Nafion-resin-modified mesocellular silica foam catalyst for 5-hydroxymethylfurfural production from D-fructose

Article abstract of DOI:10.1002/cssc.201200967

Catalytic dehydration of D-fructose to 5-hydroxymethylfurfural (HMF) was investigated over a series of Nafion-modified mesocellular silica foam (MCF) materials. By using an impregnation method, Nafion resin was highly dispersed in the ultra-large pores of the MCFs. Highly efficient and selective dehydration of D-fructose to HMF was achieved in dimethyl sulfoxide solvent; an 89.3 % HMF yield with 95.0 % selectivity was obtained in the presence of the Nafion(15)/MCF catalyst. The effects of reaction temperature, reaction time, and solvent on the dehydration of D-fructose were systematically investigated. The catalyst could be regenerated through an ion-exchange method and a high yield was retained after being used five times. As a heterogeneous catalytic process, a possible reaction mechanism for the dehydration of D-fructose over Nafion-modified MCF catalysts was proposed. Valuable sugars: Nafion-resin-functionalized mesocellular silica foam (MCF) is a strong solid acid that can be prepared through an impregnation method. By using this method, Nafion is well-dispersed in the ultra-large cells of MCFs. This material can easily be reused and exhibits high efficiency and selectivity for the acid-catalyzed dehydration of D-fructose to 5-hydroxymethylfurfural under mild conditions. Copyright

Full text of DOI:10.1002/cssc.201200967

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201200967
Nafion-Resin-Modified Mesocellular Silica Foam Catalyst
for 5-Hydroxymethylfurfural Production from d-Fructose
[
a]
Zhen Huang, Wenya Pan, Haibo Zhou, Feng Qin, Hualong Xu, and Wei Shen*
Catalytic dehydration of d-fructose to 5-hydroxymethylfurfural
HMF) was investigated over a series of Nafion-modified meso-
reaction temperature, reaction time, and solvent on the dehy-
dration of d-fructose were systematically investigated. The cat-
alyst could be regenerated through an ion-exchange method
and a high yield was retained after being used five times.
As a heterogeneous catalytic process, a possible reaction
mechanism for the dehydration of d-fructose over Nafion-
modified MCF catalysts was proposed.
(
cellular silica foam (MCF) materials. By using an impregnation
method, Nafion resin was highly dispersed in the ultra-large
pores of the MCFs. Highly efficient and selective dehydration
of d-fructose to HMF was achieved in dimethyl sulfoxide sol-
vent; an 89.3% HMF yield with 95.0% selectivity was obtained
in the presence of the Nafion(15)/MCF catalyst. The effects of
Introduction
Replacing the diminishing petroleum-based resources and con-
trolling global warming caused by the overuse of fossil fuels
are becoming critical challenges for the current world. To ach-
ieve this, renewable biomass can serve as promising alterna-
tives for the sustainable supply of valuable chemical intermedi-
charides, such as subcritical, supercritical, and microwave-as-
sisted systems, were adopted to improve the yield of
[6,16–19]
HMF.
However, these processes involve expensive ionic
liquids and solvents, high-cost and complicated separation
processes, and, in some cases, high energy consumption
owing to the high temperatures that are required, which
become barriers to commercial applications. In contrast, heter-
ogeneous catalytic processes have advantages such as
straight-forward catalyst separation and catalyst regeneration,
as well as a comparatively low cost, which make them more
[
1]
ates and fuels. For example, 5-hydroxymethylfurfural (HMF)
can be selectively oxidized into 2, 5-furandicarboxylic acid,
which is suggested as an alternative raw material to the ter-
ephthalic or isophthalic acids that are used in the manufacture
[
2]
of polymers. In this context, catalytic processes for the con-
version of biomass has become an attractive research issue
applicative in industrial production. Therefore, many heteroge-
[3–5]
[16,20–22]
over the last decade.
Considering that the manufacturing of
neous catalysts such as acidic ion-exchange resins,
H-
[23,24]
[25,26]
fine chemicals is derived from petroleum processing, the cata-
lytic transformation of sugars or nonedible lignocellulosic ma-
terials into value-added compounds and liquid fuels has great
form zeolites,
sulfated zirconia,
and supported hetero-
[27]
polyacids have been studied for the catalytic dehydration of
hexoses. However, relatively high reaction temperatures or
ionic liquids as co-solvents are necessary to help improve the
catalytic performance of these solid catalysts. Herein, there are
more essential needs for a catalyst that can express effective
catalytic performance under mild conditions and in simple re-
action systems.
[
6–10]
significance for society in the future.
As reported in the lit-
erature, many acidic catalysts have been used for the synthesis
[
1]
of HMF from fructose and glucose. Dumesic et al. and Raines
[
8]
et al. used homogeneous acids such as HCl and H SO to cat-
2
4
alyze the dehydration of d-fructose with a moderate HMF yield
of 40–80% in different solvents [e.g., dimethyl sulfoxide
Nafion perfluorosulfonic acid resin is a strong solid acid
polymer with an acid strength comparable to that of pure sul-
(DMSO), methyl isobutyl ketone (MIBK), and N,N-dimethylaceta-
[28]
mide (DMAc)]. In recent years, more attention has focused on
furic acid.
