Microwave assisted rapid conversion of fructose into 5-HMF over solid acid catalysts

Abstract

In this work, we investigated the dehydration of fructose into 5-hydroxymethylfurfural using solid catalysts with microwave assistance in dimethyl sulfoxide. Five kinds of solid catalyst ZrO₂, WO_x/ZrO₂, MoO_x/ZrO₂, SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ were prepared and characterized by XRD, UV-DRS, FTIR, XPS, TEM, TPD, BET, and Raman and the surface acid amount was obtained by combining pyridine adsorption and UV spectrometry. The SO₄²⁻/WO_x–ZrO₂ catalyst was found be the most active catalyst, and a fructose conversion of 95.80% with 83.90% 5-HMF yield was obtained at 150 °C in a relatively short reaction time of 5 min. The value of the activation energy was comparable with previous values reported in the literature, implying that SO₄²⁻/WO_x–ZrO₂ was efficient for the dehydration of fructose to 5-HMF and that microwave radiation heating had a remarkable accelerating effect on the fructose conversion.

---

1. Introduction

The decrease of fossil fuel reserves and the deterioration of the environment are driving people to seek sustainable resources. As an abundantly available, cheap, renewable and low sulfur content resource, biomass can not only act as fuel, but can also be converted into chemical intermediates for the production of various chemicals and liquid fuels. In the conversion of biomass, the study of the dehydration of C₆-sugars into 5-hydroxymethylfurfural (5-HMF) has received considerable attention.^1,2 5-HMF has a furan ring structure, an aldehyde group, a hydroxyl group and a conjugated diene, and therefore it can through hydrolysis, selective oxidation, hydrogenation and esterification methods, etc., be used to obtain a series of derivatives, such as 2,5-diformylfuran, levulinic acid,³ 2,5-furandicarboxylic acid,⁴ etc., which are widely used in the areas of drugs, fuels, and polymeric materials.^5,6

In earlier times, many types of acid catalyst such as mineral acids (like H₂SO₄, HCl, H₃PO₄),⁷ organic acids (such as formic acid and acetic acid),⁸ and strong acid cation exchange resins^9,10 were used for the dehydration of fructose. According to the earlier literature,^11,12 the reaction mechanism for obtaining 5-HMF from fructose over an acid catalyst is shown in Scheme 1. In addition, ionic liquids have served as both solvents and catalysts for the dehydration of fructose into HMF^13,14 and were proven effective, but their synthesis process is complex and they have serious drawbacks in terms of separation and recycling. In contrast, heterogeneous catalysts showed superior behavior in terms of easy recovery and recyclability for the dehydration of sugar.¹⁵ R. Kourieh et al. studied the surface acidic properties of WO_x/ZrO₂ for the aqueous hydrolysis of cellobiose, revealing that WO_x/ZrO₂ catalysts, well known for their high acid surface chemistry and the presence of acidic sites of different natures (i.e. Lewis acid sites from zirconia and both Brönsted and Lewis sites from WO_x), could potentially be good candidates for the near-boiling water phase biomass reaction.¹⁶

Scheme 1 The reaction mechanism for 5-HMF from fructose.

Several approaches have been reported for the dehydration of carbohydrates: Simeonov et al. synthesized 5-HMF by dehydration of fructose under either batch or flow chemistry conditions;¹⁷ Jadhav et al. carried out the dehydration reaction of sugar in superheated water in a stainless tubular reaction cell;¹⁸ and De et al. used AlCl₃ as a catalyst for fructose dehydration under microwave-assisted heating conditions.¹⁹ Above all, microwave technology has been applied by more and more researchers, it has a unique heating mode which could make organic reaction speeds faster than traditional heating methods and reaction yields are high as there are fewer by-products. Qi et al. studied the catalytic conversion of fructose into 5-HMF by microwave heating and compared conventional sand bath heating with microwave heating, revealing that the latter had a remarkable accelerating effect not only on fructose conversion but also on 5-HMF yield; fructose conversion and HMF yields by microwave heating (91.7% and 70.3%, respectively) were higher than those by sand bath heating (22.1% and 13.9% respectively).¹⁰

In this work, we introduced W and Mo atoms into ZrO₂ and prepared WO_x/ZrO₂ and MoO_x/ZrO₂ catalysts. Moreover, we prepared highly catalytic SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ solid acid catalysts by wet impregnation. These catalysts were characterized via certain methods, such as X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), nitrogen adsorption–desorption measurement (BET), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and ammonia-temperature programmed desorption (NH₃-TPD), and the surface acid amount was obtained by combining pyridine adsorption and UV spectrometry. Furthermore, we investigated the dehydration of fructose into 5-HMF using these catalysts under microwave assistance in dimethyl sulfoxide (DMSO) in a very short reaction time.

