Hierarchically macro-/mesoporous polymer foam as an enhanced and recyclable catalyst system for the sustainable synthesis of 5-hydroxymethylfurfural from renewable carbohydrates

Article abstract of DOI:10.1002/cplu.201500357

Renewable and abundant carbohydrates are being strongly focused upon as a green and sustainable alternative for the production of valuable chemicals and biofuels. In this study, we chemically integrated acid-base bifunctionalized mesoporous silica nanoparticles (MSNs-SO₃H-NH₂) and a macroporous polymer foam poly(HIPE) (PH) matrix that is derived from water-in-oil (W/O) high internal phase emulsion (HIPE) templating. After the subsequent sulfonation process, a SPHs@MSNs-SO₃H-NH₂ catalyst with a hierarchical porous structure and bifunctional sites was prepared and used for the highly efficient synthesis of top value-added 5-hydroxymethylfurfural (HMF) from renewable cellulose in an ionic liquid (i.e., 1-ethyl-3-methylimidazolium chloride, [EMIM]Cl) based system. Evaluation of the catalytic activity revealed that the designed macropores were favorable for ready mass transfer, whereas the high surface area for active-site anchoring provided by the mesoporous structure was beneficial in enhancing the catalyst performance. A theoretical study also suggested that acidic and basic active sites synergistically work for the transformation reaction. Under the optimized conditions, a remarkably high HMF yield (44.5 %) was obtained efficiently. Furthermore, the as-prepared catalyst was recycled in four consecutive cycles with a total loss of only 4.9 % activity. Sweet success: Hierarchically macro-/mesoporous polymer foam has been successfully synthesized and used for the sustainable production of 5-hydroxymethylfurfural (HMF; see figure) from renewable cellulose. The as-prepared catalyst was also effective for the conversion of glucose and fructose to HMF in a DMSO/H₂O system.

Full text of DOI:10.1002/cplu.201500357

DOI: 10.1002/cplu.201500357
Full Papers
Hierarchically Macro-/Mesoporous Polymer Foam as an
Enhanced and Recyclable Catalyst System for the
Sustainable Synthesis of 5-Hydroxymethylfurfural from
Renewable Carbohydrates
[
a]
[b]
[a]
[a]
[a]
Yunlei Zhang, Yao Chen, Yating Shen, Yongsheng Yan,* Jianming Pan,*
[
a]
[a]
Weidong Shi, and Longbao Yu
Renewable and abundant carbohydrates are being strongly fo-
cused upon as a green and sustainable alternative for the pro-
duction of valuable chemicals and biofuels. In this study, we
chemically integrated acid–base bifunctionalized mesoporous
silica nanoparticles (MSNs-SO H-NH ) and a macroporous poly-
um chloride, [EMIM]Cl) based system. Evaluation of the catalyt-
ic activity revealed that the designed macropores were favora-
ble for ready mass transfer, whereas the high surface area for
active-site anchoring provided by the mesoporous structure
was beneficial in enhancing the catalyst performance. A theo-
retical study also suggested that acidic and basic active sites
synergistically work for the transformation reaction. Under the
optimized conditions, a remarkably high HMF yield (44.5%)
was obtained efficiently. Furthermore, the as-prepared catalyst
was recycled in four consecutive cycles with a total loss of only
4.9% activity.
3
2
mer foam poly(HIPE) (PH) matrix that is derived from water-in-
oil (W/O) high internal phase emulsion (HIPE) templating. After
the subsequent sulfonation process, a SPHs@MSNs-SO H-NH2
3
catalyst with a hierarchical porous structure and bifunctional
sites was prepared and used for the highly efficient synthesis
of top value-added 5-hydroxymethylfurfural (HMF) from renew-
able cellulose in an ionic liquid (i.e., 1-ethyl-3-methylimidazoli-
Introduction
Recently, the diminishing availability of traditional fossil resour-
ces and growing environmental problems have been the impe-
tus behind burgeoning efforts to develop technologies that
use renewable resources for the sustainable development of
biofuels and platform chemicals. As the only sustainable
source of organic carbon and a primary energy carrier, abun-
dant biomass has evolved as the most promising alternative to
provide valuable intermediates through a carbon-neutral way
of HMF by current approaches has never been implemented
commercially because the utilization of edible carbohydrates
competes with the food supply.
Accordingly, employment of the most abundant organic
compound of inedible cellulose for the production of HMF
[
5]
would be preferable. Moreover, the utilization of cellulose,
present in lignocellulosic materials such as wood, straw, grass,
municipal solid waste, and crop residues, does not compete
[1]
[6]
in the chemical industry. It is noticeable that the versatile and
top value-added 5-hydroxymethylfurfural (HMF), which can be
easily produced from renewable carbohydrates by the acid-cat-
alyzed dehydration of hexoses, has received fruitful advance-
ments in academia owing to its wide applications in the syn-
with the food supply. In general, the transformation of cellu-
lose into HMF involves multiple steps (i.e., hydrolysis, isomeri-
[
7]
zation, and finally dehydration to HMF). Therefore, Qi et al.
developed a two-step catalytic system including a strong
acidic cation-exchange resin for cellulose hydrolysis and CrCl3
for glucose degradation, and an excellent HMF yield was ob-
[2]
[3]
[4]
thesis of fine chemicals, polymer precursors, fuels, and so
forth. Nevertheless, the sustainable and large-scale production
[
8]
tained. Compared with multiple reactions, a one-pot synthe-
sis is an attractive methodology with respect to energy conser-
vation and time-saving. To this end, intense research activity
should focus on exploring efficient catalytic systems for the
[
a] Y. Zhang, Y. Shen, Prof. Y. Yan, Prof. J. Pan, Prof. W. Shi, Prof. L. Yu
School of Chemistry and Chemical Engineering
Jiangsu University
[9]
one-pot synthesis of HMF directly from renewable cellulose.
Xuefu Road 301#
During the past few years, different types of solid acids have
been considered as good candidates for biomass transforma-
tion because they are generally associated with a lower pro-
duction of waste, the use of fewer toxic reagents, and easier
Zhenjiang 212013 (P. R. China)
E-mail: zhangchuanqing035@126.com
[
b] Y. Chen
School of the Environment and Safety Engineering
Jiangsu University
Xuefu Road 301#
[10]
recycling over conventional liquid acids. In particular, fine-
particle-based catalysts with very small size and mesoporous
structure have attracted a surge of interest owing to their high
surface area. For example, a mesoporous carbonaceous solid
Zhenjiang 212013 (P. R. China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201500357.
ChemPlusChem 2016, 81, 108 – 118
108
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
(
CS) catalyst was successfully synthesized by Wang’s group,
Herein, we chemically integrated acid–base bifunctionalized
mesoporous silica nanoparticles (MSNs-SO H-NH ) and a macro-
and the catalyst was proved to have high activity for the syn-
3
2
[11]
thesis of HMF from fructose. Mazzotta et al. prepared meso-
porous polymer foam poly(HIPE) (PH) matrix that was derived
from water-in-oil (W/O) HIPE templating. After a subsequent
sulfonation process, a SPHs@MSNs-SO H-NH catalyst with hier-
porous sulfonated carbonaceous TiO and the obtained cata-
2
lyst also showed good catalytic activity for the dehydration of
3
2
[
12]
glucose to HMF. However, it was difficult for the large poly-
mer molecules of cellulose to diffuse deeply inside the meso-
porous solid through the small pore diameters, which limited
the accessibility of active sites and negated the merits of the
archical porous structure and acid–base bifunctional sites was
successfully prepared. As illustrated in Scheme 1, the as-pre-
[
13]
high surface area. In addition, it was difficult to maintain the
small particle morphology in the reaction process, and the sep-
aration of catalysts from the reaction mixture was also incon-
venient. These shortcomings strongly hinder the practical ap-
plication of mesoporous materials in the field of cellulose
transformation. More recently, hierarchical porous materials
with a bimodal macro-/mesoporous structure have been very
[14]
attractive in a wide range of catalytic applications. Different
pore sizes integrated in one body was efficient for cellulose
transformation reactions, in which the macropores ensure the
rapid transport of the reactant and the mesopores provide
Scheme 1. Catalyst behavior in one-pot cellulose-to-HMF conversion.