However, the application of unswollen Nafion
[11–15]
ionic liquids for their excellent catalytic performance.
In addition, some special conditions for the dehydration of sac-
resin is limited because it has a low surface area, which is
2
ꢀ1
about 0.02 m g , whereas the acid density of Nafion is about
ꢀ
1 [28,29]
1
.0 mmolg .
Commercialized Nafion NR50, with a pellet
diameter of 0.6–1.5 mm, was used as a catalyst for the prepara-
[
a] Z. Huang, W. Pan, H. Zhou, F. Qin, Prof. H. Xu, Prof. W. Shen
Department of Chemistry
Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials
and Laboratory of Advanced Materials
Fudan University
Shanghai, 200433 (PR China)
Fax: (+86)21-65641740
E-mail: wshen@fudan.edu.cn
tion of HMF from fructose, and a moderate HMF yield (45%–
[22,30]
68%) was obtained.
Accordingly, many materials with
large surface areas have been used as carriers for Nafion. Over
the past twenty years, Nafion resin has been widely applied in
many organic reactions. Harmer et al. developed a Nafion
resin/silica nanocomposite by using a porous silica network in
[31]
which Nafion particles were entrapped.
The activity of
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201200967.
Nafion resin was enhanced in this material, which could be at-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1063

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
tributed to the increased density
of available acid sites. Similar
advantages
were
observed
when SBA-15, SBA-16, MCM-41,
and other mesoporous silica ma-
terials were used to synthesize
catalysts containing Nafion resin
through impregnation or a one-
[
32–36]
pot synthesis.
In our previ-
ous studies, Nafion-modified
mesoporous materials exhibited
good activity as a catalyst for
the alkylation of isobutane with
1-butene. The materials with
three-dimensional structures
outperformed the ones contain-
ing a one-dimensional channel,
which could be attributed to
the improved diffusion ability of
molecules in a three-dimension-
Figure 1. Nitrogen adsorption–desorption isotherms (left) and SAXS patterns (right) of the MCF and Nafion-modi-
fied MCF materials with different Nafion loadings.
[
33,34]
al mesostructure system.
Among the numerous types of
silica-based porous materials, mesocellular silica foams (MCFs),
with well-defined ultra-large mesopores connected by uniform-
ly sized windows, have unique advantages as catalyst sup-
materials, deduced from the nitrogen isotherms, are summar-
ized in Table 1. MCF showed a steep H1-type hysteresis loop at
high relative pressure, indicating a mesoporous material with
large pore size and narrow pore-size distribution, which was
[
37]
ports.
For example, large molecules are able to diffuse
[46]
through the pores (up to 50 nm in diameter) of MCFs, and the
mass transfer of reactants and products can easily be achieved
substantiated by the SAXS pattern of MCF.
The narrow
[38,39]
because of its open continuous three-dimensional system.
Table 1. Textural properties for Nafion modified MCF mesostructured ma-
terials.
The structural stability and reusability of MCFs have led to
their wide application as catalyst carriers in recent years; how-
[
40–45]
[b]
[a]
[a,b]
[a]
[b]
[a,b]
ever, they are rarely used in acidic catalysis processes.
Catalyst
S
[
BET
2
D
pore
V
pore
D
win
S’BET
V’pore
ꢀ1
3
ꢀ1
2
ꢀ1
3
ꢀ1
m g
]
[nm]
[cm g
]
[nm] [m g
]
[cm g
]
Herein, based upon the acidic properties of Nafion resin and
the ultra-large mesoporous MCFs, a series of Nafion-resin-sup-
ported MCF solid acid catalysts with different Nafion loadings
are reported, which were prepared by using an impregnation
method. Nitrogen adsorption and TEM–energy dispersive X-ray
MCF
323
41.8
28.1
29.8
28.0
2.24
1.63
1.31
0.86
15.8
15.6
15.7
15.5
323
322
326
315
2.24
1.92
1.87
1.56
Nafion(15)/MCF 298
Nafion(30)/MCF 228
Nafion(45)/MCF 173
[
a] Total pore diameter (Dpore) and Vpore were calculated by using the Bar-
(EDX) studies indicated very good dispersion of Nafion nano-
rett–Joyner–Halenda (BJH) adsorption branch of the nitrogen isotherm,
and the window diameters (Dwin) were calculated by using the BJH de-
sorption branch. [b] S‘BET and V‘pore were calculated per gram of MCF,
whereas SBET and Vpore were calculated per gram of materials (i.e., MCF+
Nafion).
spheres in the ultra-large mesopores. These catalysts showed
excellent catalytic performance for the dehydration of d-fruc-
tose to HMF. The effects of Nafion loading, reaction tempera-
ture, and solvent on the catalytic activity were investigated.
