1366′s Campaign to Make Better, Cheaper Solar Cells Gets Boost from Department of Energy

2/5/09Follow @wroush

1366 Technologies, the Lexington, MA-based MIT spinoff working on ways to manufacture cheaper, more efficient photovoltaic (PV) cells, said today that the Department of Energy’s Solar America Initiative has selected the company for a performance-based award worth up to $3 million. 1366 president and CEO Frank van Mierlo says the company will get the money in $100,000 chunks, assuming that it can meet specific technical milestones as it attempts to scale up its proprietary process for building PV cells from multicrystalline silicon wafers.

The focus of the Solar America Initiative, administered by the DOE’s National Renewable Energy Laboratory in Golden, CO, is to bring down the cost of electricity from solar installations until it’s competitive with electricity generated by coal-powered plants. There’s a long way to go toward that goal, mainly because it’s so expensive to process silicon wafers, and because their low efficiency at converting sunlight into electric current means it takes acres of cells to produce a usable amount of power.

“To get the SAI grant you must show demonstrable proof, in the form of a prototype or measurable results, that you can take a proof-of-concept and scale it up to a 3-megawatt production facility,” says van Mierlo. “Being in this program is a big deal—only the top 15 percent of applicants are selected.” And the $3 million could make a big difference to 1366, which started out last March with a venture pot of just $12.4 million.

While many photovoltaic startups these days are out to reinvent the basics of PV manufacturing—for example, by shifting to flexible substrate materials such as plastic ribbons, as Marlborough, MA-based Evergreen Solar is doing—1366 has a more modest, and perhaps more achievable, goal. Co-founder and chief technology officer Emanuel “Ely” Sachs (who is a professor of mechanical engineering at MIT and also founded Evergreen) says the company’s goal is to bring the efficiency of multicrystalline PV cells closer to that of monocrystalline cells, the type used in satellites and other high-end applications, which are more efficient but much more expensive to make.

“We’re doing that by addressing the two main limitations on efficiency of multicrystalline cells, which are light trapping and metallization,” