Print

Making cellulose more accessible to enzymes

By Lisa Gibson | May 31, 2010
Posted July 15, 2010, at 3:10 p.m. CST

Wisconsin-based Cellulose Sciences International has received a $150,000 Small Business Innovation & Research grant from the National Science Foundation to put toward scale-up and optimization of its proprietary cellulose-conditioning pretreatment. The process allows faster biomass conversion and more economic cellulosic ethanol production by making the cellulose much more accessible to the enzymes, according to the company.

The pretreatment is based on work started in the 1970s by Rajai Atalla, now CSI's CEO, chief scientfici officer and chemical engineer. "Now what we're doing is taking know-how from back in the day and sculpting it so it's a better fit for the biorefineries," said Tony Hartmann, who does business development for CSI. The process involves stirring in a proprietary pretreatment solution, followed by separation of the solids from the solution. Then, there is a series of wash cycles that use only solvent and water, according to CSI president Rowan Atalla. "That's it," Atalla said. "No cooking necessary." The procedure could take place inside any large stirred vessel, he added, and essentially swells the cellulose fibrils and opens them up.

Lignin and hemicelluloses are removed, almost always at elevated temperatures, in the first step of conventional treatment of lignocellulosic feedstocks. "As lignin and hemicellulose is removed, the cellulose aggregates much more tightly and it becomes much more difficult for enzymes to hydrolyze it," Atalla said. "Our process opens the structure back up and makes it more accessible to hydrolytic enzymes." It's almost identical to the gelatinization step in the corn-to-ethanol process, Hartmann said.

The initial intent with the experiment was to reduce the residence time with the enzymes, Atalla said, but the researchers discovered that the process could also reduce the amount of enzymes needed. CSI has seen a 40 percent greater conversion rate with only one-quarter the amount of enzymes used in the control group, he cited. The company has gotten best results with corn stover, but has used cellulose derived from woody biomass as well as herbaceous feedstocks.

Besides greater conversion and faster cellulosic ethanol production, the pretreatment has other benefits. It can be done at room temperature with ambient pressure and uses no exotic re-agents. "There's nothing really outlandish in there," Atalla said. "It's basically caustic and solvent." The process can also bring the costs of enzymes for cellulosic ethanol down from about 50 cents per gallon to only five cents per gallon, Hartmann said.

Future scale-up of the pretreatment from bench scale to demonstration will be located at Didion Ethanol's Cambria, Wis., 50-million-gallon corn-ethanol plant. Feedstocks will include corn stover and low-value side streams from the Didion plant. "The scale-up and timeline are all guesswork," Hartmann said. "But we know this technology works now."

The $150,000 is only a portion of scale-up costs and will most likely be spent in the next six months inside the lab, before the system is moved to Didion's plant, Hartmann said. He declined to speak about project finance in detail, but said, "Our best customers are going to be people who are already making ethanol."

Project partners include enzyme giant Novozymes, Great Lakes Ag Energy of Middleton, Wis., Didion, and pulp and paper industry consultants.

CSI's process makes economic sense, which is crucial, Hartmann emphasized, as nothing gets off the ground unless it's cheap enough. "[The pretreatment] is really economic and it stems from a really good understanding of cellulose and cellulosic properties," he said.
 

0 Responses

     

    Leave a Reply

    Biomass Magazine encourages encourages civil conversation and debate. However, we reserve the right to delete comments for reasons including but not limited to: any type of attack, injurious statements, profanity, business solicitations or other advertising.

    Comments are closed