Preparation of 5-hydroxymethylfurfural from glucose catalyzed by silica-supported phosphotungstic acid heterogeneous catalyst

Fuel 226 (2018) 417–422

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Preparation of 5-hydroxymethylfurfural from glucose catalyzed by silicasupported phosphotungstic acid heterogeneous catalyst

T

⁎

Fangmin Huang , Yuwen Su, Yu Tao, Wei Sun, Weiting Wang School of Chemistry and Materials Science, Jiangsu Key Laboratory of Green Synthetic Chemistry for Functional Materials, Jiangsu Normal University, Xuzhou 221116, China

G R A P H I C A L A B S T R A C T

A R T I C LE I N FO

A B S T R A C T

Keywords: 5-Hydroxymethylfurfural Glucose Phosphotungstic acid Silica gel

Conversion of carbohydrates into 5-hydroxymethylfurfural (HMF) is a valuable reaction for biomass efficient utilization. The silica-supported phosphotungstic acid (PTA) heterogeneous catalyst (SiO2-ATS-PTA) was prepared and characterized by XRD, FT-IR, pyridine-FTIR and SEM. The SiO2-ATS-PTA was used to catalyze the conversion of glucose, and it afforded 78.31% yield of HMF at 160 °C for 140 min. The SiO2-ATS-PTA could be reused for five runs without significant loss of catalytic activity. Furfural was the main byproduct. A kinetic analysis was carried out and the values of the activation energy and the pre-exponential factor for the reaction were 30.25 kJ mol−1 and 1.08 × 107 min−1, respectively.

1. Introduction Recently, biomass is attracting much attention as promising new resource due to their availability and abundance [1,2]. Practical pathways were developed to transform biomass-derived carbohydrates into a broad range of value-added compounds [3]. 5-Hydroxymethylfurfural (HMF) has received significant attention as one of the priority chemicals on the list of “ten bio-based chemicals” published by the U.S. Department of Energy [4–6]. Rapid progress in the development of

⁎

efficient catalytic approaches for HMF production has been witnessed over past few years [7–10]. Homogeneous catalysts, such as inorganic acid [11], CrCl2 and CrCl3·6H2O etc. [12,13] usually bring excellent yield of HMF while they also cause environmental pollution, equipment corrosion and hard catalyst separation, which limit the industrial application process of them. Recently, heterogeneous catalysts draw researches’ attention with basic characteristic of easy separation [14,15] and varies of heterogeneous catalysts were investigated [16], such as mesoporous

Corresponding author. E-mail address: [email protected] (F. Huang).

https://doi.org/10.1016/j.fuel.2018.03.193 Received 23 November 2017; Received in revised form 28 February 2018; Accepted 30 March 2018 0016-2361/ © 2018 Elsevier Ltd. All rights reserved.

Fuel 226 (2018) 417–422

F. Huang et al.

Scheme 1. The structure diagram of target catalyst SiO2-ATS-PTA.

performed using a USA PE Frontier FT-IR Spectrometer. The concentrations of the Brønsted and Lewis acidic sites on the samples were determined using the FT-IR spectra of adsorbed pyridine; the sample was pressed into a self-supporting wafer (10–15 mg/cm−2, diameter = 10 mm) and was inserted into a measurement cell with KBr windows that was connected to a vacuum apparatus. The wafer was treated at 350 °C under vacuum for 2 h and was subsequently cooled to room temperature to collect the background spectra. The pyridine adsorption was performed by equilibrating the wafer for 30 min at room temperature. To calculate the weak acidic sites and the medium and strong acidic sites, the IR spectra for the samples were recorded after degassing for 60 min at 473 K. The specific surface areas of the target catalyst were examined via nitrogen adsorption at 77 K (Quantachrome Instruments, Quadrasorb SI, America). The specific surface areas were calculated using the Brunauer–Emmett–Teller (BET) method.

