Catalytic Conversion of Glucose to 5-Hydroxymethylfurfural and Furfural by a Phosphate-Doped SnO2 Catalyst in γ-Valerolactone-Water System

Article abstract of DOI:10.1007/s10562-020-03227-7

Abstract: 5-Hydroxymethylfurfural (5-HMF) and furfural are produced from glucose by using a novel phosphate-doped SnO₂ catalyst in GVL/water system. Comparing with the low yield (19.3% 5-HMF, 7.2% furfural) of SnO₂ as catalyst, the loading of 15 wt% phosphate on the SnO₂ can enhance the yield of 5-HMF (39.2%) and furfural (12.5%) from glucose under same reaction conditions. Furthermore, under the optimal conditions, the results indicate that the phosphated SnO₂ catalyst, which contains 5 wt% phosphate, can result in 46.4% yield of 5-HMF and 18.9% yield of furfural at 180 °C for 90 min. The phosphated SnO₂ catalysts are characterized by using XRD, TEM, Py-IR, XPS and TPD to reveal their structural, surface, and acid properties. And the enhanced 5-HMF and furfural yield could be attributed to the higher acidity, and the incorporation of phosphate into the framework of SnO₂. Graphic Abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-020-03227-7

Catalysis Letters
https://doi.org/10.1007/s10562-020-03227-7
Catalytic Conversion of Glucose to 5‑Hydroxymethylfurfural
and Furfural by a Phosphate‑Doped SnO2 Catalyst
in γ‑Valerolactone‑Water System
Wenzhi Li1,2 · Tao Yang1 · Mingxue Su1 · Yang Liu1
Received: 31 December 2019 / Accepted: 20 April 2020
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
5-Hydroxymethylfurfural (5-HMF) and furfural are produced from glucose by using a novel phosphate-doped SnO2 catalyst
in GVL/water system. Comparing with the low yield (19.3% 5-HMF, 7.2% furfural) of SnO2 as catalyst, the loading of 15
wt% phosphate on the SnO2 can enhance the yield of 5-HMF (39.2%) and furfural (12.5%) from glucose under same reac-
tion conditions. Furthermore, under the optimal conditions, the results indicate that the phosphated SnO2 catalyst, which
contains 5 wt% phosphate, can result in 46.4% yield of 5-HMF and 18.9% yield of furfural at 180 °C for 90 min. The phos-
phated SnO2 catalysts are characterized by using XRD, TEM, Py-IR, XPS and TPD to reveal their structural, surface, and
acid properties. And the enhanced 5-HMF and furfural yield could be attributed to the higher acidity, and the incorporation
of phosphate into the framework of SnO2.
Graphic Abstract
Keywords Glucose · 5-Hydrohymethylfurfural · Furfural · Bi-functional catalyst · Phosphated SnO2
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s10562-020-03227-7) contains
supplementary material, which is available to authorized users.
1 Introduction
It has been known that lignocellulosic biomass is the most
abundant biomass source, which is estimated about 2×1011
metric tons every year on earth [1]. The lignocellulosic bio-
mass is mainly composed of three components including
cellulose (40–50%), hemicellulose (25–35%) and lignin
(15–20%) [2, 3]. Cellulose, as the main component of ligno-
cellulosic biomass, is consisted of d-glucose units that linked
via β-1,4-glycosidic bonds and can be converted to glucose
* Tao Yang
yt171030@mail.ustc.edu.cn
1
Laboratory of Basic Research in Biomass Conversion
and Utilization, Department of Thermal Science and Energy
Engineering, University of Science and Technology
of China, Hefei 230026, People’s Republic of China
2
National Synchrotron Radiation Laboratory, University
of Science and Technology of China, Hefei 230026,
People’s Republic of China
Vol.:(0123456789)
1 3

W. Li et al.
monomers [4, 5]. Thus, the transformation of glucose to
5-HMF and furfural is important because glucose is the most
abundant sugar and 5-HMF and furfural are key chemical
intermediates [6]. Specifcally, as a bio-based chemical plat-
form, 5-HMF can be used to synthesize a series of chemicals
such as 2,5-furandicarboxylic acid, 5-hydroxymethylfuroic
acid, 2,5-dimethylfuran, levulic acid and so on [7]. And
furfural can also be converted into valuable chemicals, for
example, dicarboxylic acid, furfural alchol and teteahydro-
furfuryl alcohol [8, 9].
