TEXT OF STORY
KAI RYSSDAL: Oil settled above $94 a barrel today. A government report on lower inventories set speculators buying. When we talk about oil it's usually a variety called light, sweet crude. It's the easiest to refine into things consumers can use. Down on the other end of the scale, heavy oil reserves are estimated to be in the trillions of barrels worldwide. Getting that black gold out is a huge and expensive problem because basically you have to melt it first. But researchers say they might have discovered a cheaper way.
From the Marketplace Innovations Desk at North Carolina Public Radio, Janet Babin reports.
JANET BABIN: No one really knows how much oil is left on Earth. But experts do know that at least half of the oil already discovered is trapped in the ground. This heavy oil, as it's called, is what's left over after companies take out conventional oil. But scientists think they've found a way to get at this stuff using bacteria.
Jennifer Adams: Micro-organisms actually eat oil, and in the process of eating it they generate methane.
Jennifer Adams at the University of Calgary worked on the research. She says it's possible to accelerate bacteria's natural processes, capture the methane they generate, and then turn it into energy.
Adams says the method could be used near oil fields that require a lot of energy to exploit, like Canada's oil sands.
These are tar-like petroleum deposits, buried in sand. Tom Wallen with Energy Intelligence says traditional oil production's dropped in Canada. But the research could fuel a surge in alternative oil production.
Tom Wallen: Tar-sands production is coming on so fast that it's more than completely displacing the decline in the old fields, and Canadian exports to the U.S. are growing significantly.
Wallen says the research could be a huge boon, even if it only increases the amount of oil extracted by 2 or 3 percent.
Fadel Gheit with the Oppenheimer and Company brokerage firm is skeptical, though. He's seen discoveries come and go:
Fadel Gheit: Things that work in the lab ... not necessarily guaranteed to work commercially.
Researchers says the process is about five years from commercialization.
I'm Janet Babin for Marketplace.