last updated 4/26/12

The conversion of biomass into fuels and/or chemicals represents one way to reduce dependence upon imported crude oil in a sustainable manner. Cellulose is an abundant form of biomass that is not used for food. Chemical catalysts for cellulose hydrolysis offer high rates and low cost, but their selectivity is not satisfactory. Some fraction of the cellulose is always converted into undesirable humic materials, also called humins.

We are studying reaction pathways that lead to the formation of humins. The goals are to use the understanding that develops to minimize the amount of humins that form. If humin formation cannot be completely suppressed, a second goal is to develop ways to convert the humins into materials with a higher value, either in-situ as they form or via additional post-hydrolysis processing.

The hydrolysis of cellulose involves a series of reactions that sequentially produce glucose, perhaps fructose and HMF as key intermediates leading to levulinic acid. Levulinic acid is stable at reaction conditions; it can be recovered and used as a platform chemical for the production of many other chemicals, fuel additives and fuels. It is not clear whether humins form directly from each of these intermediates (and/or others not mentioned), or whether humin formation proceeds via a single intermediate. It is quite clear, however, that humins do form if one begins with HMF. Consequently, our initial studies examined the acid-catalyzed conversion of HMF.

The acid-catalyzed conversion of HMF follows two parallel pathways. The desired pathway leads to equal amounts of levulinic acid and formic acid. The second pathway leads to humins. Horvat suggested that HMF is converted to 2,5-dioxo-6-hydroxy-hexanal (DHH) which then polymerizes or otherwise reacts in an unspecified way to produce humins. Our work suggests aldol addition/condensation is a route for the conversion of DHH to humins. We have also shown that the same aldol addition/condensation reactions can be used to modify the humins that form by adding new functional groups to them. This could represent a significant step toward the valorization of humins.

When glucose or fructose is used as the starting reactant, the IR spectra of the resulting humins are quite similar. There are a few significant differences from the IR spectrum of humins formed from HMF. Specifically, the humins formed from the hexoses show two sharp IR peaks that correspond to a carbonyl group conjugated to a carbon-carbon double bond. This feature can be explained by consideration of the relative concentrations of aldehydes and ketones that are present in the reacting system. In addition, the IR spectrum of humins formed from cellobiose are essentially indistinguishable from those formed from glucose. Overall, the results suggest that humins do not form directly from cellobiose, glucose or fructose. Instead, these starting reagents must first be converted to HMF and then to DHH before humins start to form.

This project is currently unfunded.

“Formation and Growth of Humins via Aldol Addition and Condensation during Acid-Catalyzed Conversion of HMF,” S. Patil and C. R. F. Lund, Energy Fuels, 25 (10), 4745-4755 (2011). [more info]

Presentations Resulting from this Project

“Mechanism of humin formation during catalytic conversion of cellulose-derived carbohydrates,” Michigan Technological University, Materials Science and Engineering Department Seminar, Houghton, MI, April 20, 2012.

“Mechanism of humin formation and growth during acid-catalyzed conversion of glucose, fructose and HMF,” 243^rd ACS National Meeting, San Diego, CA, March 26, 2012.

“Reducing Waste Byproducts of Cellulose Hydrolysis,” 2011 ESW National Conference, Buffalo, NY, October 21, 2011.

“Growth of Humins during Acid-Catalyzed Carbohydrate Conversion,” AIChE Annual Meeting, Minneapolis, MN, October 16-21, 2011.

“Characterization of humins formed during acid-catalyzed hydrolysis of glucose, fructose and HMF,” 22^nd North American Catalysis Society Meeting, Detroit, MI, June 5-10, 2011.

“Characterization of Humins formed during Hydrolysis of C6 Sugars,” Pittsburgh-Cleveland Catalysis Society Meeting, Pittsburgh, PA, May 16-17, 2011.

Varun Gajiwala, “Reactivity of Levulinic Acid during Acid Catalyzed Hydrolysis of Glucose, Fructose or HMF,” M. S. Thesis, University at Buffalo, SUNY, Dept. of Chemical and Biological Engineering (2012) [more info]

Unpublished DFT Findings for HMF

We studied HMF, Figure 1, and it's protonation using DFT (B3LYP, 6-311++G(2d,p), water solvent via pcm). Six low energy conformations were identified, corresponding to three rotational orientations of the hydroxymethyl group and two rotational orientations of the aldehyde group. Figure 2 shows calculated energies as the aldehyde group is rotated with the hydroxymethyl in each of its three low energy positions. Figure 3 shows calculated energies as the hydroxymethyl group is rotated with the aldehyde group in each of its two low energy positions. From the figures, the barrier for aldehyde rotation is ~60 kJ/mol and that for hydroxymethyl rotation is ~4 to 9 kJ/mol.

We also examined the protonation of HMF and complexes it might form with H₃O⁺. Figure 4 shows the relative energy of protonated forms of each of the six low energy conformations of HMF. The protonation positions on the figure correspond to the atom lables in Figure 1. Protonation at some locations introduces chirality, so there are more values shown for those cases than for ones without chirality. The results for complex formation show that H₃O⁺ complexes most strongly with the aldehyde O atom (~-90 kJ/mol), almost as strongly with the hydroxymethyl O atom (~-65 kJ/mol) and relatively weakly with the ring O atom (~-4 kJ/mol).