Yan et al. prepared the sulfated zirconia catalysts to synthe-
size 5-HMF from glucose and 47.6% 5-HMF yield can be
obtained within 4 h at 130 °C [24]. Nakajima et al. reported
that glucose can be transformed into 5-HMF with the aid of
niobic acid which is an efcient water-tolerant Leiws acid
catalyst [25]. Hou et al. prepared a series of metal oxides and
metal phosphates, which were used for glucose conversion,
and a high 5-HMF yield (58.3%) were obtained in the ionic
liquid 1-ethyl-3-methylimidazolium bromide (EMIMBr)
[26]. Li et al. studied the synergistic efect of SnPCP and
MnO2-PDA on glucose dehydration and achieved 5-HMF
yields of 55.8% in DMSO [27]. Marianou et al. have
studied the efect of Snδ+ species for glucose conversion
in biphasic DMSO-H2O media [28]. The highest 5-HMF
molar yield of 27.5% with complete glucose was achieved
by SnO2/γ-Al2O3. It has been shown that a limitation of the
SnO2 catalyst in previous report is that only shows Lewis
acid sites, while enough evidence has exhibited that Lewis
and Brønsted acid sites can achieve good 5-HMF yield than
catalysts with either Lewis or Brønsted acid only.
Thus, rapid developments for converting glucose to
5-HMF in the respect of catalytic systems including cata-
lysts and solvents have been designed in recent years. For
example, Mazzotta et al. prepared the Glu–TsOH–TiO2 cata-
lyst by a one-pot hydrothermal method [10]. 59% and 99%
yield of 5-HMF were obtained from glucose and fructose
respectively in a bi-phasic solvent system. According to
their studies, it has been shown that the stability of glucose
makes it more difcult than fructose for preparing 5-HMF.
However, glucose is a promising raw materials for the pro-
duction of 5-HMF owing to glucose as the most abundant
monosaccharide and cheapest hexose in the natural. And it
also has been reported that conversion of glucose to 5-HMF
may occur isomerization of glucose to fructose and followed
by the dehydration to 5-HMF in the presence of Brønsted
acid catalysts. Generally, isomerization is a vital process for
achieving high 5-HMF yield via enzymes, base or lewis acid
catalysts [11]. Thus, some homogeneous inorganic acids
and metal chlorides catalysts such as AlCl3, SnCl4, CrCl3
and RuCl3 [12–16] or organic acid such as maleic acid with
AlCl3 [17] have been reported. However, the poor recycla-
bility of homogenous catalysts and environmental pollution
cause hinder their use.
Thus, a novel catalyst was prepared by combing phos-
phate with SnO2 nanoparticles, which can be as a more
efective catalyst for preparing 5-HMF and furfural from
glucose. Phosphate doped SnO2 possesses two functional
groups, SnO2 acts as Lewis acid, catalyzes the glucose
isomerization to fructose, while phosphate act as a Brøn-
sted acid, dehydrates the fructose to 5-HMF and furfural
(Scheme 1). The efect of water had also been reported in
many other publications [29–32], and they concluded that
reaction in aqueous media could achieve better rate accel-
eration and higher selectivity by comparing with those in
organic solvents. Thus, the efect of water contents on the
reaction was also investigated in this work and the results
indicated that the presence of water could improve the selec-
tivity of product. Furthermore, the ratio of SnO2 and phos-
phate in bi-functional catalyst are important for increasing
5-HMF yield by selectively improving the desirable reac-
tions (isomerization, dehydration) and suppress the undesir-
able reactions such as rehydration to levulic acid and polym-
erization to humics. Moreover, the interactions among the
glucose, solvents, catalysts, reaction time and temperature
were also investigated in this paper.
Recently, in most publications, catalysts were usually
used for glucose conversion by combining Brønsted acids
with Lewis acids. Lewis acids can catalyze the glucose
isomerization to fructose, while Brønsted acids usually
accelerate the dehydration of fructose to 5-HMF. Early
reports are focused on combing with protic acids and Lewis
acids via a simple physically mixing. For example, Amber-
lyst 70 and TiO2-ZrO2 were reported by Atanda et al. and
highest 5-HMF yield (74%) was obtained in a water-THF
biphasic reaction system [18]. Takagaki et al. also reported
that the combination of Amberlyst-15 and hydrotalcite can
catalyze glucose into 5-HMF [19]. Now, a promising trend
in the synthesis of 5-HMF from glucose is using a hetero-
geneous catalysts with bi-functional catalytic sites, such as
modifed zeolites [20, 21], aluminum- or boric acid-doped
silica gel catalysts [22], sodium modifed Nb, Ti, Zr and Al
phosphates [23]. Especially, much attention has been paid
to 5-HMF production from glucose with kinds of metallic
oxide and metal phosphates catalysts owing to the low cost,
easy preparation and better thermal stability. For example,
2 Experimental
2.1 Materials
d-(+) Glucose, furfural (99%), 5-HMF (99%), SnO2, ZrO2
and TiO2 were purchased from Sigma-Aldrich (Shanghai,
China). Ammonium dihydrogen ,phosphate (NH4H2PO4),
dimethyl sulfoxide (DMSO), and isopropanol were all
purchased from Sinopharm Chemical Reagent Co., Ltd.
1 3

Catalytic Conversion of Glucose to 5-Hydroxymethylfurfural and Furfural by a Phosphate-Doped…
Scheme 1 The conversion of glucose to 5-HMF and furfural in a water-GVL system with P-doped SnO2
(Shanghai, China). γ-valerolactone (95%) was supplied
by LangFang Hawk Technology and Development Co.,
Lta. All chemicals were used without further purification.
