Phosphotungstic acid heterogenized by assembly with pyridines for efficient catalytic conversion of fructose to methyl levulinate

Article abstract of DOI:10.1039/c8ra02278j

Solid acid-catalyzed sugar degradation has been considered to be an efficient approach to synthesize alkyl levulinates (which can be used as fuel additives and surfactants). However, those catalytic processes typically involve harsh reaction conditions and high cost for catalyst preparation. We prepared a series of phosphotungstic acid organic hybrids by a simple solvothermal method, and used them as heterogeneous catalysts for the selective degradation of fructose to methyl levulinate (ML) in methanol with high efficiency under mild reaction conditions. The catalysts were characterized systematically, and the effects of different substituents in pyridine, reaction temperature/time, catalyst dose, and fructose concentration studied. The 3-FPYPW hybrid prepared from 3-fluoropyridine and phosphotungstic acid exhibited superior catalytic activity for the synthesis of ML (82.5%) from fructose (97.8%). A possible reaction pathway was proposed. In addition, the catalyst could be separated from the reaction mixture readily, and reused without remarkable loss of reactivity.

Full text of DOI:10.1039/c8ra02278j

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Phosphotungstic acid heterogenized by assembly
with pyridines for efficient catalytic conversion of
fructose to methyl levulinate†
Cite this: RSC Adv., 2018, 8, 16585
a
a
a
a
a
Chengjiang Fang, Yan Li, Wenfeng Zhao, Weibo Wu, Hu Li, Chao Heb
and Song Yang
*
a
*
Solid acid-catalyzed sugar degradation has been considered to be an efficient approach to synthesize alkyl
levulinates (which can be used as fuel additives and surfactants). However, those catalytic processes
typically involve harsh reaction conditions and high cost for catalyst preparation. We prepared a series of
phosphotungstic acid organic hybrids by
a simple solvothermal method, and used them as
heterogeneous catalysts for the selective degradation of fructose to methyl levulinate (ML) in methanol
with high efficiency under mild reaction conditions. The catalysts were characterized systematically, and
the effects of different substituents in pyridine, reaction temperature/time, catalyst dose, and fructose
concentration studied. The 3-FPYPW hybrid prepared from 3-fluoropyridine and phosphotungstic acid
exhibited superior catalytic activity for the synthesis of ML (82.5%) from fructose (97.8%). A possible
reaction pathway was proposed. In addition, the catalyst could be separated from the reaction mixture
readily, and reused without remarkable loss of reactivity.
Received 15th March 2018
Accepted 23rd April 2018
DOI: 10.1039/c8ra02278j
rsc.li/rsc-advances
for the avoring and fragrances or as substrate for different
types of condensation and addition reactions.13–16 Therefore,
1. Introduction
the development of efficient catalytic systems for direct
conversion of fructose to ML is important.
With the depletion of fossil resources and progressive envi-
ronmental damage, abundant and renewable resources have
been explored to replace non-renewable petroleum sources for
the sustainable supply of fuel and chemicals.1–5 In contrast to
other renewable energies such as solar, wind and hydro power,
biomass is the only feedstock that provides sustainable organic
carbon resources for both biofuels and chemicals.6–10 In this
respect, fructose, as an ample and inexpensive six-carbon sugar
molecule, has been employed widely for the synthesis of bio-
based products. For example, the loss of three water mole-
cules from fructose catalyzed by an acid can yield 5-hydrox-
ymethylfurfural, followed by alcoholysis to give alkyl
levulinates. Among these levulinates, methyl levulinate (ML)
has been identied as an important bio-platform compound
with several excellent features, such as non-toxicity, high
lubricity, ashpoint stability and good ow properties.11,12 ML
has been used as an additive for gasoline and diesel, as well as
Various acidic catalysts (e.g., SO42ꢀ/TiO2, H-USY(6), HPW,
sulfated TiO2–ZrO2 and 5-Cl-SHPAO) have been reported to be
efficient for the synthesis of ML from fructose.17–21 In partic-
ular, phosphotungstic acid (HPW) exhibits pronounced cata-
lytic activity for the conversion of fructose to ML. Nevertheless,
HPW has drawbacks, such as complete solubility in polar
solvents, difficulty in separation, equipment corrosion, and
environmental pollution. In general, the immobilization of
HPW to prepare heterogeneous counterparts has been
demonstrated to be an effective approach to overcome these
drawbacks. For instance, a ML yield of 73.7% has been ob-
tained from fructose in methanol using Fe-HPW-1 as a catalyst
at 130 ꢁC and 2 MPa for 2 h, and the ML yield dropped to 60%
aer repeat use of the catalyst ve times.22 Lai and co-workers
reported on the synthesis of ML from fructose in methanol at
180 ꢁC for 1 h in the presence of 4-HPWFe-MMTSi catalyst: the
ML yield reached 74%, and this catalyst exhibited good reus-
ability.23 Although they have excellent catalytic performance,
these catalysts have their own drawbacks, such as high
temperature and pressure, the high cost of preparation, and
several synthetic steps.
