Cascade catalysis via dehydration and oxidation: one-pot synthesis of 2,5-diformylfuran from fructose using acid and V2O5/ceramic catalysts

Article abstract of DOI:10.1039/c6ra27678d

2,5-Diformylfuran (2,5-DFF) is an important biomass chemical with broad application prospects. A vanadium pentoxide (V₂O₅)/ceramic powder catalyst (V-CP) was designed and synthesized in this study, and its structure was characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and specific surface area analysis (BET). The catalytic properties of the V-CP catalyst were evaluated. The results show that 5-hydroxymethylfurfural (5-HMF) can be completely converted to 2,5-DFF at 140 °C under atmospheric pressure in the presence of oxygen as the oxidant in dimethyl sulfoxide (DMSO) for 5 h, and the yield of 2,5-DFF was 85.7%. After recycling the catalyst 5 times, the catalytic effect decreased slightly, and the catalyst showed good recovery performance. Furthermore, a one-pot method for the preparation of 2,5-DFF was proposed by using fructose as the raw material and V₂O₅/ceramic-dilute sulfuric acid as the catalyst. The fructose was dehydrated to form 5-HMF, and 2,5-DFF was generated by the V-CP catalytic oxidation of 5-HMF. The 2,5-DFF yield was 68.4%. The results of this study provide a valuable reference for the efficient one-pot synthesis of biomass-based furan chemicals.

Full text of DOI:10.1039/c6ra27678d

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Cascade catalysis via dehydration and oxidation:
one-pot synthesis of 2,5-diformylfuran from
fructose using acid and V O /ceramic catalysts†
Cite this: RSC Adv., 2017, 7, 7560
2
5
a
b
acd
acd
a
Mei Cui, Renliang Huang,* Wei Qi,*
Rongxin Su
and Zhimin He
2,5-Diformylfuran (2,5-DFF) is an important biomass chemical with broad application prospects. A
vanadium pentoxide (V )/ceramic powder catalyst (V-CP) was designed and synthesized in this study,
2 5
O
and its structure was characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD), X-ray
photoelectron spectroscopy (XPS), and specific surface area analysis (BET). The catalytic properties of the
V-CP catalyst were evaluated. The results show that 5-hydroxymethylfurfural (5-HMF) can be completely
ꢀ
converted to 2,5-DFF at 140 C under atmospheric pressure in the presence of oxygen as the oxidant in
dimethyl sulfoxide (DMSO) for 5 h, and the yield of 2,5-DFF was 85.7%. After recycling the catalyst 5
times, the catalytic effect decreased slightly, and the catalyst showed good recovery performance.
Furthermore, a one-pot method for the preparation of 2,5-DFF was proposed by using fructose as the
Received 2nd December 2016
Accepted 17th January 2017
raw material and V
2
5
O /ceramic-dilute sulfuric acid as the catalyst. The fructose was dehydrated to form
5
6
-HMF, and 2,5-DFF was generated by the V-CP catalytic oxidation of 5-HMF. The 2,5-DFF yield was
DOI: 10.1039/c6ra27678d
8.4%. The results of this study provide a valuable reference for the efficient one-pot synthesis of
www.rsc.org/advances
biomass-based furan chemicals.
can be synthesized through the hydrogenation, oxidation,
polymerization, and hydrolysis of 2,5-DFF.
Generally, 2,5-DFF can be prepared by the catalytic oxidation
1
. Introduction
23–25
With the reduction of fossil resources and deterioration of the
global environment, the development of biomass-based liquid of 5-HMF. The main types of catalysts are noble metals,
fuels and high-value chemicals has become the focus of manganese-based catalysts, vanadium-based catalysts, and
2
6
academic and industrial concerns. Among them, the chemical polymetallic catalysts. For noble metals, Zhu et al. used MnO
furan has become an important platform compound,
2
1
–5
microspheres as the carrier for Au catalyst, achieving a 5-HMF
2,5-diformyl- conversion of 82% and a 2,5-DFF selectivity of 99% using DMF
as the solvent. In addition, Nie et al. used different carriers for
etc. These compounds can be Ru in the catalytic oxidation of 5-HMF. The results showed that
6
–11
including 5-hydroxymethylfurfural (5-HMF),
12–15
16–20
27
furan (2,5-DFF),
2,5-furandicarboxylic acid (FDCA),
21,22
2,5-dimethylfuran (DMF),
prepared by the dehydration, oxidation, or reduction of glucose when active carbon was used, the yield of 2,5-DFF can be as high
ꢀ
and fructose generated from the hydrolysis of biomass. There as 96% in toluene at 110 C and 2.0 MPa of O
2
. For manganese-
28
are two aldehyde groups in the 2,5-DFF, which make the based catalysts, Nie et al. investigated the effect of the exis-
chemical very active. Therefore, a variety of useful compounds tence of different forms of manganese on the catalytic oxidation
and polymer materials, such as plastics, fuels, and additives, of 5-HMF, and using manganese oxide octahedral molecular
sieve (OMS-2) as the catalyst and DMF as the solvent aer 1 h at
ꢀ
1
10 C under 0.5 MPa O , a 5-HMF to 2,5-DFF conversion of
2
1
00% and a selectivity of 97.2% were achieved. Furthermore,
a
State Key Laboratory of Chemical Engineering, School of Chemical Engineering and
29
Yadav et al. synthesized a composite catalyst Ag-OMS-2 (silver
content of 15%). This catalyst was used in isopropanol at 165 C
Technology, Tianjin University, Tianjin 300072, China. E-mail: qiwei@tju.edu.cn;
Fax: +86 22 27407599; Tel: +86 22 27407799
ꢀ
b
under an air pressure of 1.52 MPa for 4 h, giving a 5-HMF to 2,5-
Tianjin Engineering Center of Bio Gas/Oil Technology, School of Environmental
Science and Engineering, Tianjin University, Tianjin 300072, China. E-mail: tjuhrl@ DFF conversion of 99% and a selectivity of 100%.