2.3 Procedure for the Conversion of Glucose
into 5‑HMF and Furfural
Typically, the experiments were fnished in a 15 mL sealed
thick-wall glass tube with 50 mg catalyst, 50 mg glucose and
5.25 mL solvent (5 mL GVL, 0.25 mL distilled water). The
above mixture was heated in a preheated oil bath with a given
temperature and then kept for corresponding residence time.
With the ending of the reaction, the tub was cooling to room
temperature (under natural cooling). Finally, the products were
fltered from the mixture and stored in a 4 °C refrigerator for
further analysis.
2.2 Preparation and Characterization
of the Catalyst
A series of phosphated SnO2 catalysts with differ-
ent phosphate loading (5 wt%–30 wt%) were prepared
by impregnation of SnO2 with the required amount of
NH4H2PO4 aqueous solution. A typical synthesis of 15
wt% phosphate loading on SnO2 (15%P–SnO2) was as fol-
lows: the 0.15 g NH4H2PO4 and 1 g SnO2 were dissolved
in 30 mL distilled water, then the mixture was stirred at
room temperature for 6 h. Next, the above solution was
dried in an oven at 80 °C overnight. Finally, phosphated
SnO2 materials were obtained after calcination at 600 °C
for 4 h with a heating rate of 3 °C/min.
2.4 Quantifcation Analysis for Products
The products (5-HMF, furfural) were detected by high perfor-
mance liquid chromatography (HPLC, LC-2010AHT, SHI-
MADZU) equipped with an ultraviolet detector (SPD-10A,
SHIMADZU) and a C18 column (XORBAX Eclipse XDB-
C18, Agilent), and the detection wavelength is set to 280 nm.
The mobile phase is methanol/water (2/3 v/v) system with
a fow rate of 0.4 mL/min and the column temperature was
held at 30 °C. The glucose concentration is also determined
by HPLC and the system is equipped with a HPX-87H col-
umn (Bio-Rad, USA) and diferential refractive index detector
(Waters 2414, USA) [33]. The mobile phase is 5 mM H2SO4
systems with a fow rate of 0.6 mL/min and the detector tem-
perature was held at 63 °C with 50 °C column.
Transmission electron microscopy (TEM) images
were recorded on a JEM 2011 instrument (JEOL, Japan).
The content of phosphorus and tin of the catalyst was
measured by TEM-EDS (JEOL, Japan). X-ray diffraction
(XRD) patterns were recorded in the range of 2θ = 20–-
80° on a PAN analytical X’Pert PRO MPD, using a
Cu-Kα monochromated radiation source and a Ni fil-
ter. The acidity measurement of P-SnO2 was performed
by using NH3-Temperature Programmed Desorption
(NH3-TPD) (Quantachrome Instruments, American). The
IR spectra of the pyridine-adsorbed of 15%P–SnO2 was
obtained with a Nicolet 8700 instrument. The sample was
pretreated at 200 °C for 3 h under vacuum, the skeletal
spectra was collected after the sample cooling to 30 °C.
Then the pyridine vapor was introduced into the sample
for adsorption with N2 and the spectra of 15P%–SnO2
was recorded. The chemical state of O, Sn and P were
investigated by X-ray photoelectron spectroscopy (XPS;
ESCALAB 250 Xi, Thermo Fisher Scientific).
Glucose conversion, 5-HMF, furfural yield and selectivity
were calculated based on the following equations:
moles of glu cos e in products
glu cos e conversion = (1 −
) × 100%
moles of starting glu cos e
moles of 5 − HMF produced
moles of starting glu cos e
5−HMF yield =
× 100%
1 3

W. Li et al.
The crystalline phases of SnO2 and P doped SnO2 nano-
particles were investigated by XRD, as shown in Fig. 1. The
difraction patterns of all samples can be indexed to tetrago-
nal SnO2 (JCPDS No. 41-1445). The difraction pattern of
pure SnO2 showed sharp and intense, which implied high
crystallinity of pure SnO2. In contrast, the crystallinity of
SnO2 reduced when the pure SnO2 was treated with ammo-
nium dihydrogen phosphate and which is indicated by peak
broadening and reduction in peak intensity. And this result
suggests the phosphorus has been incorporated into SnO2
framework [37]. The TEM image of SnO2 (Fig. 2a) shows a
plate-like morphology. In contrast to pure SnO2, phosphate
doped SnO2 showed a little particle aggregation and the par-
ticle increased than pure SnO2.