[15]
a high surface area for the reaction to take place. The main
shortcomings of mesoporous catalysts (i.e., limited accessibility
of cellulose with active sites) could also be overcome by using
pared catalyst was employed for the highly efficient synthesis
of HMF from renewable cellulose involving sequential catalytic
processes in an ionic liquid [EMIM]Cl-based system. The cata-
lytic activity was evaluated in detail, and the regeneration per-
formance of the catalyst was also investigated. This study is
part of the continued effort to fabricate and apply multifunc-
tional catalysts in the one-pot conversion of cellulose to HMF
in excellent yield.
[
16]
hierarchical porous catalysts. Moreover, Takagaki et al. and
[
17]
Peng et al. made a significant contribution to improving the
catalytic performance in the one-pot synthesis of HMF directly
from cellulose. In their study, an acid–base bifunctional catalyst
was successfully designed and firstly applied to the cellulose
transformation with high HMF yields generated. It has been
demonstrated that a catalytic system that contains both acidic
and basic active sites is more beneficial for HMF production
from cellulose than acidic catalysts alone, as basic catalysts
show advantageous properties for the isomerization process Results and Discussion
[18]
during the cellulose transformation reaction. Consequently,
exploring simple methods for the fabrication of catalysts with
hierarchically macro-/mesoporous structures and acid–base bi-
functional active sites would be helpful for one-pot cellulose-
to-HMF conversion.
Morphology and hierarchical porous structure of the sam-
ples
In this study, PHs with tailored pore sizes and interconnectivity
were fabricated by tailoring the volume fraction of the internal
phase from 75 to 90%, as described in the Experiment Section.
Figure 1A–D exhibit scanning electron microscopy (SEM)
images of PHs1, PHs2, PHs3, and PHs4, respectively. SEM
images reveal that an open cellular structure exists in all cases,
and the voids are highly interconnected by means of spherical
pore throats. However, in the cases of PHs1 and PHs2, a clear
bimodal cavity-like structure is observed in Figure 1A and B.
The observed bimodal cavity-size distribution was most proba-
bly caused as a consequence of the reduced emulsion stability
and subsequent occurrence of coalescence before the gel
Recently, high internal phase emulsion (HIPE) templating,
a simple and versatile tool, has drawn widespread attention in
a wide range of preparations of cellular macroporous materi-
[19]
als. HIPEs are known as emulsions with an internal phase
volume fraction >74%) and external phase. Poly(HIPEs) with
(
low density and open-cell structure are usually synthesized by
polymerizing the monomers in the external phase of HIPEs
and removing the internal phase. Moreover, the pore structure
and interconnectivity of the poly(HIPEs) can be easily tailored
[20]
by varying parameters in the HIPE template. By using this
method, macroporous polymerized solid acid with an open-
cell structure, interconnecting pores, and strong acidity was
[
22]
point of polymerization. It has been proven that increasing
the volume fraction of the internal phase leads to a highly vis-
cous emulsion, and the emulsion stability should be increased,
which results in an increasing number of smaller droplets, and
[21]
synthesized in our previous study.
Owing to its porosity,
high permeability, and stability in the reaction mixture, the cat-
alyst exhibited comparable activity with a homogeneous cata-
lyst of equal acidity in one-pot cellulose-to-HMF conversion,
and can be reused at least five times without significant loss of
catalytic activity.
[
23]
hence increased droplet contact. Since poly(HIPE) is a replica
of the emulsion templating at the gel point of the polymeri-
zation, it was not surprising that on further increasing the
volume fraction of the internal phase to 85 and 90%, microcel-
lular foam-like PHs3 and PHs4 with relatively uniform pore size
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
109
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 1. SEM images of poly(HIPE) synthesized from emulsion templates
containing internal phases of (A) 75, (B) 80, (C) 85 (inset: magnified SEM
image), and (D) 90 vol%.
Figure 2. TEM images of (A) MSNs (inset: magnified TEM image of the corre-
sponding MSNs, scale bar: 20 nm), (B) MSNs-SO
D) SEM image of PHs3@MSNs-SO H-NH
3 3 2
H, and (C) MSNs-SO H-NH ;
(
3
2
.
distribution were obtained (see inset in Figure 1C). Moreover, it
can be concluded that the average diameter of the large void
which indicates the successful grafting of 3-(mercaptopropyl)-
trimethoxysilane (MPTMS) and (aminopropyl)trimethoxysilane
(APTMS) onto the surface of MSNs. It should also be noted
that particle agglomeration appears in Figure 2C, which was
attributed to the grafted hydrophilic organosilane (i.e., MPTMS
(
D) and pore throats (d) decreased significantly, whereas the
degree of interconnectivity was dramatically enhanced on in-
creasing the volume fraction of the internal phase as shown in
Table 1.
and APTMS) on the MSNs surface. MSNs-NH with only basic
2
active sites and an average diameter at (125Æ5) nm was also
obtained by us as shown in Figure S3 in the Supporting Infor-
mation. In practical applications, the introduction of smaller
pores into a macroporous structure is important. This is be-
cause the presence of macropores allows the mass transfer of
the bulk of the substances like liquids and gases, whereas the
smaller pores create a large internal surface area that greatly
Table 1. Composition of emulsion templates and the parameters of the
resulting poly(HIPE).
[
c]
Entry HIPE Vol. fract. Surf.
Poly(HIPE)
D
[mm]
d
[mm]
I
[a]
[b]
[d]
[e]
[f]
[
%]
[wt%]
[%]
1
2
3
4
Hs1
Hs2
Hs3
Hs4
75
80
85
90
1.0
1.0
1.0
1.0
PHs1
PHs2
PHs3
PHs4
29.3
21.3
14.7
14.2
2.8
3.2
4.3
4.0
9.1
11.3
34.2
28.6
[
24]
enhances the catalytic properties. In this study, PHs3 was
employed as a matrix to integrate with acid–base bifunctional
mesoporous MSNs-SO H-NH . As shown in Figure 2D, the syn-
[
[
a] Volume fraction of the internal phase in the emulsion templates.
b] Surfactant amount with respect to the continuous phase. [c] Relevant
3
2
thesized composite is composed of silica particles and PHs ma-
trixes, thereby establishing that the MSNs-SO H-NH was suc-
poly(HIPE) that was prepared by curing the HIPE for 24 h at 658C.
d] Average void diameter of the poly(HIPE) estimated from the SEM
3
2
[
cessfully introduced onto the framework of macroporous ma-
trixes. In addition, SNs with a diameter of (124Æ6) nm, acid–
base bifunctionalized SNs-SO H-NH with a diameter of (164Æ
images. [e] Average interconnecting pore diameter directly measured
from the SEM images. [f] Interconnectivity of as-prepared poly(HIPE) cal-
culated by using n”(d/D) ”100, in which n is the average number of in-
2
3
2
terconnecting pores per void.
6
) nm, and PHs3@SNs-SO H-NH composite were also success-
3 2
fully obtained (Figure S4 in the Supporting Information).
FTIR spectra of the MSNs, MSNs-NH , MSNs-SO H-NH , PHs3
2
3
2
To strengthen the catalytic activity of the final hierarchical
porous catalysts, MSNs and their acid-functionalized MSNs-
SO H and acid–base bifunctionalized MSNs-SO H-NH were fab-
matrixes, and SPHs3@MSNs-SO H-NH composite are displayed
3 2
in Figure 3A. On comparing Figure 3Aa and c, the existence of
MPTMS and 3-aminopropyltriethoxysilane (APTES) on the sur-
face of MSNs-SO H-NH can be established. The appearance of
3
3
2
ricated. Figure 2A–C shows the corresponding transmission
3
2
À1
electron microscopy (TEM) images of MSNs, MSNs-SO H, and
the vibration peaks assigned to SiÀOH around 955 cm and
3
À1
MSNs-SO H-NH , respectively. Clearly, MSNs with a diameter of
ÀOH at approximately 3200–3500 cm in the FTIR spectrum
3
2
(
100Æ8) nm and random pore structure were synthesized by
of MSNs can be observed (Figure 3Aa), whereas these charac-
teristic peaks disappear in Figure 3Ac owing to the chemical
reaction between the SiÀOH groups of MSNs and alkoxy
groups of organosilanes. Thus, the SiÀOH bonds of the MSNs
surface were changed into new Si-O-Si bonds. The chemical
structure of the PHs3 macroporous monolith before and after
using the ammonia-catalyzed hydrolysis of tetraethyl orthosili-
cate (TEOS) in an aqueous solution (seen inset in Figure 2A).