2. Experimental

2.1. Materials

All chemicals used were of analytical grade: fructose (99%, Aladdin®), 5-HMF (98%, Tianjin Fuchen Chemical Reagents Factory), n-butanol (99%, Beijing Chemical Works), zirconium(IV) propoxide (70 wt%, solution in 1-propanol, Sigma-Aldrich), cyclohexane and pyridine (99.5%, Xilong Chemical Co., Ltd.), ammonium molybdate (99%) and ammonium tungstate (85% ∼ 90%, Sinopharm Chemical Reagent Co., Ltd), cetyltrimethylammonium bromide (CTAB) and DMSO (99%, Tianjin Guangfu Fine Chemicals Research Institute), and deionized water (H₂O, home-made).

2.2. Catalyst synthesis

The synthesis process for the zirconia support was as follows. First, 2.5 g of CTAB was dissolved in 20 mL n-butanol with stirring at 40 °C for 10 min to obtain a clear solution, and then 0.02 mol of the zirconium(IV) propoxide solution was added to this solution with stirring for about 10 min. In another container, 0.002 mol of the transition-metal source (W, Mo) was dissolved in 10 mL of water, and the formed solution was then added to the above-mentioned solution under vigorous magnetic stirring at room temperature. After stirring for 24 h, the precipitate was transferred to a Teflon lined autoclave, and crystallization was carried out at 80 °C for 24 h, then filtered, and then washed with bi-distilled water several times. The obtained hydroxide precipitate was dried at 100 °C for 5 h and finally calcined at 550 °C for 4 h, the heating rate was controlled at 2 °C min⁻¹.

As for SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂, these catalysts were prepared similarly to the above-mentioned method. The only difference was that the obtained hydroxide precipitate was dried at 100 °C for 5 h and then impregnated with 1 mol L⁻¹ H₂SO₄ for 30 min (sample/H₂SO₄ = 1 g/10 mL), then filtered, washed with hot distilled water at 80 °C several times, dried at 100 °C, and then calcined at 550 °C for 4 h.

2.3. Catalyst characterization

XRD analyses were conducted on a Rigaku D/MAX2500 instrument using Cu Kα radiation. The scattering angle, 2θ, was varied from 10° to 80°, with a step length of 0.02°. Fourier transform infrared (FT-IR) spectra were recorded on KBr pellets by a PerkinElmer (Spectrum Two) infrared spectrometer with a wavenumber from 4000 to 500 cm⁻¹. The morphologies of the catalytic materials were obtained by a transmission electron microscope (TEM: JEM-2000EX, JEOL). In situ high pressure Raman measurements were obtained using a spectrometer (focal length, 500 mm) combined with a liquid nitrogen-cooled CCD (Acton SP-2500 and PyLoN:100B, Princeton Instruments). A single-mode DPSS laser (power output, 50 mW) at 532 nm was used as the excitation light source. XPS analyses were conducted on a Theromo ESCALAB 250Xi instrument. The surface acid amounts of the samples were quantitatively measured via UV spectrometry with adsorbates of pyridine (Py) in cyclohexane solution on a Cary 5000 UV-vis-NIR. The detailed steps have been described previously by Siqian Zhang et al.²⁰ NH₃-TPD was performed on a Chembet Pulsar TPR/TPD. In a typical experiment for the TPD measurement, the sample was pretreated at 450 °C for 30 min in helium and then cooled to 100 °C. A flow of 10% NH₃/He was then passed over the pretreated materials for 60 min. Following this ammonia adsorption procedure, the reactor was purged with helium for 60 min to remove residual/physisorbed ammonia, and ammonia desorption was carried out at 700 °C with a heating rate of 10 °C min⁻¹. Nitrogen physisorption studies were performed using a Micromeritics ASAP 2010 model static volumetric adsorption instrument. The samples were dried in an oven at 80 °C overnight prior to degassing. Prior to the adsorption experiments, the catalysts were outgassed at 350 °C for 8 h. The pore size distribution was calculated using the Barrett–Joyner–Halenda (BJH) method.

