A Molecular Screening Process for Improving Solar Cells: IBM Joins in

By Michael Kanellos

Developing new solar technologies isn’t that different from finding new drugs, say IBM and Harvard.

The two are forming a new World Community Grid project to find new molecules for harvesting power from the sun. Under the project, IBM will offer up computing cycles on its cloud computer to researchers to test out the efficacy of various molecules.

This, in a nutshell, is how drugs are developed. Researchers examine billions of different combinations of proteins to see how they interact. Running the mind-boggling number of combinations takes repetitive, brute-force computing power. The same is true when it comes to materials for organic solar. A lot of elements and materials can harvest energy from the sun, but the trick is finding the exact combination that will do it efficiently. The scope of the project would take 22 years on an ordinary grid. It might take only two years on the cloud as proposed under this project.

Of course, even if they find a solution, there is no guarantee that organic solar cells will work, or establish a large market. Reliability, durability and cost have been some of the problems associated with photovoltaic dyes. Konarka has spent tens of millions of VC funds and spent several years trying to get organic solar cells to mass production and mass acceptance. It’s still on the fringe. It may turn out that thin film technologies actually work better. But who knows? This cloud project will give researchers a huge catalog of information on how molecules react with the sun.