Moreover, the catalysts could be recycled and regenerated
through an ion-exchange method, and a high yield was re-
tained after being used for five times.
type IV hysteresis loops and the SAXS patterns of all samples
were maintained after the impregnation of Nafion resin with
different loadings, indicating that the structure of the MCF was
retained after modification. As an increasing amount of the
Nafion resin was introduced into the network of the MCF, the
Brunauer–Emmett–Teller surface area (SBET) and the pore
volume (Vpore) gradually decreased. However, when SBET and
Vpore were recalculated per gram of MCF, it was observed that
Results and Discussion
Preparation and characterization of the catalysts
In the present work, materials were prepared with three differ-
ent Nafion loadings [Nafion(X)/MCF, in which X=15, 30, or
3
ꢀ1
45 wt%] by using the impregnation method, which is de-
Vpore decreased from 2.24 to 1.56 cm g , whereas S was
BET
scribed in detail in the Experimental Section. The nitrogen ad-
sorption–desorption isotherms and small angle X-ray scattering
almost unchanged, as the Nafion loading increased from 15 to
45 wt%. From the pore-size-distribution analysis, the most
probable pore diameter decreased from 42 to about 29 nm,
whereas the size window remained at about 16 nm. It was pre-
(SAXS) patterns of MCF and Nafion/MCF materials are depicted
in Figure 1. The textural parameters of the Nafion-modified
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1064

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
viously reported that Nafion resin can disperse within the sili-
cate lattice in the form of nanospheres with a diameter of 15–
[
36]
2
0 nm. In our studies, the results of the nitrogen sorption
and SAXS analysis indicated that Nafion resin was dispersed
inside the cells of the MCF as nanoparticles, and windows of
MCF remained open even at a relatively high Nafion loading
(e.g., 45 wt%), which resulted in unchanged SBET and decreased
Vpore per gram of MCF. When a cell holds a Nafion nanoparticle,
it is unlikely to accommodate another one. Thus, the peak of
pore-size distributions (Figure S1) in the MCF was centered at
about 29 nm after modification with Nafion resin. With in-
creased Nafion loading, the ratio of pores with 29 nm in diam-
eter increased simultaneously. Compared with Nafion resin/
silica nanocomposite, the MCF had the ability to hold more
Nafion molecules and its unique cells with windows in three
directions could provide unobstructed space for molecular dif-
fusion.
TEM images of Nafion-modified MCF materials in Figure 2
show the open mesostructured network interconnected with
cells of uniform pore size. The structure of the MCF was pre-
served after impregnation of Nafion resin. In addition, the size
of the mesocellular was well matched with that calculated
from nitrogen sorption isotherms. Elemental mapping TEM–
EDX images (Figure 2e–h) of Nafion(30)/MCF clearly showed
that fluorine and sulfur were perfectly dispersed with silicon in
the structure of the synthesized mesoporous materials, which
was also in accordance with the results deduced from nitrogen
adsorption–desorption studies.
The acidity-related properties of the Nafion-modified MCF
materials and several commercial solid acid reference materials
are detailed in Table 2. The Nafion loading was evaluated by
using the elemental analysis method, which demonstrated an
incorporation yield that referred to the percentage of Nafion
resin deposited on MCF. The high incorporation yield, which
ranged from 92% to 94%, was superior to that of Nafion/SBA-
[
34]
1
6 materials. The open mesocellular framework of the MCF
enabled a more efficient diffusion of Nafion nanoparticles. The
large-sized cells with a relatively narrow diameter window al-
lowed independent and stable positioning of these nanoparti-
+
cles. However, the H concentration, determined by using
Figure 2. TEM images of a) MCF, b) Nafion(15)/MCF, c) Nafion(30)/MCF, and
d) Nafion(45)/MCF. EDX images of Nafion(30)/MCF: e) EDX map, and f) sulfur,
g) silicon, and h) fluorine mapping.
Table 2. Acidic-related properties of Nafion-modified catalysts.
Sample
Acidity
catꢀ1 [a]
Nafion incorporation
+
catꢀ1 [b]
[c]
[
mmolH g
]
[mmolSg
]
[wt%]
a potentiometric titration, was similar to the concentration of
sulfur, revealing that the active sites provided by the Nafion
resin were highly accessible.