After the functionalization process of MSNs with organosilane,
the diameter increased to (130Æ5) nm with MSNs-SO H (Fig-
3
ure 2B) and (150Æ5) nm with MSNs-SO H-NH (Figure 2C),
3
2
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
110
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. The porosity properties of the samples.
Entry
Samples
S
BET
D
[nm]
BJH
V
t
[cm g
2
À1 [a]
[b]
3
À1 [c]
[cm g
]
]
1
2
3
4
5
MSNs
195.5
141.8
97.1
0.06
45.4
8.0
14.4
22.8
–
0.21
MSNs-SO
MSNs-SO
SPHs3
3
H
0.14
0.10
–
3
H-NH
2
SPHs3@MSNs-SO
3
H-NH
2
22.6
0.09
[
a] BET specific surface area. [b] Pore diameter calculated from the desorp-
Figure 3. (A) FTIR spectra of (a) MSNs, (b) MSNs-NH
d) PHs3 matrix, and (e) SPHs3@MSNs-SO H-NH composite; (B) elemental
contents of MSNs-SO H-NH and SPHs3@MSNs-SO H-NH composite.
2 3 2
, (c) MSNs-SO H-NH ,
tion branch of the corresponding nitrogen isotherms, by applying the
BJH method. [c] Total pore volume determined at the relative pressure of
(
3
2
3
2
3
2
0.99.
introduction of MSNs-SO H-NH was compared with that of
3
2
Figure 3Ad and e. It can be clearly seen that the characteristic
NH are listed in Table 2. Clearly, the Brunauer–Emmett–Teller
2
À1
2
À1
peak of epoxy groups at 910 cm disappeared after grafting
(BET) surface area (195.5 cm g ) and pore volume
3
À1
MSNs-SO H-NH , which was mainly attributed to the chemical
(0.21 cm g ) of MSNs are higher than those of MSNs-SO H
3
3
2
2
À1
3
À1
2
À1
reaction of ÀNH with epoxy groups. These results also con-
(141.8 cm g , 0.14 cm g ), MSNs-SO H-NH2 (97.1 cm g ,
3
2
3
À1
2
À1
firmed that MSNs-SO H-NH was chemically integrated onto
0.10 cm g ), and SPHs3@MSNs-SO H-NH2 (45.4 cm g ,
3
3
2
3
À1
the PHs3 matrix rather than mere physical integration. After
0.09 cm g ), respectively. Nevertheless, the Barrett–Joyner–Ha-
lenda (BJH) pore diameter (8.0 nm) of MSNs is lower than that
of MSNs-SO H (14.4 nm) and MSNs-SO H-NH (22.8 nm). These
À1
the sulfonation process, a clear new peak around 570 cm as-
sociated with the CÀS bond was clearly observed in Fig-
3
3
2
ure 3Ae, which indicates more sulfonic groups have been suc-
results were mainly attributed to the introduction of organosi-
lane that blocked the mesopores of MSNs and particle agglom-
[25]
cessfully incorporated. Results of elemental analysis of the
MSNs-SO H-NH2 and SPHs3@MSNs-SO H-NH composite are
eration, thereby resulting in larger pores in MSNs-SO H and
3
3
2
3
shown in Figure 3B. Clearly, both of the samples possessed the
elements C, H, S, and N, which suggests the successful chemi-
cal combination of MSNs-SO H-NH with the PHs3 matrix.
MSNs-SO H-NH . After the integration of MSNs-SO H-NH and
3
2
3
2
the PHs3 matrix, BET surface area of the final SPHs3@MSNs-
3
À1
SO H-NH catalyst sharply increased from 0.06 to 45.4 cm g ,
3
2
3
2
3
À1
The nitrogen adsorption–desorption isotherms and the
with a pore volume of 0.09 cm g and mesopores of approxi-
mately 22.6 nm in diameter. The as-prepared SPHs3@MSNs-
SO H-NH catalyst with such a hierarchically macro-/mesopo-
pore-size distribution curves of MSNs (A, a), MSNs-SO H (B, b),
3
MSNs-SO H-NH (C, c), and SPHs3@MSNs-SO H-NH (D, d) are
3
2
3
2
3
2
displayed in Figure 4. Clearly, both of the samples exhibit typi-
cal type IV isotherms with a step capillary condensation in the
relative pressure range of 0.1–0.9, which confirms the meso-
rous structure was particularly intriguing in cellulose-to-HMF
conversion, in which the interconnected macropores would
allow ready mass transfer, and the mesopores along the mac-
roporous walls might have a specific large surface area and
greatly enhance the catalytic activities.
[26]
porosity in these samples. Moreover, the porosity properties
of MSNs, MSNs-SO H, MSNs-SO H-NH , and SPHs3@MSNs-SO H-
3
3
2
3
Chemical composition of the catalysts
The SEM-equipped energy-dispersive X-ray spectroscopy (EDS)
analyses of MSNs-SO H-NH2 and SPHs3@MSNs-SO H-NH2 are
3
3
given in Figure 5. Peaks corresponding to C, N, and S are ap-
parent, which suggests the successful introduction of MPTMS
and APTES onto the surface of MSNs, in accord with the results
of elemental analysis. After the integration of the MSNs-SO H-
3
NH and PHs3 matrix, and the subsequent sulfonation process,
2
the final catalyst of SPHs3@MSNs-SO H-NH gives a more obvi-
3
2
ous S peak as shown in Figure 5B, which was attributed to
more sulfonic groups being successfully introduced. As previ-
ously reported, the strengthened acidic active sites in
SPHs3@MSNs-SO H-NH2 should have beneficial properties in
3
one-pot cellulose-to-HMF conversion due to the catalyst acidity
playing a key role in the cellulose hydrolysis and monosacchar-
[
27]
ide dehydration processes.
The condensed structural formula of SPHs3 (Figure 6A) was
Figure 4. Top: Nitrogen adsorption–desorption isotherms of (A) MSNs,
B) MSNs-SO H, (C) MSNs-SO H-NH , and (D) SPHs3@MSNs-SO H-NH . Bot-
tom: Pore-size distribution curves of (a) MSNs, (b) MSNs-SO H, (c) MSNs-
SO H-NH , and (d) SPHs3@MSNs-SO H-NH
(
3
3
2
3
2
13
determined by using C cross-polarization (CP)/magic-angle
spinning (MAS) NMR spectroscopy as shown in Figure 6D. The
3
3
2
3
2
.
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
111
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
3 2 3 2
Figure 5. EDS of (A) MSNs-SO H-NH and (B) SPHs3@MSNs-SO H-NH .
Figure 6. Top: The condensed structural formulas of (A) SPHs3, (B) MSNs-
Figure 7. XPS survey spectrum in the binding-energy range of 0–1000 eV (A)
and high-resolution spectra of C1s (B, C), N1s (D), S2p (E), and O1s (F) of
13
NH
SPHs3@MSNs-NH
C7,8,9,10 with catalyst SPHs3@MSNs-NH
2
, and (C) MSNs-SO
3
H; (D) C solid-state NMR spectra of SPHs3 and
H; (E) NMR spectra curve-fitting result of signals
-SO H.
2
-SO
3
3 2 3 2
MSNs-SO H-NH and SPHs3@MSNs-SO H-NH .
