Mesoporous niobium phosphate: an excellent solid acid for the dehydration of fructose to 5-hydroxymethylfurfural in water

Abstract

By using cetyltrimethylammonium bromide (CTAB) as the template, a series of mesoporous niobium phosphates were synthesized at different pH values in an aqueous solution. Techniques such as small-angle X-ray diffraction, transmission electron microscopy (TEM) and N₂ sorption technique were employed to characterize the mesoporous structures of thus-synthesized materials, EDAX to detect the composition, FTIR and solid state ³¹P MAS NMR to investigate the framework information, while their acidic properties were analyzed using NH₃-TPD and pyridine-FTIR. Samples prepared at neutral to acidic conditions exhibited high surface area (213–297 m² g⁻¹), narrow pore size distribution (3–4 nm) and a great number of strong Lewis and Brönsted acid sites. These materials exhibited excellent activity in the dehydration of fructose to 5-hydroxymethylfurfural (HMF) in water. The maximum HMF yield reached 45% under 130 °C with a reaction time of 0.5 h and the yield slightly decreased to 32% after five cycles and the five-cycled catalyst can be almost regenerated by calcination at 500 °C with the yield of 40%. The excellent catalytic activity obtained in the aqueous phase can be attributed to its high acid site density and the tolerance to water.

---

Introduction

Niobium compounds and related materials have been considered with great interest in recent years in the field of catalysis, especially treated as solid acid.^1,2 Several important heterogeneous reactions involving water as a reactant, product, or solvent, such as hydration, hydrolysis, or esterification, could be readily catalyzed by niobium-based materials. In fact, only very few solid acids can offer required activity, stability, and most importantly water tolerance when catalyzing these reactions.

Niobium phosphate is an example of solid acid which was considered to be a very important water tolerant catalyst despite the fact that not much is known concerning its surface acidity in protic media.^3–5 Niobium phosphate exhibited unique acidic properties, such as Lewis–Brönsted acids, tunable acid site density and high thermal stability. In order to achieve high performance in catalysis, porous niobium phosphate materials have been synthesized by several groups with high surface area and uniform pore size distribution. For example, Fujiwara et al.^6,7 have synthesized porous niobium phosphate by using hexadecyl amine (HDA) or hexylamine as the template, but the pores were in the micropore region, which restricted its further application for bulky molecules. Sarkar and Pramanik⁸ reported the synthesis of mesoporous niobium oxophosphate by using a long chain alkyl surfactant, tetradecyl trimethylammonium bromide, as the structure-directing agent and large pore size was obtained (3.35 nm). These reports mainly focus on the synthesis of novel structured materials without exploration of resultant materials as a solid acid. So efforts are being made here to develop high surface area mesoporous niobium phosphate and investigate its acidic properties as well as its application in catalysis.

On the other hand, dehydration reaction is the most important catalytic application of niobium-containing solid acid.⁹ Many efforts have been made to produce 5-hydroxymethylfurfural (HMF) from the dehydration of fructose. In biomass conversion processes HMF has been regarded as a versatile and key platform molecule which can be obtained from biomass-based carbohydrates and further converted into high quality chemicals and fuels. Thus various heterogeneous catalysts have been explored,^10–16 but most reactions were carried out in the organic phase¹¹ or ionic liquid¹² up to now. The disadvantages such as environmental pollution risk, huge energy input and high cost owing to the necessary separation process, expensive solvents, and material corrosion strongly hamper the industrialization. In contrast, water is a valuable solvent with many advantages: it is natural, and hence readily and widely available, its disposal is regarded as benign with little effect on the environment. Therefore, some researchers have focused on the partial substitution of organic solvent with water.^17,18 But there was no report on the dehydration of fructose to HMF by using pure water as the solvent, because in water, or very highly polar medium, very few solid acids can maintain their desirable activity and stability without the deactivation of their acid sites.¹⁹ This is because solid acids generally deactivate because of their violent reaction with anything that has an electron pair, in consequence, it will absorb and even react vigorously with water.

Herein, we developed a simple sol–gel hydrothermal approach to synthesize mesoporous niobium phosphate, in which the influence of pH towards the structure and acidic property was investigated. Thus-prepared mesoporous niobium phosphate possesses excellent catalytic activity for the dehydration of fructose to HMF using pure water as the solvent. This can be ascribed to its unique stability and water-tolerant acidic properties. This also provides an alternative approach to obtain HMF in high yield in an aqueous system.

