Sep 27th 2007 | EMERYVILLE, REDWOOD CITY AND SAN CARLOS, CALIFORNIA
From The Economist print edition
[All emphasis added.]
SOMETIMES you do things simply because you know how to. People have known how to make ethanol since the dawn of civilisation, if not before. [...] Yet such things do not stop ethanol from being a lousy fuel. To solve that, the biotechnologists argue, you need to make a better fuel. [...]
The first step on the road has been butanol. This is also a type of alcohol that can be made by fermenting sugar (though the fermentation is done by a species of bacterium rather than by yeast), and it has some advantages over ethanol. It has more carbon atoms in its molecules (four, instead of two), which means more energy per litre—though it is still only 85% as rich as petrol. It also has a lower tendency to absorb water from the atmosphere.
A joint venture between DuPont, a large American chemical company, and BP, a British energy firm, has worked out how to industrialise the process of making biobutanol, as the chemical is commonly known when it is the product of fermentation. Although BP plans to start selling the stuff in the next few weeks (mixed with petrol, to start with), the truth is that butanol is not all that much better than ethanol. The interesting activity is elsewhere.
One route might be to go for yet-larger (and thus energy-richer) alcohol molecules. Any simple alcohol is composed of a number of carbon and hydrogen atoms (like a hydrocarbon such as petrol) together with a single oxygen atom. In practice, this game of topping up the carbon content to make a better fuel stops with octanol (eight carbon atoms) as anything bigger tends to freeze at temperatures that might be encountered in winter. But living things are familiar with alcohols. Their enzymes are geared up to cope with them. This makes the biotechnologists' task that much easier.
The idea of engineering enzymes to make octanol was what first brought Codexis, a small biotechnology firm based in Redwood City, California, into the field. [...] Codexis controls most of the important patents for what is known as molecular evolution. [...] It creates lots of variations on a theme, throws away the ones it does not want, and shuffles the rest in a process akin to sex. It then repeats the process on the survivors until something useful emerges. [...] The result, according to Codexis's boss, Alan Shaw, is enzymes that can perform chemical transformations unknown in nature. [...]
Codexis's attention [is now focused] on molecules even more chemically similar to petrol. The twist that Codexis brings is that unlike petrol, of which each batch from the refinery is chemically different from the others (because the crude oil from which it is derived is an arbitrary mixture of hydrocarbon molecules), biopetrol could be turned out exactly the same, again and again, and thus designed to have the optimal mixture of properties required of a motor fuel.
Exactly which molecules Codexis is most interested in these days, Dr Shaw is not yet willing to say. But Amyris Biotechnologies, which is also based in California, [...] has been working on a type of isoprenoid (a class of chemicals that include rubber).
Unlike Codexis, which deals in purified enzymes, Amyris employs a technique called synthetic biology, which turns living organisms into chemical reactors by assembling novel biochemical pathways within them. [Amyris founder] Dr Keasling and his colleagues scour the world for suitable enzymes, tweak them to make them work better, then sew the genes for the tweaked enzymes into a bacterium that thus turns out the desired product. [...]
Isoprenoids have the advantage that, like alcohols, they are part of the natural biochemistry of many organisms. Enzymes to handle them are thus easy to come by. They have the additional advantage that some are pure hydrocarbons, like petrol. With a little judicious searching, Amyris thinks it has come up with isoprenoids that have the right characteristics to substitute for petrol.
The third Californian firm in the business, LS9 of San Carlos, [...] also uses synthetic biology, but it has concentrated on controlling the pathways that make fatty acids. Like alcohols, fatty acids are molecules that have lots of hydrogen and carbon atoms, and a small amount of oxygen (in their case two oxygen atoms, rather than one). Plant oils consist of fatty acids combined with glycerol—and these fatty acids (for example, those from palm oil) are the main raw material for the biodiesel already sold today.
LS9 has used its technology to turn microbes into factories for fatty acids containing between eight and 20 carbon atoms—the optimal number for biodiesel. But it also plans to make what it calls “biocrude”. In this case the fatty acids would have 18-30 carbon atoms, and the final stage of the synthetic pathway would clip off the oxygen atoms to create pure hydrocarbons. This biocrude could be fed directly into existing oil refineries, without any need to modify them...