aState Key Laboratory Breeding Base of Green Pesticide & Agricultural Bioengineering,
Key Laboratory of Green Pesticide
& Agricultural Bioengineering, Ministry of
Education, State-Local Joint Laboratory for Comprehensive Utilization of Biomass,
Center for Research & Development of Fine Chemicals, Guizhou University, Guiyang,
Guizhou 550025, China. E-mail: jhzx.msm@gmail.com; hli13@gzu.edu.cn; Fax:
+86(851)8829-2170; Tel: +86(851)8829-2171
As a promising alternative, the assembly of HPW with basic
organic species is an effective way to solidify the homogeneous
acid and “tune” its acidity. For example, a series of acidic ionic
bSchool of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou
510006, P. R. China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c8ra02278j
liquid catalysts (e.g., [MIMBS]3PW12O40, [MIMPS]3PW12O40
,
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 16585–16592 | 16585

View Article Online
RSC Advances
Paper
Fig. 1 XRD patterns of (a) HPW, (b) HPYPW, (c) 3-MeOPYPW, (d) 3-
MeFPYPW and (e) 3-FPYPW.
2.2 Catalyst preparation
Scheme 1 Preparation of various acidic hybrids (PYPWs) by assembly
of HPW with pyridine under solvothermal conditions (schematic).
A 3-FPYPW sample was prepared using a solvothermal method.
In a typical procedure, HPW (1 mmol) was added to absolute
ethanol (32 mL) in a 100 mL stainless-steel autoclave containing
a Teon™-lined reactor, and stirred until complete dissolution
at room temperature to obtain HPW solution, followed by
addition of 3-FPY (1 mmol) with continuous stirring for 1.5 h at
room temperature. The resultant precipitate was aged for 12 h
and [3$2H]3[PW12O40]2) were developed for the heterogeneous
catalytic conversion of fructose into alkyl levulinates,
5-(ethoxymethyl)furfural and 5-hydroxymethylfurfural (HMF),
in high yields (70–99%).24–26 In our previous study, an acidic
hybrid catalyst (PYPW-1) was obtained by self-assembly of
pyridine (PY) with HPW at room temperature in ethanol
solution. It exhibited good reactivity in the catalytic conversion
of HMF to 5-ethoxymethylfurfural (EMF) with 90% yield, but
was nonselective for catalytic degradation of fructose (up to ca.
40% selectivity toward EMF).27
ꢁ
at 90 C, followed by ltration, washing with absolute ethanol,
and drying at 80 ꢁC overnight to give the solid acidic catalyst 3-
FPYPW. For comparison, 3-MeFPYPW, 3-MeOPYPW, and
HPYPW were also synthesized from HPW and the correspond-
ing pyridine with different substituent groups (3-MeFPY, 3-
MeOPY and PY) using the same method, respectively.
In the present work, a series of novel substituted pyridine
phosphotungstates (PYPWs) were prepared using a renewed
but simple solvothermal method (Scheme 1), wherein the
type of substituted groups was correlated directly with the
acid properties of the catalysts. The catalytic performance of
these catalysts was evaluated, and found to be efficient for the
direct conversion of fructose to ML in a one-pot process
under relatively mild reaction conditions. The effects of
different types of catalysts, catalyst dose, and reaction
temperature/time on the conversion fructose to ML, as well
as catalyst reusability, were investigated. Moreover, the
possible pathway on the transformation of fructose into ML
was proposed.
2.3 Catalyst characterization
The X-ray diffraction (XRD) patterns of the catalysts were
recorded on an X-ray diffractometer (D8 Advance) with Cu Ka
radiation (l ¼ 0.1548 nm) and a 2q ranging from 5ꢁ to 80ꢁ.
Fourier transform infrared spectroscopy (FT-IR) was recorded
on
a FT-IR spectrometer (NICOLET iS50) in KBr disks
2. Experimental section
2.1 Materials
3-Fluoropyridine (3-FPY) (98.0%), methyl nicotinate (3-MeFPY)
(99.0%), 3-methoxypyridine (3-MeOPY) (98.0%), pyridine (PY)
(98.0%), fructose (99.0%) and phosphotungstic acid (HPW)
(99.0%) were purchased from Beijing InnoChem Science &
Technology. Other reagents used in this work were purchased
from Beijing Maya Reagents; they were of analytical grade and
not puried, unless stated otherwise.