aluminum doped MCM-41 silica [9], zeolite [17], amorphous Cr2O3, SnO2, SrO [7], and modified tin oxide [18] etc. However, severe reaction conditions were usually employed in heterogeneous catalytic reaction process [19]; ionic liquid and polar organic solvent were usually used as solvent, which would cause environmental pollution [20,21]. So, researchers have turned their attentions to more promising heterogeneous catalysts with sustainability and environmental friendliness, to explore mild, effective and green catalytic reaction processes. Heteropoly acid was studied extensively because its chemical properties can be varied considerably due to the structural characteristics [22,23]. Heteropoly acid could be converted into heterogeneous catalyst when it was supported on certain supporter [24]. Gomes [25] discovered that the heterogeneous catalyst with Phosphotungstic acid (PTA) supporting on MCM-41 was effective for HMF production from fructose. However, silica gel was suggested to be a more suitable supporter for PTA than MCM-41 [26]. In this work, the catalyst SiO2-ATSPTA with PTA supporting on silica gel was synthesized and used to catalyze the conversion of glucose, which is considered as a more ideal starting material than fructose, while it is simultaneously more difficult to be conversed because of its molecular structure [27]. The catalytic performance of SiO2-ATS-PTA for HMF formation from glucose was studied and a kinetic analysis was carried out.

2.2. Catalyst preparation The target catalyst SiO2-ATS-PTA was prepared as: 4.00 g SiO2 was added into 40 mL HNO3 aqueous (30%) under 70 °C with water reflux for 70 min, then the reaction mixture was filtered, washed with water for three times, dried under 65 °C in a vacuum oven, collected and denoted as SiO2-A; 3.00 g SiO2-A was added in 6 mL toluene with 1 g aminopropyltrimethoxysilane (ATS) as adhesive, then was placed in an ultrasound generator under 80 MHz for 10 min, followed by being heated for 70 min at 70 °C with stirring at 500 rpm, dried under 65 °C in a vacuum oven, collected and denoted as SiO2-ATS; 1 g SiO2-ATS was mixed with 1.5 g PTA and 6 mL methanol was used as solvent. The reaction mixture was stirred for 2000 rpm under 30 °C for 20 h, then was filtered, washed with methanol for three times, dried at 65 °C in a vacuum oven, denoted as SiO2-ATS-PTA. The structure diagram of the catalyst SiO2-ATS-PTA was shown in Scheme 1. The mass proportion of SiO2:ATS:PTA in the 1.75 g target catalyst SiO2-ATS-PTA was 0.75 g:0.25 g:0.75 g.

2. Experimental 2.1. Materials and methods All chemicals were of analytical grade and used as received without any further purification. X-ray diffraction (XRD) of the catalyst SiO2ATS-PTA were recorded with Cu K < alpha > (D2 PHASER) radiation (λ = 0.1541 nm) using a PANalyticalXpert Pro instrument (BRUKER) at 25 °C with a silicon mono-crystal sample holder at step size of 0.017°. The intensity (Miller indices) as a function of 2θ was measured while the angle range was 5–40°. Fourier-transform infrared (FTIR) spectrum (4000–500 cm−1) of the catalyst SiO2-ATS-PTA was recorded by a Bruker Vertex 80 V FTIR vacuum spectrometer (Ettlingen, Germany) with a resolution of 2 cm−1 and 32 scans per sample. Field emission scanning electron microscope (FESEM; Hitachi SU8010, accelerated voltage: 15 kV) was used to study the morphology of the catalysts. Pyridine Fourier-transform infrared (Py-FTIR) experiments were

2.3. Typical procedure for the catalytic conversion of glucose into HMF Qi et al. proved the dehydration of HMF was restricted and the steady of HMF increased in water when acetone was used as co-solvent with water [28]. So acetone/water mixture solution 418

Fuel 226 (2018) 417–422

F. Huang et al.

(Vacetone:Vwater = 1:5) was used as solvent in this paper. The proportion of Vacetone:Vwater in the acetone/water mixture solvent used in this paper has been optimized, as shown in Fig. S1. In a typical run, a acetone/water mixture solution (Vacetone = 1 mL; Vwater = 5 mL), 0.25 g glucose and 0.20 g SiO2-ATS-PTA were added to a sealed 25 mL thick-walled glass reactor. The reaction mixture was heated under different temperatures for a given reaction time with being stirred at 500 rpm. The reaction was quenched by introducing the reactor into a cooled water bath after certain reaction time. The sample was then diluted, filtered, and further used for product analysis. For catalyst recycling test, the SiO2-ATS-PTA was separated from the reaction mixture by filtration and washed with deionized water for several times. Subsequently, the recycled SiO2-ATS-PTA was dried in a vacuum oven at 60 °C for 12 h.