Nafion(15)/MCF 0.142
Nafion(30)/MCF 0.283
Nafion(45)/MCF 0.432
0.141
0.282
0.415
0.135
–
94
94
92
–
–
–
SAC-13
0.146
3.897
0.339
0.129
Amberlyst-15
2
4
ꢀ
SO
/ZrO
2
–
–
Catalytic performance of Nafion-modified MCF catalysts
Nafion(15)/
–
[
d]
MCF
Nafion-modified MCF materials and several reference solid acid
catalysts were used for the catalytic dehydration of d-fructose
into HMF. Their catalytic performances are summarized in
Table 3. Among these results, an 89.3% HMF yield with 95.0%
selectivity (entry 1) was obtained for Nafion(15)/MCF in DMSO
at 908C after a reaction time of 2 h, which was much better
[a] Titrated potentiometrically with 0.1m NaOH after filtration and thor-
ough washing (see the Experimental Section). [b] Calculated from ele-
mental analysis. [c] Determined by comparing the amount of sulfur de-
posited and introduced. [d] Recycled and regenerated Nafion(15)/MCF
catalyst after five uses.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1065

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
MCF is shown in Figure 3A. The HMF yield increased rapidly to
60% in 30 min, whereas the conversion of d-fructose reached
over 70%. After a reaction time of 2 h, the HMF yield and d-
fructose conversion gradually reached 87% and 96%, respec-
Table 3. Catalytic performance of different catalysts for the dehydration
of d-fructose to HMF in DMSO.
[
a]
[
b]
Entry
Catalyst
HMF yield
Selectivity
[%]
STY
ꢀ
1
ꢀ1
[%]
[molmol h ]
1
2
3
4
5
6
7
8
Nafion(15)/MCF
Amberlyst-15
89.3
26.3
32.3
73.7
78.6
72.4
70.1
95.0
76.2
64.9
81.3
93.1
94.7
89.8
–
16.4
3.7
5.5
11.4
28.2
26.6
25.2
–
2ꢀ
SO
4
/ZrO
2
SAC-13
[
[
[
c]
c]
c]
Nafion(15)/MCF
Nafion(30)/MCF
Nafion(45)/MCF
none
<2
[
0
a] Reaction conditions unless otherwise stated: catalyst (acid amount:
.06 mmol H ), d-fructose in DMSO (3 wt%, 12.5 g), 908C reaction tem-
+
perature, and 2 h reaction time. [b] STY: HMF molar yield per molar unit
+
of H of the catalyst per unit time. [c] Reaction conditions were the same
+
except that the acid amount was 0.03 mmol H .
2
ꢀ
than that of Amberlyst-15, SO4 /ZrO , and SAC-13 (entries 2–4)
2
under the same conditions. Amberlyst-15 is a sulfonic acidic
ion-exchange resin with nanoscale pores (ꢁ25 nm in diame-
2
ꢀ
ter) and comparatively weak acid strength. SO4 /ZrO2 is
a strong solid acid with an amorphous pore structure, and
SAC-13 is a Nafion resin/silica nanocomposite with 13–16 wt%
Nafion entrapped within the amorphous silica porous network.
This result indicated that the higher acid strength and the
three-dimensional, open, ultra-large mesoporous system could
improve the dehydration of d-fructose to HMF in DMSO. The
structure of the MCF materials could promote efficient mass
transfer of the reactants to the active sites and the products
out of the pore system, which resulted in Nafion/MCF materials
of high efficiency in the process.
Figure 3. a) Effect of the reaction time, and b) effect of reaction temperature
on the dehydration of d-fructose. The reaction conditions were similar to
that of Table 3, entry 1 with Nafion(30)/MCF as the catalyst.
tively, and finally reached 90% and 97% after 3 h. This yield
was higher than that of many other solid acid catalysts under
similar conditions, even at a higher temperature. The catalytic
performance of Nafion/MCF for the dehydration of d-fructose
to HMF was outstanding compared to that of the ionic-liquid
[47,48]
CrCl /EMIMCl system.
The effect of reaction temperature
2
with Nafion(30)/MCF is presented in Figure 3B. It can be seen
that the HMF yield and selectivity gradually increased with
temperature when the temperature was below 908C. A high
HMF yield of 88% with a selectivity of 91% was achieved at
908C, and it was nearly unchanged up to 1208C. This reflected
that the Nafion/MCF strong solid acid was an efficient catalyst,
which could dehydrate d-fructose under mild conditions.
The catalytic performances of Nafion/MCF catalysts with dif-
ferent Nafion loadings were also investigated. A lower acid
content (0.03 mmol) was used so that there was a significant
difference. As a result, Nafion/MCF with 15 wt% Nafion loading
was superior to the other catalysts, which could be attributed
to the improved Nafion distribution on Nafion(15)/MCF.
