The Bottomless Well?

Harvard chemist and Nobel laureate Dudley Herschbach describes the common recipe for oil this way: "You cook dinosaurs and swamp ferns and God knows what for a very, very long time at modest temperatures and pressures near the Earth's surface," and out comes oil. This is the biological theory, but not the only one. A Russian theory dating to the 1860s gained cult status after Thomas Gold's 1999 book "The Deep Hot Biosphere." It holds that oil is the product not of biological decay but of chemical reactions sped up by the intense pressures and temperatures deep inside the Earth. Gold thought it was impossible to test this idea--which to its devotees raises the hope of endlessly regenerating oilfields--in a lab. But using a modern diamond anvil that can re-create the conditions of inner Earth, a team at the Carnegie Institution in Washington, encouraged by Herschbach, did something close. In a 2004 study they produced methane, a lower form of hydrocarbon than oil, and are now pushing the research forward. Herschbach spoke to NEWSWEEK's Karla Bruning. Excerpts:

HERSCHBACH: It's been a minority view that oil could be largely nonbiological in origin, even in Russia. The Western geologists long ago decided oil was a fossil fuel, partly because it contains remnants of the kinds of molecules that you really think come only from living matter, as well as enrichment of the light isotope of carbon like that which occurs in living matter, which have membranes through which light isotopes diffuse faster. Gold said the evidence of biological material could be from microbes feeding on the oil for the last few miles. That's a cogent argument. As we know, microbes love oil: at every major spill, bacteria eat up more of it than anticipated. He also points out that if hydrocarbons have diffused a long way through rock pores, that could also result in enriching the lighter isotope. So I think that is an open question.

Our experiment offered evidence that you can make methane at pressures and temperatures that correspond to deep Earth--50 to 100 miles or more. Of course, that tells you nothing about whether that's a major way to account for methane, or oil, or how long it might take this petroleum to migrate up to reservoirs. So I would emphasize we wouldn't want to claim too much from this experiment. It's suggestive, but much more needs to be done.

It seems unlikely that we can drill deeply enough to reach large quantities of nonbiological petroleum. If anybody wanted to launch a company on the basis of pumping nonbiological hydrocarbon fuels, I wouldn't invest. Gold persuaded the Swedes to drill a six-mile-deep hole in just solid, igneous rock, not the sedimentary rock in which oil is normally found, and they found very, very little methane. So that kind of disillusioned people. Gold didn't feel that maybe they'd gone deep enough. But I think it's hard to drill much more than six miles.

For all we know, a lot of oil that's been found already was abiological in its ultimate origin. If Gold is right, it just had bubbled up over time, and now it's found. If analysis of that oil found less evidence of biological material than in oil from other sources, then that might lend credence to its abiological origin.

The Western geologists, I think, are just totally on the side of thinking that it's a fossil fuel. I don't think they pay much attention to the Russian school, or part of the Russian school--it's probably not a uniform view in Russia either. It's curious. Gold had a reputation of being an iconoclastic-type guy, known for being contrary to majority ideas, so you have to take that into account. Personalities matter in politics, science and everything else.

We now have some funding, and we're going to continue this study. I'm hoping that it's only a matter of two, three years before we have further evidence.

That's sheer guesswork. It may be that you have to go to pressures and temperatures 500 miles down in the Earth, and there's no prospect of drilling. It may take 1,000 years to percolate to the surface. I would not be surprised if that's the actual outcome. It would be charming if, after our generation has burned up all the oil, 1,000 years from now our successors, if there are any, say, "Gee, all of a sudden we have all this petroleum again. Now we can have cars again. It's nice that there are a few left in the museums." But if that's the case, it's not a practical use for humanity in the foreseeable future.