
,
116 mmol gꢁ1 and 69 mmol gꢁ1 for the acidic strengths of 2ꢁ510 ꢀC,
PDVB-SS-SO3H (i.e. respective acidic amounts of 49 mmol gꢁ1
378 ꢀC and 557 ꢀC) was less than that of 313 mmol g for
,
95 mmol gꢁ1 and 169 mmol gꢁ1 for the acidic strengths of 285 ꢀC,
369 ꢀC and 518 ꢀC), and 237 mmol gꢁ1 for HC-PDVB-SS-SO3H (i.e.
respective acidic amounts of 104 mmol gꢁ1 and 133 mmol gꢁ1 for
Y ¼ ꢁ666.19 + 9.36T + 52.94t + 1.47m ꢁ 0.3Tt ꢁ 4.96
ꢂ 10ꢁ3Tm ꢁ 0.04tm ꢁ 0.03T2 ꢁ 3.37t2 ꢁ 6.34 ꢂ 10ꢁ3m2 (2)
ꢀ
ꢀ
the acidic strengths of 604 C and 706 C). Moreover, it could
also be simply proved that the results of total acidic amount
were identical with the amounts of –SO3H groups resulting from
TGA analysis. Furthermore, the acidic amount provides many
acidic sites, which can help to improve the efficiency for the
conversion of cellulose into HMF.34 Therefore, 185 mmol gꢁ1 of
strong acid sites at 378 ꢀC and 557 ꢀC for PDVB-SO3H, 264 mmol
gꢁ1 of strong acid sites at 369 ꢀC and 518 ꢀC for PDVB-SS-SO3H,
and 237 mmol gꢁ1 of super-strong acid sites at 604 ꢀC and 706 ꢀC
for HC-PDVB-SS-SO3H were effective to catalyze the conversion
of cellulose into HMF. Furthermore, the fact that strong acid
sites played a key factor for cellulose conversion was tested in
our recent work.8 Thus, HC-PDVB-SS-SO3H would
possess predominant catalytic activity over PDVB-SO3H and
PDVB-SS-SO3H, owing to the number of strong acid sites. These
results supported that HC-PDVB-SS-SO3H was selected in the
further catalytic activity studies.
The 3D response surface and the 2D contour plots showing
the effects of three variables for HC-PDVB-SS-SO3H catalyst on
HMF yield are displayed in Fig. 9. The interaction between the
amount of catalyst and temperature on HMF yield under
constant reaction time (1.75 h) are shown in Fig. 9a and b. With
an increase in the amount of catalyst and temperature, it is
obvious that HMF yield showed a trend of rapid increase in the
beginning stages, and then reaching the maximum, and nally
reduced with more catalyst and higher temperature used. The
phenomenon may be attributed to the following reasons. First,
the increasing reaction temperature and amount of catalyst
reduced the viscosity of ionic liquids and increased contact area
between catalyst and the reactants, which had the benet to
accelerate the degradation of cellulose.35 Second, the produced
HMF covered the acid sites of catalysts, and the conversion of
HMF into other tarry substances, such as insoluble humines
Fig. 8 NH3-TPD curves of PDVB-SO3H (a), PDVB-SS-SO3H (b) and HC-PDVB-SS-SO3H (c).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 59175–59184 | 59181

View Article Online
RSC Advances
Paper
Fig. 9 Optimization of reaction conditions of the conversion of cellulose to HMF using HC-PDVB-SS-SO3H as the catalyst through response
surface methodology. The effects of (a) and (b) amount of catalyst and temperature for 1.75 h; (c) and (d) reaction time and amount of catalyst at
125 ꢀC; (e) and (f) reaction time and temperature with 45 mg of HC-PDVB-SS-SO3H.
and soluble polymers by the polymerization and cross- which indicate that the bidirectional interaction between vari-
ꢀ
polymerization of HMF were enhanced with the further ables, such as reaction time and amount of catalyst at 125 C,
increasing reaction temperature and amount of catalyst.36 In reaction time and temperature with 45 mg catalyst, were also
addition, Qu's groups also presented similar results.37 More- important terms in the production of HMF from cellulose.
over, several researchers also reported the fact that the by- Moreover, the sequence of the three variables affecting HMF
products formic acid and levulinic acid could be examined by yield followed the order, temperature > reaction time > amount
HLPC in this process.38,39 Interestingly, the same phenomenon of catalyst, which was easily obtained according to the number
was also observed in Fig. 9c and d, as well as in Fig. 9e and f, of crosslinking of the contour plots with the coordinate axis of
59182 | RSC Adv., 2014, 4, 59175–59184
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 9b, d and f, respectively. The optimal condition achieved by
the response surface methodology was 0.10 g cellulose, 45 mg of
catalyst, temperature of 125 ꢀC, and reaction time of 1.75 h. The
optimal HMF yield of 28.9% was closer to the experimental
value of 29.2%, indicating that the predications were reasonable
and credible.
In order to discuss the inuence of the performance of
materials on the conversion of cellulose into HMF, the optimal
HMF yield of PDVB-SO3H (22.3%) and PDVB-SS-SO3H (25.2%)
under the same conditions were obtained (50 mg of catalyst,
2.0 h of reaction time and 130 ꢀC of reaction temperature) using
the same analysis method, which are also listed in Table 2. It
was clearly determined that HC-PDVB-SS-SO3H obtained a
higher HMF yield compared to the other two catalysts.
BET specic surface and super-strong acid sites were
considered to be responsible for the excellent performance of
HC-PDVB-SS-SO3H. First, higher BET specic surface area
helped to increase contact frequency between reaction
substrates and catalytic acid sites. Second, having more
super-strong acid sites was more conducive to improve the
catalytic effect of the conversion of cellulose to HMF than other
types of acid sites. Moreover, the HMF yield of PDVB-SS-SO3H
was slightly higher compared to PDVB-SO3H owing to the main
advantages of BET specic surface area and total acidity.
According to the above discussion about the performance of
catalysts, it could be concluded that higher BET specic surface
area and more super-strong acid sites of HC-PDVB-SS-SO3H, as
the remarkable effects, effectively facilitated the conversion of
cellulose to HMF.
Fig. 10 Reusability of HC-PDVB-SS-SO3H catalyst in the dehydration
of cellulose into HMF in [EMIM]Cl under the optimized conditions:
125 ꢀC, 45 mg of catalyst based on 0.10 g of cellulose, 1.75 h, [EMIM]Cl
2.0 g.
stable acidic active sites. The superior recyclability of
HC-PDVB-SS-SO3H occurs from their excellent chemical and
mechanical stability of both acidic sites and polymer network,
which were very important for their widely practical
applications. These results, to some extent, suggested that the
heterogeneous HC-PDVB-SS-SO3H catalyst has the potential to
be effectively recycled and reused for the dehydration reaction.
4. Conclusion
3.4 Regeneration of HC-PDVB-SS-SO3H catalyst
In this work, porous solid acid catalyst (HC-PDVB-SS-SO3H) with
high surface area was derived from pickering HIPEs template
and a hypercrosslinking process. Then, the production of HMF
from cellulose using HC-PDVB-SS-SO3H as a catalyst in the
presence of [EMIM]Cl was proved to be feasible, and obtained a
yield of HMF at 29.2% for HC-PDVB-SS-SO3H at the optimized
conditions. Moreover, the fact that the BET specic surface area
and super-strong acid site of catalyst are the key factors over the
other three factors, including temperature, time and catalyst
loading, was identied by ANOVA. This work was the continu-
ation of the efforts to prepare a stable polymeric solid acid, and
the results from this work could be valuable for future research
towards the development of new reactors, especially their wide
application for biomass and bioenergy production.