2
3
(
285.4 eV), which were mainly attributed to the successful in-
signal detected at d=143.1 ppm corresponds to carbon atoms
bonded to sulfonic acid groups. And the as-prepared
SPHs3@MSNs-SO H-NH composite reveals almost the same
troduction of organosilane onto the surface of MSNs and inte-
gration of MSNs-SO H-NH and the PHs3 matrix. Notably, as
3
2
shown in Figure 7C for the partial enlarged view of C1s with
SPHs3@MSNs-SO H-NH catalyst, a new peak at 287.8 eV was
3
2
13
signals in the C NMR spectra except for two new signals as
3
2
shown in Figure 6E. First, the signal detected at d=44.9 ppm
observed and assigned to the C=O group, derived from the
polymerized glycidyl methacrylate monomer in the PHs3
is assigned to carbon atoms of CÀNH (C9 in Figure 6B), which
2
[
28]
indicates the appearance of other acid functionalities.
matrix. As shown in Figure 7D, the peak at 400.9 eV corre-
Second, another signal, detected at d=55.0 ppm in the
sponds to N1s, which suggests that basic amino groups exist-
13
C NMR spectrum, was assigned to the carbon atoms of CÀ ed in MSNs-SO H-NH . However, the signal of N 1s in
3
2
SO H (C10 in Figure 6C), thereby establishing the presence of
SPHs3@MSNs-SO H-NH shifts from 400.9 to 400.58 eV, which is
3 2
3
[29]
the acid functionality. These observations confirmed the suc-
cessful synthesis of the acid–base bifunctional and hierarchical-
ly meso-/macroporous SPHs3@MSNs-SO H-NH catalyst, which
ascribed to the chemical reaction of ÀNH with epoxy groups.
2
Additionally, the S2p spectra consisting of two peaks located
at 167.8 and 152.8 eV, which correspond to S2p1/2 and S2p3/2
in Figure 7E, respectively, establish the presence of sulfonic
groups in MSNs-SO H-NH due to the binding energy of S2p
3
2
was consistent with the results obtained from FTIR spectro-
scopic analysis.
3
2
[
30]
The chemical composition of MSNs-SO H-NH2 and
being sensitive to the acidic strength. High-resolution XPS
spectra of O1s for MSNs-SO H-NH2 and SPHs3@MSNs-SO H-
3
SPHs3@MSNs-SO H-NH and their chemical status were ana-
3
2
3
3
lyzed by X-ray photoelectron spectroscopy (XPS), and the re-
sults are shown in Figure 7. The survey spectrum gave signals
mainly associated with C1s, O1s, N1s, S2p, and Si2p species
for both MSNs-SO H-NH and SPHs3@MSNs-SO H-NH , which
NH was located at a binding energy of 531.8 eV (Figure 7F).
2
Acidic and basic properties of the catalysts
3
2
3
2
indicated that the catalysts were composed of C, O, N, S, and
Si elements. The high-resolution XPS spectra of the C 1s region
of MSNs-SO H-NH and SPHs3@MSNs-SO H-NH are shown in
Temperature-programmed FTIR spectroscopy using pyridine as
a Lewis base is one of the most important analytical tools for
characterizing the acidic properties of catalysts. Figure 8 pres-
ents the pyridine-desorption FTIR spectra of MSNs-SO H-NH ,
3
2
3
2
Figure 7B,C. Both of the two samples possess characteristic
3
2
peaks of CÀC (284.6 eV), C=C (285.0 eV), CÀO, and CÀN
SPHs3, and SPHs3@ MSNs-SO H-NH at 25, 50, 150, and 2508C,
3 2
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
112
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
3 2
Figure 8. Pyridine-desorption FTIR spectra of (A) MSNs-SO H-NH , (B) SPHs3,
Figure 9. NH
3
-TPD curves of (A) MSNs-SO
3
H-NH
2
and (B) SPHs3@MSNs-SO
3
H-
and (C) SPHs3@MSNs-SO
3
H-NH
2
at (a) 25, (b) 50, (c) 150, and (d) 2508C.
NH ; CO -TPD curves of (C) MSNs-SO
2
2
3
H-NH
2
and (D) SPHs3@MSNs-SO H-NH .
3
2
À1
respectively. A band at 1450 cm , which is assignable to ab-
sorbed pyridine at the Brønsted acidic sites, is observed for all
three catalysts. Upon increasing the temperature to 2508C,
SPHs3 and SPHs3@MSNs-SO H-NH exhibit a decrease in inten-
SPHs3@MSNs-SO H-NH had
a
higher total acidity of
3
2
À1
À1
3.10 mmolg and basicity of 0.45 mmolg , respectively. Cata-
lyst SPHs3@SNs-SO H-NH was calculated to possess a total
3
2
À1
À1
acidity of 3.0 mmolg and basicity of 0.41 mmolg , which
was comparable to that of PHs3@MSNs-SO H-NH (Figure S6 in
3
2
sity of this band caused by desorption of surface-bound pyri-
dine from the catalyst surface. This behavior indicates the pres-
ence of weakly chemisorbed pyridine, which was hydrogen-
3
2
the Supporting Information).
[
31]
bonded to the Brønsted acid sites of the ÀSO H groups.
3
Conversion of cellulose to HMF
In addition to the pyridine IR spectroscopy results, which
show evidence of Brønsted acid sites in the catalyst, the cata-
lyst acid and alkaline densities were evaluated by temperature-
programmed desorption (TPD) of NH and CO . Ammonia and
The performances of catalysts SPHs1, SPHs2, SPHs3, and SPHs4
were determined by evaluating the HMF yields from cellulose
in a one-pot reaction. All reaction steps were repeated three
times as depicted in the Experimental Section, and the average
yield of HMF was calculated. First, SPHs3 (40 mg) was used to
degrade cellulose to HMF at temperatures ranging from 110 to
1508C. Reaction times ranging from 10 to 60 minutes were
used to optimize the reaction conditions and to maximize the
HMF yield. As shown in Figure 10A, the HMF yields gradually
improve as the temperature increases from 110 to 1508C
within the first 10 minutes. The highest HMF yield of 32.4%
was obtained at 1208C within 10 minutes. This excellent cata-
lytic performance was mainly attributed to the high porosity
and permeability of catalyst SPHs3. However, when the reac-
tion time was further extended to 60 minutes, the HMF yields
at all temperatures decreased. Furthermore, it was evident that
the highest HMF yield of 37.5% was obtained at 1208C and
3
2
carbon dioxide desorption experiments were performed at
a linear heating rate of 108C per minute up to a temperature
of 4008C as the thermogravimetric analysis (TGA) of the cata-
lyst indicated a significant mass loss at higher temperature
(
Figure S5 in the Supporting Information). Figure 9 lists the
typical NH -TPD and CO -TPD curves of MSNs-SO H-NH and
3
2
3
2
SPHs3@MSNs-SO H-NH , respectively. Desorption peaks are ob-
3
2
served at various temperatures. Among the desorbed NH or
3
CO molecules, those appearing at ꢀ1508C correspond to the
2
physically adsorbed NH or CO , whereas those appearing at
3
2
higher temperatures correspond to molecules released from
[
32]
acidic or basic sites. Each desorption peak (except for those
appearing at ꢀ1508C) in Figure 8 was integrated to measure
the corresponding acid or alkaline density as described in the
Experimental Section. Table 3
shows the corresponding acidic
Table 3. The acidic and basic features of as-prepared catalysts.
and basic features of the as-pre-
pared catalysts in detail. It is
evident that after the function-
alization of MSNs with MPTMS
and APTES, the obtained MSNs-
Entry
Sample
S content
Acidity
[mmolg
N content
Basicity
[mmolg ]
À1 [a]
À1 [b]
À1 [a]
À1 [c]
[mmolg
]
]
[mmolg
]
1
2
3
4
MSNs-SO
MSNs-SO
SPHs3
3
H
0.36
0.18
3.04
3.12
0.45
0.23
2.86
3.10
–
1.04
–
–
1.06
–
3
H-NH
2
SO H-NH2 possessed
a total
3
SPHs3@MSNs-SO
3
H-NH
2
0.49
0.45
À1
acidity of 0.23 mmolg and ba-
À1
[a] Calculated by elemental analysis. [b] Measured by using the NH
CO -TPD method.