tju.edu.cn
Moreover, vanadium-based catalysts are also used in the
catalytic oxidation of 5-HMF for the preparation of 2,5-DFF due
to their superior catalytic performance and low cost. In homo-
c
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin),
Tianjin 300072, China
d
Tianjin Key Laboratory of Membrane Science and Desalination Technology, Tianjin
24
geneous catalysis, Ma et al. investigated the homogeneous
University, Tianjin 300072, China
oxidation of 5-HMF in acetonitrile using VOSO and Cu(NO )
4
3 2
†
Electronic supplementary information (ESI) available: Supplementary gures.
as catalysts. It provided a 5-HMF to 2,5-DFF conversion of 99%
See DOI: 10.1039/c6ra27678d
7560 | RSC Adv., 2017, 7, 7560–7566
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
and a selectivity of 99%. In heterogeneous catalysis, S ´a daba
30
et al. prepared a vanadium pentoxide (V O )/zeolite catalyst for
2
5
2
,5-DFF preparation, and an 83.2% yield was achieved in
31
dimethylformamide. Antonyraj et al. used a V O /active
2
5
carbon catalyst for the selective oxidation of 5-HMF to 2,5-DFF,
which provided a conversion of 95% and a selectivity of 96%
aer 4 h of reaction under 0.28 MPa of oxygen in methyl isobutyl
ketone. Overall, high yields and selectivities for 2,5-DFF
production were obtained in some studies, specially with high
pressure of O2.
In recent years, some researchers have tried to synthesize
Scheme
1
Preparation of
V O
2 5
/ceramic catalysis and one-pot
2,5-DFF directly from fructose or glucose. Fructose can be synthesis of 2,5-DFF from fructose.
directly dehydrated to prepare 5-HMF, and therefore, the
preparation of 2,5-DFF using fructose as raw material has been
32–34
32
extensively studied.
SBA–SO H for the catalyzed dehydration of fructose to prepare analysis (BET). 2,5-DFF was prepared from 5-HMF using V-CP as
-HMF, followed by K-OMS-2 catalytic oxidation of 5-HMF to a catalyst and O2 as an oxidizing atmosphere under normal
For example, Yang et al. used Fe
3
O
4
–
ray photoelectron spectroscopy (XPS), and specic surface area
3
5
prepare 2,5-DFF. That study used a two-step method for the pressure. The effects of the reaction temperature, reaction time,
preparation of 2,5-DFF from fructose and generated a 2,5-DFF V O content, and solvent on the catalytic performance were
2
5
33
yield of 80%. Liu et al. synthesized a new type of molybdenum- investigated, and the recovery performance of V-CP was also
based polyoxometalates that has dual catalytic functions of evaluated. Furthermore, a one-pot method for the synthesis of
dehydration and oxidation and converted fructose to 2,5-DFF 2,5-DFF from fructose was developed using V-CP and dilute
using a one-pot method. The 2,5-DFF yield reached 69.3% sulfuric acid as the catalyst under normal pressure (Scheme 1).
under the optimum conditions. In the research of glucose as The changes in the contents of fructose, 5-HMF, and 2,5-DFF in
35
raw material, Takagaki et al. studied the preparation of 2,5- the reaction system were analyzed by high performance liquid
DFF from glucose and fructose using cation exchange resin chromatography (HPLC).
(Amberlyst-15) and ruthenium-hydrotalcite (Ru/HT) as cata-
lysts, generating 2,5-DFF yields of 25% and 49%, respectively.