moles of furfural produced
moles of starting glu cos e
furfural yield =
× 100%
× 100%
× 100%
5 − HMF yield
5−HMF selectivity =
furfural selectivity =
glu cos e conversion
furfural yield
glu cos e conversion
3 Results and Discussion
3.1 Characterization of Catalysts
XPS measurements were used to analyze the composi-
tions and chemical states of SnO2 and P doped SnO2. C1s
peak of the carbon contamination was used as reference to
calibrate the binding energies of peaks. The XPS spectrum
of the O 1s of SnO2 (Fig. 3a) shown two peaks which indi-
cated that two oxygen species are present in the SnO2. The
oxygen peak observed at 530.7 eV corresponds to the bulk
lattice oxygen of SnO2, and the peak at 532.2 eV corre-
sponds to the chemisorbed oxygen ions in SnO2. Comparing
with SnO2, an additional peak in 15P%–SnO2 (Fig. 3b) can
be observed at 532.1 eV, which is assigned to oxygen species
of P–O bond linkages. As shown in Fig. 3c, the XPS spectra
of Sn 3d energy state of SnO2 revealed two peaks at 495.2
and 486.7 eV which are corresponded to Sn 3d3/2 and Sn
3d5/2, respectively. Furthermore, there are a shift of lattice
oxygen in the O1 s region and Sn 3d3/2 of 15P%–SnO2 the to
higher binding energy, which may be caused by the partial
electron transfer from Sn to P due to the incorporation of P
into SnO2 lattice. Thus, the electron density on Sn decreased
and this result is similar with Guo [38] and Atanda et al.
Energy Dispersive Spectrometer (EDS) was used to ana-
lyze the elements of diferent P doped SnO2 materials (Fig.
S1). The result was listed in Table 1, which confrmed the
existence of P and Sn species and the ratio of P and Sn
were calculated. The total acidity measurement of SnO2 and
phosphate doped SnO2 were calculated by using NH3-TPD
and the result is summarized in Table 1. It has been shown
that all P doped SnO2 showed high acidity than pure SnO2,
furthermore, the total acidity increased from 42.2 to 84.1
μmol/g with the increase of phosphate from 5 wt% to 15
wt%. The Py-IR of 15P%–SnO2 was displayed in Fig. S2, the
band at 1450 cm−1, which was assigned to Lewis acid sites,
the band at 1500 cm−1 was assigned to Lewis and Brøn-
sted acid sites, while the band at 1550 cm−1 was assigned to
Brønsted acid sites [34]. It was demonstrated that the cata-
lyst possesses BrØnsted and Lewis acid sites on the basis of
results of Py-IR of 15%P–SnO2 (Fig. S2), then, the possible
structure of phosphate doped SnO2 catalyst (Scheme S1) was
proposed combining with XPS results. During the glucose
conversion, we deduct that the Sn4+ could be regarded as
Lewis acid sites that was active for glucose isomerization to
fructose [35, 36], then, the presence of BrØnsted acid sites
(phosphate) would accelerate the dehydration of fructose
to 5-HMF. The combination of SnO2 and phosphate would
show a better catalytic performance in glucose conversion.
Table 1 Compositional, and acid properties of SnO2 and P-SnO2
Entry
Sample
Sn/P moles
ratioa
Total acidity
(µmmol/g)b
1
2
3
4
SnO2
–
12.9
42.2
62.3
84.1
5%P–SnO2
10%P–SnO2
15%P–SnO2
37.1
9.3
8.1
aDetermined from TEM-EDS analysis
bDetermined from NH3-TPD analysis
Fig. 1 XRD patterns of SnO2, 5P%, 10P% and 15P%–SnO2
1 3

Catalytic Conversion of Glucose to 5-Hydroxymethylfurfural and Furfural by a Phosphate-Doped…
Fig. 2 TEM images of a SnO2,
b 5P%–SnO2, c 10P%–SnO2,
and d 15P%–SnO2
Fig. 3 XPS spectra of a O1 S of SnO2, b O1 S of 15%P–SnO2, c Sn 3d of SnO2 and 15%P–SnO2, d P 2p of 15%P–SnO2
1 3

W. Li et al.
work [37]. Moreover, as shown in Fig. 3d, the P 2p spec-
tra showed a single peak at 134.1 eV that demonstrated the
existence of phosphorus in the pentavalent oxidation state
[38, 39]. Thus, the XPS results can prove that the phospho-
rus has been incorporated into the SnO2 framework success-
fully and the results are consistent with the XRD results.
investigate the catalytic activity of P doped SnO2, some
organic solvents and water were studied during the glucose
dehydration and the results are listed in Table 3. Compared
with water as solvent (entry 5), the conversion and yield
enhanced remarkably by using organic solvents (entry 2,
4). Meanwhile, using single organic solvent such as GVL
(Table 3, entry 2) show a low yield of 5-HMF (16.7%) and
furfural (10.6%) yield than GVL/water (Table 3, entry 3)
solvents (39.2% 5-HMF yield, 12.5% furfural yield), which
indicated that the presence of water in organic system is vital
for glucose dehydration. The main reason may be caused by
low solubility for glucose in organic solvents [40], moreover,
water can also became as a proton shuttle to promote the
reaction and reduce the energy barriers [32]. Furthermore,
the presence of water can also afect the catalytic activity
of acid catalyst [41, 42]. And the same conclusions can be
acquired by phosphate doped ZrO2 catalyst (Table 3, entry
6, 7).