PTA SiO2

Intensity(a.u.)

SiO2-ATS-PTA

10

2.4. Determination of the products

Glucose conversion:x g = [(Cg-0−Cg-t )/Cg-0] ∗100

(1)

Yield of HMF:yp = [(Cp-t−Cp-0 )/ Cg-0] ∗100

(2)

40

50

60

70

80

Fig. 2. XRD pattern of commercial PTA, SiO2, and SiO2-ATS-PTA.

patterns of the commercial HPW, the support SiO2 and the SiO2-ATSPTA. Only broad and diffuse diffraction peaks were observed in the XRD pattern of the target catalyst SiO2-ATS-PTA and no diffraction peak connected to PTA species were discovered, which indicates that the SiO2-ATS-PTA was amorphous, and it could be inferred that the PTA species were finely dispersed on the surface of supporter SiO2. The element mapping photos of the target catalyst SiO2-ATS-PTA are shown in Fig. 3, which indicates that element W and P were both finely dispersed on the surface of supporter SiO2, and it further proved that the PTA species were finely dispersed on the supporter SiO2. The FTIR spectrum of pyridine adsorption and desorption of the catalyst SiO2-ATS-PTA was shown in Fig. 4. The catalyst SiO2-ATS-PTA contains Brønsted and Lewis acid sites, as evidenced by the absorbance peaks at about 1538 cm−1 and 1445 cm−1 in the desorption spectrum under 200 °C, respectively. The concentrations of Brønsted and Lewis acid site of the catalysts SiO2-ATS-PTA were calculated as 14.73 μmol/g and 27.16 μmol/g, respectively. The Brønsted acid sites are obviously due to the presence of HPW, and the Lewis acid sites could be possible due to the interaction between HPW with framework SiO2 [23], which need to be further evidenced.

Where Cg and Cp corresponds to the molar concentration of glucose and product (HMF), respectively, and the subscripts 0 and t correspond to t = 0 and reaction time t, respectively. 3. Results and discussion 3.1. Synthesis and characterization of the catalysts The target catalyst SiO2-ATS-PTA was synthesized and the structure confirm was studied by FT-IR, XRD and pyridine-FTIR. As shown in Fig. 1, obvious absorption peaks around 949 cm−1, 894 cm−1 and 858 cm−1 were observed in the FT-IR spectrum of SiO2-ATS-PTA. And these peaks were assigned to PTA species, which demonstrate that the PTA species were loaded on the supporter SiO2. Fig. 2 shows the XRD

Transmittance(%)

30

Theta(2 )

HMF concentration was determined using high-performance liquid chromatography (HPLC, Agilent 1200) using a column (Zorbax SB-C18) with a UV detector to analyze HMF yield and the column was maintained at a column temperature of 30 °C, using water-methanol (Vmethanol:Vwater=15:85) as the mobile phase at a flow rate of 0.4 mL·min−1. Glucose conversion was determined with a refractive index detector and an Carbomix H-NP column maintained at a column temperature of 55 °C, using H2SO4 solution (25 mmol·L−1) as the mobile phase at a flow rate of 0.6 mL·min−1.