Xiang et al. used CrCl
composite catalysts to prepare 2,5-DFF from glucose using
2
3 3
/NaBr and vanadate NaVO as the
. Experimental
.1 Materials and reagents
36
2
a one-pot two-step method, with a 2,5-DFF yield of 55%. In 5-HMF (>99%), 2,5-DFF (>99%), and fructose (>99%) were ob-
general, the product yield and selectivity using catalysts with tained from the Sigma-Aldrich (Beijing, China). Dimethyl sulf-
dual functionality or multiple catalysts for 2,5-DFF preparation oxide (DMSO, >99.8%) was purchase from the Concord
directly from fructose or glucose, still need further improve-
ment, and difficulty also exists in the reuse of multiple catalysts
in the two-step process.
Technology Co. Ltd. (Tianjing, China). V
silicate (Na SiO , >98%), magnesium oxide (MgO, >98%),
calcium oxide (CaO, >98%), silicon dioxide (SiO , >95%),
aluminum oxide (Al , >98%), and concentrated sulfuric acid
2 5
O (>98%), sodium
2
3
2
Ceramic is one of the ideal carriers for industrial catalytic
2 3
O
reactions due to its excellent physical properties, such as a high were purchased from the Guangfu Institute of Fine Chemicals
melting point, high hardness, and high wear resistance as well as (Tianjin, China).
37
its excellent mechanical properties. It also has the advantages of
corrosion resistance to acid and alkali, oxidation resistance, easy
preparation, low cost, etc. Therefore, ceramics have been chosen
2
.2 Preparation of the V
2 5
O /ceramic catalyst
SiO , Al O , Na SiO , V O , MgO, CaO, and water were mixed in
a mass ratio of 10 : 5 : 5 : 2.6 : 1 : 1 : 1. The mixture was then
dried at 150 C for 2 h. The temperature was increased at
a heating rate of 4 C min , and the mixture was calcinated at
38,39
2
2
3
2
3
2 5
as the catalyst matrix/carrier by many researchers.
Previously,
our group had synthesized various ceramic used as enzyme
ꢀ
38–40
carriers, which exhibited good catalytic performance.
In
ꢀ
À1
addition, our group has been committed to the synthesis of
biomass-based furan chemicals, especially 5-HMF synthesis,
ꢀ
8
50 C for 4 h. Aer cooling down to room temperature and
soaking in a 3% sulfuric acid solution for 2 h, the catalyst was
washed and dried to obtain the V-CP with 10 wt% of V
To compare and examine the effect of the catalyst composi-
tion on the properties of the catalysts, ceramic catalysts with the
V O contents of 0 (ceramic matrix, CP), 1%, 5%, 15%, and 20%
8,11,14,41
regarding which much work has been carried out.
The
2 5
O .
heterogeneous catalyst, which was prepared by doping ceramics
with metal or metal oxide, has the catalytic properties of metal
elements while retaining the advantages of ceramics. It should be
an excellent catalyst for the synthesis of furan chemicals.
2
5
were prepared using the same process as described above.
In this paper, a new V
prepared by co-sintering catalytic active V
matrix component. The chemical composition and micro-
structure of the catalyst were systematically characterized by X-ray diffraction (XRD, D/MAX-2500, scanning range from 10–
2
O
5
/ceramic catalyst (V-CP) was
2 5
O
with a ceramic
2
.3 Characterization of catalysts
ꢀ
ꢀ
À1
scanning electron microscopy (SEM), X-ray diffraction (XRD), X- 90 at a scanning speed of 6 min , Rigaku, Japan) was used to
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 7560–7566 | 7561

View Article Online
RSC Advances
Paper
characterize the structure of blank ceramic matrix (CP) and V- (SEM), as shown in Fig. 1. SEM images showed that the CP and
CP. Scanning electron microscopy (SEM, S4800, Hitachi, V-CP morphologies were similar. A large number of hexagonal
Japan) images of the catalysts were acquired with an acceler- prisms can be clearly seen. In addition, there were some
ating voltage of 5 kV to examine their morphologies. The amorphous components attached to the surfaces of the hexag-
specic surface area of the catalyst was measured using a BET onal prisms.
nitrogen adsorption apparatus (F-sorp 2400, Beijing Jin Aipu
XRD was used to verify the crystal structures of CP and V-CP.
Technology Co. Ltd.). The elemental composition and chemical Fig. 2 shows that both CP and V-CP have obvious diffraction
ꢀ
valance on the surface of the catalyst were analyzed by X-ray peaks, and particularly, there is a strong absorption peak at 27 .
photoelectron spectroscopy (XPS, PHI-5000 Versa Probe, Analysis using MDI Jade soware showed that the XRD patterns
ULVAC-PHI, Japan).
of the two were very close, and the characteristic peaks of SiO2
hexagonal crystals exist. These results indicated that the basic
lattice structure of the ceramic was retained aer the incorpo-
2
.4 Catalytic performance test
.4.1 Preparation of 2,5-DFF by the catalytic oxidation of 5-
HMF. Four milliliters (mL) of DMSO, 63 mg of 5-HMF, and
00 mg of V-CP were added to the reaction vessel and placed in an
ration of V
aer the incorporation of V
However, SiO and Al , as the main constituents of ceramic,
2
O
5
. The peak intensities of MgO and CaO decreased
2
2 5
O
due to their content decrease.