Although the MCF has a large volume to hold the Nafion resin,
some Nafion could adsorb onto the external surface of the
framework. As the loading increased, the relative content of
Nafion on the outside of the framework also increased, leading
to inferior activities. Consequently, the HMF yield dropped
from 78.6% to 70.1% when the Nafion loading was increased
from 15 to 45 wt% (entries 5–7), whereas the space–time yield
Effect of solvent on the activity of d-fructose dehydration
A series of solvents with different boiling points, including iso-
propanol, diacetone alcohol, MIBK, DMAc, and DMSO, were
used to study the effect of the solvent on the activity of d-fruc-
tose dehydration. From the results (Table 4), it was demonstrat-
ed that the best fructose conversion (96.4%) and HMF yield
(87.6%) were achieved in DMSO by using Nafion(30)/MCF and
a reaction time of 2 h at 908C, whereas MIBK presented a mod-
erate HMF yield of 16.4% under the same conditions. Very
poor yields were obtained in isopropanol, diacetone alcohol,
and DMAc. As reported in the literature, DMSO is suitable for
(
STY) demonstrated that the efficiency of catalyst was also de-
ꢀ
1
ꢀ1
creased from 28.2 to 25.2 molmol h . For comparison,
a blank experiment was performed without any catalyst and
the yield of HMF was less than 2% (entry 8). Finally, Nafion/
MCF was tested with a high initial concentration of d-fructose
to simulate a practical application (Table S2). It could be seen
that the HMF yield reached 78.9% and 69.5% in 10 and
[49]
the production of HMF from sugars. It acted as both an elec-
tron acceptor and an electron donator to improve the dehy-
dration of fructose. In isopropanol and diacetone alcohol,
a stable intermediate b-d-fructopyranoside was formed under
the acidic conditions, which could not be further dehydrated
20 wt% fructose solutions, respectively. The advantages of the
Nafion/MCF material highlighted its potential applied value for
the biomass transformation.
The effect of the reaction time on the conversion and HMF
yield for the dehydration of d-fructose catalyzed by Nafion(30)/
[50]
into HMF. DMAc, as a stable and water-miscible aprotic sol-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1066

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[
a]
Table 4. Dehydration of d-fructose in different solvents.
Entry
Solvent
d-fructose
conversion [%]
HMF yield
[%]
1
2
3
4
5
6
isopropanol
diacetone alcohol
MIBK
88.6
74.4
56.8
18.7
96.4
94.7
5.3
8.8
16.4
4.4
87.6
48.5
DMAc
DMSO
MIBK/DMSO
[
b]
[a] Reaction conditions: Nafion(30)/MCF (0.2 g), d-fructose (0.375 g), sol-
vent (12.125 g), at 908C for 2 h. [b] Volume ratio of MIBK/DMSO=5:1.
+
vent, could bind the free H in the solvent (control experi-
ments conducted, see Table S1). It could adsorb on the acid
sites of the catalyst, especially on the strongly acidic Nafion
resin, to inhibit or block the reaction, which resulted in a 4.4%
HMF yield. In contrast, MIBK resulted in a better yield; howev-
er, the dehydration of fructose could be limited by the solubili-
ty of d-fructose in MIBK under relatively mild conditions.
Considering that MIBK was primarily used as an extraction sol-
vent and DMSO was difficult to separate from the products,
a two-component solvent system including MIBK and DMSO
was used for the dehydration reaction, which aimed to opti-
mize the isolation process of HMF by reducing the amount of
DMSO used. The influence of MIBK/DMSO ratios on fructose
conversion and HMF yield was investigated, and when 86% of
DMSO was replaced with MIBK, a moderate HMF yield (48.5%)
was achieved (see Table S4).
Figure 4. Recycling of the Nafion(15)/MCF catalyst in the dehydration of d-
fructose. The reaction conditions were same as that of Table 3, entry 1.
ditions we adopted. As a heterogeneous catalysis process, the
reaction was not only affected by the acidic properties and
structure of the catalyst, but also by the interaction between
the reactant and the solid catalyst. The diffuse reflectance in-
frared Fourier transform (DRIFT) analysis (Figure S7) showed
that the interactions between fructose and the MCF, such as
electrostatic effects and hydrogen bonds from the polar sila-
nol-rich surface of the MCF, could weaken the stability of the
hydroxyl group and CꢀH bonds in fructose. This influence fa-
cilitated the dehydration of d-fructose and would have been
greatly enhanced when Nafion, which is of strong electron-at-
tractive ability, was used in the cells of the MCF. Thus, the de-
hydration of d-fructose was largely promoted on Nafion/MCF
and the catalytic performance of Nafion-modified MCF materi-
als was much better than the other strong solid acidic materi-
als that were investigated. A possible mechanism for the dehy-
dration of d-fructose catalyzed by Nafion-modified MCF com-
posite materials was proposed according to a literature refer-
Investigation on the catalyst recycling
Nafion resin could be recycled easily and regenerated by using
an ion-exchange method. The Nafion/MCF catalyst was used
five times to study the stability of the catalyst; control experi-
ments with SAC-13 as the catalyst were performed under the
same conditions. The method of catalyst regeneration is de-
scribed in detail in the Supporting Information. The results are
presented in Figure 4, and it could be seen that the yield of
HMF decreased slightly (about 6%) over catalyst Nafion(15)/
MCF after being reused five times, whereas the fructose con-
version and HMF yield on SAC-13 were noticeably deceased.
This indicated that Nafion/MCF was a stable solid acid, which
benefited from the high dispersion of nanoparticles in the cells
and the open continuous three-dimensional structure. The un-
obstructed molecule transfer route in the open mesocellular
framework of the MCF enabled Nafion/MCF to be easily regen-
erated. Generally, the decrease in catalytic activity could be at-
tributed to minor leaching of the Nafion resin (less than 10%,
see Table 2) or acid site occupation by the byproducts that
were produced after each run.