3.2. Catalytic performance of catalysts The target catalyst SiO2-ATS-PTA and several control catalysts were tested for glucose conversion, and then the glucose conversion (xg) and HMF yield (yp) of these samples were calculated and shown in Table 1. Under the given conditions, blank test gave a low HMF yield of 5.3% (Table 1, entry1). A low HMF yield was obtained when using SiO2 as catalyst (Table 1, entry 2), suggesting the supporter SiO2 is inactive for the glucose conversion. After loading ATS on the supporterSiO2, the SiO2-ATS also gave a HMF yield of 4.98% (Table 1, entry 3), suggesting the SiO2-ATS is also inactive for the conversion of glucose. PTA and the target catalyst SiO2-ATS-PTA were also employed as catalysts for glucose conversion, and the results were shown in entry 4 and entry 5, respectively. The target catalyst SiO2-ATS-PTA leads a HMF yield 78.31%, which is much higher than that for PTA. The result indicates that the catalytic activity of the target catalyst SiO2-ATS-PTA was much higher than the unsupported PTA and it probably because the specific surface area of SiO2-ATS-PTA (290.55 m2/g) is much higher than that of PTA (2.48 m2/g). The catalytic performance of SiO2-ATS-PTA is compared with some previously reported works as shown in Table 2. The HMF yield of 78.31% obtained with the catalyst SiO2-ATS-PTA in the present work is much higher than that over the reported heterogeneous catalysts as FePO4 [29], Sn-Beta/HCl [30], Al-MCM [31], SnPO [32], Beta-Cal750 [33], TaPO [34] and SO4/ZrO2 [35]. This reaction result is even better

858 cm -1 949 cm -1

SiO2 894 cm-1

20

SiO2-ATS-PTA

900

600

Wavenumber(cm-1) Fig. 1. FTIR pattern of S iO2 and SiO2-ATS-PTA. 419

Fuel 226 (2018) 417–422

F. Huang et al.

Fig. 3. Element mapping of SiO2-ATS-PTA. Table 2 Comparisons of catalytic behavior of SiO2-ATS-PTA and representative catalytic systems for conversion of glucose to HMF.

Absorbance(a. u.)

Adsorption 200 °C 350 °C

1650

1600

1550

1500

1450

1400

Catalyst

Solvent

Loada

T (°C)

t/min

yp/%

Ref.

SiO2-ATS-PTA FePO4 Sn-Beta/HCl Al-MCM41 SnPO Beta-Cal750b TaPOc SO4/ZrO2 FeCl3 CrCl3

Water/acetone Water/THF Water Water/MIBK [EMIM]Br Water/DMSO Water/MIBK Water [EMIM]Cl [EMIM]Cl

4.0% 2.5% 10.0% – 10.0% 2.4% 3.0% 0.5% 6.0% 6.0%

160 140 140 195 120 180 170 100 80 80

140 60 120 150 180 60 60 360 180 180

78.31 22.60 7.92 31.30 54.90 20.20 18.50 10.00 10.00 70.00

This work [29] [30] [31] [32] [20] [33] [34] [13] [13]

a Load = mglucose: (mglucose + msolvent), in which mglucose is the quantity of glucose and msolvent is the quantity of the solvent. b Beta-Cal750; Beta-NH4 was calcined in air at 750 °C to obtain the protonform samples. c TaPO is the abbreviation of tantalum phosphate.

1350

Wavenumber(cm-1) Fig. 4. The Pyridine-adsorption FT-IR spectra of the catalyst SiO2-ATS-PTA.

100 130

Table 1 Glucose conversion and yield of HMF of different catalysts for glucose dehydration. Solvent

Glucose conversion/%

Yield/%

Blank SiO2 SiO2-ATS PTA SiO2-ATS-PTA SiO2-ATS-PTA

Water/solvent Water/solvent Water/solvent Water/solvent Water/solvent Water

19.21 27.64 28.67 75.38 99.87 99.77

5.25 5.95 4.98 52.27 78.31 52.21

The HMF yield(%)

1 2 3 4 5 6

Catalyst

140 80

Condition: 160 °C, 140 min, 0.2 g SiO2-ATS-PTA; 0.0857 g PTA, 0.1143 g SiO2ATS, 0.0857 g SiO2, 500 rpm, acetone/water (1 mL/5 mL).

150 160

60

40

20

than the one offered by the reported homogeneous catalyst, such as FeCl3 (10.0%) and CrCl3 (70.0%), which employed ionic liquid as solvent [13].