2
2 3
O
1
still exhibit very high diffraction peak intensities in the V-CP
oil bath at a set temperature. The reaction was initiated by
passing oxygen at 40 mL min into the reactor. Aer completion
of the reaction, the catalyst was collected by centrifugation and
washed three times using water. The reaction solution and
washing solution were collected, and the volume was adjusted to
pattern. This is consistent with other ceramic catalysts synthe-
À1
39
sized in our laboratory. In addition, the diffraction peaks of
V2O5 were not detected in the XRD pattern of V-CP, which
indicated that V O existed mainly in an amorphous form.
2
5
2 5
The XPS spectra of the V O /ceramic catalysts and the rela-
50 mL. The product was analyzed using high-performance liquid
tive atomic ratios, as well as the binding energies of the major
elements, are shown in Fig. 3 and Table 1. The results showed
that the binding energies of the major elements, Si, Al, Mg, Ca,
and Na, in CP and V-CP are basically the same, which indicated
that each element has the same chemical valence in CP and V-
CP. From Fig. 3b, it can be seen that there are characteristic
peaks at 517.7 eV and 524.2 eV corresponding to V2p3/2 and
V2p1/2 orbitals, respectively, which indicates that the vanadium
chromatography (HPLC). The effects of the solvent composition,
reaction temperature, reaction time, ow rate of O , and V O
2
2 5
loading amount on the conversion and yield were investigated.
.4.2 Preparation of 2,5-DFF by the one-pot catalytic
2
conversion of fructose
Route 1. 4 mL of DMSO, 90 mg of fructose, 11 mL of concen-
trated sulfuric acid, and 100 mg of V-CP (10 wt%) were added to
the reaction vessel, which was placed in an oil bath at 140 C to
initiate a dehydration reaction. Aer 20 min, the oxidation reac-
ꢀ
exists as +5 in V
results from Antonyraj et al., who used active carbon as
2 5
O . This is consistent with the XPS spectra
31
À1
tion was conducted by passing oxygen at 40 mL min into the
reactor. Aer completion of the reaction, the catalyst was collected
by centrifugation and washed three times using water. The reac-
tion solution and washing solution were collected, and the volume
was adjusted to 50 mL. The product was analyzed using HPLC.
Route 2. 4 mL of DMSO, 90 mg of fructose, and 11 mL of
concentrated sulfuric acid were added to the reaction vessel,
ꢀ
which was placed in an oil bath at 140 C to start the dehy-
dration reaction. Aer 20 min, 100 mg of V-CP (10 wt%) was
added to the vessel, and the oxidation reaction was initiated by
À1
passing oxygen at 40 mL min into the reactor. The following
process was the same as in route 1.
2.4.3 Analytical approaches. The concentrations of fruc-
tose, 5-HMF, and 2,5-DFF were measured using HPLC (Agilent,
USA) equipped with an a refractive index (RI) detector. An HPX-
8
7H ion exclusion column (300 Â 7.8 mm, Bio-Rad Co. USA) was
ꢀ
used with a column oven temperature of 65 C, a mobile phase
of 0.005 M H SO4 at a ow rate of 0.7 mL min , and an
À1
2
injection volume of 100 mL. All samples were ltered through
a 0.22 mm membrane prior to injection.
3. Results and discussion
3
.1 Characterization of the catalyst structure
The morphologies of the ceramic matrix (CP) and V
catalyst (V-CP) were observed by scanning electron microscopy catalyst.
2 5
O /ceramic
2 5
Fig. 1 SEM images of (a) ceramic matrix (CP) and (b) V O /ceramic
7562 | RSC Adv., 2017, 7, 7560–7566
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
heating rate, and not using a pore-forming agent). Under these
conditions, the surface uidity of the base material is signi-
cant, which results in the planarization of CP and V-CP surfaces.
SEM images also showed no obvious pores for both CP and V-
CP, which is consistent with the BET results.
