Glucose is an isomer of fructose and since it occurs as the monomeric unit in cellulose, it can be considered to be the most abundant monosaccharide in nature. Therefore, glucose is more appropriate than fructose as starting material for 5-HMF. Efficient routes for converting glucose to 5-HMF are an active research topic. However, glucose has been shown to be difficult to convert to 5-HMF (yields <30 %) with solvents such as water [22], organics [39] and organic-water mixtures

[40] . The reason for this is apparently because glucose tends to form a stable six-membered pyranoside structure that has a low enolization rate [25]. Since enolisation rate is the rate-determining step for 5-HMF formation, glucose will react much slower than fructose. Thus, glucose is more likely to undergo cross­polymerization with reactive intermediates and 5-HMF, since it can form true oligosaccharides that contain reactive reducing groups [25]. Because of this limi­tation in the fundamental chemistry, there were no efficient processes for the selective dehydration of glucose into 5-HMF, until a major breakthrough came in 2007 when Zhao et al. [41] published a method for transforming glucose into 5-HMF with an ionic liquid solvent (1-ethyl-3-methylimidazolium chloride, [EMIM] [Cl]) and CrCl2 catalyst. In that work, a 5-HMF yield of 68 % was obtained at a temperature of 100 °C for a reaction time of 3 h. In this reaction, CrCl3~ anion is thought to not only play a key role in proton transfer that facilitates mutarotation of glucose in [EMIM][Cl], but also to play a critical role in the isomerization of glucose to fructose by a formal hydride transfer. Once fructose is formed, it is rapidly dehydrated to 5-HMF in the ionic liquid in the presence of the catalyst. Inspired by this work, a series of papers were reported for the conversion of glucose to 5-HMF using chromium chloride as catalyst [19, 42—45].

Yong et al. [45] studied the production of 5-HMF from fructose and glucose in 1-butyl-3-methylimidazolium chloride ([BMIM][Cl]) using CrCl2 as catalyst, and 5-HMF yields of 96 % and 81 %, respectively for reaction conditions at 100 °C for 6 h reaction time achieved. Those authors considered that NHC/CrClx (NHC=N — heterocyclic carbene) complexes played the key role in glucose dehydration in [BMIM][Cl]. Additionally, in the CrCl2/EMIM system, a NHC/Cr complex could be formed under the reaction conditions and therefore serves as a catalyst [45]. Remarkably, 5-HMF yields were approximately 14 % higher for the reaction carried out in air than that conducted in argon. Zhang et al. [46] studied the production of 5-HMF from glucose catalyzed by hydroxyapatite supported chro­mium chloride in an ionic liquid (1-butyl-3-methylimidazolium chloride), and a maximum 5-HMF yield of 40 % was obtained. In the work of Zhao et al., glucose conversions and 5-HMF yields were lower when CrCl3 was used instead of CrCl2

[41] . However, subsequent studies suggested that there are only minor differences in the catalytic activity of bivalent and trivalent chromium salts [19, 47]. Compared with the strongly reductive Cr (II), the trivalent form, Cr (III), possesses higher stability in the environment, and Cr (III) is essential for mammals in removing glucose from the bloodstream [48]. Binder and Raines [44] made an extensive investigation on glucose conversion in dimethylacetamide (DMA) with the addition of halide salts. Addition of 10 wt% LiCl or LiBr along with CrCl2, CrCl3, or CrBr3 resulted in 5-HMF yields up to 80 %.

A zero-valent Cr(CO)6-based catalyst system was found to be effective for the conversion of glucose to 5-HMF in ionic liquid [EMIM][Cl], even at low catalyst loadings [49]. Through in-situ, ex-situ, and quantitative poisoning experiments, it was demonstrated that small, uniform Cr0-nanoparticles, either preformed via microwave irradiation (3.6-0.7 nm) or generated in-situ via thermolysis (2.3-0.4 nm) during the reaction, are active species responsible for the observed catalysis when using Cr(CO)6 as the precatalyst [49]. In view of some of the advantages of the Cr(CO)6-derived nanoparticles catalyst system, including the relatively low cost and air-stability of the precatalyst as well as its ability to maintain high efficiency at low catalyst loadings, the results should provide a new method to develop more effective metal-nanoparticles catalysts for glucose or related biomass conversion processes.

Two analogous chromium catalysts other than chromium chloride have been reported to be effective for conversion of glucose to 5-HMF [32]. Han and co-workers [50] used 1-ethyl-3-methylimidazolium tetrafluoroborate ([EMIM] [BF4]) with SnCl4 as catalyst, and obtained a 5-HMF yield of ca. 60 % at 100 °C for 3 h. They screened metal chlorides and ILs, for which only chromium(III), aluminum(III), and tin(IV) chlorides exhibited activity, and tin(IV) was concluded to be the most active catalyst. Out of the eight ILs examined, [EMIM][BF4] was found to be most favorable. Those authors proposed that the formation of a five — membered-ring chelate complex consisting of Sn and two neighboring hydroxyl groups in glucose was a probable intermediate in the formation of 5-HMF (Fig. 9.3). Their 1H-NMR measurements showed that chloride in SnCl4 was transferred and interacted with hydrogen atoms, and Sn atoms interacted with oxygens to promote the formation of a straight-chain glucose required for transfor­mation to the enol intermediate and formation of 5-HMF [50]. Stanlberg et al. [51] examined ionic liquids with lanthanide catalyst, and obtained a 5-HMF yield of 24 %. Lanthanide(III) salts have also been tried for the dehydration of glucose to 5-HMF [51]. The use of the strongest Lewis acid Yt(OTf)3 resulted in 24 % of 5-HMF yield, and the catalytic effect increased with increasing atom number in the lanthanide series. Furthermore, the 5-HMF yield was observed to increase with increasing the chain length of the alkyl groups on the imidazolium cation, that is, 1-octyl-3-imidazolium chloride ([OMIM][Cl]) had a significantly higher yield than [EMIM][Cl]. This phenomenon has not been observed with other catalyst systems where [EMIM][Cl] has been superior or equivalent to other methylimidazolium chlorides [51].

Although the catalysts such as CrCl2, CrCl3, Cr(CO)6 and SnCl4 are effective for the dehydration of glucose into 5-HMF, they are poisonous, difficult to recycle and have high environmental risk. The use of inherently nonhazardous catalysts and solvent systems are needed for application in today’s society. Considering that the dehydration of glucose to 5-HMF involves two steps, namely, isomerization of glucose into fructose through base catalysis that is followed by dehydration of

Fructose

furanose

Fig. 9.3 Proposed mechanism for glucose dehydration to 5-HMF catalyzed by SnCl4 in [EMIM] [BF4] (Reproduced with permission from reference [50]. Copyright © 2009 Royal Society of Chemistry) fructose by acid catalysis to give 5-HMF [22], Qi et al. developed a method for glucose conversion into 5-HMF with ionic liquid-water mixture without using chromium-analogous catalysts [52]. They found that the addition of an appropriate amount of water into the ionic liquid has a synergistic effect on the glucose conversion to 5-HMF, and promoted the formation of 5-HMF from glucose com­pared with that in either pure water or in the pure ionic liquid solvent. In the proposed reaction system, a 5-HMF yield of 53 % could be obtained in 50:50 w/w% 1-hexyl-3-methyl imidazolium chloride-water mixture in 10 min reaction time at 200 °C in the presence of ZrO2. It was confirmed that 1,3-dialkylimidazolium ILs with Cl_ and HSO4~ anion were effective for 5-HMF formation from glucose in IL-H2O mixture. The addition of the other protic solvents such as methanol and ethanol into the ionic liquid had a similar synergistic effect as water and promoted fructose and 5-HMF formation [52].