[51]
ence (Scheme 1).
The protonation of the hydroxyl group
caused by protonic acidity and the influences on chemical
bonds were both taken into account. Once the fructofuranoses
were formed and moved into the cells of the MCF, the proto-
nation of the hydroxyl group on the saturated carbon could
occur quickly. Because of the strong interaction with Nafion
and silanol groups, loss of the first water molecule could
become more efficient, leading to the formation of enediol in-
termediates. After deprotonation, under the influence of elec-
trostatic forces and/or hydrogen bonding on the CꢀO and Cꢀ
H bonds, the intermediate could continue to lose two water
molecules to form HMF with high efficiency and selectivity.
Moreover, the structural space limitation of MCF was helpful to
inhibit the side reactions, which formed soluble polymers and
insoluble humins in this system.
Possible mechanism for dehydration of d-fructose over
Nafion/MCF catalysts
Conclusions
Based on the above results, the HMF yield in the dehydration
Nafion-resin-modified mesocellular silica foams are firstly pre-
pared through an impregnation method. Within the open
of d-fructose was determined by many factors under the con-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1067

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Scheme 1. Possible mechanism for the dehydration of d-fructose over Nafion-modified MCF materials.
[
52]
three-dimensional MCF framework, Nafion nanoparticles are
highly dispersed in the ultra-large mesopores. In addition, an
Zhao et al.
1
In
a
typical synthesis, pluronic P123 (8.0 g,
.38 mmol) was dissolved in HCl (1.6m, 300 mL) solution, followed
by addition of TMB (8.0 g, 67 mmol) and NH4F (0.092 g,
.48 mmol). After increasing the reaction temperature to 313 K,
TEOS (17.6 g, 84 mmol) was added. The solution was stirred at
08C for 20 h before it was transferred into an autoclave for aging
treatment at 1308C for 24 h. Next, the white precipitates were col-
lected through filtration and dried at room temperature for a fur-
ther 24 h. The template P123 was removed from the as-synthesized
MCF by using a Soxhlet extractor and ethanol over 48 h.
89.3% HMF yield with a high selectivity of 95.0% is obtained
2
from the dehydration of d-fructose when Nafion(15)/MCF is
used as a strong solid acid catalyst in DMSO at 908C for 2 h.
The outstanding catalytic performance is not only ascribed to
the strong acid strength of Nafion but also to the open ultra-
large pore system of the MCF. DSMO provides a more suitable
environment and presents a better performance in the dehy-
dration of d-fructose to HMF. Moreover, the Nafion/MFC cata-
lyst can be easily recycled and regenerated. Based on the re-
sults, a possible mechanism for d-fructose dehydration to HMF
over Nafion-modified MCF solid acid is proposed.
4
The Nafion-modified catalysts were prepared by impregnating
Nafion in the above prepared MCF. The solid was dried at 808C for
2 h before it was mixed with ethanol and Nafion resin solution.
The mixture was heated at 508C to gradually remove the solvents;
the water and alcohols were subsequently evaporated under
vacuum at 608C for 12 h. The resultant materials were denoted as
Nafion(X)/MCF, in which X indicates the theoretical weight percent
of Nafion loading (in this work, three Naifon loadings were studied,
that is, X=15, 30 or 45).
1
Experimental Section
Reagents and materials: Triblock copolymer pluronic P123
(
EO PO EO , M =5800), HMF (used as the standard sample), Am-
20 70 20 W
berlyst-15, SAC-13, and all solvents used in this work were pur-
chased from Sigma–Aldrich. Naifon resin solution (5 wt% Nafion in
water-alcohol solution) was purchased from Dupont. Tetraethyl or-
thosilicate (TEOS), 1,3,5-trimethylbenzene (TMB), d-Fructose, zirco-
nium hydroxide, anhydrous ethanol, concentrated sulfuric acid,
and aqueous HCl were purchased from Shanghai Chemical Co. All
chemicals were used directly without further treatment. Pure water
was obtained through an ultra-pure water system (Millipore,
Catalyst characterization: Nitrogen sorption isotherms were per-
formed by using a Quantachrome Quadrasorb SI analyzer. SAXS
measurements were performed on a NanoSTAR SAXS system with
CuK radiation (l=0.154 nm, 40 kV, 35 mA). TEM images and TEM-
a
EDX element mapping were obtained on a JEOL 2100F microscope
operated at 200 kV. The DRIFTs spectra were recorded on a Thermo
Nicolet 6700 Fourier-transform infrared spectrometer equipped
with a highly sensitive mercury cadmium telluride (MCT) detector,
cooled by using liquid nitrogen. The ion-exchange capacities of
Nafion-modified materials (corresponding to acid site concentra-
tion) were determined through potentiometric titrations with
2ꢀ
1
8.2 MWcm at 298 K). The catalyst SO /ZrO was synthesized as
4 2
[26]
previously reported.