3
.2 Preparation of 2,5-DFF by the catalytic oxidation of 5-
HMF using V-CP
,5-DFF was prepared by the catalytic oxidation of 5-HMF using
as the oxidant for the catalytic performance evaluation of
/ceramic catalysts. The effects of the reaction temperature,
time, V content, and solvent composition on the 5-HMF
2
O
2
2 5
V O
2 5
O
conversion and 2,5-DFF yield were systematically investigated,
as shown in Fig. 4. The 2,5-DFF selectivity is 85.7–95.3% when
ꢀ
the temperature is in the range of 100–140 C, which indicates
that the V-CP has high selectivity, especially under the lower
temperature. However, only a small amount of 5-HMF (<23%)
was converted to 2,5-DFF aer 5 h of reaction at temperatures
ꢀ
below 120 C (Fig. 4a), which may due to the low reaction rate at
low temperature. The conversion of 5-HMF and yield of 2,5-DFF
gradually increased as the temperature increased. The 5-HMF
ꢀ
conversion was 100% and the 2,5-DFF yield was 85.7% at 140 C
Fig. 2 XRD patterns of CP and V-CP.
aer 5 h of reaction. The selectivity gradually decreased (from
95.3% to 85.7%) with the increase of the temperature, which is
likely due to formation of by-products. The HPLC results show
that the main by-products are levulinic acid (LA), formic acid
ꢀ
(
FA) and 2,5-furandicarboxylic acid (FDCA). At 140 C and 5 h,
the yield of LA, FA and FDCA is 5.7%, 4.5% and 3.4%, respec-
tively. In this case, the conversion of 5-HMF is 100% and the
total yield of 2,5-DFF and by-products is 99.4%. Compared with
other catalysts (see ESI Table S1†), the yield of 2,5-DFF in this
paper is higher or comparable, but the value is lower than those
4 3 2
of other homogeneous catalytic systems (e.g., VOSO /Cu(NO ) ,
ESI Table S1†).
ꢀ
At 140 C, the conversion of 5-HMF to 2,5-DFF versus time is
shown in Fig. 4b. The conversion of 5-HMF and yield of 2,5-DFF
increased rapidly in the rst three hours, and then, the change
in the trend became slow, reaching 100% and 87.5%, respec-
tively, at 5 h. The selectivity of 2,5-DFF decreased as the reaction
progresses, which is due to the gradually formed by-products as
the reaction proceeds. For example, the generated 2,5-DFF can
be further oxidized and generate FDCA.
Fig. 3 XPS spectra for V-CP.
4
2
a carrier for V
oxide as a carrier for V
In addition, the BET specic surface areas of the blank
ceramic matrix and V O /ceramic catalyst were also measured.
2
O
5
, and Adil et al., who used a nickel manganese
2 5
O
to prepare the catalyst.
2
5
Fig. 4c shows the effects of different V O loadings on the
2
5
The results showed that the BET specic surface area of CP is
conversion of 5-HMF and yield of 2,5-DFF. With the V
loading increasing from 1% to 10%, the conversion of 5-HMF
2 5
O
2
À1
2
À1
10.8 m g and that of V-CP is 17.9 m g . Both materials have
very small specic surface areas, which is determined by the
ceramic preparation process (including the temperature,
(
8
from 4.8% to 100%) and the yield of 2,5-DFF (from 4.3% to
5.7%) both increased dramatically at 140 C and 5 h of
ꢀ
Table 1 Relative atomic ratios and binding energies of the main elements in CP and V-CP
C1s
13.8
284.8
16.51
284.8
O1s
52.6
531.5
51.4
Na1s
4.7
1071.9
3.5
Mg2p
Al2p
Si2p
Ca2p
V2p1/2
V2p3/2
CP
atm%
Binding energy (eV)
atm%
2.6
49.7
3.1
8.8
74.2
4.9
15.1
102.4
17.2
2.3
347.1
2.2
—
—
1.2
524.2
—
—
1.2
517.7
V-CP
Binding energy (eV)
531.9
1072.2
45.0
74.2
102.7
347.2
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 7560–7566 | 7563

View Article Online
RSC Advances
Paper
Fig. 4 The effects of different reaction conditions on the preparation of 2,5-DFF by the catalytic oxidation of 5-HMF using V-CP ((a) reaction
temperature, (b) time, (c) V content, and (d) DMSO/water volume ratio). If not specified, the default reaction conditions were as follows: 5-
ꢀ
5 2
HMF, 63 mg; V-CP, 100 mg; V O content, 10%; solvent, DMSO; volume, 4 mL; O , 40 mL min ; temperature, 140 C; and reaction time, 5 h.