Fig. 2 FT-IR spectra of (a) HPW, (b) HPYPW, (c) 3-MeOPYPW, (d) 3-
MeFPYPW and (e) 3-FPYPW.
16586 | RSC Adv., 2018, 8, 16585–16592
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
TWIN system (FEI TECNAI) operating at 300 kV. The W
elemental composition of 3-FPYPW catalyst was analyzed by
inductively coupled plasma-optical emission spectrometry (ICP-
OES) on an Optima 5300 DV instrument (PerkinElmer). The
pore size of the 3-FPYPW sample was measured at 77 K on an
ASAP 2460 analyzer (Micromeritics).
Fig. 3 SEM of images 3-FPYPW catalyst.
2.4 Analyses of the acid density of catalysts
The total acid density of solid catalysts was determined by the
sodium ion-exchange method.28 Briey, 0.05 g of solid acid
catalyst was added to 2 mol Lꢀ1 NaCl solution (15 mL) followed
by ultrasonic oscillation for 0.5 h. The resulting mixture was
ltered and washed thrice with 2 mol Lꢀ1 NaCl solution.
Aerwards, phenolphthalein and 0.1 mol Lꢀ1 NaOH solutions
were employed as indicator and titrant, respectively. When the
solution changed from colorless to light-red for 30 s, the
endpoint of titration was reached. An expression for the accu-
rate quantity of acid could be written as:
Fig. 4 TEM images of 3-FPYPW catalyst.
c(H+) ¼ (c(OHꢀ) ꢂ DV)/m
(1)
where c(H+) denotes the acid quantity of the catalyst (mol gꢀ1),
c(OHꢀ) denotes the concentration of the NaOH solution (mol
Lꢀ1), DV denotes the volume of the NaOH solution consumed in
titration (L), and m represents the quality of the catalyst samples
(g). The content of H+ was obtained using the following formula:
containing a specic amount of solid sample (ca. 10 mg).
Scanning electron microscopy (SEM) was carried out using an
electron microscope (FESEM XL-30; Philips). Transmission
electron microscopy (TEM) images were obtained on a G2 F20
(FEI TECNAI) system. Thermogravimetric analysis (TGA) was
conducted using an STA409 instrument in dry air at a heating
. Scanning transmission electron
microscopy-high-angle annular dark-eld (STEM-HAADF)
imaging was done using an aberration-corrected G2 F30 S-
n(H+) ¼ c(H+) ꢂ m
(2)
rate of 10 ꢁC minꢀ1
where n(H+) denotes the molar content of H+ (mmol), c(H+)
denotes the acid quantity of the catalyst (mmol gꢀ1) and m
represents the amount of the catalyst in the reaction system (g).
Fig. 5 (a) STEM-HAADF image and (b) W, (c) P, (d) N and (e) F elemental mappings for 3-FPYPW catalyst.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 16585–16592 | 16587

View Article Online
RSC Advances
Paper
Table 1 Conversion of fructose to ML using various catalystsa
Fructose
(Conv. %)
Catalyst
ML (Y, %)
HMF (Y, %)
DMMF (Y, %)
MMF (Y, %)
TONb
HPW
83.5
82.5
56.4
36.5
16.4
0
0
0.5
2.3
2.0
0
0
8.3
10.8
11.2
0
0
11.6
22.7
11.5
98.3
97.8
87.8
81.6
55.7
8.9
8.1
6.6
5.1
3.6
3-FPYPW
3-MeFPYPW
3-MeOPYPW
HPYPW
a
Reaction conditions: fructose (0.2 mol Lꢀ1), catalyst (90 mg), 5 mL methanol, 120 ꢁC, 10 h. b Turnover number (TON) was evaluated on the basis of
the molar amount of ML/mole effective acid sites.