2.4. Procedure for catalytic degradation of fructose

The catalytic reaction for fructose dehydration to HMF was performed in a CEM Discover SP microwave reactor. 5 mL of fructose solution and catalyst were charged in a microwave tube. The tube was placed in the microwave reactor. The microwave power was set to 100 W. The desired temperature and time were set. After the reaction, the tube was allowed to cool at room temperature. The conversion of fructose was analyzed with a LC98IIRI GPC with ZORBAX NH₂ column and a refractive index detector. The UV-visible spectrum of pure HMF solution has a distinct peak at 284 nm. The yield of HMF was determined by measuring the absorbance of the HMF product solution at 284 nm using the standard curve method. Repeated measurement of the same solution showed that the percentage of error associated with this measurement was ±0.3%.^21,22

3. Results and discussion

3.1. Characterization of the catalyst

3.1.1. XRD patterns and FTIR measurement. Fig. 1 shows the XRD patterns of different catalysts, the samples showed diffraction peaks for 2θ = 30.25°, 35.5°, 50.3° and 60°, it was estimated that zirconia essentially presented in the tetragonal form²⁴ and it had been generally accepted that the tetragonal phase was more important for the formation of acidic centers.²³ The bare nanosized ZrO₂ mainly existed in the monoclinic phase (m-ZrO₂), a stable phase under relatively high temperature; however, once WO_x was added, ZrO₂ dominantly crystallized in the tetragonal phase (t-ZrO₂), a metastable structure at low temperature.^24,25 In addition, there was no diffraction line for the crystalline WO_x phase in the XRD patterns, implying that WO_x in these samples exists in a highly dispersed state on the ZrO₂ surface and so do other samples elsewhere.¹⁶ The absorption spectra for UV-vis DRS are illustrated in the ESI Fig. 1,† and the band-gap energy (E_g) values of the samples are included in Table 1. The E_g of the synthesized samples were lower than ZrO₂, indicating that domain size became bigger and the transition-metal source existed in ZrO₂,^26,27 although the existence of the transition-metal could not be analyzed by XRD.

Fig. 1 XRD patterns of different samples.

Table 1 The physical properties of the samples

Catalyst	Surface area (m^2 g^−1)	Pore size (nm)	Pore volume (m^2 g^−1)	Qi (μmol g^−1)	Eg (eV)
ZrO2	48.19	14.08	0.17	29.60	5.10
WOx–ZrO2	139.70	7.59	0.22	131.42	3.75
MoOx–ZrO2	64.23	4.74	0.08	76.36	3.19
SO4^2−/WOx–ZrO2	159.55	8.18	0.32	380.34	3.58
SO4^2−/MoOx–ZrO2	82.75	5.35	0.11	256.71	2.91

---

The FT-IR spectra of the different catalysts are shown in Fig. 2. The band at 3438 cm⁻¹ was attributed to the O–H stretching vibrations of the physically absorbed water molecules, the bands at 1632 cm⁻¹ and 1388 cm⁻¹ were attributed to the HOH bending of water molecules associated with the sulfate group.²⁸ The band at 979 cm⁻¹ is characteristic of the stretching vibrations of the W=O bond.²⁹ The band at 762 cm⁻¹ was assigned to Mo=O stretching modes.³⁰ The sulfated samples exhibited strong bands at 1056, 1120, and 1247 cm⁻¹, which showed bidentate sulfate ions coordinated to the metal.³¹ These bands were absent in unsulfated samples. The TEM images of the samples, displayed in ESI Fig. 2,† revealed a globular structure. Average sized particles were obtained after calcination at 550 °C. The size of the particles in the MO_x–ZrO₂ and SO₄²⁻/MO_x–ZrO₂ (M=W, Mo) samples were smaller than ZrO₂.

Fig. 2 FT-IR spectra of different samples.