50

100

150

200

Time(min) Fig. 5. Variation of HMF yield with reaction temperature. Conditions: 0.2 g SiO2-ATS-PTA; 500 rpm; acetone/water (1 mL/5 mL).

3.2.1. Effect of reaction temperature and reaction time Effect of reaction temperature on the yield of HMF from glucose was studied. Several parallel experiments under different reaction temperatures as 130 °C, 140 °C, 150 °C and 160 °C were carried. The yields of HMF in these samples were calculated and the results were shown in Fig. 5. The reaction time required to attain a HMF yield of 49.21% was only 70 min at 160 °C, while 115 min was required to achieve 49.52% yield of HMF at 140 °C and only achieve 25.2% yield of HMF for 70 min at 130 °C. At 160 °C, the HMF yield increased rapidly in the initial reaction stages, and the maximum yield of 78.31% was obtained after 140 min; while at 140 °C, the maximum yield of 60.2% was obtained for

160 min reaction time. Besides, in all cases, the HMF yield decreased on further extending the reaction time and it decreased more sharply under higher reaction temperature. It is possible because the side reactions increased and unidentified soluble polymers and humin formed while reaction time extended [35]. 3.2.2. Effect of dosage The effectiveness of catalyst SiO2-ATS-PTA dosage on glucose conversion and HMF yield was also studied. The catalyst dosage was varied 420

Fuel 226 (2018) 417–422

F. Huang et al.

120

Glucose conversion HMF yield

120

Glucose conversion HMF yield

80

%

%

80

40

40

0

0

0.08

0.12

0.16

0.20

1

0.24

2

Dosage(g)

Table 3 Reaction rate constant of glucose conversion at different reaction temperature.

1 2 3 4

k (min−1)

Temperature (°C) 130 140 150 160

0.010 0.017 0.026 0.043

± ± ± ±

9.14 × 10 1.50 × 10−3 6.72 × 10−3 3.96 × 10−3

5

constant k increased with reaction temperature, indicating that higher temperature accelerate the glucose conversion rate [36]. An Arrhenius plot was generated using the value of rate constant. The kinetic parameters of glucose conversion to HMF catalyzed by the SiO2-ATS-PTA are summarized in Table 4. The value of activation energy was lower while the value of pre-exponential factor were much higher than those reported by Qu et al. (2012) (55.77 kJ mol−1 and 1.6 × 104 min−1, using room temperature ionic liquid as solvent and [C2OHMIM]BF4 as catalyst). It was possible because the conversion reaction rate was accelerated by higher diffusion velocity in water/ acetone solvent than that in room temperature ionic liquid solvent.

Correlation coefficient −4

4

Fig. 8. The reuse performance of SiO2-ATS-PTA. Conditions: 500 rpm, acetone/ water (1 mL/5 mL), 160 °C, 140 min, 0.2 g SiO2-ATS-PTA was added for the first run.

Fig. 6. Variation of HMF concentrations with the catalyst dosage. Conditions: 500 rpm, acetone/water (1 mL/5 mL), 160 °C, 140 min.

Entry

3

Reuse times

0.9812 0.9848 0.9756 0.9798

Conditions: 0.2 g SiO2-ATS-PTA; 500 rpm; acetone/water (1 mL/5 mL). Table 4 Kinetic parameters for conversion of glucose.

as 0.08 g, 0.12 g, 0.16 g, 0.20 g and 0.24 g. The glucose conversion and yield of HMF were calculated and shown in Fig. 6, which indicates that both glucose conversion and HMF yield reached the maximum value (99.89%, 78.31%) when the catalyst dosage was 0.20 g. The decreasing of HMF yield on adding more catalysts can be explained by assuming that more catalyst species promotes the side reaction.