3.2 Production of 5‑HMF and Furfural from Glucose
Under Various Catalyst
Various metallic oxides and phosphate modifcation metallic
oxide catalysts were prepared and used to produce 5-HMF
and furfural from glucose at 180 °C, 90 min in GVL/Water
solvent system, and the result are listed in Table 2. Notably,
15%P–SnO2 exhibited encouraging catalytic performance
than SnO2 (Table 2, entries1, 4). Comparing with SnO2
(19.3% 5-HMF yield, 7.2% furfural yield), the enhanced cat-
alytic activity of 15%P–SnO2 (39.2% 5-HMF yield, 12.5%
furfural yield) can be attributed to the presence of phosphate
(Brønsted acid sites) and SnO2 (Leiws acid sites). Thus, it
is signifcant to modify the SnO2 to enhance their catalytic
activity by a simple method. Moreover, the same results can
also be obtained according to Table 2 (entries 2, 5). 27.5%
5-HMF yield and 9.5% furfural yield are achieved by ZrO2
as catalyst; while 15%P–ZrO2 was used as catalyst, 5-HMF
and furfural yield increased to 39.8% and 13.5%, respec-
tively. Although ZrO2 showed better catalytic activity than
SnO2 in the same reaction systems, phosphate doped ZrO2
and SnO2 have a similar catalytic activity (Table 2, entries
4, 5), which can be demonstrated that the importance of
phosphate on SnO2. Moreover, P doped TiO2 also exhibited
more efcient catalytic activity than pure TiO2 (Table 2,
entry 3, 6).
3.4 Efect of Water Contents on 5‑HMF and Furfural
Yields in GVL and DMSO
Thus, based on above results, in order to probe the detailed
efect about water, the water contents varying from 0 to 100
wt% in diferent solvents (GVL and DMSO) system have
been studied, and the results are given in Fig. 4. In the case
of glucose conversion, the water contents exhibited a marked
efect on the reaction with respect to 5-HMF selectivity. The
catalytic results (Fig. 4a, b) refected that 5-HMF selectiv-
ity increased with the increase of water contents, with val-
ues close to 60% 5-HMF selectivity in pure water system,
indicating that higher water contents benefted the 5-HMF
reaction selectivity. Furthermore, when the volume ratio of
H2O/GVL or H2O/DMSO equal to 5%, catalyst presented
3.3 HMF and Furfural Production from Glucose
in Diverse Solvents
Table 3 Efect of solvents on the conversion of glucose and the yields
of 5-HMF and furfural
The reaction solvent plays a vital role on the process of
distribution, isolation and degradation of products. To
Entry
Solvent
Glucose con-
version (%)
5-HMF
Furfural
yield (%)
yield (%)
Table 2 Catalytic ability of diferent metallic oxide and P modifca-
1
2
3
4
5
6
7
MSO-watera
GVLb
87.6
>99
97.8
78.5
21.5
>99
>99
36.1
16.7
39.2
30.4
13.6
21.9
37.9
–
tion metallic oxide catalysts for glucose dehydration
10.6
12.5
3.3
–
16.7
17.1
GVL-watera
Isopropanola
Waterb
Entry
Catalyst
Glucose con-
version (%)
5-HMF
Furfural
yield (%)
yield (%)
1
2
3
4
5
6
SnO2
ZrO2
TiO2
15%P–SnO2
15%P–ZrO2
15%P–TiO2
85.1
97.1
83.2
97.8
>99
86.3
19.3
27.7
15.6
39.2
39.8
25.2
7.2
9.5
5.8
12.5
13.5
8.5
GVLc
GVL-waterc
Reaction conditions
a50 mg glucose, 5 mL solvent, 0.25 mL water, 50 mg 15%P–SnO2
b50 mg glucose, 5 mL solvent (GVL or Water) 50 mg 15%P–SnO2
c50 mg glucose, 5 mL solvent (GVL or Water) 50 mg 15%P–ZrO2,
180 °C, 90 min
Reaction conditions: 50 mg glucose, 50 mg catalyst, 5 mL GVL, 0.25
mL water, 180 °C, 90 min
1 3

Catalytic Conversion of Glucose to 5-Hydroxymethylfurfural and Furfural by a Phosphate-Doped…
Fig. 5 Efect of phosphate content on the catalytic of SnO2. Reaction
conditions: 50 mg glucose, 5 mL GVL, 0.25 mL water, 50 mg cata-
lyst, 180 °C, 90 min
increasing of the water contents further, the yield of furfural
is lower than the water contents equaling to 0% and the pos-
sible reason is that the side reaction of 5-HMF conversion
to furfural could be inhibited.