Technologies) system tted with a Lichrospher C18 column and
UV detector (l ¼ 284 nm). The column temperature was set at
25 ꢁC, and the mobile phase used was acetonitrile/0.1 wt%
acetic acid aqueous solution (30 : 70, v/v) at a ow rate of 1.0
mL minꢀ1. The concentration of fructose was determined by
HPLC using an Aminex HPX-87H column (Bio-Rad Laborato-
ries) and a refractive index (RI) detector. The column tempera-
ture was set at 65 ꢁC, whereas the mobile phase was CH3CN/H2O
(80/20, v/v) at a ow rate of 1.0 mL minꢀ1. ML was analyzed by
gas chromatography using an 7890B (Agilent Technologies)
system tted with a HP-5 column (30 m ꢂ 0.320 mm ꢂ 0.25 mm)
and a ame ionization detector (FID); naphthalene (10 mg) was
added into liquid products as an internal standard. The amount
of byproducts (MMF, DMMF) was measured by gas
Fig. 6 Effect of H+ content on the synthesis of ML from fructose.
chromatography-mass spectrometry (GC-MS) using a 6890N GC/
Reaction conditions: fructose (0.2 mol Lꢀ1), 5 mL methanol, 120 ꢁC,
5973 MS system (Agilent Technologies), and quantied by area
and 10 h.
normalization. The conversion rate (X, mol%) of fructose, and
HMF or ML yield (Y, mol%) were calculated using the following
equations:
2.5 Catalytic conversion of fructose into ML
X (%) ¼ [1 ꢀ (molar concentration of fructose in products)/(molar
The synthesis of ML from fructose was conducted in a 15 mL
concentration of initial fructose)] ꢂ 100%
(3)
Ace pressure tube equipped with a magnetic stirring bar and
placed in an oil bath. In a general synthetic process, 1 mmol
fructose, 90 mg 3-FPYPW catalyst (containing 0.1 mmol H+),
and 5 mL methanol were added to the tube while the reaction
temperature and time were regulated within the desired range.
Aer the reaction, the tube was cooled to room temperature
with owing water. The reaction solution was ltered to remove
solid catalyst, and transferred into a volumetric ask and
diluted with methanol for subsequent quantitative analyses.
Y (%) ¼ (molar concentration of product)/(molar concentration
of initial fructose) ꢂ 100%
(4)
3. Results and discussion
3.1 Catalyst characterization
The XRD patterns of HPW, HPYPW, 3-MeOPYPW, 3-MeFPYPW
and 3-FPYPW are provided in Fig. 1. HPW displayed the typical
cubic secondary structure of a Keggin anion with characteristic
diffraction peaks positioned at 9.9ꢁ, 14.1ꢁ, 17.6ꢁ, 20.3ꢁ, 22.7ꢁ,
24.8ꢁ, 29.1ꢁ and 34.1ꢁ.29 The XRD patterns of 3-FPYPW, 3-MeF-
PYPW, 3-MeOPYPW, and HPYPW exhibited a similar structure
to that of HPW: cubic. The diffraction peaks of 3-FPYPW, 3-
MeFPYPW, 3-MeOPYPW and HPYPW appeared at relatively low
peak intensity when the H+ of H3PW12O40 was substituted by 3-
FPY, 3-MeFPY, 3-MeOPY and PY, respectively. For 3-FPYPW, 3-
MeFPYPW, 3-MeOPYPW and HPYPW catalysts, the intensity of
the diffraction peaks decreased gradually with the change of
2.6 Catalyst recycling
Recycling of 3-FPYPW catalyst was examined in methanol (5
mL), with 3-FPYPW (0.1 mmol H+) and fructose (1 mmol) for
10 h at 120 ꢁC. Aer each cycle of the reaction, the catalyst was
separated by centrifugation from the reaction mixture, followed
by washing thrice with absolute ethanol, and drying for 24 h at
80 ꢁC. The recovered catalyst was used directly for the next cycle
under identical reaction conditions.
2.7 Analytical method
The HMF concentration was analyzed by high-performance substituents, and the order of intensity was 3-FPYPW > 3-MeF-
liquid chromatography (HPLC) using an 1100 (Agilent PYPW > 3-MeOPYPW > HPYPW. The 2q values shied towards
16588 | RSC Adv., 2018, 8, 16585–16592
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Fig. 7 (A) Effect of reaction temperature on fructose conversion. Reaction conditions: 0.2 mol Lꢀ1 fructose, 5 mL methanol, 3-FPYPW (0.1 mmol
H+), and 10 h. (B) Effect of reaction time on fructose conversion. Reaction conditions: 0.2 mol Lꢀ1 fructose, 5 mL methanol, 3-FPYPW (0.1 mmol
H+), and 120 ꢁC.
a lower angle upon introduction of organic ligands (3-FPY, 3- TEM image of 3-FPYPW (Fig. 4(b)), and showed that 3-FPYPW
MeFPY, 3-MeOPY and PY) into HPW, thereby implying an catalyst was crystalline, data that matched closely with XRD
electrostatic interaction between HPW and organic species.