3.2.4. Byproducts analysis In order to confirm the side reactions along with the conversion of glucose to HMF, the products in these reaction samples for glucose conversion in water/acetone solvent was determined by HPLC-MS. It is notable that furfural was discovered as the main byproduct, as shown in Fig. S4. The concentration of furfural was negligible in these samples below 140 min reaction time while it became assignable when the reaction time kept increasing. The concentration of furfural shows an increasing trend with the reaction time increasing. The generation path of furfural in this reaction mixture was suggested and shown as path (2) in Fig. 7. The acetone was considered to contribute the formation of furfural and at the same time to inhibit the other two side reaction paths (path (1) and (3)).

3.2.3. Kinetic analysis of the dehydration of glucose The kinetic analysis of the conversion of glucose in water/acetone in the presence of SiO2-ATS-PTA was performed. Value of ln (1 − xg) (where xg is glucose conversion) was plotted against reaction time t under different temperature aiming to obtain the value of rate constant k (as shown in Table 3). The results indicate that the value of rate

3.2.5. Catalysis recycle The reusability performance is important to evaluate the efficiency of a heterogeneous catalytic system according to the principles of green and sustainable chemistry. In this work, 0.2 g SiO2-ATS-PTA was added in the first run, and the recovered SiO2-ATS-PTA (According Part 2.3 in this paper) was used for the next reaction run without adding any new

Parameter

Value

Reaction order, n Activation energy, Ea (kJ·mol−1) Pre-exponential factor, A (min−1) Correlation coefficient

1 30.25 1.08 × 107 0.9992

Humin (1)

OH O

HO

HO

O OH OH

HO OH

OH

OH

HO

O

O

O (2)

O

O

O HO

OH

O

(3)

H3C

OH

HCOOH O Fig. 7. The generation path of the main byproduct. 421

Fuel 226 (2018) 417–422

F. Huang et al.

SiO2-ATS-PTA, repeat this operation till the fifth run. The recycling performance of catalyst SiO2-ATS-PTA was studied and shown in Fig. 8. The glucose conversion and the HMF yield decreased 6.51% and 9.96% when the SiO2-ATS-PTA was reused the fifth times. A reason for the HMF yield decreasing was possible the loss of catalyst in the recycling process. Besides, some active site could be covered by the generated humin during the reaction process [35]. A BET determination of the recovered SiO2-ATS-PTA was carried and the result indicates that the specific surface area of SiO2-ATS-PTA decreased from 290.55 m2/g to 234.38 m2/g after five runs. The decreasing of the specific area was another reason for the catalyst deactivation during the recycling process.