Catalyst preparation: The pure siliceous as-synthesized MCF sample
was prepared following the synthesis procedure described by
[53]
aqueous solutions of NaCl and NaOH.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1068

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Catalytic performance: In a typical reaction protocol for the dehy-
dration of d-fructose, a 3 wt% fructose/DMSO solution containing
d-fructose (0.37 g) was charged into a 50 mL flask, followed by the
addition of the catalyst (0.2 g). The reaction was performed at
[19] A. Dutta, A. K. Patra, S. Dutta, B. Saha, A. Bhaumik, J. Mater. Chem. 2012,
2
2, 14094–14100.
[
[
20] C. Lansalot-Matras, C. Moreau, Catal. Commun. 2003, 4, 517–520.
21] X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Jr., Ind. Eng. Chem. Res. 2008,
4
7, 9234–9239.
22] X. Qi, M. Watanabe, T. M. Aida, J. R. L. Smith, Green Chem. 2009, 11,
327–1331
9
08C for 2 h with magnetic stirring. The point of zero time was
[
[
taken to be when the reaction temperature reached the set point.
After the reaction, the mixture was cooled and placed in a centri-
fuge for 15 min at a rate of 8000 rpm. The upper, clean solution
was extracted and diluted 20 times with water before HPLC analy-
sis. Before each reaction, a control sample for HPLC analysis was
taken from the 3 wt% fructose/DMSO solution and diluted
1
23] C. Moreau, R. Durand, S. Razigade, J. Duhamet, P. Faugeras, P. Rivalier, P.
Ros, G. Avignon, Appl. Catal. A 1996, 145, 211–224.
[24] R. O’Neill, M. N. Ahmad, L. Vanoye, F. Aiouache, Ind. Eng. Chem. Res.
2009, 48, 4300–4306.
[
[
[
25] H. Yan, Y. Yang, D. Tong, X. Xiang, C. Hu, Catal. Commun. 2009, 10,
558–1563.
26] X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Jr., Catal. Commun. 2009, 10,
771–1775.
27] M. A. Schwegler, P. Vinke, M. van der Eijk, H. van Bekkum, Appl. Catal. A
992, 80, 41–57.
1
2
0 times with water.
Product analysis and HPLC measurement: The composition analysis
of the reaction mixture was conducted on a Shimadzu HPLC
system equipped with refractive index detector. The saccharide
and HMF were quantified by using HPLC with a Shodex SH1011
sugar column (300ꢁ8.0 mm, 6 mm), sulfuric acid (0.0005m in
1
1
[
[
28] M. A. Harmer, W. E. Farneth, Q. Sun, Adv. Mater. 1998, 10, 1255–1257.
29] M. Hara, T. Yoshida, A. Takagaki, T. Takata, J. N. Kondo, S. Hayashi, K.
Domen, Angew. Chem. 2004, 116, 3015–3018; Angew. Chem. Int. Ed.
2004, 43, 2955–2958.
ꢀ1
water) as the mobile phase at a flow rate of 0.8 mLmin , and
a column temperature of 508C. Herein, the yield of HMF and the
conversion of d-fructose were obtained by using an external stan-
dard method and then following Equation 1:
[
30] K. Yamaguchi, T. Sakurada, Y. Ogasawara, N. Mizuno, Chem. Lett. 2011,
4
0, 542–543.
31] M. A. Harmer, W. E. Farneth, Q. Sun, J. Am. Chem. Soc. 1996, 118, 7708–
715.
[32] F. Martꢃnez, G. Morales, A. Martꢃn, R. van Grieken, Appl. Catal. A 2008,
47, 169–178.
33] W. Shen, Y. Gu, H. Xu, D. Dubꢄ, S. Kaliaguine, Appl. Catal. A 2010, 377,
–8.
[
7
HMF yield
d-fructose conversion
HMF selectivity ¼
ð1Þ
3
[
[
[
1
34] W. Shen, Y. Gu, H. Xu, R. Che, D. Dubꢄ, S. Kaliaguine, Ind. Eng. Chem.
Res. 2010, 49, 7201–7209.
35] M. Fujiwara, K. Kuraoka, T. Yazawa, Q. Xu, M. Tanaka, Y. Souma, Chem.
Commun. 2000, 1523–1524.
Acknowledgements
This work was supported by STCSM (08DZ2270500,
9DZ2271500).
0
[36] A. Ciftci, N. A. Sezgi, T. Dogu, Ind. Eng. Chem. Res. 2010, 49, 6753–6762.
[
37] P. Schmidt-Winkel, W. W. Lukens, D. Zhao, P. Yang, B. F. Chmelka, G. D.
Stucky, J. Am. Chem. Soc. 1999, 121, 254–255.
Keywords: dehydration
mesoporous materials · nafion · silica foams
·
heterogeneous
catalysis
·
[38] D. Trong On, S. Kaliaguine, J. Am. Chem. Soc. 2003, 125, 618–619.