The regeneration of catalyst was very important for the practical
production of HMF, which was accorded with the demand of
green and sustainable chemistry. Therefore, to evaluate the
reusability of HC-PDVB-SS-SO3H catalyst, recycling tests were
conducted at the optimized reaction conditions. The results for
the regeneration experiment are shown in Fig. 10. The solid acid
catalysts HC-PDVB-SS-SO3H were recovered by simple suction
ltration using a Hirsch funnel with the mixture products
diluted with deionized water aer the end of the rst reaction,
washing them several times with a mixture of n-hexane and
methanol, and drying at 30 ꢀC for 24 h in a vacuum oven.
Subsequently, recovered HC-PDVB-SS-SO3H was used to cata-
lyze the conversion of cellulose into HMF for four times, and the
yields of HMF were 28.7%, 27.5%, 27.4%, and 26.5%, respec-
tively. It was determined that the HMF yield slowly reduced. It
was probably because the formation of by-product on the
catalytic centers of HC-PDVB-SS-SO3H could not wash out into a
Acknowledgements
mixture of n-hexane and methanol, which limited the accessi- This work was nancially supported by the National Natural
bility of new reactant molecules to the sulfonic acid sites. Science Foundation of China (no. 21107037, no.21176107, no.
However, the HC-PDVB-SS-SO3H catalyst still had good 21306013), Natural Science Foundation of Jiangsu Province
activity aer reusing it four times. Moreover, the FTIR spectra (no.BK20131223), National Postdoctoral Science Foundation
of regenerated HC-PDVB-SS-SO3H was similar with (no. 2013M530240), Special National Postdoctoral Science
HC-PDVB-SS-SO3H, especially the intensity of the peaks of Foundation (no. 2014T70480), Postdoctoral Science Foundation
Si–O–Si and –SO3H groups (Fig. 4), indicating that the of Jiangsu Province (no. 1202002B) and Programs of Senior
HC-PDVB-SS-SO3H catalyst possessed a rigid structure and Talent Foundation of Jiangsu University (no. 12JDG090).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 59175–59184 | 59183

View Article Online
RSC Advances
Paper
21 B. V. O. Ikem, A. Menner, T. S. Horozov and A. Bismarck, Adv.
Mater., 2010, 22, 3588–3592.
References
22 M. G. Schwab, I. Senkovska, M. Rose, N. Klein, M. Koch and
J. Pahnke, So Matter, 2009, 5, 1055–1059.
1 J. C. Serrano-Ruiz and J. A. Dumesic, Energy Environ. Sci.,
2011, 4, 83–99.
ˇ
ˇ
23 U. Sevsek, J. Brus, K. Jerabek and P. Krajnc, Polymer, 2014,
55, 410–415.
2 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem.,
Int. Ed., 2007, 46, 7164–7183.
¨
24 W. Stober, A. Fink and E. Bohn, J. Colloid Interface Sci., 1968,
3 A. A. Rosatella, S. P. Simeonov, R. F. M. Frade and
C. A. M. Afonso, Green Chem., 2011, 13, 754–793.
4 F. W. Lichtenthaler, Acc. Chem. Res., 2002, 35, 728–737.
5 K. I-Jung, S. Norihiro, Y. Yusuke and C. W. W. Kevin, RSC
Adv., 2013, 3, 2028–2034.
6 F. G. Calvo-Flores and J. A. Dobado, ChemSusChem, 2010, 3,
1227–1235.
7 Y. L. Zhang, J. M. Pan, M. Y. Gan, H. X. Ou, Y. S. Yan,
W. D. Shi and L. B. Yu, RSC Adv., 2014, 4, 11664–11672.
8 Y. L. Zhang, J. M. Pan, Y. S. Yan, W. D. Shi and L. B. Yu, RSC
Adv., 2014, 4, 23797–23806.
26, 62–69.
25 H. H. Wei, Y. Y. Lee, W. H. Peng and K. C. W. Wu, Catal.
Today, 2011, 174, 65–69.
26 B. P. Binks and S. Lumsdon, Phys. Chem. Chem. Phys., 2000,
2, 2959–2967.
27 A. Menner and A. Bismarck, Macromol. Symp., 2006, 242, 19–
24.
