Coskata has also improved the last steps of the process, in which the ethanol is separated from the water. Ordinarily, this is done using distillation, which is expensive and consumes 30 percent as much energy as burning the ethanol will release. Coskata instead uses a modified version of an existing technology called vapor permeation. Vapor permeation uses hydrophilic membranes to draw off the water, leaving pure ethanol behind. It also consumes half as much energy as distillation per gallon of fuel. Vapor permeation is difficult to use with most biological manufacturing processes, Tobey says, because biomass fed to the microörganisms washes out with the water and can clog up the system. But in Coskata’s process, the bacteria feed only on syngas, not on biomass. So no extra filtration is required to make vapor permeation work.
Coskata continues working on its bacteria, trying to increase the amount of ethanol they can produce. The company now uses varieties of Clostridium, a genus that includes a species that make botulism toxin and another that processes manure on farms. Coskata has started building an automated system for screening new strains of Clostridium according to their ability to make ethanol. Along the way, it has had to develop techniques for protecting its bacteria from being exposed to oxygen; the bacteria are anaerobic, and oxygen kills them at about the same concentrations at which carbon monoxide kills humans. The automated system should allow the company to sort through 150,000 new strains a year, up from a few thousand now.
The researchers can go only so far by sorting through random variations, however. Eventually, Tobey hopes to begin manipulating the microbes’ genes directly, activating only those that improve ethanol production. Such engineering is fairly common now, but the Clostridium bacteria that Coskata uses haven’t been studied much. So although Tobey knows what chemical steps the bacteria use to transform syngas into ethanol, he doesn’t yet know the details of how genes regulate this process, and what role these genes play in the general processes that keep the bacteria alive. What’s more, effective ways of manipulating the genes in these particular bacteria haven’t yet been developed.
Even as Coskata continues to improve its microbes, it is planning to move the fuel production process out of the lab and scale it up to the commercial level. With the help of GM and other partners, the company will build a facility that’s able to produce 40,000 gallons of ethanol per year. Coskata representatives say construction will begin within the year. The company’s bioreactors should make it easy to adapt the technology to a larger scale, Tobey says; they can simply be lined up in parallel to achieve the needed output volumes. The next two or three years will reveal whether Coskata’s process can start to replace significant amounts of gasoline with cheap ethanol.