3
-TPD method. [c] Measured by using the
sicity of 1.06 mmolg . More-
2
over, the final catalyst of
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
113
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 10. (A) Effect of reaction time at different temperatures on HMF
yields from cellulose catalyzed by SPHs3, and (B) HMF yields obtained with
varied catalyst loading.
Figure 11. (A) Effect of reaction time at different temperature on HMF yields
from cellulose catalyzed by MSNs-SO H-NH , and (B) HMF yields obtained
3
2
with varied catalyst loading.
3
0 minutes. This observation suggested that an increase in re-
SO H-NH could be mainly attributed to the cooperative cataly-
3 2
action time or temperature may result in the formation of
some undesired byproducts such as soluble polymers and in-
sis of both the acid and base functional groups in the catalyst.
Until now, the process of cellulose-to-HMF conversion has
been thought to consist of four main stages, cellobiose-to-oli-
gomer dissolution, oligomer-to-glucose depolymerization, glu-
cose-to-fructose isomerization, and fructose-to-HMF dehydra-
[
33]
soluble humins. Although these byproducts were difficult to
indentify using the present techniques, their generation was
confirmed by the color changes of the reaction mixture (Fig-
ure S7 in the Supporting Information), which coincided with
[
7]
tion. Cellulose, glucose, and fructose were adopted as sub-
strates to directly assess the effects of the catalyst acid and
base strengths on HMF yields in different reaction systems. Re-
ports in the literature suggested that introduction of dimethyl
sulfoxide (DMSO) into aqueous solution as a cosolvent can sig-
[11]
the results of Wang’s study. Therefore, subsequent experi-
ments were performed over 1208C and 30 minutes.
Figure 10B shows the optimized amount of catalysts SPHs1
through SPHs4 in terms of the maximum yield of HMF. The cat-
alyst loading amounts varied from 10 to 50 mg, while the
other reaction parameters were constant: T=1208C, t=30 mi-
nutes. The results suggest that higher interconnectivity led to
better catalytic performance. For instance, SPHs3 produced the
greatest HMF yield of 37.5% with the least amount of catalyst
[
35]
nificantly increase the HMF yield in carbohydrate conversion.
Thus, DMSO/H O (2.0 mL, 1:1 v/v) was employed as the solvent
2
with glucose or fructose as the substrate, while the other reac-
tion parameters were held constant (T=1208C, t=30 minutes,
catalyst amount=40 mg). It is evident from Figure 12A that
the HMF yields and fructose conversion are greatly promoted
in the presence of SPHs3, compared with the catalytic system
containing no catalyst. However, a comparable HMF yield of
20.6% and fructose conversion percentage of 35.3% obtained
in the blank catalytic system were attributed to the facile con-
(
40 mg) owing to its high interconnectivity. The relative poor
catalytic performance of SPHs1 and SPHs2 was caused primari-
[34]
ly by their broad pore-size distribution. Therefore, PHs3 was
employed as the matrix to integrate acid–base bifunctional
mesoporous MSNs-SO H-NH .
3
2
Cellulose was used directly as a reactant in the cellulose-to-
HMF conversion carried out in the ionic liquid (IL)-based
system in the presence of MSNs-SO H-NH catalyst. Figure 11A
version of fructose to HMF in the DMSO/H O system at high
2
temperature. Other groups have reported the same observa-
[
17]
tion. Clearly, the HMF yield and selectivity obtained in the
3
2
reveals the influence of the reaction time at different tempera-
presence of acid–base bifunctional SPHs3@MSNs-SO H-NH2
3
tures on HMF yields from cellulose catalyzed by MSNs-SO H-
(88.0% HMF yield and 88.8% HMF selectivity) and acid SPHs3
(85.1% HMF yield and 86.7% HMF selectivity) were similar to
each other. These results indicated that an acid catalyst was
necessary for the dehydration of fructose to produce HMF and
the presence of base made no sense in promoting HMF yield
and selectivity.
3
NH . The same trend was seen with catalyst SPHs3, and HMF
2
yields gradually improved as the temperature increased from
110 to 1408C within the first hour. However, as the time was
further extended to 60 minutes, the HMF yields at all tempera-
tures decreased owing to the formation of byproducts. The
highest HMF yield with 31.2% was obtained at 1208C and two
The yields of HMF converted from glucose in the presence
hours. The optimized reaction temperature with MSNs-SO H-
of acid–base bifunctional SPHs3@MSNs-SO H-NH2 and acid
3
3
NH was longer than that of macroporous SPHs3, which was
SPHs3 catalysts are shown in Figure 12B. It is clearly seen that
the cases with SPHs3@MSNs-SO H-NH catalysts exhibited the
2
mainly attributed to the lower permeability of the mesoporous
structure.
3
2
highest HMF yield (66.9%) and HMF selectivity (72.3%), in con-
trast to those of SPHs3 (HMF yield 51.4% and selectivity
62.2%). The conversion of glucose to HMF involves two steps:
glucose-to-fructose isomerization and fructose-to-HMF dehy-
dration. From the results of fructose-to-HMF, we have con-
firmed that HMF can be easily converted from fructose in our
reaction system in the presence of acid catalysts. Therefore,
Figure 11B shows the optimal catalyst amounts of acid-,
base-, acid–base-functionalized, or unmodified MSNs for the
maximum yields of HMF. The catalyst loading amounts varied
from 10–50 mg, while the other reaction parameters were con-
stant: T=1208C, t=2.0 h. Clearly, the acid–base bifunctional-
ized MSNs-SO H-NH2 catalyst gave a higher HMF yield at
3
3
1.2% than that obtained with acidic, basic, or blank MSNs cat-
the improvement of SPHs3@MSNs-SO H-NH2 catalytic per-
3
alysts, respectively. The enhanced HMF yield (31.2%) of MSNs-
formance can be explained by the existence of basic active
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
114
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 13. (A) Recyclability of SPHs3@MSNs-SO
cellulose-to-HMF; (B) SEM images of reused SPHs3@MSNs-SO
five runs; (C) XPS spectrum of SPHs3@MSNs-SO H-NH after regeneration,
and (D) elemental analysis of reused SPHs3@MSNs-SO H-NH after five runs
inset: the acidity and basicity of reused SPHs3@MSNs-SO H-NH calculated
by elemental analysis).
3
H-NH
2
in the conversion of
3
H-NH after
2
Figure 12. HMF yield, substrate conversion, and HMF selectivity with cata-
lysts SPHs3 and SPHs3@MSNs-SO H-NH in (A) one-pot fructose-to-HMF,
B) glucose-to-HMF, (C) fructose-to-HMF transformation.
3
2
3
2
3
2
(
(
3
2
sites, which are beneficial for the isomerization of glucose to
[
18]
fructose.
maintained during the catalytic process. The superior recycla-
bility of SPHs3@MSNs-SO H-NH was the result of the superior
Previous study on fructose-to-HMF and glucose-to-HMF con-
versions confirmed that acidic active sites play an important
role in the dehydration of fructose to HMF and that basic
active sites enhance the isomerization of glucose to fructose.
Acid catalysts can facilitate the oligomer-to-glucose depolyme-
3
2
thermal stability of its acidic sites and polymer network, which
is an important property in practical applications. The above
results suggest that, to some extent, the heterogeneous cata-
lysts have great potential to be effectively separated and
reused for dehydration reactions.
[36]
rization step in the cellulose-to-HMF process.
Therefore,
SPHs3@MSNs-SO H-NH with both acidic and basic active sites
To establish the structural characteristics of the recycled cat-
3
2
was more beneficial for HMF production from cellulose (HMF
yield of 44.5% and HMF selectivity of 49.6%) than acidic cata-
lyst SPHs3 alone, as shown in Figure 12C. Notably, SPHs3@SNs-
alyst, an XPS spectrum of SPHs3@MSNs-SO H-NH2 that had
3
been used for four cycles is recorded as shown in Figure 13C.