28 D. C. Niu, Z. J. Liu, Y. S. Li, F. Luo, J. Y. Zhang, J. P. Gong and
J. L. Shi, Adv. Mater., 2014, 29, 4947–4953.
29 C. Koibeck, M. Killian, F. Maier, N. Paape, P. Wasserscheid
¨
and H. P. Steinruck, Langmuir, 2008, 24, 9500–9507.
9 N. L. Torad, M. Naito, J. Tatami, A. Endo, S. Y. Leo,
S. Ishihara, K. C. W. Wu, T. Wakihara and Y. Yamauchi,
Chem.–Asian J., 2014, 9, 759–763.
10 K. Na, M. Choi and R. Ryoo, Microporous Mesoporous Mater.,
2013, 166, 3–19.
11 Y. Y. Lee and K. C. W. Wu, Phys. Chem. Chem. Phys., 2012, 14,
13914–13917.
12 I. I. Ivanova and E. E. Knyazeva, Chem. Soc. Rev., 2013, 42,
3671–3688.
13 I. J. Kuo, N. Suzuki, Y. Yamauchi and K. C. W. Wu, RSC Adv.,
2013, 3, 2028–2034.
14 C. V. McNeffa, D. T. Nowlana, L. C. B. McNeffa and
R. L. Fedieb, Energy Environ. Sci., 2012, 5, 7559–7574.
15 W. H. Peng, Y. Y. Lee, C. Wu and K. C. W. Wu, J. Mater.
Chem., 2012, 22, 23181–23185.
16 F. J. Liu, A. M. Zheng, I. Noshadi and F. S. Xiao, Appl. Catal.,
B, 2013, 136, 193–201.
17 I. Pulko, J. Wall, P. Krajnc and N. R. Cameron, Chem.–Eur. J.,
2010, 16, 2350–2354.
30 L. Wang, H. Wang, F. J. Liu, A. M. Zheng, J. Zhang, Q. Sun,
J. P. Lewis, L. F. Zhu, X. J. Meng and F. S. Xiao,
ChemSusChem, 2014, 7, 402–406.
31 F. J. Liu, X. J. Meng, Y. L. Zhang, L. M. Ren, F. Nawaz and
F. S. Xiao, J. Catal., 2010, 271, 52–58.
32 J. M. Pan, B. Wang, J. D. Dai, X. H. Dai, H. Hang, H. X. Ou
and Y. S. Yan, J. Mater. Chem., 2012, 22, 3360–3369.
33 D. Liu, P. Yuan, H. M. Liu, J. G. Cai, D. Y. Tan, H. P. He,
J. X. Zhu and T. H. Chen, Appl. Clay Sci., 2013, 80, 407–412.
34 P. Kalita, B. Sathyaseelan, M. A. Javaid, S. M. Zaidi,
M. A. Vinu and A. Chari, Chem.–Eur. J., 2010, 16, 2843–2846.
35 A. I. Torres, P. Daoutidis and M. Tsapatsis, Energy Environ.
Sci., 2010, 3, 1560–1572.
36 J. N. Chheda, Y. Roman-Leshkov and J. A. Dumesic, Green
Chem., 2007, 9, 342–350.
37 Y. S. Qu, Q. Y. Wei, H. Q. Li, P. Oleskowicz-Popiel,
C. P. Huang and J. Xu, Bioresour. Technol., 2014, 162, 358–
364.
¨
¨
38 E. I. Gurbuz, J. M. R. Gallo, D. M. Alonso, S. G. Wettstein,
W. Y. Lim and J. A. Dumesic, Angew. Chem., 2013, 52,
1270–1274.
18 H. F. Zhang and A. I. Cooper, So Matter, 2005, 1, 107–113.
19 B. H. L. Oh, A. Bismarck and M. B. Chan-Park,
Biomacromolecules, 2014, 15, 1777–1787.
39 H. B. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
2007, 316, 1597–1600.
20 A. F. Gross and A. P. Nowak, Langmuir, 2010, 26, 11378–
11383.
59184 | RSC Adv., 2014, 4, 59175–59184
This journal is © The Royal Society of Chemistry 2014