Leading the Evolution of Green, Affordable Fuels

In 2004, Frances Arnold had just finished a talk on synthetic biology—a method to build new biological systems or improve existing ones using man-made DNA or other synthesized molecules—when Vinod Khosla raised his hand to ask a question.

Khosla, a cofounder of Sun Microsystems and one of the world's leading venture capitalists, asked Arnold if synthetic biology had any useful purpose.

"He asked me what I'd do with it," recalls Arnold, the Dick and Barbara Dickinson Professor of Chemical Engineering, Bioengineering, and Biochemistry. "I said that I'd like to solve the energy crisis."

Arnold's declaration was more than a pipe dream. She was already well into a project to try to convert methane—an abundant organic compound and greenhouse gas—into methanol, a liquid that can be used as fuel. One of her big breakthroughs was developing a process called directed evolution to create new biocatalysts and metabolic pathways that are useful in a wide range of industries by making DNA mutations and screening for improvements.

After the talk, Khosla invited Arnold to his office in Palo Alto to discuss her ideas. Arnold says, "He asked me if I was interested in starting a company and offered me help to get it off the ground. Like many top venture capitalists, he said, 'I invest in people.'"

As a professor at Caltech, Arnold also invests in people—specifically, graduate students and postdocs. After her initial contact with Khosla, she asked two of her research team members—Matthew Peters, a postdoc, and Peter Meinhold, a graduate student—if they wanted to start a biofuels company with her. The company would develop and commercialize the work that they had already been doing: trying to convert methane to methanol.

"Frances called Matt and me into her office," says Meinhold. "She said, 'This billionaire wants to give me money to start a company. Do you want to run it?' We thought this would be a great opportunity. I wasn't thinking about commercial success. My rationale was that I would do the same science that I would do as a postdoc and learn a heck of a lot more on how to create value."

So in August 2005, with help from Caltech's Office of Technology Transfer (OTT), Meinhold and Peters launched Arnold's startup, renting lab space in Pasadena. Meinhold directed the research and Peters handled the business side, with Arnold advising and serving on the board of directors. "OTT helped set us up with the right people, including attorneys, and were helpful in making the licensing deal with Khosla," Meinhold says.

They called the company Methanotech, keying off the work with methane-to-methanol conversion. However, after the company's first year, progress was slow, so Arnold, Meinhold, and Peters decided to switch gears. "There's nothing wrong with the idea of converting methane to methanol," Arnold says. "But things need to move quickly when you're spending someone else's money."

Arnold knew of another technology being developed by a colleague at UCLA to produce isobutanol out of biomass-derived sugars using engineered E. coli. Isobutanol can be blended with gasoline at higher ratios than ethanol, a fuel additive made from agricultural feedstocks such as corn. It can be burned in cars and small engines, it can be converted to jet and diesel fuel, and it can be used in the manufacture of solvents, plastics, synthetic rubber, and other petroleum-based products. The company licensed the UCLA technology but then began engineering more robust and more complex yeast instead of E. coli to produce isobutanol. Through synthetic biology, metabolic engineering, and directed evolution techniques, they began working to make isobutanol commercially viable.

With the change in direction came a change in the company's name. "Vinod said, 'Come up with a two-syllable word that doesn't mean anything,'" Arnold says. She followed his first recommendation, but quietly ignored the second, settling on "Gevo." Gevo, Arnold says, stands for "green evolution." (Meinhold argues that it's actually a derivative of "genetic evolution.")

The goal for Gevo has been to produce as much isobutanol as possible, as cheaply as possible, and with the fewest byproducts. This means, among other challenges, engineering the organism and its enzymes to increase efficiency and developing an entire process that takes in biomass at one end and produces isobutanol at the other. And, perhaps most important, it must be cost-effective to produce isobutanol from biomass or no one will buy the product.

Half a year after the launch of this new approach, Meinhold and Peters hired three scientists and moved from borrowed space to their own space in Pasadena. Gevo researchers started modifying the UCLA technology to produce isobutanol at a higher yield and lower cost. Through directed evolution, scientists at Caltech and Gevo have made modifications to two of the enzymes involved in the process, dramatically improving the yield.

By 2007, Gevo had about 30 employees and had moved to Englewood, Colorado, where operating costs are lower than in California. Following his original $3 million stake, Khosla continued to support Gevo, investing $27.5 million by 2010. The Virgin Green Fund, a private equity firm backed by Virgin Group's Richard Branson, invested $14 million from 2007 to 2010, and other venture capital firms soon joined in.

Part of Gevo's strategy has been to economize by retrofitting existing fermentation facilities, rather than building facilities from scratch. Gevo first did this in September 2010, when it bought a corn ethanol plant in Minnesota. The company spent a year, from May of 2011 to May of 2012, retrofitting the plant into an isobutanol manufacturing facility. In February 2011, Gevo went public, raising $107 million and helping to pay for the retrofit. Peters left the company later that year to join a prototyping and research company.

Last summer, Gevo began producing commercial quantities of isobutanol out of the sugar derived from corn. "We had the first advanced biofuels plant making non-ethanol biofuels," Meinhold says. "We ran the plant successfully for a few months," he explains, but the cost was still too high to compete against more traditional fuels, so the decision was made to temporarily idle the plant while the technology is further optimized at Gevo's headquarters in Colorado. One benefit of the retrofit design is that the plant, if it makes economic sense, could still produce ethanol.

"To some degree you can learn in the lab how efficient a plant will be, but there are always surprises once it's running at scale," Meinhold says. "There are lots of little things that affect the efficiency. I had been warned by people who have started biotech plants that these problems are normal. You run into issues. This is reality."

Meinhold expects that Gevo—which now employs 120 people—will return to producing isobutanol sometime in 2013 and will eventually start using sugar derived from biomass rather than corn. Other companies, including DuPont and BP, have launched isobutanol or related renewable-fuel ventures, but Meinhold is confident that Gevo's technology will prove the most successful.

"I want to establish commercially viable isobutanol technology from corn and then produce commercially viable technology for cellulosic biomass so that we can move from corn to cellulosics," he says. "At the same time, we want to develop other markets. So, whether it's fuel or plastics, we'll pursue these options and decide what makes the most commercial sense.

"Our mission is to replace petroleum-derived fuels and chemicals with biomass," says Meinhold. "This is important because using biomass is greenhouse neutral. We intend to allow farmers and engineers to make fuel that will decrease the demand for foreign-imported oil, we intend to create jobs locally, and we plan to make money doing it."