Compared with the XPS spectrum of the fresh catalyst shown
in Figure 7A, strong C 1s, O 1s, N 1s, S 2p, and Si 2p signals are
observed, which indicates the superior chemical stability of as-
prepared SPHs3@MSNs-SO H-NH during the repeated catalytic
SO H-NH2 with a macroporous structure produced HMF in
3
a yield of 40.8% from cellulose under the optimized condi-
tions, lower than that of SPHs3@MSNs-SO H-NH , thereby es-
3
2
3
2
tablishing that the designed hierarchically macro-/mesoporous
structure played a key role in improving the HMF yield. In
a control experiment, an IL-based reaction system without cat-
alyst produced an HMF yield of only 1.2%, which indicated the
high catalytic performance of the as-prepared hierarchically
macro-/mesoporous SPHs3@MSNs-SO H-NH catalyst.
process. Elemental analysis for the catalyst after the fourth run
shows that the S content decreased by 0.42% (Figure 13D).
À1
The acidity and basicity remained at 3.09 and 0.37 mmolg ,
respectively. Results of the inductively coupled plasma-atomic
emission spectrometry (ICP-AES) measurements demonstrate
that the sulfur concentration in the final products after the
3
2
fifth run with catalyst SPHs3@MSNs-SO H-NH2 was
3
À1
0
.02 mgmL , which resulted from the leaching of ÀSO H
3
Recyclability of the catalysts
groups.
To make an evaluation of the catalyst recyclability, the cellu-
lose-to-HMF conversion was studied over four cycles using
SPHs3@MSNs-SO H-NH . All reaction conditions were the same
Conclusion
3
2
as those of the fresh catalyst experiments mentioned in this
study. Figure 13A shows that the mole yields of HMF remained
at about 39.6%, which was a value comparable to that of the
fresh catalyst, until the fourth run. No significant decrease in
HMF yield was observed till the fourth run, which means that
the immobilized ÀSO H and ÀNH functional groups were not
A hierarchically macro-/mesoporous SPHs@MSNs-SO H-NH2
3
catalyst containing both acidic and basic sites has been suc-
cessfully synthesized through the chemical combination of
MSNs-SO H-NH with macroporous PHs derived from water-in-
3
2
oil (W/O) HIPE templating, followed by a sulfonation process.
The newly developed catalyst was used for the sustainable
synthesis of HMF from renewable cellulose in an IL-based sol-
vent system, and an enhanced HMF yield of 44.5% was ob-
tained under optimized conditions. Moreover, the as-prepared
3
2
leached extensively during the process. SEM images of reused
SPHs3@MSNs-SO H-NH after four runs (Figure 13B) reveals
3
2
that catalysts with a microcellular foamlike structure were well
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
115
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
catalyst was also effective for the conversion of glucose and
Synthesis and functionalization of mesoporous silica nano-
particles
fructose to HMF in a DMSO/H O system, which enabled maxi-
2
mum HMF yields of 66.9 and 88.0%, respectively. The en-
hanced catalytic activity toward one-pot carbohydrate conver-
sion for HMF synthesis was attributed to the designed hier-
archically macro-/mesoporous structure and acid–base bifunc-
tions. This study represents a continuation of efforts to make
stable polymeric solid acids and provides results that will be
valuable for future research directed toward the development
of new reactors and their application in biomass conversion
and bioenergy.
Mesoporous silica nanoparticles (MSNs) with random pore struc-
ture were synthesized by using the ammonia-catalyzed hydrolysis
of TEOS in aqueous solution. Briefly, CTAB (0.91 g) and ammoni-
[37]
um fluoride (NH F, 1.5 g) were dissolved in DI water (250 mL). This
4
solution was heated to 808C, and of TEOS (4.5 mL) was slowly
added to it. The mixture was stirred continuously at 808C until the
hydrolysis was completed. The prepared white powder was further
dispersed into the solution of ethanol (200 mL) and hydrochloric
acid (HCl, 37.5%; 4.0 mL) and heated at reflux at 908C for 24 h.
This procedure was repeated two times to make sure that the sur-
factants were completely removed. The obtained products were
centrifuged and washed with DI water. To functionalize MSNs with
acid and base groups, the organosilane, that is, MPTMS and
APTMS, were grafted onto the surface of MSNs. The typical proce-
dure was carried out as follows: as-prepared MSN (1.0 g) samples
were dehydrated at 708C for 4.0 h under a nitrogen flow, and
heated at reflux with APTES (1.2 mL) at 1208C in toluene (20 mL)
for 16 h. The obtained products (MSNs-SH) were filtered, washed
with a mixture of diethyl ether and dichloromethane (1:1, v/v), and
dried at 508C for 24 h in air. Then, MSNs-SH was oxidized to
Experimental Section
Chemicals and reagents
Divinyl benzene (DVB), glycidyl methacrylate (GMA), azodiisobutyr-
onitrile (AIBN), tetraethyl orthosilicate (TEOS, 98%), 3-(mercapto-
propyl)trimethoxysilane (MPTMS, 99%), 3-aminopropyltriethoxysi-
lane (APTES, 99%), hexadecyltrimethyl ammonium bromide (CTAB,
[
38]
become MSNs-SO H by using a modified published procedure.
3
9
9%), dichloromethane (CH Cl ), diethyl ether, N,N-dimethylforma-
2 2
Typically, MSNs-SH (0.5 g) was added to the mixture of 30% hydro-
gen peroxide (10 mL), DI water (10 mL), and methanol (10 mL). The
mixture was stirred at room temperature for 12 h. After that, the
resultant precipitate was collected by filtration, washed with DI
water several times, and dried under vacuum. The resulting sample
mide (DMF), pure HMF, and 1-ethyl-3-methyl imidazolium chloride
[EMIM]Cl) were purchased from Aladdin (Shanghai, China). Hyper-
(
mer 2296 was kindly supplied by Foshan Excel Chemical Co., Ltd.
À1
(
Fujian, China). Deionized water (DI, 18.2 mLcm ) was obtained
through a Milli-Q water-purification system. All other chemicals
and solvents were of reagent grade or better, and purchased from
Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).
was named MSNs-SO H. For the preparation of bifunctional MSNs
3
exhibiting both acid and base groups (denoted as MSNs-SO H-
3
NH ), APTMS was grafted onto the presynthesized MSNs-SO H by
2
3
using the same grafting process as previously mentioned. In the
comparison experiments, silica nanoparticles (SNs) and acid–base
bifunctionalized SNs-SO H-NH2 were also prepared in this study
3
Preparation of macroporous poly(HIPE) (PH) matrixes
(the synthesis procedure is shown in the Supporting Information).
As shown in Scheme 2, DVB (3.0 mL), GMA (1.0 mL), AIBN (0.05 g),
and Hypermer 2296 (0.05 g) were dissolved in toluene (1.0 mL) in
a round-bottomed flask with gentle stirring to form a continuous
phase. After the dropwise addition of DI water (28.33 mL) as the
dispersing phase, a stable W/O HIPE was obtained (Figure S1 in the
Supporting Information). The obtained HIPE was then transferred
to a sealed glass mold, and polymerized in a 508C oven for 24 h.
Thereafter, the polymerized monolith was removed from the
molds, dried in a vacuum at 508C for 24 h, and extracted in Soxhlet
apparatus with acetone to remove any impurities. Finally, the resul-
tant polymer foams were dried to a constant weight in a vacuum
at 508C. In this study, a series of PHs that differed in their volume
fraction of the internal phase in the emulsion templates was pre-
pared (see Table 1, entries 1–4 for the detailed composition of the
emulsion templates).
Integration of bifunctional silica particles and PH matrixes
The integration of bifunctional silica particles and PH matrixes was
completed through the chemical reaction between amine and
epoxy groups. Briefly, MSNs-SO H-NH or SNs-SO H-NH (0.2 g) was
3
2
3
2
added to DMF (100 mL), and then macroporous PHs monoliths
(1.0 g) were added. The mixture was stirred at 458C for 24 h under
a nitrogen flow. After cooling to room temperature, the mixture
was filtered and washed with ethanol and DI water twice. After
drying at 508C for 48 h, hierarchically macro-/mesoporous
PHs@MSNs-SO H-NH or PHs@SNs-SO H-NH was obtained. In the
3
2
3
2
subsequent sulfonation process, PHs@MSNs-SO H-NH
or
3
2
PHs@SNs-SO H-NH2 (0.5 g) was dispersed in 98% sulfuric acid
3
(50 mL). After stirring for 12.0 h at 708C, the sample was washed
with a large amount of water until the filtrate was neutral and
then dried at 508C for 24 h. The obtained catalysts were denoted
as SPHs@MSNs-SO H-NH2 and SPHs@SNs-SO H-NH , respectively.
3
3
2
For comparison, sulfonated poly(HIPE) (SPH) was also synthesized
by using the same sulfonation process, with all other parameters
kept the same.
Catalyst characterization
A confocal laser scanning microscope (CLSM, Leica, TCSSP5II, Ger-
many) was used for fluorescence testing. Morphology investiga-
tions were analyzed by means of field emission SEM, which was
Scheme 2. Schematic illustration for the synthesis of macroporous matrixes
by high internal phase emulsion polymerization.
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
116
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
carried out with an electron microscope equipped with a field
emission electron gun and EDS. The morphology of the MSNs was
characterized by TEM. BET specific surface areas, total pore vol-
umes, and the average pore widths of the catalysts were deter-
polymer, which is an essential step for the conversion of cellulose
and opens up the possibility for subsequent catalytic transforma-
[39]
tions. Experiments were carried out in a 25 mL graduated Pyrex
glass tube immersed in an oil bath preheated at the required tem-
perature using a hot plate with a digital magnetic stirrer. The cellu-
lose conversion included two steps of pretreatment and catalytic
reaction. In a typical experiment for pretreatment, cellulose (0.1 g)
was added to [EMIM]Cl (2.0 g), and the whole mixture was heated
mined from the adsorption and desorption isotherms of N using
2
a Micromeritics TriStar II 3020 instrument. Prior to characterization,
the samples were degassed under a N stream for 12.0 h. The spe-
2
cific surface area of the samples was calculated by applying the
À1
BET theory to the nitrogen-adsorption data within the P/P range.
at 1208C and stirred at 800 rmin for 0.5 h, so as to dissolve the
0
Pore-size distributions were calculated from the desorption branch
of the corresponding nitrogen isotherms, by applying the BJH
method. TGA of samples was performed with a Netzsch Simultane-
ous Thermal Analyzer STA 449C (crucibles: aluminum from
cellulose. For a typical reaction step, the catalyst (40 mg) was
added into the cellulose/[EMIM]Cl solution while maintaining heat-
ing and stirring at 800 rmin under the optimized conditions. The
conversion process for glucose and fructose was the same as that
for cellulose except exclusion of the pretreatment step, and the
À1
Netzsch) under a nitrogen atmosphere and a heating rate of
À1
1
08Cmin from 30 to 8008C. The organic functional groups of the
solvent was replaced by DMSO/H O (1:1 v/v).
2
catalyst were characterized with solid-state nuclear magnetic reso-
nance spectroscopy (Bruker DSX-400WBNMR spectrometer, NMR/
Solid 400). The XPS measurements were carried out on a Thermo
ESCALAB 250 with an accuracy of 0.3 eV. AlKa radiation was used as
the radiation source (1486.6 eV), and the recorded spectra were
calibrated by the characteristic binding energy (BE) peak at
Analysis methods
HMF was analyzed by a 1200 Agilent high-performance liquid chro-
matography (HPLC) instrument equipped with an Agilent TC-
C18(2) column (4.6”250 mm, 5.0 mm) and UV detector at 283 nm.
During this process, the temperature of the column remained con-
stant at 258C, and the mobile phase was methanol/water (7:3, v/v)
2
84.9 eV belonging to the C 1s region. For the Fourier transform
infrared (FTIR) experiment, a mixture of the sample and the
vacuum-dried IR-grade KBr with a weight ratio of 1:100 was press-
ed into a disk, and then recorded with a transmission mode using
a Nicolet NEXUS-470 FTIR apparatus (USA) equipped with a DTGS
À1
at a flow rate of 0.7 mLmin , and 22.5 mL of each sample was in-
jected manually. The concentration of HMF was calculated based
on the standard curve obtained with pure HMF (Figure S2 in the
Supporting Information). Fructose and glucose were analyzed by
using the 1200 Agilent HPLC instrument fitted with a refractive
index detector and Biorad column HPX 87H (7.8 mm”300 mm).
During this process, the column temperature remained constant at
658C, while the applied mobile phase was 0.001m H SO at a flow
À1
detector using 64 scans and a resolution of 4 cm in the range
À1
from 400 to 4000 cm . The elemental composition was recorded
by using a vario EL III elemental analyzer (Elementar, Hanau, Ger-
many). The concentration of sulfur in the final products was mea-
sured by ICP-AES analysis. For pyridine-IR studies, the sample was
saturated with pyridine vapor in a closed vessel at 258C for 2.0 h,
and desorption spectra were recorded at elevated temperatures.
TPD of NH or CO was carried out on a TP 5000-II multiple adsorp-
2
4
À1
rate of 0.55 mLmin . The product yield, selectivity, and conversion
of the substrate were calculated based on Equations (2), (3), and
(4) as follows:
3
2
tion apparatus (Tianjin Xianquan Corporation of Scientific Instru-
ments, China). During the NH -TPD operation, the sample (100 mg)
3
was pretreated in a helium atmosphere at 408C for 1.0 h. When
the mass baseline was stable, the helium flow was stopped, and
NH was introduced until adsorption of the samples was saturated.
Then, purging with helium was performed to remove residual NH3
from the surface of the sample. Subsequently, the sample was
moles of HMF obtained
3
HMF yield ½mol%Š ¼
 100
moles of initial cellulose
ð2Þ
moles of reacted substrate
moles of initial substrate
heated from 40 to 4008C under flowing helium at a rate of
Substrate conversion ½%Š ¼
 100
À1
1
08Cmin for NH3 desorption. The acidity of the catalyst was
measured by the amounts of desorbed NH , which were calculated
based on Equation (1)
ðmoles of initial substrateÞ À ðmoles of final substrateÞ
3
¼
 100
ð3Þ
moles of initial substrate
A
AT
Q ¼ N Â Â 100%
ð1Þ
Q
moles of HMF obtained
HMF selectivity ½%Š ¼
 100
ð4Þ
moles of reacted substrate
À1
in which Q [mmolg ] is the total amount of desorbed NH , and A
N
3
is the corresponding deconvolution area of one NH -desorption
3
peak, A is the total deconvolution area of all the NH -desorption
T
3
À1
peaks, and Q [mmolg ] is the amount of desorbed NH3 corre-
sponding to the TPD peak of relative area A. The basic feature was
Recyclability of catalysts
determined by means of CO -TPD using the same operation pro-
2
cess as that described for NH -TPD above except NH was replaced
with CO2.
3
3
The recycling efficiency of the catalyst was determined according
to the dehydration of cellulose to HMF. After the reaction, the flask
was cooled to room temperature, and the catalysts were separated
from the reaction mixture by filtration and centrifugation, washed
Typical procedure for the catalytic conversion of cellulose
into HMF
1
0 times with a mixture of deionized water and ethanol, and dried
at 808C for 24 h in a vacuum oven. Subsequently, the obtained re-
cycling catalysts were reused for four consecutive cycles to pro-
duce HMF from cellulose under reaction conditions that were the
In this study, the ionic liquid (IL) [EMIM]Cl was employed in the cel-
lulose hydrolysis process to break the b-1,4-glycosidic bonds of the
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
117
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
same as that of the fresh experiments. The HMF yield was deter-
mined from each run.
[14] a) S. H. Lu, Y. Liu, Y. Q. Wang, Chem. Commun. 2010, 46, 634–636;
b) T. Z. Ren, Z. Y. Yuan, A. Azioune, J. J. Pireaux, B. L. Su, Langmuir 2006,
2
2, 3886–3894.
15] a) Z. J. Wang, S. Ghasimi, K. Landfester, K. A. I. Zhang, J. Mater. Chem. A
014, 2, 18720–18724; b) I. I. Ivanova, E. E. Knyazeva, Chem. Soc. Rev.
[
2
Acknowledgements
2013, 42, 3671–3688.
16] A. Takagaki, M. Ohara, S. Nishimura, K. Ebitani, Chem. Commun. 2009,
[
[
6276–6278.
This study was financially supported by the National Natural Sci-
ence Foundation of China (nos. 21107037, A21176107, 21306013),
Natural Science Foundation of Jiangsu Province (nos. BK2011461,
BK2011514), National Postdoctoral Science Foundation (no.
17] W. H. Peng, Y. Y. Lee, C. N. Wu, K. C. W. Wu, J. Mater. Chem. 2012, 22,
23181–23185.
[18] L. Chen, G. Zhu, D. Zhang, H. Zhao, M. Guo, W. Shi, S. Qiu, J. Mater.
Chem. 2009, 19, 2013–2017.
19] I. Capron, B. Cathala, Biomacromolecules 2013, 14, 291–296.
20] V. O. Ikem, A. Menner, T. S. Horozov, A. Bismarck, Adv. Mater. 2010, 22,
3588–3592.
[
[
2013M530240), Postdoctoral Science Foundation funded Project
of Jiangsu Province (no. 1202002B), and Programs of Senior
Talent Foundation of Jiangsu University (no. 12JDG090).
[21] J. M. Pan, H. P. Gao, Y. L. Zhang, J. Zeng, W. D. Shi, C. H. Song, Y. S. Yan,
L. B. Yu, D. D. Chang, RSC Adv. 2014, 4, 59175–59184.
[
[
[
[
[
22] N. R. Cameron, D. C. Sherrington, L. Albiston, D. P. Gregory, Colloid
Polym. Sci. 1996, 274, 592–595.
23] V. O. Ikem, A. Menner, A. Bismarck, Angew. Chem. Int. Ed. 2008, 47,
Keywords: biomass · mesoporous materials · nanoparticles ·
polymers · sustainable chemistry
8277–8279; Angew. Chem. 2008, 120, 8401–8403.
24] F. Iskandar, A. B. D. Nandiyanto, K. M. Yun, C. J. Hogan, K. Okuyama, P.
Biswas, Adv. Mater. 2007, 19, 1408–1412.
25] J. Scaranto, A. P. Charmet, S. Giorgianni, J. Phys. Chem. C 2008, 112,
[
1] a) G. Tsilomelekis, T. R. Josephson, V. Nikolakis, S. Caratzoulas, ChemSus-
Chem 2014, 7, 117–126; b) S. Dutta, K. C.-W. Wu, Green Chem. 2014, 16,
9443–9450.
4
615–4626.
2] J. M. R. Gallo, D. M. Alonso, M. A. Mellmer, J. A. Dumesic, Green Chem.
013, 15, 85–90.
3] P. T. Do, J. R. McAtee, D. A. Watson, R. F. Lobo, ACS Catal. 2013, 3, 41–
6.
26] D. Y. Zhao, J. L. Feng, Q. S. Huo, N. Melosh, G. H. Fredrickson, B. F.
Chmelka, G. D. Stucky, Science 1998, 279, 548.
27] S. Dutta, S. De, I. Alam, M. M. Abu-Omar, J. Catal. 2012, 288, 8–15.
28] Y. Yang, X. Liu, X. B. Li, J. Zhao, S. Y. Bai, J. Liu, Q. H. Yang, Angew. Chem.
Int. Ed. 2012, 51, 9164–9168; Angew. Chem. 2012, 124, 9298–9302.
29] B. Rµc, A. Molnµr, P. Forgo, M. Mohai, I. Bertóti, J. Mol. Catal. A: Chem.
[
[
2
[
[
4
[
[
4] M. Balakrishnan, E. R. Sacia, A. T. Bell, Green Chem. 2012, 14, 1626–1634.
5] a) M. C. Li, Q. L. Wu, K. L. Song, Y. Qing, Y. Q. Wu, ACS Appl. Mater. Inter-
faces 2015, 7, 5006–5016; b) S. M. Coman, M. Verziu, A. Tirsoaga, B.
Jurca, C. Teodorescu, V. Kuncser, V. I. Parvulescu, G. Scholz, E. Kemnitz,
ACS Catal. 2015, 5, 3013–3026.
[
[
[
[
[
2
006, 244, 46–57.
30] A. T. Rodriguez, M. Chen, Z. Chen, C. J. Brinker, H. Fan, J. Am. Chem. Soc.
006, 128, 9276–9277.
31] R. W. Stevens, S. S. C. Chuang, B. H. Davis, Appl. Catal. A 2003, 252, 57–
4.
2
[
[
6] D. Klemm, B. Heublein, H. P. Fink, A. Bohn, Angew. Chem. Int. Ed. 2005,
7
4
4, 3358–3393; Angew. Chem. 2005, 117, 3422–3458.
32] C. Ravindra Reddy, Y. S. Bhat, G. Nagendrappa, B. S. Jai Prakash, Catal.
Today 2009, 141, 157–160.
33] C. Flora, R. Franck, P. Catherine, C. Amandine, G. Emmanuelle, E. Nadine,
Appl. Catal. B 2011, 105, 171–181.
34] D. J. Mihalcik, W. B. Lin, ChemCatChem 2009, 1, 406–413.
35] a) J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. Int. Ed. 2007,
7] a) S. Dutta, S. De, B. Saha, I. Alam, Catal. Sci. Technol. 2012, 2, 2025–
2
1
036; b) J. B. Binder, R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979–
985.
[
[
8] X. H. Qi, M. Watanabe, T. M. Aida, R. L. Smith Jr., Cellulose 2011, 18,
327–1333.
9] a) I.-J. Kuo, N. Suzuki, Y. Yamauchi, K. C.-W. Wu, RSC Adv. 2013, 3, 2028–
[
[
1
4
6, 7164–7183; Angew. Chem. 2007, 119, 7298–7318; b) Y.-Y. Lee,
2
3
034; b) Y.-Y. Lee, S. Dutta, K. C.-W. Wu, ChemSusChem 2014, 7, 3241–
246.
K. C. W. Wu, Phys. Chem. Chem. Phys. 2012, 14, 13914–13917; c) Y. Yang,
C. W. Hu, M. M. Abu-Omar, Bioresour. Technol. 2012, 116, 190–194.
36] Y. Yang, C. W. Hu, M. M. Abu-Omar, Green Chem. 2012, 14, 509–513.
37] W. Stçber, A. Fink, E. Bohn, J. Colloid Interface Sci. 1968, 26, 62–69.
38] C. Gill, B. Price, C. Jones, J. Catal. 2007, 251, 145–152.
[
10] a) F. Zhang, C. Liang, X. T. Wu, H. X. Li, Angew. Chem. Int. Ed. 2014, 53,
498–8502; Angew. Chem. 2014, 126, 8638–8642; b) R. Rinaldi, F.
[
[
[
[
8
Schüth, Energy Environ. Sci. 2009, 2, 610–626; c) S. Dutta, H.-M. Kao,
K. C.-W. Wu, APL Mater. 2014, 2, 113314–113320.
39] J. X. Long, X. H. Li, B. Guo, F. R. Wang, Y. H. Yu, L. F. Wang, Green Chem.
[
[
[
11] J. J. Wang, W. J. Xu, J. W. Ren, X. H. Liu, G. Z. Lu, Y. Q. Wang, Green Chem.
2012, 14, 1935–1941.
2
011, 13, 2678–2681.
12] M. G. Mazzotta, D. Gupta, B. Saha, A. K. Patra, A. Bhaumik, M. M. Abu-
Omar, ChemSusChem 2014, 7, 2342–2350.
13] B. K. Gorityala, J. M. Ma, K. K. Pasunooti, S. T. Cai, X. W. Liu, Green Chem.
Manuscript received: August 6, 2015
Accepted Article published: August 25, 2015
Final Article published: September 18, 2015
2011, 13, 573–577.
ChemPlusChem 2016, 81, 108 – 118
www.chempluschem.org
118
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim