Exceptionally high yields of furfural from assorted raw biomass over solid acids

Abstract

Development of stable and recyclable solid acid catalysts in the efficient valorisation of hemicellulose to yield C₅ sugars and furfural is vital to boost the prospects of using lignocelluloses for chemicals synthesis. Using an silicoaluminophosphate, namely SAPO-44, as a catalyst, an environmentally benign process of furfural synthesis from diversified real substrates (without any treatment or the need for separation of its components) is shown. In an efficient one-pot methodology, at 443 K and in the presence of a biphasic solvent system, selective conversions of hemicelluloses from raw biomass (bagasse, rice husk and wheat straw) to extraordinarily high yields of furfural of about 93% were attainable. Under similar reaction conditions, an 82% yield of furfural is also achievable directly from isolated hemicellulose within 10 h. Hydrophilic SAPO-44, having higher hydrothermal stability, showed similar activity for all the substrates for a minimum of up to 8 times in recycling runs. Various physico-chemical characterizations (X-ray diffraction, thermogravimetric analysis-derivative thermogravimetry, temperature programmed desorption-NH₃, N₂ sorption, solid-state nuclear magnetic resonance spectroscopy) of fresh and spent catalysts were used to improve SAPO-44 stability.

---

Introduction

Replacement of non-renewable fossil feedstock with renewable biomass feedstock for the efficient production of chemicals is a great challenge in the area of green chemistry.¹ The setting-up of a bio-based economy is most important as it will provide us with a favourable environment by minimizing global warming and it has a potential to fulfil societal needs for chemicals and energy. The use of lignocellulosic biomass for this purpose is ideal because it is non-edible (for humans) and it is available abundantly in the form of crop or forest waste. The bio-refinery product furfural is a key platform chemical used in the synthesis of various useful organics.^2–6

Conventional methods use mineral acid technology for the production of furfural^7,8 from C₅ sugars and hemicelluloses; however, these invariably create environmental issues and several other problems (corrosiveness, toxicity, handling hazards, recovery, generation of neutralization waste and so on). Thus, to avoid these inconveniences, use of non-toxic, recoverable heterogeneous catalysts having acidic functionality (solid acid catalysts) have recently been reported for the production of furfural, and primarily from xylose, a C₅ sugar.^9–24 These reactions are mainly carried out in water; but to improve the yields of furfural, use of polar aprotic solvents and biphasic solvent systems are also reported.^11,12,16,25

Although solid acids are known to convert xylose into furfural, use of xylose as a substrate in the synthesis of furfural is not an economical option as xylose needs to be obtained in the pure form from hemicellulose in yet another reactor. The two-pot methods for the production of C₅ sugars and furfural are described in the literature.^26–28 Recently, a two-pot method for the conversion of hemicellulose to furfural was demonstrated where wood was first treated with hot water to convert its hemicellulose to yield soluble products and then in the next reactor these products, in the presence of H-mordenite (HMOR) as a catalyst, yielded furfural at the 74% level.²⁸ Similarly, in another report it is shown that soluble products (xylo-oligosaccharides) having a degree of polymerisation of 1–14 (obtained from wood chips after hot-water treatment) can yield ca. 90% furfural using mineral acid (HCl, H₂SO₄) and NaCl as catalyst.²⁹ To overcome the use of the two-pot method and use of mineral acids, recently a one-pot method for the conversion of hemicellulose into C₅ sugars and furfural using solid acid catalysts was disclosed.^30,31 Although in these communications it is reported that the zeolite, HUSY (Si/Al = 15), could give a yield of ca. 54% furfural, unfortunately the catalyst underwent morphological changes during the reactions.³⁰ Consequently, it becomes crucial to develop stable solid acid catalysts for the efficient conversion of hemicellulose into furfural in a one-pot fashion (Scheme 1). Additionally, it is also essential to improve the furfural yields in these reactions to make the overall method attractive. In yet another report, microwave heating of hemicellulose and bagasse in the presence of mineral acid and a chloride salt is claimed to give good yields of furfural.³²

Scheme 1 One-pot conversion of hemicellulose to furfural using a solid acid catalyst.

In the current study the foremost intention was to introduce a stable catalytic system for the selective production of furfural in high yields directly (preferably) from raw (real) biomass. Silicoaluminophosphates (SAPOs) are known to have higher hydrothermal stability^33,34 and thus it was thought that those may be stable under the hemicellulose reaction conditions. Further incentive for choosing SAPO catalysts was its expected higher hydrophilicity arising from the presence of ‘P’,³⁵ which may limit the catalyst in the aqueous layer and thereby enhance the prospects of accomplishing higher furfural yields. Here we show that the use of SAPOs in the hemicellulose conversion improves the furfural yields in the absence of any pH-modifying reagents.

Results and discussion

One-pot processing of isolated hemicellulose

Fig. 1 presents the data on the yields obtained for the products in the conversion of isolated softwood-derived hemicellulose (xylan), when reactions were carried out in a water + toluene solvent system at 443 K for 8 h. Under these reaction conditions, a non-catalytic reaction yields only 26% furfural, along with 8% oligomers and 8% xylose + arabinose. It is evident from Fig. 1 that the SAPO-44 catalyst provides the best activity amongst all the other catalysts evaluated in this study. With SAPO-44 we could achieve a furfural yield of 63% directly from the isolated hemicellulose, whereas the zeolites showed furfural yields within a lower range (44–56%). The lower acid amount and higher hydrophobicity of zeolites compared with SAPO-44 is responsible for the lower activity (Table S1, ESI†). Other SAPO catalysts (SAPO-46, SAPO-5 and SAPO-11) were also active in this reaction; however, these showed slightly lower furfural yields (51%, 40% and 35%, respectively). Again, this is because of either the lower amount of total acid sites present on these catalysts, or the absence of strong acid sites (Table S1, ESI†). Although, Amberlyst®-15 (61%), Nafion® SAC-13 (52%) and niobium pentoxide (55%) yield furfural in good quantities, these were not stable – as is evident from catalyst recycling studies (see below) and other earlier work.^30,31 Next, instead of toluene we used methyl-iso-butyl ketone (MIBK) and p-xylene as organic solvents together with water (1 : 1 v/v). Results show that, compared with toluene, these solvents (MIBK: 53%, p-xylene: 55%) are less active in the production of furfural from hemicellulose.

Fig. 1 Evaluation of various solid acid catalysts for furfural synthesis from softwood hemicellulose in a one-pot method. Reaction conditions: softwood hemicellulose (0.3 g), catalyst (0.075 g), water + toluene = 60 mL (1 : 1 v/v), 443 K, 8 h.

To understand this difference in the activity with the change in solvent system, the partition coefficient of furfural in these solvents was calculated (Fig. S1, ESI†). Additionally the miscibility of organic solvents in water was also studied. It was found that the higher partition coefficient of furfural in the water + toluene system, as compared with the other two solvent systems, is responsible for a higher furfural extraction in toluene and thus a better furfural yield was achieved through use of the water + toluene system. A lower furfural yield in water + MIBK compared with water + p-xylene is observed – although the water + MIBK system has a higher partition coefficient compared with water + p-xylene. This latter is because of the higher miscibility of MIBK in water (19.1 g L⁻¹ at 293 K) compared with the miscibility of p-xylene in water (0.18 g L⁻¹ at 293 K), which leads to the availability of furfural in both of the solvents, which then leads in turn to the degradation reaction.

At this juncture, to boost the furfural yield using the SAPO-44 catalyst further, reactions were performed for a longer time at 443 K. A furfural yield of 63% observed after 8 h was increased to 71% after 10 h, but a further increase in time (to 12 h) did not show any significant improvement in the furfural yield (which was then 72%).

To further advance the furfural yield, instead of using a substrate/catalyst (S/C) ratio of 4 (w/w) the ratio was increased to 8 and with this a furfural yield of 57% could be obtained with an 11% yield for xylose + arabinose after 8 h reaction time at 443 K. The marginally lower furfural yields might be because of the non-availability of active sites on the catalyst. In a subsequent reaction, an S/C ratio of 2 was maintained and a similar result (furfural yield of 63%) was observed to the one seen with the S/C ratio of 4. In order to investigate the effect of substrate concentration, reactions were caused while maintaining the S/C ratio at 4. Results demonstrate that an increase in hemicellulose concentration up to 5 wt% gives similar amounts of furfural (63 ± 1%), whereas it decreases to 42% when 10 wt% hemicellulose concentrations are used. At high hemicellulose concentrations, formation of a by-product (humin) leads to inferior furfural yield, as is evident from the formation of a dark-coloured reaction solution.

It is expected that by increasing the extracting-solvent-to-furfural ratio, furfural yields will be enhanced since furfural will be extracted into the extracting solvent and thus are unavailable for further reactions (in water). The water-to-toluene ratio was thus altered from 1 : 1 to 1 : 2 and 1 : 4 (v/v) by keeping the substrate/water and substrate/SAPO-44 ratios constant. By employing a 1 : 2 ratio for 8 h reaction time, a superior yield of furfural (75%) (C₅ sugar = 7%) was obtained; but when a 1 : 4 ratio was used, the yield was decreased to 49% (C₅ sugar = 4%). In comparison with these results with a 1 : 1 ratio, a 63% furfural yield was achieved after 8 h reaction. Earlier results showed us that by carrying out the reaction for a longer time the furfural yield can be improved, so it was decided to perform the reactions with a 1 : 2 ratio for a longer time. The results illustrate that after 10 h, a furfural yield of 82% was accomplished, as compared with 75% after 8 h. With a further increase in time to 12 h, an even greater improved yield of 85% for furfural was feasible – although a subsequent increase in time (13 h and 14 h) did not yield any additional furfural. To the best of our knowledge, this is probably the first report where such a high yield of furfural was obtained directly from hemicellulose in a one-pot method.

By fine tuning the reaction parameters, it was revealed that in order for there to be an efficient conversion of hemicelluloses into furfural, the optimized reaction conditions were: use of 1 or 5 wt% substrate solutions, SAPO-44 catalyst, S/C ratio of 4, water + toluene at 1 : 2 (v/v), tempertaure of 443 K, reaction time of 10–12 h. Under these conditions, 82%/85% furfural yield was possible with >99% conversion of softwood hemicellulose. To verify the robustness of the catalyst, recycling studies were carried out with the recovered SAPO-44 catalyst from the previously mentioned reaction (10 h, 82% yield). The catalyst was simply washed with distilled water and then subjected to the next reaction (for details see Experimental section, ESI†). Fig. S2 (ESI†) shows exceedingly good recycling activity of the SAPO-44 catalyst up to the 8^th catalytic run, with similar furfural yields (83 ± 1%). Together with furfural, the formation of oligomers (2–3%), sugars (14–16%) and 5-hydroxymethyl furfural (HMF, 2%) accounts for the 100 ± 2% carbon balance. Furthermore, to find out the loss of catalyst after all these reactions, after the 8^th run, the SAPO-44 catalyst was recovered from the reaction mixture (by centrifugation), washed with distilled water, dried (in oven and vacuum) and calcined (in air) at 823 K for 12 h. From the calculation of weight difference between the initially charged catalyst (1^st reaction) and the catalyst recovered after the 8^th run (calcined), minimal weight loss (ca. 10%) was observed.

The catalyst recycling studies were also performed with ion-exchange resins and niobium pentoxide and these showed a decrease in the activity in recycling studies (Amberlyst®15: 61% and 51%, Nafion® SAC-13: 52% and 47%, niobium pentoxide: 56% and 46%). These results strengthen our idea behind the use of hydrothermally stable SAPO.

To confirm the formation of furfural in this reaction, we isolated the furfural from the reaction mixture (toluene) by evaporating the solvent in a rotary evaporator and then the semi-solid obtained was characterized by proton nuclear magnetic resonance (¹H-NMR) and ¹³C-NMR analysis. The NMR data presented in Fig. S3a and b (ESI†) confirms the formation of furfural in a pure form. Later, we calculated the isolated yield of furfural as 79% with an error of ±2% (in repeated attempts), when considering the total furfural yield in the reaction as 100%.

To establish the catalytic system, isolated hemicelluloses from a variety of sources, such as softwood (oat spelt and beechwood) and hardwood (birchwood), were used. When hardwood hemicellulose was processed at 443 K over SAPO-44 for 10 h in water + toluene (1 : 2 v/v), yields of 69% furfural with 4% oligomers and 13% xylose + arabinose were obtained (total carbon recovery = 86%). Under similar reaction conditions, conversion of hemicellulose derived from oat spelt and beechwood showed similar activity (ca. 82% furfural yield, ca. 100 ± 2% total carbon recovery).

One-pot processing of un-treated raw biomass

It is evident from the previous work described above that SAPO-44 is a highly active catalyst for converting isolated hemicelluloses derived from different sources. Subsequently, it was desired to process the real lignocellulose materials (containing cellulose, hemicelluloses and lignin) using an optimized catalytic system in a one-pot method without any pre-treatment (Fig. S4, ESI†). The figure also explains the different sizes of raw materials used in the study. Annually, India produces vast amounts of agricultural products: chiefly sugarcane (3.42 × 10⁸ Mt, in 2011), rice (1.58 × 10⁸ Mt, in 2011) and wheat (8.69 × 10⁷ Mt, in 2011); more generally, worldwide production of these crops is enormous³⁶ (for details see Table S2a, b and c, ESI†). So it inspired us to exploit the wastes generated from these crops (bagasse, rice husk and wheat straw) in our work. In our study, real biomass substrates were collected from different parts of India and were used to develop a uniform methodology for the conversion of waste to furfural. All the raw biomass collected were analysed by a Technical Association of the Pulp and Paper Industry (TAAPI) method and inductively coupled plasma – an optical emission spectroscopy (ICP-OES) technique to determine the composition. The results are summarized in Table 1. All the experiments were repeated 3 times and the error in the results is ±2% (based on the actual value) and the results are comparable with those results found in the literature.³⁷

Table 1 Composition analysis of raw biomass

Parameters^a	Substrates
	Bagasse (I)	Bagasse (II)	Bagasse (III)	Rice husk (I)	Rice husk (II)	Rice husk (III)	Wheat straw
a TAAPI method is used for composition analysis.b ICP-OES analysis; calculations were done based on 1 g of raw biomass.
Inorganic and organic composition in %
Ash	2.1	2.8	3.3	15.7	17.0	18.6	12.4
Pentosan	30.0	24.1	21.6	15.9	11.2	12.4	21.4
Holocellulose	63.2	71.1	68.3	60.5	52.4	52.6	63.2
α-Cellulose	38.5	41.2	39.9	37.1	28.5	34.3	36.7
β-Cellulose	12.9	15.2	15.3	9.9	11.7	6.8	17.1
γ-Cellulose	11.8	14.7	13.1	13.5	12.2	11.5	9.4
Total lignin	15.2	20.4	21.6	24.1	21.7	22.7	17.8
Nutrient composition^b in mmol g^−1
Na	0.0	0.0	0.0	0.0	0.0	0.0	0.0
K	0.04	0.03	0.02	0.07	0.05	0.07	0.12
Ca	0.03	0.03	0.07	0.02	0.01	0.03	0.03
Mg	0.03	0.02	0.03	0.01	0.04	0.04	0.02
Al	0.01	0.01	0.02	0.01	0.01	0.02	0.02
P	0.02	0.01	0.01	0.01	0.05	0.05	0.01

---

Fig. 2 illustrates that from real substrates a remarkably high yield of 93% is achievable for furfural when the reactions were performed at 443 K for 8 h. In particular, when bagasse is used a 92 ± 1% yield is possible, while for other substrates an 86–92% yield is realizable. In all these reactions 100 ± 5% mass balance is achieved. The results demonstrate the capability of the SAPO-44 catalyst to process unpurified (non-isolated) hemicellulose from raw biomass to furfural in high yields. It is essential to mention here that in these reactions, formation of a minimal amount of glucose, fructose and HMF was also seen. The origin of these compounds is attributed to the fact that under similar reaction conditions cellulose is undergoing conversion (18%). This was further confirmed with the X-ray diffraction (XRD) analysis of fresh and recovered bagasse (with a catalyst) after reaction. A decrease in intensity for cellulose was observed in the XRD pattern (Fig. S5, ESI†), which proves that part of the cellulose is converted under these reaction conditions. It is also evident that some amorphous and crystalline parts of the cellulose are still not converted (2θ = 12.5–18.5° and 19.2–26.7°). However, a close look at the pattern reveals that because of the overlapping of the peaks of SAPO-44 by cellulose (bagasse), few peaks of SAPO-44 are visible (2θ = 9.5°, 20.6°, 21.8°, 30.9°). But it is certain that SAPO-44 does not undergo any structural changes, as was validated by the XRD of the calcined samples.

Fig. 2 Utilization of non-isolated hemicellulose from raw biomass. Reaction conditions: substrate (0.67 g), SAPO-44 (0.05 g), water + toluene = 60 mL (1 : 2 v/v), 443 K, 8 h.

In order to get an insight on the capability of SAPO-44 used in raw biomass work, a catalyst recycling study was undertaken. It is apparent from Fig. 3 that the catalyst showed more or less matching activity in at least eight cycles, even with different substrates. This clearly emphasizes the fact that the SAPO-44 catalyst is very robust and stable under the reaction conditions.

Fig. 3 Recycling study of the SAPO-44 catalyst used in raw biomass reactions. Reaction conditions: substrate (0.67 g), SAPO-44 (recycling quantity approx. 0.05 g), water + toluene = 60 mL (1 : 2 v/v), 443 K, 8 h.

Stability and other properties of SAPO-44 catalyst

The superior activity of the SAPO catalyst compared with zeolite(s) might be explained in terms of its hydrophilic properties and its hydrothermal stability. To discover the differences in the hydrophilicity of the catalysts, those were dispersed in a water-organic solvent mixture (Fig. S6, ESI†). It was found that when both water and toluene were present, SAPOs are predominantly present in an aqueous layer while HMOR is present in both the layers. This property facilitates the separation of the SAPO catalyst from the furfural extracted in the toluene layer, and thus helps in suppressing the condensation and degradation reactions.³⁵ To confirm this, a TGA-DTG study was performed and the profiles are presented in Fig. S7 (ESI†). The profile for fresh SAPO-44 catalyst is displayed in Fig. S7a (ESI†), which shows 7.9% mass loss corresponding to desorption of water at a lower temperature (365.7 K). A further loss of 0.8% at 673 K is because of the decomposition of the structure-directing agent. No subsequent loss – even after heating the sample up to 1273 K – implies that the SAPO-44 material is thermally stable. To discover what might happen under the reaction conditions where SAPO-44 and HMOR are suspended in water, TGA-DTG experiments were carried out after exposing the catalysts to water vapour (for details see ESI†). Interestingly, it was apparent that for the SAPO-44 catalyst at 365.5 K a mass loss of 17.2% was possible because of the elimination of adsorbed water (Fig. S7b, ESI†). The difference of 9.3% of mass loss between the SAPO-44 catalyst exposed to water or not exposed to water indicates adsorption of water on the SAPO-44 surface. Furthermore, the HMOR catalyst shows a slow but steady mass loss up to ca. 673 K for both activated (8.4%) and water-exposed (12.5%) catalytic material (Fig. S7c and d, ESI†). For the HMOR catalyst, almost a 4.1% difference in mass loss was observed between the catalyst exposed to water and one not exposed to water. Based on these results it is safe to claim that SAPO-44 is more hydrophilic. We propose that these properties are helpful in achieving higher yields and stability under reaction conditions.

The XRD analysis for SAPO establishes the presence of a crystalline phase in the synthesized material (Fig. 4).^38–41 XRD patterns for spent SAPO-44 catalysts recovered from the isolated hemicellulose reactions indicate the presence of all the peaks corresponding to CHA topology, as was seen with the fresh SAPO-44 catalyst, with the intensity of the peaks being almost the same. This data confirms the structural stability of SAPO-44. Furthermore, we have checked the XRD patterns for spent (calcined) SAPO-44 used in raw biomass reactions (during calcination, cellulose and lignin are removed, which is confirmed by CHNS elemental analysis). It was observed that bagasse does not have much effect on the crystallinity of SAPO-44 but, because of the presence of a high SiO₂ content (ash) in rice husk and wheat straw (Table 1), a slight hump is observed when 2θ is in the range 16–29°, due to the presence of amorphous silica. This concept is also supported by a NH₃₋ temperature-programmed desorption study of fresh and spent SAPO-44, where Table S1 (ESI†) shows that both the amount of total acid sites and the acid site distribution remain similar in both fresh and spent SAPO-44 catalysts used in either isolated hemicellulose or real biomass reactions.

Fig. 4 X-ray diffraction patterns for SAPO.

Moreover, a nitrogen-sorption study confirmed that the surface area of SAPO-44 was approximately the same even after use in either isolated hemicellulose or real biomass reactions (Table S3, ESI†). Consequently, an almost similar pore size was also observed for SAPO-44 in all the cases (Table S3, ESI†). This data further proves that the SAPO-44 is a very stable catalyst under the reaction conditions.

Solid-state ²⁹Si, ²⁷Al and ³¹P magic-angle spinning NMR spectra further corroborate the structural stability of SAPO-44 (Fig. S8a–c; ESI†). In ²⁹Si-NMR, both fresh and spent SAPO-44 (used in the isolated hemicellulose reaction) shows that the peaks corresponding to Q⁰, Q¹, Q², Q³ and Q⁴ species at −92.4, −96.2, −101.7, −107.7 and −112.8 ppm, respectively, have comparable intensity. The presence of a sharp single peak with similar intensity at 38.2 ppm in ²⁷Al-NMR is indicative of the tetrahedral Al environment in both fresh and spent SAPO-44. Besides this sharp peak, another small peak at −15 ppm due to the presence of octahedral Al is observed in both fresh and spent catalysts. ³¹P-NMR spectra also support the fact that the SAPO-44 structural stability is the only tetrahedral environment for P, which is observed in both fresh and spent catalyst (used in the isolated hemicellulose reaction). The peak corresponding to −31.5 ppm is due to the existence of a P[4Al] environment in the SAPO-44 framework. The presence of additional SiO₂ in raw biomass affects the solid-state NMR spectra of the spent catalyst, so it was difficult to confirm the structural stability of SAPO-44 in these reactions. But recycling study data may suggest that the catalysts are stable. In comparisons given in earlier communications, it was shown that zeolites undergo morphological changes.³⁰

Conclusions

In summary, we have demonstrated that the solid acid catalyst SAPO-44 is capable of processing all varieties of abundantly and cheaply available isolated hemicelluloses (derived from softwood and hardwood) and that the biphasic system (water + toluene) with 1 : 2 v/v ratio works well to give 85% furfural yield in a one-pot method at 443 K. Also, a highly efficient one-pot pathway is shown to process the non-isolated hemicellulose part of the raw biomass (bagasse, rice husk and wheat straw) directly to furfural with an even higher yield (86–93%), thus invalidating the need to isolate the hemicellulose from the raw biomass and also the need to obtain the xylose in a different reactor from the hemicellulose. The hydrophilic character of the SAPO-44 catalyst helps in achieving higher furfural yields because a catalyst that prefers to remain in the aqueous layer is not available to catalyse further reactions of furfural. It is proved that SAPO-44 catalysts are highly stable under reaction conditions by subjecting them to various physico-chemical characterizations, and this showed consistent activity in at least eight recycling runs.

It is suggested that these catalysts will be useful for converting a variety of substrates derived from biomass, which need to be processed in the presence of water.

Acknowledgements

We acknowledge Department of Science & Technology (DST), India for funding. P. B. thanks the University Grants Commission (UGC), India for the Research Fellowship.

Notes and references

1. P. N. R. Vennestrom, C. M. Osmundsen, C. H. Christensen and E. Taarning, Angew. Chem., Int. Ed., 2011, 50, 10502.
2. J. C. Serrano-Ruiz, R. Luque and A. Sepulveda-Escribano, Chem. Soc. Rev., 2011, 40, 5266.
3. R. M. West, Z. Y. Liu, M. Peter, C. A. Gartner and J. A. Dumesic, J. Mol. Catal. A: Chem., 2008, 296, 18.
4. M. G. Kim, G. Boyd and R. Strickland, Holzforschung, 1994, 48, 6.
5. J. P. Lange, W. D. van de Graaf and R. J. Haan, ChemSusChem, 2009, 2, 437.
6. C. J. Barrett, J. N. Chheda, G. W. Huber and J. A. Dumesic, Appl. Catal., B, 2006, 66, 111.
7. J. N. Chheda, Y. R. Leshkov and J. A. Dumesic, Green Chem., 2007, 9, 342.
8. R. Weingarten, J. Cho, J. W. C. Conner and G. W. Huber, Green Chem., 2010, 12, 1423.
9. V. Choudhary, A. B. Pinar, S. I. Sandler, D. G. Vlachos and R. F. Lobo, ACS Catal., 2011, 1, 1724.
10. R. O'Neill, M. N. Ahmad, L. Vanoye and F. Aiouache, Ind. Eng. Chem. Res., 2009, 48, 4300.
11. C. Moreau, R. Durand, D. Peyron, J. Duhamet and P. Rivalier, Ind. Crops Prod., 1998, 7, 95.
12. J. Lessard, J. F. Morin, J. F. Wehrung, D. Magnin and E. Chornet, Top. Catal., 2010, 53, 1231.
13. E. Lam, J. H. Chong, E. Majid, Y. Liu, S. Hrapovic, A. C. W. Leung and J. H. T. Luong, Carbon, 2012, 50, 1033.
14. S. R. Lima, M. Pillinger and A. A. Valente, Catal. Commun., 2008, 9, 2144.
15. J. Zhang, J. Zhuang, L. Lin, S. Liu and Z. Zhang, Biomass Bioenergy, 2012, 39, 73.
16. A. S. Dias, M. Pillinger and A. A. Valente, J. Catal., 2005, 229, 414.
17. E. Lam, E. Majid, A. C. W. Leung, J. H. Chong, K. A. Mahmoud and J. H. T. Luong, ChemSusChem, 2011, 4, 535.
18. X. Shi, Y. Wu, P. Li, H. Yi, M. Yang and G. Wang, Carbohydr. Res., 2011, 346, 480.
19. I. Sadaba, S. R. Lima, A. A. Valente and M. L. Granados, Carbohydr. Res., 2011, 346, 2785.
20. T. Suzuki, T. Yokoi, R. Otomo, J. N. Kondo and T. Tatsumi, Appl. Catal., A, 2011, 408, 117.
21. A. S. Dias, S. R. Lima, D. Carriazo, V. Rives, M. Pillinger and A. A. Valente, J. Catal., 2006, 244, 230.
22. A. S. Dias, S. R. Lima, M. Pillinger and A. A. Valente, Carbohydr. Res., 2006, 341, 2946.
23. A. S. Dias, M. Pillinger and A. A. Valente, Microporous Mesoporous Mater., 2006, 94, 214.
24. S. R. Lima, A. Fernandes, M. Antunes, M. Pillinger, F. Ribeiro and A. Valente, Catal. Lett., 2010, 135, 41.
25. S. Kim, S. You, Y. Kim, S. Lee, H. Lee, K. Park and E. Park, Korean J. Chem. Eng., 2011, 28, 710.
26. S. B. Kim and Y. Y. Lee, Biotechnol. Bioeng. Symp., 1985, 15, 83.
27. Y. Kim, R. Hendrickson, N. Mosier and M. R. Ladisch, Energy Fuel, 2005, 19, 2189.
28. E. I. Gurbuz, J. M. R. Gallo, D. M. Alonso, S. G. Wettstein, W. Y. Lim and J. A. Dumesic, Angew. Chem., Int. Ed., 2013, 52, 1270.
29. R. Xing, W. Qi and G. W. Huber, Energy Environ. Sci., 2011, 4, 2193.
30. R. Sahu and P. L. Dhepe, ChemSusChem, 2012, 5, 751.
31. P. L. Dhepe and R. Sahu, Green Chem., 2010, 12, 2153.
32. R. Carrasquillo-Flores, M. Kaldstrom, F. Schuth, J. A. Dumesic and R. Rinaldi, ACS Catal., 2013, 3, 993.
33. B. M. Lok, C. A. Messina, R. L. Patton, R. T. Gajek, T. R. Cannan and E. M. Flanigen, J. Am. Chem. Soc., 1984, 106, 6092.
34. P. Bhaumik and P. L. Dhepe, ACS Catal., 2013, 3, 2299.
35. P. Bhaumik and P. L. Dhepe, RSC Adv., 2013, 3, 17156.
36. Food and Agriculture Organization of the United Nations, http://faostat3.fao.org/faostat-ateway/go/to/download/Q/QC/E.
37. H. Kobayashi and A. Fukuoka, Green Chem., 2013, 15, 1740.
38. S. Ashtekar, S. V. V. Chilukuri and D. K. Chakrabarty, J. Phy. Chem., 1994, 98, 4878.
39. J. Das, S. P. Lohokare and D. K. Chakrabarty, Indian J. Chem., Sect. A: Inorg., Bio-inorg., Phys., Theor. Anal. Chem., 1992, 31A, 742.
40. D. Young and M. E. Davis, Zeolites, 1991, 11, 277.
41. W. Kong, W. Dai, N. Li, N. Guan and S. Xiang, J. Mol. Catal. A: Chem., 2009, 308, 127.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed description of materials used, catalyst preparations, characterization techniques, catalytic reaction and calculation, recycling study of SAPO-44 in isolated hemicellulose reaction, worldwide distribution of crops, hydrophilicity–hydrophobicity study, TGA-DTG analysis, TPD-NH₃ analysis, solid state NMR studies. See DOI: 10.1039/c4ra04119d

---