2
O
5
À1
2
reaction, with an insignicant change in the 2,5-DFF yield. A
To evaluate the recovery performance of V-CP, the reaction
further increase in the V O loading to 15% also increased the solution was ltered and washed with double distilled water
2
5
2,5-DFF yield and selectivity slightly. However, the 2,5-DFF yield three times aer the rst reaction. The dried V-CP was then
and selectivity decreased slightly when the V O loading subjected to 5-HMF oxidation under the same conditions, and
2
5
increased to 20%, which indicated that the higher content of the results are shown in Fig. 5. Aer being recycled ve times,
V
2
O
5
also promoted side reactions.
the 5-HMF conversion decreased gradually from 100% to 94%,
on 2,5-DFF and the 2,5-DFF yield decreased from 85.7% to 70.6% at 140 C
ꢀ
We investigated the effect of the ow rate of O
2
synthesis. As shown in Fig. S1,† with increasing the ow rate of and 5 h of reaction. In the recovery process, the ltration,
À1
O
2
from 10 to 40 mL min , the conversion of 5-HMF increased transfer, and other experimental operations resulted in the loss
from 69.7% to 100% and the yield of 2,5-DFF increased form of V-CP, while the active ingredient V O in the catalyst may also
2
5
5
7.9% to 85.7%. A further increase in the oxygen ow rate to 60 have leaked slowly. Therefore, the 2,5-DFF yield and selectivity
À1
mL min , there is no signicant change in the 2,5-DFF yield.
both decreased slightly. Overall, this catalyst showed good
Moreover, the effects of different solvents on the efficiency of recovery performance.
the catalytic reaction were also investigated in this paper.
Fig. 4d shows that the conversion of 5-HMF and the yield/
selectivity of 2,5-DFF were the highest in DMSO solvent,
which indicated that DMSO can effectively inhibit the occur-
rence of side reactions and promote 2,5-DFF formation. Water
is the greenest solvent for liquid phase reactions. In this paper,
different volume ratios of DMSO–H O mixed solutions were
2
used as solvents, and the results showed that the 5-HMF
conversion and 2,5-DFF yield decreased signicantly with an
increase in the water content. No oxidation of 5-HMF occurred
2
and no 2,5-DFF was formed in pure water and DMSO/H O 1 : 3
systems. Additionally, we also investigated the effect of the
different organic solvents (acetonitrile, tetrahydrofuran, dime-
thylformamide) on the 2,5-DFF production. The results show
that yield of 2,5-DFF is lower than 10% in these cases, which
indicated that the DMSO has a signicant effect on the prepa-
ration of 2,5-DFF by 5-HMF oxidation.
Fig. 5 Recycling and reuse of V-CP for 2,5-DFF production. Reaction
conditions: 5-HMF, 63 mg; V-CP, 100 mg; V
2
O
5
content, 10%; DMSO,
À1
ꢀ
4
2
mL; O , 40 mL min ; reaction temperature, 140 C; and reaction
time, 5 h.
7564 | RSC Adv., 2017, 7, 7560–7566
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
.3 Preparation of 2,5-DFF by the one-pot catalytic
RSC Advances
3
to 67.6% rapidly from 20 min to 140 min. The 2,5-DFF yield
reached the maximum of 68.4% aer 200 min. Compared with
another catalytic system (see ESI Table S2†), the yield of the
H SO /V-CP system in this paper is higher or comparable. For
conversion of fructose
Fructose is a sustainable biomass raw material. Research has
shown that 5-HMF can be prepared by acid-catalyzed dehydra-
tion using fructose. In this paper, fructose was used as the raw
material to prepare 2,5-DFF by a one-pot method with diluted
2
4
example, in the case of molybdenum-based polyoxometalates
33
(
Cs0.5
H
2.5PMo12), the yield of 2,5-DFF is 69.3% at 433 K using
air as oxidant. Formic acid and levulinic acid were detected as
2 5
sulfuric acid and V O /ceramic catalysts to catalyze the dehy-
+
43
the main by-products; for V–g-C
dration reaction was conducted under N
3
N
4
(H )/V–g-C
3
N
4
,
the dehy-
dration of fructose and oxidation of 5-HMF. In the experiment,
two experimental procedures were used to carry out the reaction
according to the time of addition of the V O /ceramic catalyst.
ꢀ
2
(0.1 MPa) at 120 C for
2
1
h and then the V–g-C N was added under O (0.1 MPa) at
3 4 2
2
5
ꢀ
30 C for 6 h, achieving the highest yield of 63%; for Fe O –
3
4
The rst one included adding V-CP together with the diluted
sulfuric acid into the reaction system. The oxygen was intro-
duced aer 20 min of dehydration for oxidation. Fig. 6a shows
that the 5-HMF yield increased from 0 to 67.2% in the rst
32
SBA–SO H/K-OMS-2, the 2,5-DFF yield was up to 80% by
3
stepwise addition of catalysts. However, the coexistence of these
two catalyst in the initial stage of reaction resulted in low 2,5-
DFF yield, which is probably due to the formation of humins or
the oxidation of fructose. In this study, further reactions
resulted in slightly lower yields of 2,5-DFF, which was mainly
due to the further reaction of 2,5-DFF.
10 min of reaction with the rapid conversion of fructose.
