X

Searching for cheap solar cells in computer models

A Harvard University-led research project taps IBM computers and idle computers on the Internet to screen candidates for cheap organic solar cell materials.

Martin LaMonica Former Staff writer, CNET News
Martin LaMonica is a senior writer covering green tech and cutting-edge technologies. He joined CNET in 2002 to cover enterprise IT and Web development and was previously executive editor of IT publication InfoWorld.
Martin LaMonica
2 min read

Screening candidates for organic solar cells in computers has yielded a promising molecule that could lead to cheap, flexible solar cells.

Researchers have developed a way to find novel solar cell materials: throw computers at the problem.

In a paper published this week in Nature Communications, the researchers said their method of sifting through millions of possible molecules has yielded a compound that holds promise as a material for organic solar cells.

The Harvard University-led project, which started more than two years ago, is a collaboration with IBM to manage and supply the computing resources for the World Community Grid, where people supply idle PC time to contribute to research projects lacking sufficient compute resources.

Traditional solar cells are made from silicon and several companies are making thin-film solars with alternate materials. The goal of this research project is to find organic molecules that can act as a semiconductor in a solar cell. These organic solar cells won't be as efficient at converting sun to light, but would be flexible and potentially cheaper since they would not require extensive processing or expensive materials.

Alan Aspuru-Guzik, a theoretical chemist at Harvard, and his colleagues have created a computer model that has screened 3.5 million molecules for promising characteristics. Compounds that meet the right criteria are passed to researchers at Stanford University to synthesize the molecules and test the properties, according to an article in Nature News.

"It's how the pharmaceutical people do it: the theorists give a ranking to the experimentalists," Aspuru-Guzik told Nature News. "We're trying to save experimental time."

In its paper, the researchers said its computer-based screening process allowed it to locate a molecule that had electrical characteristics that few organic semiconductors to date have shown. "The study suggests that a computational screening approach can lead to the informed synthesis and characterization of novel organic materials for electronics applications," it concludes. The researchers expect to publish the 100 most-promising molecules, according to IBM.

The computing power for the screening is coming in part from 1.8 million PCs owned by 600,000 volunteers around the world. IBM said that its World Community Grid has already sped up research for AIDS drugs, specific grains of rice, and water-filtering techniques.