Sugar supply: A worker at Renmatix pumps sugar produced by the company’s new supercritical water process into a storage container.

However, working with supercritical water comes with challenges. The materials that can be used with supercritical water are limited—it will dissolve glass, for example. The extremely fast reactions also make it difficult to ensure that the chemistry doesn’t go too far and produce undesirable by-products. In past attempts, the supercritical water has dehydrated some of the sugar produced, resulting in compounds that can poison the yeast used to convert sugar to ethanol. Typically, the process also yields a relatively small amount of sugar from a given amount of biomass.

Fred Moesler, Renmatix’s vice president of process technology, says the company has overcome these problems. The company hasn’t said how it does this, but Gary Aurand, a research scientist at the University of Iowa who is familiar with the company from its early days (when it was known as Sriya Innovations), suggests the company may be using supercritical water in only part of its process.

Turning biomass into sugar using supercritical water involves first grinding biomass into small particles, then dissolving cellulose in water. Without dissolving it, only the cellulose molecules at the surface of the particles will be broken down, resulting in low sugar production. After the cellulose is dissolved, further exposure to high temperatures and pressure will break the cellulose molecules down into sugar.

Aurand says that water only needs to be supercritical for the dissolving step. If Renmatix could engineer a system to move the dissolved material into an area of lower temperature and pressure, it could slow down the process of breaking down the cellulose into sugar, preventing the formation of the unwanted compounds.

All Renmatix has said is that it uses two steps to break down cellulose and a similar material, hemicellulose. Breaking down cellulose produces glucose, the sugar that yeast can readily use to produce ethanol. Breaking down hemicelluloses produces another sugar called xylose, which doesn’t work with conventional fermentation, but which can be used in some advanced biofuels and biochemicals processes. The economics of the process will depend on the market for xylose.

Renmatix has raised some of the money for a plant capable of producing 100,000 tons of sugar per year—large enough to show that the process has commercial potential, it says. But the company is still working to obtain the loans needed to go forward. In the past, using supercritical water to process biomass has been seen as uneconomical, so it may prove difficult to get banks to sign on. “Little is known about the technology,” says Andy Aden, manager of biorefinery analysis at the National Renewable Energy Laboratory in Golden, Colorado. Based on prior calculations, he says, “it is likely to be expensive.”