However, the yield decreased to 59.0% aer further reaction
until 20 min. The 2,5-DFF yield increased gradually from 0 to
21.8% aer the introduction of oxygen, while the 5-HMF yield
The above experimental results showed that the second
procedure is obviously superior to the rst, as higher fructose
conversion and 2,5-DFF yield were obtained. When V-CP and
acid catalysts were added into the system at the same time, V-CP
can possibly catalyze the undesired oxidation of fructose, which
resulted in the lower yield of 5-HMF using the rst method
compared to the second one (67.2% vs. 76.6%). In addition, due
to the lack of oxygen atmosphere in the system, the generated 5-
HMF cannot be rapidly oxidized to 2,5-DFF, which may also
generate some by-products (e.g., humins). These by-products
inhibited the catalytic activity of V-CP and resulted in a lower
yield of 2,5-DFF in the rst method than in the second (21.8%
vs. 68.4%).
decreased to 16.4% due to its oxidation reaction.
The second approach included adding diluted sulfuric acid

rst and then adding the V-CP catalyst with oxygen for oxidation
aer 20 min of dehydration. Fig. 6b shows that the yield of 5-
HMF reached 76.6% at 20 min of reaction time with the rapid
conversion of fructose. Aer adding V-CP and introducing
oxygen, 5-HMF was converted, and the 2,5-DFF yield increased
4. Conclusions
2 5
A new V O /ceramic catalyst was synthesized and successfully
applied to the efficient catalytic oxidation of 5-HMF to 2,5-DFF.
The results of structural characterization showed that the
doping of V
2 5
O did not change the crystal structure of the
ceramic, and V O sintered on the surface of the ceramic crystal
2
5
in the amorphous form. 2,5-DFF was prepared by catalyzing 5-
ꢀ
HMF at 140 C under atmospheric pressure in the presence of
oxygen as the oxidant in DMSO, and the yield of 2,5-DFF was
85.7%. Aer 5 times of recycling this catalyst, the 2,5-DFF yield
was still over 70%, which showed good recovery performance.
Furthermore, a new one-pot method for the preparation of 2,5-
DFF from fructose was proposed by using fructose as the raw
material and V O /ceramic-dilute sulfuric acid as the catalyst.
2
5
The conversion of fructose was 100%, and the 2,5-DFF yield was
8.4%. The catalyst and catalytic process proposed in this paper
6
provide a new method for the efficient synthesis of 2,5-DFF from
fructose.
2 5
Fig. 6 Reaction curve of 2,5-DFF preparation by the one-pot V O /
ceramic and diluted sulfuric acid co-catalysis method. (a) Adding V-CP
at the beginning of the reaction together with acid and introducing
oxygen after 20 min of reaction time. (b) Adding V-CP and oxygen after
Acknowledgements
2
0 min of reaction time. Reaction conditions: fructose, 90 mg; acid
À1
This work was supported by the Natural Science Foundation
of Tianjin, China (No. 16JCYBJC19600), the Natural Science
concentration, 0.05 M; V-CP, 100 mg; DMSO, 4 mL; O
2
, 40 mL min
;
ꢀ
and reaction temperature, 140 C.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 7560–7566 | 7565

View Article Online
RSC Advances
Paper
Foundation of China (No. 21621004) and the Beiyang Young 21 Y. Roman-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic,
Scholar of Tianjin University (2012).
Nature, 2007, 447, 982–985.
2
2 X. Xiang, J. Cui, G. Ding, H. Zheng, Y. Zhu and Y. Li, ACS
Sustainable Chem. Eng., 2016, 4, 4506–4510.
3 A. Gandini, Green Chem., 2011, 13, 1061–1083.
4 J. Ma, Z. Du, J. Xu, Q. Chu and Y. Pang, ChemSusChem, 2011,
4, 51–54.
2
2
References
1
2
3
4
5
6
7
8
9
A. A. Rosatella, S. P. Simeonov, R. F. M. Frade and
C. A. M. Afonso, Green Chem., 2011, 13, 754–793.
J. P. Lange, E. van der Heide, J. van Buijtenen and R. Price,
ChemSusChem, 2012, 5, 150–166.
25 A. Gandini and M. N. Belgacem, Prog. Polym. Sci., 1997, 22,
1203–1379.
26 Y. Zhu, M. Shen, Y. Xia and M. Lu, Catal. Commun., 2015, 64,
37–43.
T. Wang, M. W. Nolte and B. H. Shanks, Green Chem., 2014,
16, 548–572.
27 J. Nie, J. Xie and H. Liu, J. Catal., 2013, 301, 83–91.
J. Song, H. Fan, J. Ma and B. Han, Green Chem., 2013, 15, 28 J. Nie and H. Liu, J. Catal., 2014, 316, 57–66.
2619–2635.
29 G. D. Yadav and R. V. Sharma, Appl. Catal., B, 2014, 147, 293–
301.
X. Liu, J. Xiao, H. Ding, W. Zhong, Q. Xu, S. Su and D. Yin,
Chem. Eng. J., 2016, 283, 1315–1321.
J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131,
30 I. S ´a daba, Y. Y. Gorbanev, S. Kegnæs, S. S. R. Putluru,
R. W. Berg and A. Riisager, ChemCatChem, 2013, 5, 284–293.
31 C. A. Antonyraj, B. Kim, Y. Kim, S. Shin, K.-Y. Lee, I. Kim and
J. K. Cho, Catal. Commun., 2014, 57, 64–68.
32 Z. Z. Yang, J. Deng, T. Pan, Q. X. Guo and Y. Fu, Green Chem.,
2012, 14, 2986–2989.
33 Y. Liu, L. Zhu, J. Tang, M. Liu, R. Cheng and C. Hu,
ChemSusChem, 2014, 7, 3541–3547.
34 Z. Yang, W. Qi, R. Su and Z. He, Energy Fuels, 2017, 31, 533–
541.
1
979–1985.
G. Yong, Y. Zhang and J. Y. Ying, Angew. Chem., Int. Ed., 2008,
7, 9345–9348.
R. Huang, W. Qi, R. Su and Z. He, Chem. Commun., 2010, 46,
115–1117.
A. Chinnappan, C. Baskar and H. Kim, RSC Adv., 2016, 6,
3991–64002.
0 Z. M. Xue, M. G. Ma, Z. H. Li and T. C. Mu, RSC Adv., 2016, 6,
8874–98892.
4
1
6
1
1
1
1
1
1
1
1
1
1
2
9
35 A. Takagaki, M. Takahashi, S. Nishimura and K. Ebitani, ACS
Catal., 2011, 1, 1562–1565.
36 X. Xiang, L. He, Y. Yang, B. Guo, D. Tong and C. Hu, Catal.
Lett., 2011, 141, 735–741.
1 Z. Yang, W. Qi, R. Huang, J. Fang, R. Su and Z. He, Chem.
Eng. J., 2016, 296, 209–216.
2 R. Fang, R. Luque and Y. Li, Green Chem., 2016, 18, 3152–
3
157.
3 S. Wang, Z. Zhang, B. Liu and J. Li, Ind. Eng. Chem. Res., 2014,
3, 5820–5827.
37 K. Onitsuka, A. Dogan, J. F. Tressler, Q. Xu, S. Yoshikawa and
R. E. Newnham, J. Intell. Mater. Syst. Struct., 1995, 6, 447–
455.
5
4 Y. Wang, B. Liu, K. Huang and Z. Zhang, Ind. Eng. Chem. Res., 38 Z. Wu, W. Qi, M. Wang, R. Su and Z. He, J. Mol. Catal. B:
014, 53, 1313–1319.
Enzym., 2014, 110, 32–38.
5 L. F. Liao, Y. Liu, Z. Y. Li, J. P. Zhuang, Y. B. Zhou and 39 Z. Wu, W. Qi, M. Wang, Y. Wang, R. Su and Z. He, Biochem.
S. Z. Chen, RSC Adv., 2016, 6, 94976–94988.
Eng. J., 2013, 77, 190–197.
6 X. Wan, C. Zhou, J. Chen, W. Deng, Q. Zhang, Y. Yang and 40 Y. Qu, Z. Wu, R. Huang, W. Qi, R. Su and Z. He, RSC Adv.,
Y. Wang, ACS Catal., 2014, 4, 2175–2185.
2016, 6, 64581–64588.
7 H. Ait Rass, N. Essayem and M. Besson, ChemSusChem, 2015, 41 N. Jiang, R. Huang, W. Qi, R. Su and Z. He, BioEnergy Res.,
, 1206–1217.
2012, 5, 380–386.
8 Z. Zhang, J. Zhen, B. Liu, K. Lv and K. Deng, Green Chem., 42 S. F. Adil, S. Alabbad, M. Kuniyil, M. Khan, A. Alwarthan,
2
8
2
015, 17, 1308–1317.
9 L. Gao, K. Deng, J. Zheng, B. Liu and Z. Zhang, Chem. Eng. J.,
015, 270, 444–449.
N. Mohri, W. Tremel, M. N. Tahir and M. R. H. Siddiqui,
Nanoscale Res. Lett., 2015, 10, 52.
43 J. Chen, Y. Guo, J. Chen, L. Song and L. Chen, ChemCatChem,
2014, 6, 3174–3181.
2
0 Y. W. Zhang, Z. M. Xue, J. F. Wang, X. H. Zhao, Y. H. Deng,
W. C. Zhao and T. C. Mu, RSC Adv., 2016, 6, 51229–51237.
7566 | RSC Adv., 2017, 7, 7560–7566
This journal is © The Royal Society of Chemistry 2017