3.1.2. Raman spectra and XPS analysis. The Raman spectra of the different catalysts are shown in Fig. 3. All of them possessed the typical Raman signals of t-ZrO₂ at 641, 476, 312, and 274 cm⁻¹ and only ZrO₂ showed obvious bands of m-ZrO₂,³² which conformed to the ZrO₂ phase detected by XRD characterization. A big Raman band from trace amounts of crystalline WO₃ NPS at 833 cm⁻¹ was present in the spectra for the WO_x–ZrO₂ catalyst sample, which reflected the good dispersion of the tungsten oxide phase on the ZrO₂ support;³³ the Raman band from trace amounts of crystalline MoO₃ NPS at 812 cm⁻¹ was present in the spectra for the MoO_x–ZrO₂ catalyst sample,³⁴ which reflected the good dispersion of the molybdenum oxide phase on the ZrO₂ support. The band at 962 cm⁻¹ was attributed to the stretching vibration of M=O (M=W, Mo).^23,35 The Raman spectra of SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ weren’t measured, because they had a strong fluorescence effect.

Fig. 3 Raman spectra of different samples.

The XPS spectra of the different catalysts are shown in Fig. 4. The O1s, Mo3d, Zr3d, S2p, W4f and Zr4p peaks of the samples demonstrated binding energies of 530.4, 232.6, 181.9, 168.7, 36.1 and 30.9 eV, respectively.^36,37 This directly confirmed the presence of W or Mo in the materials.

Fig. 4 XPS of different samples.

3.1.3. TPD measurement and the BET surface area. The NH₃-TPD profiles of the different catalysts are depicted in Fig. 5 to evaluate the acidity of the catalysts. The NH₃-TPD profile of SO₄²⁻/WO_x–ZrO₂ has a wide region of stronger acid sites around 450 °C and weak acid sites by the peak around 200 °C. Although SO₄²⁻/MoO_x–ZrO₂ showed a peak distributed at approximately 500 °C, it was small. Nevertheless, other catalysts showed a broad peak for the maximum amount of acid sites around 200 °C. It indicated that the acid strength of SO₄²⁻/WO_x–ZrO₂ was much higher than the others. The amounts of acid sites calculated via UV spectrometry measurement are shown in Table 1, it is obvious that the sulfated samples exhibited a higher surface acid amount than unsulfated samples, corresponding to the NH₃-TPD results. The highest acid amount of SO₄²⁻/WO_x–ZrO₂ may be due to the enlargement of the surface area.^38–40 Generally, the specific surface areas of the sulfated samples were slightly larger than the unsulfated samples, this may because the interaction between SO₄²⁻ and the sample stabilized the sample against sintering producing a thermally stable ZrO₂ structure.³⁸ Nitrogen adsorption–desorption isotherms and the average pore size distribution curves of the samples are shown in ESI Fig. 3–7.† All the isotherms resemble Type IV isotherms based on the IUPAC classification with a large type H3 hysteresis loop, this indicates the presence of mesoporosity. The hysteresis loop was associated with the capillary condensation taking place in the mesopores signifying the preservation of the mesoporous structure even after crystallization at elevated temperature.⁴¹ The surface areas and pore size of the samples measured by the BET method are included in Table 1. The specific surface area increased in the following order: ZrO₂ < MoO_x/ZrO₂ < SO₄²⁻/MoO_x–ZrO₂ < WO_x/ZrO₂ < SO₄²⁻/WO_x–ZrO₂: this result proves that SO₄²⁻/WO_x–ZrO₂ may be an efficient catalyst, corresponding to the following reaction results.

Fig. 5 NH₃-TPD profiles of different samples.

All the findings presented above were consistent with the following general model (Scheme 2).^42–44 The SO₄²⁻ is an electron-withdrawing group, after it was introduced into material it could increase the formation of acidic centers.

Scheme 2 The balance of Brönsted and Lewis acids (M=W, Mo).