3.5 Efect of Diferent Phosphate Loading Catalysts
on 5‑HMF and Furfural Production
Promoted by the efect of tunable ratio of Brønsted and
Lewis acid catalyst to 5-HMF and furfural in previous stud-
ies [37], a series of phosphate-doped catalysts with difer-
ent SnO2/phosphate ratio were prepared by changing the
contents of ammonium dihydrogen phosphate. As shown
in Fig. 5, the efect of the amount of phosphate coupled
with SnO2 on the co-production of 5-HMF and furfural
from glucose were investigated. It can be shown that without
phosphate content, the yield of 5-HMF (19.3%) and furfural
(7.2%) is low. It is obviously shown that the incorporation of
phosphate sites could enhance catalytic activity, and the best
products can attribute to 5% phosphate doped SnO2, which
can result in 46.4% 5-HMF and 18.9% furfural yield. With
the increase of NH4H2PO4 contents further, the 5-HMF and
furfural decreased. Thus, in order to investigate the efects
of acidity over phosphated and bare SnO2 on catalytic activ-
ity, a comparison between products yield and catalysts total
acidity (on the basis of NH3-TPD) were concluded in Fig.
S4. The results indicated that the yield of 5-HMF increased
from 19.3% to 46.4% and the yield of furfural increased
from 7.2% to 18.8% with the increase of total acidity from
12.9 to 42.2 µmmol/g. The absence of phosphate for bare
SnO2 could be explained because of lack of BrØnsted acid
sites that resulted in low products yield. While the higher
Fig. 4 E a Efect of water contents on 5-HMF production from glu-
cose in GVL/water catalyzed by 15%P–SnO2. Reaction conditions: 50
mg glucose, 5 mL GVL, 0–0.75 mL water, 50 mg 15%P–SnO2, 180
°C, 90 min; b Efect of water contents on 5-HMF production from
glucose in DMSO/water catalyzed by 15%P–SnO2. Reaction condi-
tions: 50 mg glucose, 5 mL DMSO, 0–0.75 mL water, 50 mg 15%P–
SnO2, 180 °C, 90 min
the highest yields of 5-HMF. Interestingly, furfural did not
been detected in DMSO system, however, 12.48% furfural
yield can be obtained in GVL/water system. It means that
the species of products might depend on the type of solvents.
The efect of the amount of water in GVL solvent on the
co-production of 5-HMF and furfural from glucose was
studied detailedly and the results are shown in Fig. S3. When
5% water was used, the 39.2% 5-HMF yield and 12.5% fur-
fural yield could be achieved at 180 °C, 90 min. The result
indicated that the presence of water in the reaction system
exhibited a positive efect on the 5-HMF yield. While the
water content is lower than 5%, the presence of water also
has a positive efect on the furfural yield. However, with the
1 3

W. Li et al.
total acidity reached to 62.3 and 84.1 µmmol/g, the 5-HMF
yield slightly dropped to 44.8% and 39.2%. Consequently,
comparing with bare SnO2, the incorporation of phosphate
into SnO2 framework can enhance the acidity of catalysts
largely, which can promote catalytic activity of glucose
conversion. Furthermore, with the changing of phosphate,
a suitable catalyst acidity can also infuence the catalytic
results, implying that SnO2/phosphate ratio primarily deter-
mine 5-HMF selectivity [43].
3.6 Production of 5‑HMF and Furfural from Glucose
Catalyzed by P‑SnO2 Under Diferent Reaction
Temperatures
The efects of reaction temperature on the conversion of glu-
cose to 5-HMF and furfural were examined and the results
are shown in Fig. 6. In the case of glucose transformation,
the reaction temperature exhibited a signifcant impact on
the reaction with respect to the yield of 5-HMF and fur-
fural. The results refected that 5-HMF and furfural yields
increased with temperature in the range of 170 to 185 °C,
with values of 44.28% for 5-HMF and 16.79% for furfural
achieved at 185 °C, which indicated that higher tempera-
tures benefted the reaction process. Upon further prolong-
ing the reaction temperature beyond the peak value, 5-HMF
yields decreased, which may be attributed to the view that
higher reaction temperature resulted in side reactions. Fur-
thermore, the reaction temperature for achieving the 5-HMF
peak value (185 °C) was lower than that for the furfural peak
value (190 °C) at the same time.
3.7 Proposed Reaction Route by P Doped SnO2
in GVL/Water System
Fig. 6 Production of 5-HMF (a) and furfural (b) from glucose cata-
lyzed by P-SnO2 under diferent reaction temperatures. Reaction con-
ditions: 50 mg glucose, 5 mL GVL, 0.25 mL water, 50 mg catalyst,
175–190 °C, 90 min
It has been shown that the 5-HMF and furfural are main
products in our catalytic system. According to previous
reports, 5-HMF would lose a -CH2O group to result in the
production of furfural via decarbonylation [43]. In order to
investigate the reaction pathway, the production of furfural
was conducted via experiments using 5-HMF as initial sub-
strate. However, no obvious furfural can be detected in our
catalytic system under above conditions (Table S1, entry 1).
Thus, it could be considered that 5-HMF and furfural were
formed through the competitive reaction rather than sequen-
tial reaction. Recently, Wang et al. have investigated the
mechanism of furfural formation from fructose and glucose
by using in situ 13C NMR [44]. Then, in order to investigate
whether furfural could be achieved from fructose pathway
in our reaction systems. As shown in Table S1 (entry 2),
when fructose was used as initial materials, 5-HMF and fur-
fural yield were 13.7% and 2.1%, respectively. Furthermore,
small amounts of formaldehyde can also be detected in our
products by HPLC. Then, a plausible reaction pathway
(Scheme 2) was proposed on the basis of above experi-
mental results and previous reports [44–48]. The glucose
isomerization to fructose was accompanied by the SnO2 and
dehydration reaction of fructose resulted in the formation of
5-HMF with the presence of phosphate. Furthermore, when
furfural was produced from fructose pathway, C5-C6 bond
of fructose would split with P doped SnO2 catalyst and the
C1 compound can be transformed to formaldehyde, suggest-
ing the retro-aldol could be present during the process of
selective scission of C–C bond [44].
3.8 Catalyst Recycles
The stability and reusability of catalyst is one of the most
important factors to evaluate the catalytic performance
1 3

Catalytic Conversion of Glucose to 5-Hydroxymethylfurfural and Furfural by a Phosphate-Doped…
Scheme 2. The possible path-
way for glucose conversion to
5-HMF and furfural in GVL/
water system on 15%P–SnO2
which means that the catalytic ability can be improved
after regeneration.
4 Conclusions
The conversion of glucose to 5-HMF and furfural in the
GVL-water system was accomplished by P doped SnO2
catalysts, which were prepared by using a simple impregna-
tion method. The efect of water contents have been studied
detailedly and the results indicated that with the increase of
water up to 100 wt%, the selectivity of 5-HMF increased
from 17.0 to 63.3%, which demonstrated that the presence
of water in the solvent system can inhibit side reactions. Fur-
thermore, by changing the amount of NH4H2PO4, a series
of P-SnO2 catalyst were prepared, among which 5%P–SnO2
exhibits highest activity (46.4% 5-HMF yield, 18.9% fur-
fural yield). Moreover, the catalyst shows a promising stabil-
ity towards the conversion of glucose to 5-HMF and furfural.
And the good performance of P doped SnO2 can be attrib-
uted to the synergistic efect between SnO2 and phosphate.
Fig. 7 Reusability of 15%P–SnO2. Reaction conditions: 50 mg glu-
cose, 5 mL GVL, 0.25 mL water, 50 mg catalyst, 180 °C, 90 min
of catalyst. The reusability of the P doped SnO2 catalyst
was investigated and the results are presented in Fig. 7.
After each reaction, the catalyst was obtained by fltration,
washed with deionized-water and acetone, and then dried
at 80 °C oven. There was slightly decline (from 99.85%
to 86.54%) in the conversion of glucose for all the reac-
tion cycles, and the yield of 5-HMF and furfural show the
same reduction trend. At the same time, with the increase
of cycles, we observed that the color of the catalyst turned
dark brown and we guessed that the deposition of humin
on the catalyst surface causing the deactivation of cata-
lyst. Then we decided to regenerate the catalyst after fve
cycles and the used catalyst was calcined at 600 °C for 4 h
again. The yield of 5-HMF changed from 35.8% to 40.2%,
Acknowledgements This work was supported by the National Key
Technology R&D Program of China (No. 2018YFB1501601), the
National Natural Science Foundation of China (51976212), “Transfor-
mational Technologies for Clean Energy and Demonstration”, Strategic
Priority Research Program of the Chinese Academy of Sciences, Grant
No. XDA 21060101.
References
1. Rout PK, Nannaware AD, Prakash O, Kalra A, Rajasekharan R
(2016) Chem Eng Sci 142:318–346
2. Yang Y, Liu Q, Li D, Tan J, Zhang Q, Wang C, Ma L (2017) RSC
Adv 27:16311–16318
1 3

W. Li et al.
27. Li K, Du M, Ji P (2018) ACS Sustain Chem Eng 6:5636–5644
28. Marianou AA, Michailof CM, Pineda A, Iliopoulou EF, Trian-
tafyllidis KS, Lappas AA (2018) Appl Catal A 555:75–87
29. Hayashi Y, Sumiya T, Takahashi J, Gotoh H, Urushima T, Shoji
M (2006) Angew Chem Int 45:958–961
3. Cai Q, Liu R, He Y, Chai M, Cai J (2018) Fuel Process Technol
171:308–317
4. Shi N, Liu Q, Zhang Q, Wang T, Ma L (2013) Green Chem
15:1967–1974
5. Chen D, Zheng Z, Fu K, Zeng Z, Wang J, Lu M (2015) Fuel
159:27–32
30. Kurahashi T, Shinokubo H, Osuka A (2006) Angew Chem
118:6484–6486
6. Werpy T, Petersen G (2004) Top value added chemicals from
biomass, vol 1, Results of screening for potential candidates from
sugars and synthesis gas, US DOE Report
31. Xia Y, Liang Y, Chen Y, Wang M, Jiao L, Huang F, Liu S, Li Y,
Yu Z (2007) J Am Chem Soc 12:3470–3471
32. Zhou F, Sun X, Wu D, Zhang Y, Su H (2017) ChemCatChem
9:2784–2789
7. Iris KMY, Daniel CWT (2017) Bioresour Technol 238:716–732
8. Kang S, Zhang G, Yang X, Yin H, Fu X, Liao J, Tu J, Huang X,
Qin F, Xu Y (2017) Energy Fuels 31:2847–2854
9. Yang Y, Ma J, Jia X, Du Z, Duan Y, Xu J (2016) RSC Adv
6:51221–51228
33. Li W, Liu Q, Ma Q, Zhang T, Ma L, Jameel H, Chang HM (2016)
Bioresour Technol 219:753–756
34. Modrogan E, Valkenberg MH, Hoelderich WF (2009) J Catal
261:177–187
10. Torres AI, Daoutidis P, Tsapatsis M (2010) Energy Environ Sci
3:1560
35. Zhao X, Wen T, Zhang J, Ye J, Ma Z, Yuan H, Ye X, Wang Y
(2017) RSC Adv 7:21678–21685
11. Zhang X, Murria P, Jiang Y, Xiao W, Kenttämaa HI, Abu-Omar
MM, Mosier NS (2016) Green Chem 18:5219–5229
12. Liu F, Audemar M, De Oliveira VK, Cartigny D, Clacens JM,
Costa Gomes MF, Pádua AAH, De Campo F, Jérôme F (2013)
Green Chem 15:3205
36. Huang F, Su Y, Tao Y, Sun W, Wang W (2018) Fuel 226:417–422
37. Atanda L, Mukundan S, Shrotri A, Ma Q, Beltramini J (2015)
ChemCatChem 7:781–790
38. Guo S, Han S, Mao H, Zeng C, Sun Y, Chi B, Pu J, Li J (2013)
Mater Res Bull 48:3022–3036
13. Pagán-Torres YJ, Wang T, Gallo JMR, Shanks BH, Dumesic JA
(2012) ACS Catal 2:930–934
39. Drevet R, Dragoé D, Barthés-Labrousse MG, Chaussé A,
Andrieux M (2016) Appl Surf Sci 384:442–448
40. Zhang X, Zhang D, Sun Z, Xue L, Wang X, Jiang Z (2016) Appl
Catal B 196:50–56
14. Teong SP, Yi G, Zeng H, Zhang Y (2015) Green Chem
17:3751–3755
15. Tian G, Tong X, Cheng Y, Xue S (2013) Carbohydr Res
370:33–37
41. Su M, Li W, Zhang T, Xin H, Li S, Fan W, Ma L (2017) Catal Sci
Technol 7:3555–3561
16. Wrigstedt P, Keskiväli J, Leskelä M, Repo T (2015) Chem-
CatChem 7:501–507
42. Hara M (2014) Bull Chem Soc Jpn 87:931–941
43. Yu IKM, Tsang DCW (2017) Bioresour Technol 238:716–732
44. Wang Y, Yang X, Zheng H, Li X, Zhu Y, Li Y (2019) Mol Catal
463:130–139
17. Zhang X, Hewetson BB, Mosier NS (2015) Energy Fuels
29:2387–2393
18. Atanda L, Silahua A, Mukundan S, Shrotri A, Torres-Torres G,
Beltramini J (2015) RSC Adv 5:80346–80352
19. Takagaki A, Ohara M, Nishimura S, Ebitani K (2009) Chem Com-
mun 41:6276–6278
45. Asakawa M, Shrotri A, Kobayashi H, Fukuoka A (2019) Green
Chem 21:6146–6153
46. Cui J, Tan J, Deng T, Cui X, Zhu Y, Li Y (2016) Green Chem
18:1619–1624
20. Hu L, Wu Z, Xu J, Sun Y, Lin L, Liu S (2014) Chem Eng J
244:137–144
47. Wang L, Guo H, Xie Q, Wang J, Hou B, Jia L, Cui J, Li D (2018)
Appl Catal A 572:51–60
21. Van de Vyver S, Roman-Leshkov Y (2015) Angew Chem Int
54:12554–12561
48. Binder JB, Cefali AV, Blank JJ, Raines RT (2010) Energy Environ
Sci 3:765–771
22. Osatiashtiani A, Lee AF, Granollers M, Brown DR, Olivi L,
Morales G, Melero JA, Wilson K (2015) ACS Catal 5:4345–4352
23. Ordomsky VV, Sushkevich VL, Schouten JC, van der Schaaf J,
Nijhuis TA (2013) J Catal 300:37–46
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
24. Yan H, Yang Y, Tong D, Xiang X, Hu C (2009) Catal Commun
10:1558–1563
25. Nakajima K, Baba Y, Noma R, Kitano M, Kondo JN, Hayashi S,
Hara M (2011) J Am Chem Soc 133:4224–4227
26. Hou Q, Zhen M, Liu L, Chen Y, Huang F, Zhang S, Li W, Ju M
(2018) Appl Catal B 224:183–193
1 3