results (Fig. 1(f)). In addition, the distribution of F, N, P and W
Fig. 2 shows the FT-IR spectra of HPW, HPYPW, 3-MeO- elemental mappings of 3-FPYPW were analyzed by STEM-
PYPW, 3-MeFPYPW and 3-FPYPW. HPW and the hybrids HAADF, as shown in Fig. 5(b–e). The results revealed
exhibited four typical absorption peaks of a Keggin structure at a successful exchange of protons for 3-uoropyridine, and these
800–1100 cmꢀ1, which could be assigned to the asymmetric elements were well dispersed. Moreover, the acidity of catalysts
vibrations of the W–Oc (806 cmꢀ1), W–Ob (892 cmꢀ1), W–Od was analyzed by the sodium ion-exchange method. With
(987 cmꢀ1) and P–Oa (1082 cmꢀ1) bonds, respectively. Moreover, enhancement of the absorption capacity of the substituent
the peak at 594 cmꢀ1 was due to the bending vibration of the group, the total acid density of the catalyst also increased. The
P–O band. Absorption peaks at around 1633, 1499, 1262, 1202 acidity of the catalysts was in the order: 3-FPYPW (1.13 mmol
and 3092 cmꢀ1 were attributed to the v(C]C), v(C]N), d(C–H), gꢀ1) > 3-MeFPYPW (0.95 mmol gꢀ1) > 3-MeOPYPW (0.79 mmol
g(C–H) and C–H bands of pyridine molecules, respectively. gꢀ1) > HPYPW (0.51 mmol gꢀ1).
Correspondingly, the peak at 1364 cmꢀ1, 1717 cmꢀ1, 1252 cmꢀ1
and 3232 cmꢀ1 was assigned to the C–O band (Fig. 2(c)), C]O
3.2 Catalyst screening for the conversion of fructose to ML
band (Fig. 2(d)), C–F band (Fig. 2(e)), and surface-adsorbed
The degradation of fructose into ML in methanol was carried
out using various catalysts (Table 1). The soluble catalyst HPW
gave fructose conversion of 98.3% and ML yield of 83.5%. 3-
FPYPW exhibited the best catalytic activity, and 82.5% yield of
ML from fructose (97.8% conversion) was achieved for 10 h at
120 ꢁC. In the presence of the 3-MeFPYPW sample, fructose
conversion and ML yield were 87.8% and 56.4%, respectively.
For the 3-MeOPYPW catalyst, 81.6% fructose conversion and
36.5% ML yield were obtained. For the conversion of fructose to
ML, the range of catalytic activity was HPW (TON ¼ 8.9) > 3-
FPYPW (8.1) > 3-MeFPYPW (6.6) > 3-MeOPYPW (5.1) > HPYPW
(3.6). These results implied that an appropriate increase in the
acid density of the catalyst enhanced its catalytic efficiency for
the synthesis of ML from fructose. The effect of the molecular
structure of the various PY moieties containing different
substituents in the catalysts on the conversion of fructose to ML
was investigated. For HPYPW, 3-MeOPYPW and 3-MeFPYPW
catalysts, the yield of ML increased with the increasing steric
hindrance of substituents. The 3-FPYPW catalyst showed good
catalytic activity in the conversion of fructose to ML because the
–F substituent has strong electronegativity. A possible reason is
that the structure of the catalyst changed due to steric
hindrance or the induction effect of the substituents. To eluci-
date this phenomenon further, a series of catalysts were tested
by N2 adsorption–desorption isotherms. The results showed
water, respectively. Upon introduction of the substituents
(–OCH3, –COOCH3, –F) into pyridine, the characteristic peaks of
pyridine shied to a relatively higher wavenumber, which may
have been due to the conjugation effect or induction effect.
The morphology of the 3-FPYPW catalyst was investigated by
SEM. Fig. 3 shows that 3-FPYPW displayed a smooth poly-
hedron structure. The lattice plane spacing was observed in the
Fig. 8 Effect of fructose concentration on the synthesis of ML from
fructose. Reaction conditions: 5 mL methanol, 3-FPYPW (0.1 mmol
H+), 120 ꢁC, and 10 h.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 16585–16592 | 16589

View Article Online
RSC Advances
Paper
fructose (0.2 mol Lꢀ1), 3-FPYPW (0.1 mmol H+) and 10 h to
further optimize the yield of ML (Fig. 7(A)). When the temper-
ature was increased from 90 to 120 ꢁC, the ML yield rose
correspondingly from 11.8 to 82.5%, with the yields of inter-
mediate products decreasing to 0. However, when the reaction
temperature continued increasing to 140 ꢁC, almost no change
in the conversion of fructose was observed, but the ML yield
began to fall. This may have been due to side-reactions at high
temperature. In addition, the effect of the reaction time on the
yield of ML was investigated under optimized reaction condi-
tions of fructose (0.2 mol Lꢀ1), 3-FPYPW (0.1 mmol H+), and
ꢁ
120 C. As shown in Fig. 7(B), the yield of ML increased from
33.3 to 82.5% by prolonging the reaction time from 2 to 10 h.
Then, the yield of ML was not changed signicantly when the
reaction time reached 12 h. In addition, at a reaction temper-
ature of 140 ꢁC, different reaction times (1 h, 2 h, 4 h, 6 h and 8
h) were selected for investigation (Fig. S1†). The results indi-
cated that with extension of the reaction time from 1 to 8 h, the
yield of ML increased from 43.8 to 80.2%. The maximum ML
yield was comparable with that (82.5%) obtained for 10 h at
Fig. 9 Recyclability of 3-FPYPW in the conversion of fructose to ML.
Reaction conditions: 0.2 mol Lꢀ1 fructose, 5 mL methanol, 3-FPYPW
(0.1 mmol H+), 120 ꢁC, and 10 h.
that the pore sizes of 3-FPYPW, 3-MeFPYPW, 3-MeOPYPW and
HPYPW catalysts were 9.0 nm, 8.4 nm, 6.6 nm and 5.5 nm,
respectively, demonstrating that the molecular structure of the
catalyst was closely related to the yield of ML.
ꢁ
ꢁ
120 C. Based on energy consumption, 120 C and 10 h were,
thus, the optimum reaction temperature and time for the
conversion of fructose to ML, respectively.
3.3 Effect of catalyst dose on the synthesis of ML from
fructose
The effect of catalyst dose (evaluated by H+ content, see Section
2.4) on the conversion of fructose to ML was investigated using
3-FPYPW as catalyst (Fig. 6). The yield of ML increased from
31.4 to 82.5% by increasing the H+ content in 3-FPYPW from
0.01 to 0.1 mmol. Meanwhile, the yields of the intermediates
HMF, MMF and DMMF were reduced gradually to 0 with an
increase in catalyst dose from 0.01 to 0.1 mmol H+. Upon
further increase of the H+ content to 0.14 mmol, the yield of ML
decreased to 80.3% due to side reactions, as claried by GC-MS
(Fig. S3–S6†). Therefore, 3-FPYPW containing 0.1 mmol H+ was
chosen as the optimum amount of catalyst.
3.5 Effect of fructose concentration on the conversion of
fructose to ML
As shown in Fig. 8, the fructose concentration (0.04, 0.2, 0.4, 0.7
and 1 mol Lꢀ1) had a very important role in the synthesis of ML
from fructose. When the substrate concentration was
0.04 mol Lꢀ1, the conversion of fructose was 97.4%, and the ML
yield reached 83.1%. With a decrease in fructose concentration,
the yield of ML and conversion of fructose showed no signi-
cant increasing trends. The ML yield decreased from 82.5 to
48.0% when the fructose concentration increased from 0.2 to
1.0 mol Lꢀ1. These results suggest that relatively higher fructose
concentrations led directly to a decrease in the yield of ML,
3.4 Effect of reaction temperature/time on the conversion of
fructose to ML
The reaction temperature was an important parameter in the which could have been due to a lack of active sites in the cata-
synthesis of ML from fructose. Experiments were carried out at lyst.16,30 Herein, the optimum fructose concentration was set at
four reaction temperatures (90, 100, 120 and 140 ꢁC) using 0.2 mol Lꢀ1
.
Fig. 10 (A) FT-IR spectra and (B) XRD patterns of recycled (after the fourth run) and fresh 3-FPYPW catalysts.
16590 | RSC Adv., 2018, 8, 16585–16592
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Fig. 11 GC spectra at different reaction times for the conversion of fructose to ML.
the reaction time was 2 h, but HDF disappeared. As the reaction
time increased from 2 to 8 h, the amount of ML increased
gradually, and MMF, DMMF decreased gradually. In addition, it
has been reported that HMF is an intermediate in the degra-
dation of fructose to ML.31 According to our experimental
ndings and the literature, the reaction pathway for the acid-
catalyzed product of ML from fructose in methanol is
proposed and the results are shown in Scheme 2. First, the loss
of three water molecules from fructose via 3-FPYPW catalysis
can yield HMF. Second, the hydroxyl or carbonyl group of HMF
is attacked by a methanol molecule to form MMF or HDF, and
reacts further with methanol to form DMMF. Lastly, the inter-
mediate products MMF and DMMF are converted to ML over 3-
FPYPW. In summary, fructose is rst hydrated to HMF, followed
3.6 Catalyst recycling
To investigate the reusability of 3-FPYPW catalyst, a four-cycle
experiment was carried out under optimum reaction condi-
tions. The recovery of 3-FPYPW catalyst was 96.7%; hence the
leaching rate was 3.3% relative to the initial dose of the catalyst
aer reusing four times. To investigate further the nature of the
catalyst, the leaching experiment was undertaken according to
specic steps. Briey, the mixture was cooled to room temper-
ature and ltered aer reacting for 3 h. The ltrate was mixed
further with 90 mg fructose and reacted at the same tempera-
ture (120 ꢁC) for another 1 h. The yield of ML was increased by
only 0.6%. This result showed that only a small amount of 3-
FPYPW was leached into the mixture aer the reaction. The
ltrate mentioned above was tested further by ICP-OES, and
only 0.4% of W in 3-FPYPW was detected. Fig. 9 shows that the
ML yield decreased slightly from 82.5 to 71.2% aer four
consecutive reaction cycles. Fig. 10 shows that the trans-
mittance of FT-IR spectra and intensity of XRD patterns of the
recycled 3-FPYPW catalyst were lower than that of the “fresh”
counterpart, suggesting that the catalyst may have absorbed
organic moieties during the reaction. In addition, the TGA
patterns of the fresh and recovered 3-FPYPW catalysts were
investigated. They conrmed the deposition of organic species
onto the recycled catalyst (Fig. S2†), which possibly resulted in
a loss of catalyst activity.
3.7 Reaction pathway for the conversion of fructose to ML
To understand the reaction pathway on the transformation of
fructose to ML, experiments were undertaken at ve reaction
times (0.5 h, 2 h, 4 h, 6 h and 8 h) using fructose (0.2 mol Lꢀ1),
5 mL methanol, 3-FPYPW (0.1 mmol H+) and 120 ꢁC, and GC-MS
of the reaction mixtures undertaken (Fig. 11 and S3–S6). At
0.5 h, DMMF and 5-(dimethoxymethyl)-2-furanmethanol (HDF)
were detected in the reaction mixture by GC-MS but MMF and
Scheme 2 Proposed reaction pathway for the conversion of fructose
ML were not found. ML, MMF and DMMF were detected when to ML.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 16585–16592 | 16591

View Article Online
RSC Advances
Paper
by interaction with methanol to form ML, which is carried out
efficiently in methanol under mild conditions using 3-FPYPW
catalyst.
9 H. Li, S. Yang, S. Saravanamurugan and A. Riisager, ACS
Catal., 2017, 7, 3010–3029.
10 M. Kamran, Renewable Sustainable Energy Rev., 2018, 82,
609–617.
11 J. Zhang, S. B. Wu, B. Li and H. Zhang, ChemCatChem, 2012,
4, 1230–1237.
4. Conclusions
A series of HPW-based hybrids were prepared using a sol-
vothermal method. These hybrids were applied for the conver-
sion of fructose to ML in a single pot. The total acid density of
the catalyst was measured by the sodium ion-exchange method.
The total acid density of 3-FPYPW was higher than that of 3-
MeFPYPW, 3-MeOPYPW or HPYPW. Among these catalysts, 3-
FPYPW displayed superior catalytic activity for the synthesis of
ML from fructose, and a 82.5% yield of ML was achieved using
fructose (0.2 mol Lꢀ1), 5 mL methanol, and 0.1 mmol H+ for
10 h at 120 ꢁC, which might be ascribed to the relatively higher
acidity of 3-FPYPW. A reaction pathway for the conversion of
fructose to ML was proposed. In addition, the 3-FPYPW catalyst
could be reused for at least four cycles with only a slight
decrease in its activity.
12 L. P. Zhou, H. J. Zou, J. X. Nan, L. Wu, X. M. Yang, Y. L. Su,
T. L. Lu and J. Xu, Catal. Commun., 2014, 50, 13–16.
13 T. vom Stein, M. Meuresch, D. Limper, M. Schmitz,
¨
M. Holscher, J. Coetzee, D. J. Cole-Hamilton,
J. Klankermayer and W. Leitner, J. Am. Chem. Soc., 2014,
136, 13217–13225.
´
´
14 A. Dıaz-Rodrıguez, W. Borz˛ecka, I. Lavandera and V. Gotor,
ACS Catal., 2014, 4, 386–393.
15 D. Gupta and B. Saha, Catal. Commun., 2017, 100, 206–209.
16 X. Tang, Z. Li, X. H. Zeng, Y. T. Jiang, S. J. Liu, T. Z. Lei, Y. Sun
and L. Lin, ChemSusChem, 2015, 8, 1601–1607.
17 L. C. Peng, L. Lin, H. Li and Q. L Yang, Appl. Energy, 2011, 88,
4590–4596.
18 S. Saravanamurugan and A. Riisager, ChemCatChem, 2013, 5,
1754–1757.
19 S. Q. Zhao, G. Z. Xu, C. Chang, S. Q. Fang, Z. Liu and F. G. Du,
Catalysts, 2015, 5, 1897–1910.
20 E. C. Njagi, H. C. Genuino, C. H. Kuo, S. Dharmarathna,
A. Gudz and S. L Suib, Microporous Mesoporous Mater.,
2015, 202, 68–72.
21 F. Yu, R. Y. Zhong, H. Chong, M. Smet, W. Dehaen and
B. F. Sels, Green Chem., 2017, 19, 153–163.
22 Y. Liu, C. L. Liu, H. Z. Wu and W. S. Dong, Catal. Lett., 2013,
143, 1346–1353.
23 F. J. Lai, J. Luo, D. Jiang, T. C. Su and F. Zhang, J. Chem.
Technol. Biotechnol., 2018, 93, 557–568.
Conflicts of interest
There are no conicts to declare.
Acknowledgements
This work had nancial support from the National Natural
Science Foundation of China (21576059 & 21666008), Key
Technologies R&D Program of China (2014BAD23B01), Fok
Ying-Tong Education Foundation (161030), and Guizhou
Science & Technology Foundation ([2018]1037 & [2017]5788).
24 B. Liu, Z. H. Zhang and K. J. Deng, Ind. Eng. Chem. Res., 2012,
51, 15331–15336.
25 J. Z. Chen, G. Y. Zhao and L. M. Chen, RSC Adv., 2014, 4,
4194–4202.
26 Y. S. Qu, C. P. Huang, Y. L. Song, J. Zhang and B. H. Chen,
Bioresour. Technol., 2012, 121, 462–466.
27 Z. W. Wang, H. Li, C. J. Fang, W. F. Zhao, T. T. Yang and
S. Yang, Energy Technol., 2017, 11, 2046–2054.
28 T. T. Liu, Z. L. Li, W. Li, C. J. Shi and Y. Wang, Bioresour.
Technol., 2013, 133, 618–621.
References
1 H. Li, Z. Fang, J. Luo and S. Yang, Appl. Catal., B, 2017, 200,
182–191.
2 B. Liu and Z. H. Zhang, ACS Catal., 2016, 6, 326–338.
3 B. Liu and Z. H. Zhang, ChemSusChem, 2016, 9, 2015–2036.
4 L. T. Mika, E. Csefalvay and A. Nemeth, Chem. Rev., 2018,
118, 505–613.
5 L. Hu, J. X. Xu, S. Y. Xu, A. Y. He, X. Tang, L. Liu, J. M. Xu and
Y. J. Zhao, ACS Catal., 2018, 8, 2959–2980.
6 D. F. Li, W. X. Ni and Z. S. Hou, Chin. J. Catal., 2017, 38,
1784–1793.
7 C. Z. Li, H. L. Cai, B. Zhang, W. Z. Li, G. G. Pei, T. Dai,
A. Q. Wang and T. Zhang, Chin. J. Catal., 2015, 36, 1638–
1646.
´
´
´
´
´
29 J. G. Hernandez-Cortez, L. Martinez, L. Soto, A. Lopez,
´
J. Navarrete, M. Manrıquez and V. H. Lara, Catal. Today,
2010, 150, 346–352.
30 I. Thapa, B. Mullen, A. Saleem, C. Leibig, R. T. Baker and
J. B. Giorgi, Appl. Catal., A, 2017, 539, 70–79.
31 G. Q. Lv, L. L. Deng, B. Q. Lu, J. L. Li, X. G. Hou and
Y. X. Yang, J. Cleaner Prod., 2017, 142, 2244–2251.
8 S. N. Abbas, K. H. Mok, N. Rashid, Y. Xie, M. Ruether,
J. O'Brien and M. Akhtar, Bioorg. Chem., 2016, 66, 41–45.
16592 | RSC Adv., 2018, 8, 16585–16592
This journal is © The Royal Society of Chemistry 2018