Hot dry rock found deep underground is one of the most abundant potential sources of clean energy. Drilling holes into the rock, fracturing it, and pumping water through it to extract the heat, and then using that heat to generate electricity, could supply the world's energy needs many times over according to an MIT report. In practice, however, harvesting that energy is proving a challenge, in part because developers have located the first projects in earthquake prone regions.

In 2007 a project in Switzerland that was drilling five-kilometer-deep holes to access hot dry rock had to shut down because it set off a series of small earthquakes. Earlier this year the New York Times called into question a similar project in Northern California. The Department of Energy has decided to review the project, and may not allow it to go forward.

Meanwhile, the New York Times reports today that the project has been delayed because a bit has failed to penetrate a layer of rock close to the surface. In the amount of time it should take workers to drill a hole 12,000 feet deep, they have only reached a depth of 4,000 feet.

These aren't the first setbacks for this approach to harvesting energy, which is known as enhanced geothermal energy. Here's what we wrote in a 2006 report on the Switzerland project.

Despite its simplicity, this concept has failed several times. In the 1970s, a pioneering project initiated by Los Alamos National Laboratory demonstrated that one could fracture rock and circulate brine to extract heat. But that project could never get enough brine in -- and therefore enough heat out -- to make the process competitive with conventional power plants burning fossil fuels such as coal or natural gas. Gunnar Grecksch, a geophysicist and hot-rock fracturing expert at the Leibniz Institute for Applied Geosciences in Hanover, Germany, says follow-on efforts in the U.K. and Japan failed for the same reason: the fracturing of the rocks was never sufficient. "Flow resistance is still the key problem," he says. "In none of these projects were the flow rates in the range you need for a commercial system."