[10] Zhang L, Xi G, Chen Z, Qi Z, Wang X. Enhanced formation of 5-HMF from glucose using a highly selective and stable SAPO-34 catalyst. Chem Eng J 2017;307:877–83. [11] Fachri BA, Abdilla RM, van de Bovenkamp HH, Rasrendra CB, Heeres HJ. Experimental and kinetic modeling studies on the sulfuric acid catalyzed conversion of D-fructose to 5-hydroxymethylfurfural and levulinic acid in water. ACS Sustainable Chem Eng 2015;3:3024–34. [12] Hu L, Sun Y, Lin L. Efficient conversion of glucose into 5-hydroxymethylfurfural by chromium(III) chloride in inexpensive ionic liquid. Ind Eng Chem Res 2012;51:1099–104. [13] Zhao H, Holladay JE, Brown H, Zhang ZC. Metal chlorides in ionic liquid solvents convert sugars to 5-hydroxymethylfurfural. Science 2007;316:1597–600. [14] Morales G, Osatiashtiani A, Hernandez B, Iglesias J, Melero JA, Paniagua M, et al. Hydrothermally stable, conformal, sulfated zirconia monolayer catalysts for glucose conversion to HMF. Chem Commun 2015;5:4345–52. [15] Kourieha R, Rakicb V, Bennicia S, Auroux A. Relation between surface acidity and reactivity in fructose conversion into HMF using tungstated zirconia catalysts. Catal Commun 2013;30:5–13. [16] Huang H, Denard CA, Alamillo R, Crisci AJ, Miao Y, Dumesic JA, et al. Tandem catalytic conversion of glucose to 5-hydroxymethylfurfural with an immobilized enzyme and a solid acid. ACS Catal 2014;4:2165–8. [17] Hua L, Wu Z, Xua J, Sun Y, Lin L, Liu S. Zeolite-promoted transformation of glucose into 5-hydroxymethylfurfural in ionic liquid. Chem Eng J 2014;244:137–44. [18] Lopes M, Dussan K, Leahy JJ, Silva VTD. Conversion of d-glucose to 5-hydroxymethylfurfural using Al2O3-promoted sulphated tin oxide as catalyst. Catal Today 2017;279:233–43. [19] Zhou L, Liang R, Ma Z, Wu T, Wu Y. Conversion of cellulose to HMF in ionic liquid catalyzed by bifunctional ionic liquids. Bioresour Technol 2013;129:450–5. [20] Otomo R, Yokoi T, Kondo JN, Tatsumi T. Dealuminated beta zeolite as effective bifunctional catalyst for direct transformation of glucose to 5-hydroxymethylfurfural. Appl Catal A: Gen 2014;470:318–26. [21] Morales G, Osatiashtiani A, Hernandez B, Iglesias J, Melero JA, Paniagua M, et al. Hydrothermally stable, conformal, sulfated zirconia monolayer catalysts for glucose conversion to 5-HMF. Chem Commun 2015;5:4345–52. [22] Seaton K, Little I, Tate C, Mohseni R, Roginskaya M, Povazhniy V, et al. Adsorption of cesium on silica gel containing embedded phosphotungstic acid. Microporous Mesoporous Mater 2017;244:55–66. [23] Ma T, Ding J, Shao R, Xu W, Yun Z. Dehydration of glycerol to acrolein over WellsDawson and Keggin type phosphotungstic acids supported on MCM-41 catalysts. Chem Eng J 2017;316:797–806. [24] Zeng Z, Cui L, Xue W, Ma N. Study on adsorption behavior of 12-phosphotungstic acid on silica gel. Ind Eng Chem Res 2013;52:8070–8. [25] Gomes F, Mendes F, Souza M. Synthesis of 5-hydroxymethylfurfural from fructose catalyzed by phosphotungstic acid. Catal Today 2017;279:296–304. [26] Sebulsky RT, Henke AM. Alkylation of benzene with dodecene-1 catalyzed by supported silicotungstic acid. Ind Eng Chem Process Res Dev 1971;10(2):272–2729. [27] Román-Leshkov Y, Chheda JN, Dumesic JA. Phase modifiers promote efficient production of hydroxymethylfurfural from fructose. Science 2006;312:1933–7. [28] Qi X, Watanabe M, Aida TM, Smith Jr. RL. Catalytic dehydration of fructose into 5hydroxymethylfurfural by ion-exchange resin in mixed-aqueous system by microwave heating. Green Chem 2008;10:799–805. [29] Yang L, Yan X, Xu S, Chen H, Xia H, Zuo S. One-pot synthesis of 5-hydroxymethylfurfural from carbohydrates using an inexpensive FePO4 catalyst. RSC Adv 2015;5:19900–6. [30] Nikolla E, Román-Leshkov Y, Moliner M, Davis ME. One-pot synthesis of 5-hydroxymethylfurfural from carbohydrates using tin-beta zeolite. ACS Catal 2011;1:408–10. [31] Morales IJ, Recio MM, González JS, Torres PM, López AJ. Production of 5-hydroxymethylfurfural from glucose using aluminium doped MCM-41 silica as acid catalyst. Appl Catal B: Environ 2015;164:70–6. [32] Hou Q, Zhen M, Liu L, Chen Y, Huang F, Zhang S, et al. Tin phosphate as a heterogeneous catalyst for efficient dehydration of glucose into 5-hydroxymethylfurfural in ionic liquid. Appl Catal B: Environ 2018;224:183–93. [33] Jiménez-Morales I, Teckchandani-Ortiz A, Santamaría-González J, MairelesTorres P, Jiménez-López A. Selective dehydration of glucose to 5-hydroxymethylfurfural on acidic mesoporous tantalum phosphate. Appl Catal B 2014;144:22–8. [34] Osatiashtiani A, Lee AF, Brown DR, Melero JA, Morales G, Wilson K. Bifunctional SO4/ZrO2 catalysts for 5-hydroxymethylfufural (5-HMF) production from glucose. Catal Sci Technol 2014;4:333–42. [35] Fachri BA, Abdilla RM, van de Bovenkamp HH, Rasrendra CB, Heeres HJ. Experimental and kinetic modeling studies on the sulfuric acid catalyzed conversion of D-fructose to 5-hydroxymethylfurfural and levulinic acid in water. ACS Sustainable Chem Eng 2015;3:3024–34. [36] Qu Y, Huang C, Song Y, Zhang J, Chen B. Efficient dehydration of glucose to 5hydroxymethylfurfural catalyzed by the ionic liquid, 1-hydroxyethyl-3-methylimidazolium tetrafluoroborate. Bioresour Technol 2012;121:462–6.

4. Conclusion When the heterogeneous catalyst SiO2-ATS-PTA was used to catalyze the glucose conversion in acetone/water (1:5, v/v) solvent, high yield of HMF (78.31%) was attained under 160 °C for 140 min, which indicates that the SiO2-ATS-PTA is an effective heterogeneous catalyst for conversion of glucose to HMF. The SiO2-ATS-PTA could be reused for five times without significant loss of catalytic activity. Acknowledgments This research is funded by NSFC – China (Nos. 21503098, 21603089), TAPP, PAPD, Jiangsu Province Science Foundation for Youths (BK20160209), Natural Science Foundation of the Higher Education Institutions of Jiangsu Province (No. 15KJB150005, No. 15KJD530002, No. 16KJB150014). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fuel.2018.03.193. References [1] Catrinck MN, Ribeiro ES, Monteiro RS, Ribas RM, Barbosa MHP, Teófilo RF. Direct conversion of glucose to 5-hydroxymethylfurfural using a mixture of niobic acid and niobium phosphate as a solid acid catalyst. Fuel 2017;159:280–6. [2] Zhang Y, Wang J, Li X, Liu X, Xia Y, Hu B, et al. Efficient conversion of glucose to HMF using organo catalysts with dual acidic and basic functionalities-A mechanistic and experimental study. Fuel 2017;162:30–6. [3] Moreno-Recio M, Santamaría-González J, Maireles-Torres P. Brönsted and Lewis acid ZSM-5 zeolites for the catalytic dehydration of glucose into 5-hydroxymethylfurfural. Chem Eng J 2016;303:22–30. [4] Pagá n-Torres YJ, Wang T, Gallo JMR, Shanks BH, Dumesic JA. Production of 5hydroxymethylfurfural from glucose using a combination of Lewis and Brønsted acid catalysts in water in a biphasic reactor with an alkylphenol solvent. ACS Catal 2012;2:930–4. [5] Zhang Y, Pan J, Shen Y, Shi W, Liu C, Yu L. Brønsted acidic polymer nanotubes with tunable wettability toward efficient conversion of one-pot cellulose to 5-hydroxymethylfurfural. ACS Sustainable Chem Eng 2015;3:871–9. [6] Song J, Fan H, Ma J, Han B. Conversion of glucose and cellulose into value-added products in water and ionic liquids. Green Chem 2013;15(10):2619–35. [7] Zhang M, Su K, Song H, Li Z, Cheng B. The excellent performance of amorphous Cr2O3, SnO2, SrO and graphene oxide-ferric oxide in glucose conversion into 5HMF. Catal Commun 2015;69:76–80. [8] Chen D, Liang F, Feng D, Xian M, Zhang H, Liu H, et al. An efficient route from reproducible glucose to 5-hydroxymethylfurfural catalyzed by porous coordination polymer heterogeneous catalysts. Chem Eng J 2016;300:177–84. [9] Morales IJ, Recio MM, González JS, Torres PM, López AJ. Production of 5-hydroxymethylfurfural from glucose using aluminium doped MCM-41 silica as acid catalyst. Appl Catal B: Environ 2015;164:70–6.

422