The Price of Biofuels

Do we really have any alternative to biofuels?

Superbugs
Since the oil crisis of the 1970s, when the price of a barrel of petroleum peaked, chemical and biological engineers have chased after ways to turn the nation's vast reserves of "cellulosic" material such as wood, agricultural residues, and perennial grasses into ethanol and other biofuels. Last year, citing another of President Bush's goals--reducing U.S. gasoline consumption by 20 percent in 10 years--the U.S. Department of Energy (DOE) announced up to $385 million in funding for six "biorefinery" projects that will use various technologies to produce ethanol from biomass ranging from wood chips to switchgrass.

According to a 2005 report by the DOE and the U.S. Department of Agriculture, the country has enough available forest and agricultural land to produce 1.3 billion tons of biomass that could go toward biofuels. Beyond providing a vast supply of cheap feedstock, cellulosic biomass could greatly increase the energy and environmental benefits of biofuels. It takes far less energy to grow cellulosic materials than to grow corn, and portions of the biomass can be used to help power the production process. (The sugarcane-based ethanol produced in Brazil also offers improvements over corn-based ethanol, thanks to the crop's large yields and high sugar content.)

But despite years of research and recent investment in scaling up production processes, no commercial facility yet makes cellulosic ethanol. The economic explanation is simple: it costs far too much to build such a facility. Cellulose, a long-chain polysaccharide that makes up much of the mass of woody plants and crop residues such as cornstalks, is difficult--and thus expensive--to break down.

Several technologies for producing cellulosic ethanol do exist. The cellulose can be heated at high pressure in the presence of oxygen to form synthesis gas, a mixture of carbon monoxide and hydrogen that is readily turned into ethanol and other fuels. Alternatively, industrial enzymes can break the cellulose down into sugars. The sugars then feed fermentation reactors in which microörganisms produce ethanol. But all these processes are still far too expensive to use commercially.

Even advocates of cellulosic ethanol put the capital costs of constructing a manufacturing plant at more than twice those for a corn-based facility, and other estimates range from three times the cost to five. "You can make cellulosic ethanol today, but at a price that is far from perfect," says Christopher Somerville, a plant biologist at the University of California, Berkeley, who studies how cellulose is formed and used in the cell walls of plants.

"Cellulose has physical and chemical properties that make it difficult to access and difficult to break down," explains Caltech's Arnold, who has worked on and off on the biological approach to producing cellulosic ethanol since the 1970s. For one thing, cellulose fibers are held together by a substance called lignin, which is "a bit like asphalt," Arnold says. Once the lignin is removed, the cellulose can be broken down by enzymes, but they are expensive, and existing enzymes are not ideal for the task.

Many researchers believe that the most promising way to make cellulosic biofuels economically competitive involves the creation--or the discovery--of "superbugs," microörganisms that can break down cellulose to sugars and then ferment those sugars into ethanol. The idea is to take what is now a multistep process requiring the addition of costly enzymes and turn it into a simple, one-step process, referred to in the industry as consolidated bioprocessing. According to Lee Lynd, a professor of engineering at Dartmouth College and cofounder of Mascoma, a company based in Cambridge, MA, that is commercializing a version of the technology, the consolidated approach could eventually produce ethanol at 70 cents a gallon. "It would be a transformational breakthrough," he says. "There's no doubt it would be attractive."

But finding superbugs has proved difficult. For decades, scientists have known of bacteria that can degrade cellulose and also produce some ethanol. Yet none can do the job quickly and efficiently enough to be useful for large-scale manufacturing.

Nature, Arnold explains, offers little help. "There are some organisms that break down cellulose," she says, "but the problem is that they don't make fuels, so that doesn't do you much good." An alternative, she says, is to genetically modify E. coli and yeast so that they secrete enzymes that degrade cellulose. But while many different kinds of enzymes could do the job, "most them don't like to be inserted into E. coli and yeast."

Arnold, however, is optimistic that the right organism will be discovered. "You never know what will happen tomorrow," she says, "whether it's done using synthetic biology or someone just scrapes one off the bottom of their shoe."

She didn't quite scrape it off her shoe, but Susan Leschine, a microbiologist at the University of Massachusetts, Amherst, believes she just might have stumbled on a bug that will do the job. She found it in a soil sample collected more than a decade ago from the woods surrounding the Quabbin Reservoir, about 15 miles from her lab. The Quabbin sample was just one of many from around the world that Leschine was studying, so it was several years before she finished analyzing it. But when she did, she realized that one of its bacteria, Clostridium phytofermentans, had extraordinary properties. "It decomposes nearly all the components of the plant, and it forms ethanol as the main product," she says. "It produces prodigious amounts of ethanol."

Leschine founded a company in Amherst, ­SunEthanol, that will attempt to scale up ethanol production using the bacterium. There's "a long way to go," she acknowledges, but she adds that "what we have is very different, and that gives us a leg up. We already have a microbe and have demonstrated it on real feedstocks." Leschine says that other useful microbes are probably waiting to be discovered: a single soil sample, after all, contains hundred of thousands of varieties. "In this zoo of microbes," she says, "we can think that there are others with similar properties out there."

Blooming Prairies
Whether ethanol made from cellulosic biomass is good or bad for the environment, however, depends on what kind of biomass it is and how it is grown.

In his office in St. Paul, David Tilman, a professor of ecology at the University of Minnesota, pulls out a large aerial photo of a field sectioned into a neat grid. Even from the camera's vantage point far above the ground, the land looks poor. In one plot are thin rows of grasses, the sandy soil visible beneath. Tilman says the land was so infertile that agricultural use of it had been abandoned. Then he and his colleagues scraped off any remaining topsoil. "No farmer has land this bad," he says.

In a series of tests, Tilman grew a mixture of native prairie grasses (including switchgrass) in some of the field's plots and single species in others. The results show that a diverse mix of grasses, even grown in extremely infertile soil, "could be a valuable source of biofuels," he says. "You could make more ethanol from an acre [of the mixed grasses] than you could from an acre of corn." Better yet, in a paper published in Science, Tilman showed that the prairie grasses could be used to make ethanol that is "carbon negative": the grasses might consume and store more carbon dioxide than is released by producing and burning the fuel made from them.

The findings are striking because they suggest an environmentally beneficial way to produce massive amounts of biofuels without competing with food crops. By 2050, according to Tilman, the world will need a billion hectares more land for food. "That's the land mass of the entire United States just to feed the world," he says. "If you did a lot of biofuels on [arable] land--it is very easy to envision a billion hectares for biofuels--you will have no nature left and no reserve of land after 50 years." Instead, ­Tilman argues, it makes sense to grow biomass for fuels on relatively infertile land no longer used for agriculture.

But down the hill from Tilman's office, his colleagues in the applied-economics department worry about the practical issues involved in using large amounts of biomass to make fuel. For one thing, they point out, the technology and infrastructure that could efficiently handle and transport the bulky biomass still need to be developed. And since the plant material will be expensive to move around, biofuel production facilities will have to be built close to the sources of feedstock--probably within 50 miles.

The amount of biomass needed to feed even one medium-size ethanol facility is daunting. Eidman calculates that a facility producing 50 million gallons per year would require a truck loaded with biomass to arrive every six minutes around the clock. What's more, he says, the feedstock is "not free": it will cost around $60 to $70 a ton, or about 75 cents per gallon of ethanol. "That's where a lot of people get fooled," he adds.

Since no commercial cellulosic facility has been built, says ­Eidman, it is difficult to analyze the specific costs of various technologies. Overall, he suggests, the economics look "interesting"--but cellulosic ethanol will have to compete with corn-derived biofuels and get down to something like $1.50 a gallon. Eidman believes it will be at least 2015 before biofuels made from cellulose "are much of a factor" in the market.