3.2. Catalytic activities of various catalysts for fructose conversion and 5-HMF yield

3.2.1. The catalytic performances of different catalysts. The prepared samples were tested for fructose dehydration to 5-HMF at 140 °C with 5 wt% fructose in DMSO for 5 min. The reactions using 10 wt% (catalyst amount with respect to fructose, g g⁻¹ × 100) of catalyst are shown in Table 2. It was found that all of the catalysts were active, and furthermore that the addition of the heteroatom and impregnation with H₂SO₄ influenced the catalytic activity. The yield increased in the following order: without catalyst < ZrO₂ < MoO_x/ZrO₂ < WO_x/ZrO₂ < SO₄²⁻/MoO_x–ZrO₂ < SO₄²⁻/WO_x–ZrO₂, corresponding to the specific surface area and surface acid amount. The activity of SO₄²⁻/ZrO₂ was similar compared to SO₄²⁻/MoO_x–ZrO₂ and SO₄²⁻/WO_x–ZrO₂, but the reaction time was 20 min. The addition of the heteroatom doping could shorten the reaction time, this may be because the heteroatom helped to stabilize the tetragonal phase of zirconia in the catalyst and inhibited the transformation from the metastable tetragonal phase to the monoclinic phase at a high calcination temperature,^24,25 and the tetragonal phase was more important for the formation of acidic centers. The SO₄²⁻/WO_x–ZrO₂ catalyst was found to be the most active catalyst, this was attributed to the stable tetragonal phases and the amount of acidic sites in this catalyst. In addition, the SO₄²⁻/MoO_x–ZrO₂ even had a better catalytic effect, although the catalytic effect of SO₄²⁻/MoO_x–ZrO₂ was poorer than SO₄²⁻/WO_x–ZrO₂. Therefore, in this work we mainly studied the dehydration of fructose into 5-HMF using SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ under microwave-assistance in dimethyl sulfoxide.

Table 2 Influence of the species of catalysts on the dehydration of fructose

Catalyst	Reaction solvent	Reaction time/min	Reaction temperature/°C	Fructose conversion/%	5-HMF yield/%	5-HMF selectivity/%	Ref.
None	DMSO	5	140	43.93	15.86	36.10	This work
ZrO2	DMSO	5	140	51.13	31.02	60.67	This work
WOx–ZrO2	DMSO	5	140	67.66	53.03	78.38	This work
MoOx–ZrO2	DMSO	5	140	61.74	47.26	76.55	This work
SO4^2−/WOx–ZrO2	DMSO	5	140	87.52	70.88	80.99	This work
SO4^2−/MoOx–ZrO2	DMSO	5	140	86.05	67.11	77.99	This work
SO4^2−/ZrO2	DMSO	20	140	93.60	72.80	77.78	28
TiO2	DMSO	5	140	—	54.10	—	21
AlCl3	DMSO	5	140	—	71.30	—	19
Ion-exchange resin	Acetone–water	10	150	91.70	70.30	76.67	10

---

3.2.2. Effect of fructose concentration on the dehydration of fructose. Fig. 6 shows the effect of the fructose concentration on the dehydration of fructose. There was a negligible effect on the dehydration of fructose when the fructose concentration was below 5 wt%. The rate of fructose conversion had little effect when the fructose concentration was above 5 wt%, but the HMF yield and the HMF selectivity decreased gradually with increasing initial fructose concentration. This was probably because self-polymerization of HMF or cross-polymerization between fructose and HMF would occur easily at a higher initial concentration of fructose,¹⁰ giving rise to the formation of brown-black soluble polymers and insoluble humins.¹⁹ In addition, it was obvious that the catalytic effect of SO₄²⁻/MoO_x–ZrO₂ was poorer than SO₄²⁻/WO_x–ZrO₂, but the variation of the trend of the fructose conversion (the HMF yield and the HMF selectivity) was similar. So, in conclusion, we took 5 wt% as the initial fructose concentration in this work.

Fig. 6 Influence of fructose concentration on the dehydration of fructose. (Reaction conditions: reaction temperature = 140 °C; catalyst amount = 10 wt%; reaction time = 5 min.)

3.2.3. Influence of the catalyst amount on the dehydration of fructose. Fig. 7 shows the effect of the catalyst amount with respect to fructose conversion, 5-HMF yield and 5-HMF selectivity. The amounts of catalyst used were 5 wt%, 10 wt%, 20 wt% and 30 wt% (catalyst amount with respect to fructose, g g⁻¹ × 100) respectively. In the absence of catalyst, only a 15.86% yield of 5-HMF was attained at 140 °C for a reaction time of 5 min in DMSO, nevertheless the 5-HMF yield reached up to 61.06% (SO₄²⁻/WO_x–ZrO₂) and 54.21% (SO₄²⁻/MoO_x–ZrO₂) when 5 wt% catalyst was added. This firmly indicated that SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ showed excellent catalytic performance for the dehydration of fructose to 5-HMF. When the catalyst dosage increased from 5 wt% to 10 wt%, the 5-HMF yield still increased at 140 °C for a reaction time of 5 min because the catalytic sites increased with the increasing amount of catalyst, boosting the dehydration reaction. However, when we kept increasing the amount of catalyst, there was a smoothing in fructose conversion and 5-HMF yield, which was probably because the side reaction of the dehydration of fructose was also accelerated with the over-use of the catalyst. Given the above discussion, it was evident that 10 wt% of catalyst provided the best result for the model reaction.

Fig. 7 Influence of the catalyst amount on the dehydration of fructose. (Reaction conditions: fructose concentration = 5 wt%; reaction time = 5 min; reaction temperature = 140 °C.)

3.2.4. Influence of different reaction times and temperatures on the dehydration of fructose and kinetics analysis. Fig. 8 shows the influence of different reaction times and temperatures on fructose conversion. It was obvious that the reaction temperature and time had a significant effect on the fructose conversion. While the reaction temperature increased, the fructose conversion increased. As the reaction proceeded, the fructose conversion also increased: when prolonging the reaction time to 5 min, fructose conversion was almost constant at 160 °C. When the temperature was 140 °C, fructose conversion was almost constant for 7 min, showing that an increased reaction temperature could shorten the reaction time. Many works have reported a reaction order of one for the dehydration of fructose to 5-HMF. We performed a kinetics analysis of the dehydration of fructose catalyzed by SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ in DMSO. Values of ln(1 − x) were plotted against reaction time (t) to obtain first-order kinetics constants. With these constants, an Arrhenius plot was generated, as shown in Fig. 8. This model has been employed in most kinetics studies and was found to give reasonable levels of agreement with experimental data.^45–49 The activation energies and pre-exponential factors for acid-catalyzed fructose dehydration to 5-HMF obtained by different authors under different conditions are summarized in Table 3. The obtained activation energies and pre-exponential factors from this work were similar, perhaps because we used the same solvents and heating methods,⁴⁷ on the other hand, considering the experimental error/standard deviation, we could also know that the SO₄²⁻/MoO_x–ZrO₂ and SO₄²⁻/WO_x–ZrO₂ exhibited similar catalytic activities. Compared to the previous values reported in the literature, this showed that the present process was more efficient than previous works as evident from the shorter reaction times.

Fig. 8 Influence of reaction time and temperature on fructose conversion and Arrhenius linearity fitting curve. (a) SO₄²⁻/WO_x–ZrO₂, and (b) SO₄²⁻/MoO_x–ZrO₂. (Reaction conditions: catalyst amount = 10 wt%; fructose concentration = 5 wt%.)

Table 3 Comparison of activation energies and pre-exponential factors for acid-catalyzed fructose dehydration to 5-HMF

Ea (kJ mol^−1)	A (min^−1)	Reaction solvent	Catalyst	Ref.
68.31	1.45 × 10^8	DMSO	SO4^2−/WOx–ZrO2	This work
69.31	1.80 × 10^8	DMSO	SO4^2−/MoOx–ZrO2	This work
99.20	2.10 × 10^8	DMSO	[BMIM]OH	45
96.00	1.79 × 10^8	Subcritical water	ZnSO4	46
60.40	4.80 × 10^6	70 : 30 (w/w) acetone/DMSO	Ion-exchange resin	47
103.40	8.70 × 10^11	70 : 30 (w/w) acetone/water	Ion-exchange resin	10
80.00	3.10 × 10^8	Supercritical methanol	H2SO4	48
160.60	—	Aqueous solution	HCl	49

---

Fig. 9 shows the influence of different reaction times and temperatures on the 5-HMF yield and 5-HMF selectivity. The 5-HMF yield and selectivity increased with reaction temperature, and the temperature played a positive role in the dehydration of fructose. When the temperature increased up to 160 °C, the yield of 5-HMF was slightly decreased, but the selectivity of 5-HMF significantly declined. When the catalyst was SO₄²⁻/WO_x–ZrO_2, the fructose conversion was 94.97% with 82.55% 5-HMF selectivity in a 1 min reaction time at 160 °C. In addition, with the extending of the reaction time, the 5-HMF yield and selectivity were improved consistently. When the reaction time was extended up to 7 min and longer at 150 °C, the 5-HMF yield decreased. But when the temperature was above 160 °C, the 5-HMF selectivity decreased consistently. The variation of the trends of the HMF yield and selectivity for SO₄²⁻/MoO_x–ZrO₂ were similar to those of SO₄²⁻/WO_x–ZrO₂, they had a fructose conversion of 95.33% with 81.64% 5-HMF selectivity in a 3 min reaction time at 160 °C. In conclusion, it was generally noted that higher temperatures (>140 °C) and longer reaction times (>5 min) resulted in a loss of the yield as expected on account of by-product formation and/or decomposition of HMF.

Fig. 9 Influence of reaction time and temperature on 5-HMF yield and 5-HMF selectivity. (a) SO₄²⁻/WO_x–ZrO₂, and (b) SO₄²⁻/MoO_x–ZrO₂. (Reaction conditions: catalyst amount = 10 wt%; fructose concentration = 5 wt%.)

4. Conclusions

In this study, ZrO₂, WO_x/ZrO₂, MoO_x/ZrO₂, SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ solid catalysts were prepared and characterized by XRD, UV-DRS, FTIR, XPS, TEM, TPD, BET, and Raman and the surface acid amount was obtained by combining pyridine adsorption and UV spectrometry. These catalysts were employed in the dehydration of fructose to 5-HMF under microwave-assistance in DMSO, and the SO₄²⁻/WO_x–ZrO₂ catalyst was found be the most active catalyst, a fructose conversion of 95.80% with a 83.90% 5-HMF yield was obtained at 150 °C for a relatively short reaction time of 5 min. The catalyst amount and fructose concentration were further optimized. The value of the activation energy was comparable with previous values reported in the literature, and this implied that SO₄²⁻/WO_x–ZrO₂ and SO₄²⁻/MoO_x–ZrO₂ were efficient for the dehydration of fructose to 5-HMF under microwave assistance in DMSO.