Experimental section

Chemicals

All chemicals were purchased from Sinopharm Chemical Reagent Co. Ltd, except Nb tartrate, which was prepared in our laboratory according to the literature.⁸ All bought chemicals were of analytical grade and used as received without further purification.

Synthesis of mesoporous niobium phosphate (NbPO)

Mesoporous niobium phosphate material synthesized at different pH was denoted as NbPO-pHx. In a typical synthesis, 1.32 g (0.01 mol) of diammonium hydrogen phosphate was dissolved in 20 ml water and then the pH was adjusted to 2 using phosphoric acid. With vigorous stirring, 20 ml of 0.5 M niobium tartrate (pH = 2) was added to the above solution. Then the mixed solution was dropped into the aqueous solution of cetyltrimethyl ammonium bromide (CTAB) which was previously prepared by dissolving 1.0 g of CTAB in 13 ml of distilled water. The pH value of the final solution was about 2. Afterwards, this mixture was stirred for additional 60 min at 35 °C, and then the transparent solution was aged in a Teflon-lined autoclave for 24 h at 160 °C. After cooling down, the solid was filtered, washed with distilled water and then dried at 50 °C. Finally, the NbPO-pH2 sample was obtained by calcination at 500 °C for 5 h in air to remove organic species. Similarly, another two NbPO samples were also synthesized according to the above-mentioned procedure, but with different initial pH. For the NbPO-pH7 sample, diammonium hydrogen phosphate and niobium tartrate were just mixed together. For the NbPO-pH10 sample, firstly diammonium hydrogen phosphate and niobium tartrate were mixed together, then pH adjusted to 10 by using aqueous ammonia. Except for that, the other steps were identical to those used for sample NbPO-pH2.

Catalytic activity measurement

The mixture of fructose (0.8 g), catalyst (0.8 g) and water (10 ml) was loaded into a Teflon-lined stainless steel autoclave with a thermocouple probed detector and placed into an oil-bath with magnetic stirring. N₂ gas was used to purge air outside the reactor and kept at a pressure of about 0.8 MPa to prevent boiling. Zero time was taken to be when the temperature reached the given temperature. The reactor was raised to a certain temperature and held at this temperature for a given period of time. After reaction, the mixture was centrifuged and the supernatant liquid was analyzed by an Agilent 1200 Series HPLC based on the external standard method. After each reaction, the catalyst was collected and dried at 50 °C for 24 hours to remove the volatile impurity. Then a new reaction was started as a new cycle with the used catalyst.

Characterization of mesoporous niobium phosphate and products analysis

Powder XRD patterns were recorded on a Bruker diffractometer (D8 Focus) by using Cu Kαα (λ = 0.15406 nm) radiation. Nitrogen adsorption–desorption isotherms were measured at −96 °C on a NOVA 4200e analyzer (Quantachrome Co. Ltd). Before the measurements, all samples were outgassed at 180 °C for 12 h under vacuum to remove moisture and volatile impurities. The BET method was used to calculate the specific surface area. Pore size distribution curves were derived from the desorption branches of the isotherms and calculated by the Barrett–Joyner–Halanda (BJH) method. The total pore volume (V_t) was estimated at a relative pressure of 0.975. Transmission electron microscopy (TEM) images were recorded on a FEI Tecnai F20 s-TWIN instrument, and the electron beam accelerating voltage was 200 kV. The Fourier transform infrared (FTIR) absorption spectra were recorded on a Nicolet NEXUS 670 FT-IR spectrometer, with 32 scans at an effective resolution of 4 cm⁻¹.

Infrared (IR) spectra of pyridine adsorption were also recorded on the same instrument. The samples were pressed into self-supporting disks (13 mm diameter, 50 mg) and placed in an IR cell attached to a closed glass-circulation system. The disk was dehydrated by heating at 400 °C for 1 h under vacuum in order to remove physisorbed water. After the cell was cooled to room temperature, the IR spectrum was recorded as the background. Pyridine vapor was then introduced into the cell at room-temperature until equilibrium was reached, and then a second spectrum was recorded. Subsequent evacuations were performed at 100, 200, 300 and 400 °C, for 10 min followed by spectral acquisitions. The spectra presented were obtained by subtracting the spectra recorded before and after pyridine adsorption.

Ammonia temperature-programmed-desorption (NH₃-TPD) was carried out in a PX200 apparatus (Tianjin Golden Eagle Technology Limited Corporation) with a thermal conductivity detector (TCD). The catalyst (100 mg) was charged into the quartz reactor, and the temperature was increased from room temperature to 500 °C at a rate of 10 °C min⁻¹ under a flow of N₂ (40 ml min⁻¹), and then the temperature was decreased to 90 °C. Finally, NH₃ was pulsed into the reactor at 90 °C under a flow of N₂ (40 ml min⁻¹). When the baseline was stable, the temperature was increased from 90 °C to 800 °C at a rate of 10 °C min⁻¹.

The solid state ³¹P MAS NMR experiments were preformed on a Bruker AVANCE AV 400 spectrometer at a Larmor frequency of 161.938 MHz using a single-pulse sequence with high-power proton decoupling. A Bruker MAS probehead was used with a 4 mm zirconia rotor and chemical shifts were referenced to 85% H₃PO₄ at 0 ppm.

The analysis of the reaction products was carried out by means of a HPLC apparatus (Agilent 1200 Series) equipped with an XDB-C18 column (Eclipse USA). An auto-sampler (Agilent G1329A) was used to enhance the reproducibility. HMF was analyzed with an ultraviolet detector (Agilent G1314B) and other products were analyzed with a refractive index detector (Agilent G1362A).

Results and discussion

Chemical composition

Firstly, EDAX analysis was carried out to detect the composition of the final calcined products and the results are summarized in Table 1. It was found that the molar ratios of Nb, P, O in the final products deviated from the initial value (P/Nb = 1 in the initial solution) and were different from one another due to the different synthetic pH value. The molar ratio of Nb/P was decreased with the decrease of pH in the precursor. This can significantly influence the acidic properties as discussed later.

Table 1 Chemical composition of mesoporous niobium phosphate synthesized at different pH

Sample	Nb (at%)	P (at%)	O (at%)	Total (at%)	nNb/nP
NbPO-pH2	21.89	21.31	56.80	100	1.02/1
NbPO-pH7	23.19	17.75	59.05	100	1.31/1
NbPO-pH10	27.93	14.23	57.83	100	1.96/1

---

Characterizations of the mesostructure

The small-angle XRD patterns of the as-synthesized mesoporous niobium phosphate materials are shown in Fig. 1. All samples except NbPO-pH10 show a single broad peak at 2θ of 2.3° with no distinctive higher order peaks in their small-angle XRD patterns, indicating the disordered, wormhole-like structure. The wide-angle XRD patterns of calcined samples (Fig. 1, inset) only display two broad peaks in the 2θ range of 15–40° and 40–60°, indicating the amorphous nature of the pore walls. No crystallized NbOPO₄ and/or Nb₂O₅ phase were detected in these samples.

Fig. 1 Small-angle (before calcination) and wide-angle (inset) XRD patterns of mesoporous niobium phosphate (NbPO) samples.

Nitrogen adsorption–desorption isotherms and the pore size distributions of these materials are shown in Fig. 2. All these samples, except NbPO-pH10, are of type IV isotherms, with hysteresis loops which are intermediated between typical H1 and H2-type in the P/P_o range from 0.4 to 0.8, characteristic of mesoporous materials according to IUPAC classification. This hysteresis suggested the presence of large uniform wormlike mesopores.²⁰ The Brunauer–Emmett–Teller (BET) surface area, average pore diameter, and pore volume of these samples are summarized in Table 2. It can be seen that increasing the pH of the pre-mixture would reduce both the surface area and pore volume. The respective BET surface area and pore volume of sample NbPO-pH2 were 290.1 m² g⁻¹ and 0. 29 ml g⁻¹; of the sample NbPO-pH7 were 213.7 m² g⁻¹ and 0.27 ml g⁻¹ and were only 34.6 m² g⁻¹ and 0.06 ml g⁻¹ for the sample NbPO-pH10. Especially, sample NbPO-pH10 was almost microporous. This may be due to the fast precipitation of niobium phosphate at higher pH not allowing the interaction with surfactant and then to assemble together.

Fig. 2 N₂ adsorption–desorption isotherms of samples synthesized with different pH of precursor.