[39] Y. J. Han, J. T. Watson, G. D. Stucky, A. Butler, J. Mol. Catal. B 2002, 17,
1
–8.
[
[
[
40] E. W. Ping, J. Pierson, R. Wallace, J. T. Miller, T. F. Fuller, C. W. Jones, Appl.
Catal. A 2011, 396, 85–90.
41] G. Zhang, J. Yi, J. Shim, J. Lee, W. Choi, Appl. Catal. B 2011, 102, 132–
[
1] Y. Romꢂn-Leshkov, J. N. Chheda, J. A. Dumesic, Science 2006, 312, 1933–
1937.
[2] P. Gallezot, Chem. Soc. Rev. 2012, 41, 1538–1558.
139.
[3] R. D. Cortright, M. Sanchez-Castillo, J. A. Dumesic, Appl. Catal. B 2002,
42] C. Na-Chiangmai, N. Tiengchad, P. Kittisakmontree, O. Mekasuwandum-
rong, J. Powell, J. Panpranot, Catal. Lett. 2011, 141, 1149–1155.
3
9, 353–359.
4] Y. Romꢂn-Leshkov, C. J. Barrett, Z. Y. Liu, J. A. Dumesic, Nature 2007, 447,
82–985.
5] G. Yong, Y. Zhang, J. Y. Ying, Angew. Chem. 2008, 120, 9485–9488;
Angew. Chem. Int. Ed. 2008, 47, 9345–9348.
[
[
[43] R. Gao, W. Dai, X. Yang, H. Li, K. Fan, Appl. Catal. A 2007, 332, 138–145.
[44] R. Gao, Q. Zhu, W. Dai, K. Fan, Green Chem. 2011, 13, 702–708.
[45] Y. Liu, J. Xu, L. Wang, Y. Cao, H. He, K. Fan, Catal. Lett. 2008, 125, 62–68.
[46] P. Schmidt-Winkel, W. W. Lukens, P. Yang, D. I. Margolese, J. S. Lettow,
J. Y. Ying, G. D. Stucky, Chem. Mater. 2000, 12, 686–696.
9
[
[
6] M. Bicker, J. Hirth, H. Vogel, Green Chem. 2003, 5, 280–284.
7] A. Takagaki, M. Ohara, S. Nishimura, K. Ebitani, Chem. Commun. 2009,
[
[
[
[
47] H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007, 316, 1597–
6276–6278.
1
600.
48] Y. Zhang, V. Degirmenci, C. Li, E. J. M. Hensen, ChemSusChem 2011, 4,
9–64.
49] A. S. Amarasekara, L. D. Williams, C. C. Ebede, Carbohydr. Res. 2008, 343,
021–3024.
[8] J. B. Binder, R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979–1985.
[9] A. Takagaki, M. Takahashi, S. Nishimura, K. Ebitani, ACS Catal. 2011, 1,
5
1
562–1565.
10] S. Hu, Z. Zhang, Y. Zhou, J. Song, H. Fan, B. Han, Green Chem. 2009, 11,
73–877.
11] S. Hu, Z. Zhang, Y. Zhou, B. Han, H. Fan, W. Li, J. Song, Y. Xie, Green
Chem. 2008, 10, 1280–1283.
[
[
[
3
8
50] V. Ferriꢅres, T. Benvegnu, M. Lefeuvre, D. Plusquellec, G. Mackenzie, M. J.
Watson, J. A. Haley, J. W. Goodby, R. Pindak, M. K. Durbin, J. Chem. Soc.
Perkin Trans. 2 1999, 951–960.
51] G. R. Akien, L. Qi, I. T. Horvath, Chem. Commun. 2012, 48, 5850–5852.
52] Q. Li, Z. Wu, D. Feng, B. Tu, D. Zhao, J. Phys. Chem. C 2010, 114, 5012–
12] X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, ChemSusChem 2009, 2, 944–
[
[
946.
[
[
13] X. Tong, Y. Li, ChemSusChem 2010, 3, 350–355.
14] Q. Bao, K. Qiao, D. Tomida, C. Yokoyama, Catal. Commun. 2008, 9,
5019.
[
53] D. Margolese, J. A. Melero, S. C. Christiansen, B. F. Chmelka, G. D. Stucky,
1383–1388.
Chem. Mater. 2000, 12, 2448–2459.
[
[
15] C. Moreau, A. Finiels, L. Vanoye, J. Mol. Catal. A 2006, 253, 165–169.
16] X. Qi, M. Watanabe, T. M. Aida, J. R. L. Smith, Green Chem. 2008, 10,
7
99–805.
17] J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. 2007, 119,
298–7318; Angew. Chem. Int. Ed. 2007, 46, 7164–7183.
18] S. Dutta, S. De, A. K. Patra, M. Sasidharan, A. Bhaumik, B. Saha, Appl.
Catal. A 2011, 409–410, 133–139.
[
[
7
Received: December 17, 2012
Revised: March 14, 2013
Published online on May 13, 2013
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1063 – 1069 1069