References

1. J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539–554.
2. R. J. van Putten, J. C. van der Waal, E. de Jong, C. B. Rasrendra, H. J. Heeres and J. G. de Vries, Chem. Rev., 2013, 113, 1499–1597.
3. K. W. Omari, J. E. Besaw and F. M. Kerton, Green Chem., 2012, 14, 1480–1487.
4. S. Siankevich, G. Savoglidis, Z. F. Fei, G. Laurenczy, D. T. L. Alexander, N. Yan and P. J. Dyson, J. Catal., 2014, 315, 67–74.
5. Y. Román-Leshkov, J. N. Chheda and J. A. Dumesic, Science, 2006, 312, 1933–1937.
6. T. F. Wang, M. W. Nolte and B. H. Shanks, Green Chem., 2014, 16, 548–572.
7. Y. Takeuchi, F. M. Jin, K. Tohji and H. J. Enomoto, J. Mater. Chem., 2008, 43, 2472–2475.
8. Y. H. Li, X. Y. Lu, L. Yuan and X. Liu, Biomass Bioenergy, 2009, 33, 1182–1187.
9. Y. Li, H. Liu, C. H. Song, X. M. Gu, H. Li, W. Zhu, S. Yin and C. Han, Bioresour. Technol., 2003, 133, 247–253.
10. X. H. Qi, M. Watanabe, T. M. Aida and R. L. S. Smith Jr, Green Chem., 2008, 10, 799–805.
11. G. R. Akien, L. Qi and I. T. Horváth, Chem. Commun., 2012, 48, 5850–5852.
12. H. Ma, F. Wang, Y. H. Yu and X. H. Li, Ind. Eng. Chem. Res., 2015, 54, 2657–2666.
13. Y. Qu, Q. Wei, H. Li, P. Oleskowicz-Popiel, C. Huang and J. Xu, Bioresour. Technol., 2014, 162, 358–364.
14. X. H. Qi, M. Watanabe, T. M. Aida and R. L. S. Smith Jr, Green Chem., 2009, 11, 1327–1331.
15. A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–2502.
16. R. Kourieh, S. Bennici, M. Marzo, A. Gervasini and A. Auroux, Catal. Commun., 2012, 19, 119–126.
17. S. P. Simeonov and C. A. M. Afonso, J. Chem. Educ., 2013, 90, 1373–1375.
18. A. H. Jadhav, H. Kim and I. T. Hwang, Bioresour. Technol., 2013, 132, 342–350.
19. S. De, S. Dutta and B. Saha, Green Chem., 2011, 13, 2859–2868.
20. S. Q. Zhang, Z. B. Zhao and Y. H. Ao, Appl. Catal., A, 2015, 496, 32–39.
21. S. Dutta, S. De, A. K. Patra, M. Sasidharan, A. Bhaumik and B. Saha, Appl. Catal., A, 2011, 409–410, 133–139.
22. C. Z. Li, Z. K. Zhao, A. Q. Wang, M. Y. Zheng and T. Zhang, Carbohydr. Res., 2010, 345, 1846–1850.
23. W. D. Sun, Z. B. Zhao, C. Guo, X. K. Ye and Y. Wu, Ind. Eng. Chem. Res., 2000, 39, 3717–3725.
24. E. I. Ross-Medgaarden, W. V. Knowles, T. Kim, M. S. Wong, W. Zhou, C. J. Kiely and I. E. Wachs, J. Catal., 2008, 256, 108–125.
25. L. D. Huy, P. Laffez, P. Daniel, A. Jouanneaux, N. T. Khoi and D. Siméone, Mater. Sci. Eng., B, 2003, 104, 163–168.
26. G. R. Rao and H. R. Sahu, Proc. Indiana Acad. Sci., 2001, 113, 651–658.
27. E. Iglesia, D. G. Barton, S. L. Soled, S. Miseo, J. E. Baumgartner, W. E. Gates, G. A. Fuentes and G. D. Meitzner, Stud. Surf. Sci. Catal., 1996, 101, 533–542.
28. X. H. Qi, M. Watanabe, T. M. Aida and R. L. S. Smith Jr, Catal. Commun., 2009, 10, 1771–1775.
29. G. N. Kystova, Y. A. Chesalov, L. M. Plyasova, I. Y. Molina and A. I. Nizovskii, Vib. Spectrosc., 2011, 55, 235–240.
30. S. B. Rathod, M. K. Lande, B. R. Arbad and A. B. Gambhire, Arabian J. Chem., 2014, 7, 253–260.
31. J. R. Satam and R. V. Jayaram, Catal. Commun., 2008, 9, 1033–1039.
32. E. I. Ross-Medgaarden, W. V. Knowles, T. Kim, M. S. Wong, W. Zhou, C. J. Kiely and I. E. Wachs, J. Catal., 2008, 256, 108–125.
33. E. I. Ross-Medgaarden and I. E. Wachs, J. Phys. Chem. C, 2007, 111, 15089–15099.
34. A. Sarkar, S. Pramanik, A. Achariya and P. Pramanik, Microporous Mesoporous Mater., 2008, 115, 426–431.
35. M. Dieterle, G. Weinberg and G. Mestl, Phys. Chem. Chem. Phys., 2002, 4, 812–821.
36. N. Vaidyanathan, M. Houalla and D. H. Hercules, Langmuir, 1999, 15, 8336.
37. A. M. Azad, S. G. Mhaisalkar, L. D. Birkefeld, S. A. Akbar and K. S. Goto, J. Electrochem. Soc., 1992, 139, 2913–2920.
38. F. Yang, Y. Li, Q. Zhang, X. F. Sun, H. X. Fan, N. Xu and G. Li, Carbohydr. Polym., 2015, 131, 9–14.
39. E. Kılıç and S. Yılmaz, Ind. Eng. Chem. Res., 2015, 54, 5220–5225.
40. H. P. Yan, Y. Yang, D. M. Tong, X. Xiang and C. W. Hu, Catal. Commun., 2009, 10, 1558–1563.
41. S. J. Gregg and K. S. W. Sing, Adsorption, Surface Area and Porosity, 1982.
42. T. Jin, T. Yamaguchi and K. Tananbe, J. Phys. Chem., 1986, 90, 4794–4796.
43. E. Iglesia, S. L. Soled and M. Kramer, J. Catal., 1993, 144, 238–253.
44. B. Tyagi, M. K. Mishra and R. V. Jasra, Catal. Commun., 2006, 7, 52–57.
45. Y. S. Qu, Y. L. Song, C. P. Huang, J. Zhang and B. H. Chen, Ind. Eng. Chem. Res., 2012, 51, 13008–13013.
46. M. Bicker, J. Hirth and H. Vogel, Green Chem., 2003, 5, 280–284.
47. X. H. Qi, M. Watanabe, T. M. Aida and R. L. S. Smith Jr, Ind. Eng. Chem. Res., 2008, 47, 9234–9239.
48. M. Bicker, D. Kaiser, L. Ott and H. Vogel, J. Supercrit. Fluids, 2005, 36, 118–126.
49. F. S. Asghari and H. Yoshida, Ind. Eng. Chem. Res., 2007, 46, 7703–7710.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra22979k

---