Dramatic advancements in the saccharide to 5-(chloromethyl)furfural conversion reaction.

In 2008, we reported that glucose, sucrose, and cellulose could be converted into 5-(chloromethyl)furfural (CMF) 1 in extraordinary yields in a simple biphasic reactor. The process was conducted under mild conditions and involved nothing more sophisticated than aqueous HCl as a reagent. More recently, we have shown that the reaction can be applied equally well to raw biomass (cotton, newsprint, wood, corn stover, and straw), producing not only 1 and related, minor products 2–4 in excellent overall yield, but also converting the hemicellulose present into furfural 5 (Scheme 1). CMF is of interest because it can be easily processed into drop-in biofuel candidates, such as 5-(ethoxymethyl)furfural 6 and 5-methylfurfural 7 by reacting with ethanol and hydrogen, respectively (Scheme 2). In terms of biofuel precursors, CMF can be thought of like plant oils, which are likewise not appropriate for direct use in automobiles, but can be converted into suitable fuels by reacting with alcohols. The two are complementary in the sense that they are ultimately sourced from different biological pathways in plants. While the method described above gave, to our knowledge, the highest isolated yields to date of simple organic compounds from cellulose and cellulosic biomass, we originally noted that two of the reaction parameters would be targeted for optimization, that is, reaction time and the volume of solvent required to extract the products. Herein, we report significant advancements on these fronts and in related aspects of the process. As originally described, reactions were generally run to 30 h duration in order to obtain the maximum yields of products. 2] Although still faster than glucose fermentation to alcohols, the process was slower than many other approaches to biomass deconstruction, particularly thermal conversion and reforming methods which, despite producing a range of products of varied utility and in moderate yield, are able to process raw biomass in a matter of minutes. We had observed, in the course of the reaction, a substantial outgassing of hydrogen chloride, which correlated with a decreasing rate of conversion to products, such that it became necessary after 18 h to introduce additional concentrated HCl to the reactor to drive the process to completion. We have since determined that conducting the same reaction in a closed system results in a steep reduction in reaction time, with the substrates shown in Table 1 all being converted to CMF 1 within 3 h. Additionally, no alternative cyclization isomer 2, 5-(hydroxymethyl)furfural (HMF) 3, or levulinic acid (LA) 4 is now observed in the organic extract, the crude product being essentially pure CMF 1 The mass balance consists of a small amount of humic material and LA 4 found in the aqueous phase of the reaction mixture, and the yields of the latter are also reported (Table 1). The increase in CMF yields, compared to those previously reported, 2] appear to be the result of a reduction in side products 2 and 3. The published method also involved the continuous extraction of products 1–4 into an organic medium, thus sequestering them from the aqueous acid and preventing degradation reactions that have long plagued the acidic processing of biomass. The extraction rate used was 10 mL min 1 which, over a 30 h period, involved the distillation of ca. 18 L of solvent for the production of a few grams of product. Of course, the reactor made use of a solvent loop whereby the solvent was continuously recycled, so only 650 mL of solvent was needed to charge the reactor, and a total of volume of 1.25 L of extracts [a] Prof. M. Mascal, Dr. E. B. Nikitin Department of Chemistry and Bioenergy Research Group University of California, Davis 1 Shields Avenue, Davis, CA 95616 (USA) Fax: (+ 1) 5307528995 E-mail : mascal@chem.ucdavis.edu Scheme 1. Concurrent processing of corn stover into hexoseand pentose-derived products.