3/3/2014 Meredith Staub
Written by Meredith Staub
"Solar cells have existed in some form or other for more than a hundred years now," MechSE assistant professor Elif Ertekin said. "But they still haven’t been optimized; their performance is still not as good as it could be. So we’re trying to determine what we have to do at the materials level to manufacture devices that are better at converting sunlight to electricity."
Ertekin's research is in computational modeling and materials design, with interests that include thermoelectrics and photocatalysts as well as photovoltaics. Although her work is entirely computational, she works closely with the experimentalists who synthesize the materials and measure their properties.
"It’s a pretty exciting time for computational work," Ertekin said. "Systems that our group can model on a computer can now be grown almost exactly—down to the atomic scale—in a laboratory."
With the average efficiency of a conventional solar cell, it takes one to two years to generate the same amount of energy needed to make the silicon the cell is made of. In order to compete with other forms of energy production, the efficiency needs to be increased significantly. One potential way to do this is to "hyperdope" the silicon with impurities. Hyperdoping is giving the silicon impurities at concentrations several orders of magnitude higher than the solid solubility limit. This process creates what is called “black silicon,” a material Ertekin and her group are studying that would be capable of absorbing sunlight in the low-energy portion of the solar spectrum, and is a good model system to explore how to make silicon a better absorber of sunlight.
There are also other factors to consider, such as manufacturing costs. In order to augment the efficiency and decrease the costs, Ertekin and her group are looking at alternatives to the bulk silicon that most modern solar cells use as the active layer.
"We want a low-cost material that is still good at converting sunlight to electricity even when it's been manufactured at low temperatures, even if it's not electronic-grade pure," Ertekin said. "So the key is to figure out what it takes for a material to be defect-tolerant."
Researchers around the world have taken interest in thin-film semiconducting materials, creating what are known as "thin film solar cells." The amount of material required to make these cells is significantly less than that of bulk silicon solar cells, although their efficiencies still lag behind silicon by several percentage points. The two most popular thin film materials are cadmium telluride (CdTE) and copper indium gallium selenide (CIGS). CdTE is currently the most cost-effective, while CIGS has the highest efficiency of the thin-film materials. Ertekin and her group are focusing on another thin-film material: copper zinc tin sulfide (CZTS). Because it is made from earth-abundant elements, CZTS offers a significant advantage over many thin-film candidates; the raw material for CZTS is about five times cheaper than that of CIGS.
Most importantly, efficiencies with CZTS are already up to about 12%. The best semiconductor solar cells are about 27-29% efficient out of a laboratory, and 21-23% efficient off of a manufacturing line or industrial process. Thin film modules can range from 10% to a little under 20%. Compared to these efficiencies, 12% for a CZTS cell doesn’t seem entirely impressive—but considering the age of the technology relative to that of other solar cell materials, it’s actually quite surprising.
"It’s a very early-stage material," Ertekin said. "The research community has only been working on CZTS for a handful of years. The fact that it’s as high as 12% so early in the game makes you think there’s still a lot of room for improvement in this particular material.
"And I think that as photovoltaic technology continues to improve, as it gets more convenient and more reliable, we’ll see more and more adoption here in this country. It’s an exciting time for the field right now."