Table 2 Summary of physicochemical properties of mesoporous niobium phosphate

Sample	Surface area/m^2 g^−1	Pore size/nm	Pore volume/cm^3 g^−1
NbPO-pH2	290.1	3.8	0.291
NbPO-pH7	213.7	3.5	0.268
NbPO-pH10	34.6	35	0.057

---

Particle morphology and pore dimensions

The representative TEM images of mesoporous niobium phosphates are shown in Fig. 3. It can be seen that the pores are distributed randomly, having a disordered mesostructure at pH 2 and 7. While for sample NbPO-pH 10, the worm-like pores existed, but the surface area was extremely low, which may be due to closed pores instead of open pores in NbPO-pH2 and NbPO-pH7.

Fig. 3 TEM image of mesoporous niobium phosphate.

Framework vibrations

The FTIR spectra of mesoporous NbPO samples are shown in Fig. 4, as a comparison, the spectrum of commercial Nb₂O₅ is also presented. It can be found that all the three NbPO samples had vibration at 1025 cm⁻¹, but which was absent in Nb₂O₅, suggesting the formation of Nb–O–P networks in mesoporous NbPO samples. In the spectrum of Nb₂O₅, a broad absorption centered at 3420 cm⁻¹(νOH), a weak band at 1620 cm⁻¹(νOH) due to the absorption of water and the strong band at 630 cm⁻¹ due to the Nb–O stretching vibration mode are observed. In contrast, the FTIR spectra of the NbPO samples show a strong band due to the asymmetric vibration stretching mode of the phosphate ion with a maximum at 1025 cm⁻¹. An additional strong band is found at 630 cm⁻¹, which is assigned to Nb–O stretching modes, agreeable to that of Nb₂O₅. The band at 3420 cm⁻¹ and 1620 cm⁻¹ can be associated to the adsorbed molecular water. Furthermore, the ratio of peak area located at 1025 cm⁻¹ as Nb–O–P to the peak area of Nb–O at 630 cm⁻¹ was increased from NbPO-pH10 to NbPO-pH2. The phenomenon is associated with P contents in the mesoporous niobium phosphate samples detected in EDAX analysis. It can be seen that the peak intensity of Nb–O–P was increased with the P contents increased. The same results can be obtained from the peak intensity of Nb–O which was corresponding to the Nb amounts in the samples.^3,21,22

Fig. 4 FTIR patterns of Nb₂O₅ (a), mesoporous niobium phosphate (NbPO) sample at pH 2 (b), pH 7 (c) and pH 10 (d).

NH₃-TPD

The acidic properties of calcined NbPO samples were investigated by temperature-programmed desorption of ammonia and the results are shown in Fig. 5. Both NbPO-pH2 and NbPO-pH7 samples show two broad peaks at 300 and 600 °C, due to the presence of weak, medium and strong acid sites. Their strength and amount are comparable to zeolite β with Si–Al = 50. However, sample NbPO-pH10 just has a broad peak at a range of 100–400 °C due to the weak and medium acidic sites.^23,24 Sample NbPO-pH2 shows the highest intensity within a desorption temperature range of 200–400 °C, indicating the highest number of medium acid sites than the other two types of acid. In addition, it is also clear from Fig. 5 that the medium strength acid sites are much higher for NbPO-pH2 than those of NbPO-pH7, which is mainly due to the presence of high phosphate content within the mesoporous framework. High phosphate content could be attributed to the geminal P(OH)₂ sites and surface P-OH sites.²⁵ Furthermore, both samples of NbPO-pH2 and NbPO-pH7 show the presence of strong acid sites, as their respective desorption temperatures range from 400 to 700 °C. Obviously, the total acidic amount for NbPO samples increased when the solution in the synthesis was adjusted from basic (pH 10) to acidic medium (pH 2). This can be attributed to the high content of P, high surface area and stable structure synthesized at low pH medium.

Fig. 5 NH₃-TPD profiles of (a) NbPO-pH2, (b) NbPO-pH7 and (c) NbPO-pH10 and (d) Zeolite β with Si–Al = 50.

The pyridine-FTIR

To investigate the different types of acidic sites, the Py-FTIR spectra were recorded. Fig. 6 shows the IR spectra of adsorbed pyridine over NbPO-pH2 at different temperatures. The sharp bands at 1540 cm⁻¹ are the characteristic bands of Brönsted acid, which could be due to the presence of Nb-OH and P-OH groups on the surface.⁴ The band at 1490 cm⁻¹ could be attributed to the adsorption of pyridine in Brönsted and Lewis sites at the same time. The band at 1450 cm⁻¹, corresponding to the adsorption of pyridine at the Lewis acid sites (PyL), is also clearly observed. This Lewis acidic site could be due to the presence of a reasonable amount of octahedra NbO₆ and tetrahedra NbO₄ in the mesoporous niobium phosphate frameworks.⁵ In order to evaluate the strengths of the acid sites (Lewis and Brönsted acid sites), the spectrum was collected after evacuation at different temperatures. The weak sites are defined as the ones from which pyridine is removed by evacuation at 200 °C; the medium strength corresponds to evacuation between 200 and 400 °C and the strong sites remain adsorbing pyridine after evacuation at 400 °C. From Fig. 6, we can observe that NbPO-pH2 have much more medium acidic sites (223.9 μmol g⁻¹) rather than strong acidic sites (134.4 μmol g⁻¹), which is coincident to previous NH₃-TPD data. Quantification of medium and strong acid sites was obtained from the spectra of adsorbed pyridine using the expressions in the literature.⁴

Fig. 6 Pyridine-FTIR spectra of NbPO-pH2 obtained after evacuation at different temperatures. (a) 100 °C; (b) 200 °C; (c) 300 °C and (d) 400 °C.

Solid state ³¹P MAS NMR

Solid state ³¹P MAS NMR spectra for the three samples are depicted in Fig. 7. For sample NbPO-pH2 and NbPO-pH7, a broad signal centered at −9.6 ppm can be clearly seen. In contrast, for sample NbPO-pH10, the signal centered at −16.2 ppm was obtained, with the similar signal width. The isotropic ³¹P chemical shifts of (H₂PO₄)⁻, (HPO₄)²⁻ and (PO₄)³⁻ groups appear at around −10, −20 and −30 ppm, respectively.^26,27 Therefore, the chemical shifts at −9.6 ppm can be assigned to (H₂PO₄)⁻; in contrast, NbPO-pH10 with chemical shifts at −16.2 ppm corresponded to (HPO₄)²⁻ mostly. The existence of surface (H₂PO₄)⁻ groups in NbPO-pH2 and NbPO-pH7 samples confirms the origin of strong acidic sites and in accordance with NH₃-TPD results. The Brönsted acidic sites were possibly derived from OH groups on the surface termination of the phosphorus.

Fig. 7 ³¹P MAS NMR spectra of mesoporous niobium phosphates (a)NbPO-pH2, (b)NbPO-pH7 and (c)NbPO-pH10.

Catalytic study

It is well known that the catalytic performance of catalysts depends on their properties, while these properties are in turn determined by the thermodynamic and kinetics of the synthesis. In order to know more about the relationship between the properties and catalytic performance, these mesoporous niobium phosphates were investigated in the dehydration of fructose to HMF in the aqueous phase.

Table 3 summarises the results of the catalytic activities observed in the dehydration of fructose at 130 °C for 30 min. The yield of HMF reached 45% in the NbPO-pH2 catalyst. The conversion of fructose is not the highest in the NbPO-pH2 catalyst, but the selectivity to HMF was much higher than another two samples, which may be associated with their acidic properties. Sample NbPO-pH2 has the highest amount of acidic sites and more Brönsted sites which may be in favor for fructose conversion to HMF. In the analysis of the reaction mixture, except the major product of HMF, some other by-products, such as furfural, some soluble polymers and humins, were also formed. The concentration of furfural was too low to be quantified. For soluble polymers and humins, it was very difficult to quantify. Further investigation of the catalytic performance of mesoporous niobium phosphate in the dehydration of fructose to HMF in the aqueous phase was focused on NbPO-pH2.

Table 3 Dehydration of fructose catalyzed by different niobium phosphates

Catalyst	Conv. of fructose/%	Selec. to HMF/%	Yield of HMF/%
NbPO-pH2	57.6	78.2	45.0
NbPO-pH7	67.7	49.6	33.6
NbPO-pH10	22.3	9.8	2.18

---

Fig. 8 shows the influence of catalyst dosage on fructose conversion, selectivity to HMF, and HMF yield at a reaction temperature of 130 °C. When the mass ratio of the catalyst to fructose was 0.5, the conversion of fructose was 36.6%, the selectivity to HMF was 94.7% and the yield of HMF was 34.0%. Fructose conversion increased from 36.6 to 56.6% when the catalyst dosage increased from 0.5 to 1.0, while the selectivity to HMF decreased gradually from 94.7 to 78.2%, but the final yield of HMF also increased from 34.0 to 45.0%. When the catalyst dosage further increased to 1.2, the conversion of fructose was increased a little bit, while the selectivity to HMF decreased seriously, so the yield of HMF decreased. Therefore, the mass ratio of catalyst to fructose was kept to 1 in the following experiments if not otherwise indicated.

Fig. 8 Influence of mass ratio of catalyst/fructose on fructose conversion, selectivity to HMF and HMF yield.

Fig. 9(A) shows the effect of temperature on fructose conversion, selectivity to HMF, and HMF yield. The fructose conversion and HMF yield are positively related to reaction temperature below 130 °C although the selectivity to HMF decreased. When the temperature further increased from 130 to 150 °C, the yield was decreased from 45.0 to 29.4%. The reason might be the decomposition of HMF to soluble molecules and/or the polymerization with fructose to soluble polymers, i.e. humins, during the dehydration process under high temperature.²⁸ Therefore, a reaction temperature of 130 °C and the mass ratio of reactant to catalyst at 1 : 1 were used as the reaction conditions hereafter.

Fig. 9 Influence of reaction temperature (A) and time (B) on fructose conversion, selectivity to HMF and HMF yield.

Fig. 9(B) shows the effect of reaction time at 130 °C on the fructose conversion, selectivity to HMF and HMF yield. It can be seen that the conversion of fructose can mount to 45% only after 0.5 h reaction time, indicating that the catalyst is highly active for the dehydration of fructose. Then the conversion increased slightly as the reaction time prolonged. At the same time, the selectivity decreased from 81.7% at 15 min reaction until the minimum of 50.5% at 1.5 h. Further increase in reaction time brought about slight decrease in selectivity. This may be attributed to the further condensation of HMF into byproducts, for example, humins. So the yield of HMF reaches a maximum at 0.5 h of reaction time.

As a comparison, other common solid acid catalysts were also tested. Table 4 provides a comparative catalytic performance of common catalyst with our mesoporous niobium phosphate. The results suggested that our mesoporous niobium phosphate material is a superior catalyst for the production of HMF under aqueous conditions.²⁹ The strong acidic sulfonated copolymer resins, like Amberlyst-15, have low thermal stability and are normally used below 130 °C due to their organic frameworks.

Table 4 Comparable catalytic performance of common commercial solid acid catalysts with NbPO-pH2 (reaction conditions: 130 °C, 30 min)

Catalyst	Conv. of fructose/%	Selec. to HMF/%	Yield of HMF/%
NbPO-pH2	57.6	78.2	45.0
Ambersty-15	61.6	51.9	31.3
Beta	12.86	7.49	0.96
ZSM-5	15.72	6.36	1.00

---

The stability of the catalyst is of great importance for practical usage, so the cycle usage test of niobium phosphate was conducted and the results are presented in Fig. 10. It can be seen that the catalytic activity and selectivity of the catalyst for dehydration of fructose to HMF continuously decrease. But after regeneration at 500 °C for 3 h, the yield of HMF can be returned to the level of fresh catalyst although the conversion and selectivity changed little, indicating a good reusability of our catalyst. The decrease of the catalytic activity is due to two reasons: one is the deposition of insoluble humins on the catalyst surface, it was confirmed by the weight loss of used catalyst (five cycles) from TG-DTA (unshown here), it was as high as 20%. Another reason is the leaching of P species. The leaching of P species really occurred on the used catalyst. After being used five times, the molar ratio of Nb/P increased from 1.0 in the fresh sample to 1.3. But from the catalytic performance of the regenerated sample, the activity can be recovered even after P species leaching occurred.

Fig. 10 Cycle usage and regeneration of mesoporous niobium phosphate catalyst.

Conclusion

In conclusion, we have synthesized a new kind of mesoporous niobium phosphates and investigated the influence of the synthetic pH on mesoporous structure and acidity. TEM and N₂-adsorption–desorption studies proved the formation of mesoporous structure at neutral to acidic conditions. NH₃-TPD analysis confirmed the presence of acid sites. Pyridine adsorbed FTIR spectra identified the presence of Lewis and Brönsted acid sites as well as the distribution of acid strength. This kind of catalyst has high activity in the dehydration of fructose to HMF in aqueous solution, HMF yield can reach 45% under optimal reaction conditions. Moreover, this mesoporous niobium phosphate can be recycled more than five times with only a little deactivation and can be regenerated by calcination at 500 °C.

Acknowledgements

This project was supported financially by the 973 Program of China (2010CB732300), the NSFC of China (No. 21101063 and 20973058), the Science and Technology Commission of Shanghai Municipality (No. 10XD1401400; 10dz2220500) and the Fundamental Research Funds for the Central Universities, China.

Notes and references

1. I. Nowak and M. Ziolek, Chem. Rev., 1999, 99, 3603.
2. K. Tanabe and S. Okazaki, Appl. Catal., A, 1995, 133, 191.
3. M. H. C. de La Cruz, Â. S. Rocha, E. R. Lachter, A. M. S. Forrester, M. C. Reis, R. A. S. San Gil, S. Caldarelli, A. D. Farias and W. A. Gonzalez, Appl. Catal., A, 2010, 386, 60.
4. A. S. Rocha, A. M. S. Forrester, M. H. C. de la Cruz, C. T. da Silva and E. R. Lachter, Catal. Commun., 2008, 9, 1959.
5. K. Nakajima, Y. Baba, R. Noma, M. Kitano, J. N. Kondo, S. Hayashi and M. Hara, J. Am. Chem. Soc., 2011, 133, 4224.
6. N. K. Mal and M. Fujiwara, Chem. Commun., 2002, 2702.
7. N. K. Mal, A. Bhaumik, P. Kumar and M. Fujiwara, Chem. Commun., 2003, 872.
8. A. Sarkar and P. Pramanik, Microporous Mesoporous Mater., 2009, 117, 580.
9. K. Tanabe, Catal. Today, 2003, 78, 65.
10. Y. R. Leshkov, J. N. Chheda and J. A. Dumesic, Science, 2006, 312, 1933.
11. J. J. Wang, W. J. Xu, J. W. Ren, X. H. Liu, G. Z. Lu and Y. Q. Wang, Green Chem., 2011, 13, 2678.
12. H. B. Zhao, J. E. Holladay and H. Brown, Science, 2007, 15, 1597.
13. F. Benvenuti, C. Carlini, P. Patrono, A. M. R. Galletti, G. Sbrana, M. A. Massucci and P. Galli, Appl. Catal., A, 2000, 193, 147.
14. A. Takagaki, M. Ohara, S. Nishimura and K. Ebitani, Chem. Commun., 2009, 6276.
15. X. Qi, M. Watanabe, T. M. Aida and R. L. Smith Jr, Catal. Commun., 2009, 10, 1771.
16. X. Qi, M. Watanabe, T. M. Aida and R. L. Smith Jr, Green Chem., 2008, 10, 799.
17. T. S. Hansen, J. Mielby and A. Riisager, Green Chem., 2011, 13, 109.
18. F. L. Yang, Q. H. Liu, X. F. Bai and Y. G. Du, Bioresour. Technol., 2011, 102, 3424.
19. T. Okuhara, Chem. Rev., 2002, 31, 3641.
20. S. K. Das, M. K. Bhunia, A. K. Sinha and A. Bhaumik, ACS Catal., 2011, 1, 493.
21. C. C. M. Pereira and E. R. Lachter, Appl. Catal., A, 2004, 266, 67.
22. M. H. C. De la Cruz, J. F. C. da Silva and E. R. Lachter, Catal. Today, 2006, 118, 379.
23. M. Paul, N. Pal, B. S. Rana, A. K. Sinha and A. Bhaumik, Catal. Commun., 2009, 10, 2041.
24. P. Lauriol-Garbey, J. M. M. Millet, S. Loridant, V. Bellière-Baca and P. Rey, J. Catal., 2011, 281, 362.
25. M. K. de Pietre, L. C. P. Almeida, R. Landers, R. C. G. Vinhas and F. J. Luna, React. Kinet., Mech. Catal., 2010, 99, 269.
26. H. Nakayama, T. Eguchi, N. Nakamura, S. Yamaguchi, M. Danjyo and M. Tsuhako, J. Mater. Chem., 1997, 7, 1063.
27. M. P. Kapoor, S. Inagki and H. Yoshida, J. Phys. Chem. B, 2005, 109, 9231.
28. R. Karinen, K. Vilonen and M. Niemela, ChemSusChem, 2011, 4, 1002.
29. T. Armarolia, G. Buscaa and C. Carlini, J. Mol. Catal. A: Chem., 2000, 151, 233.

---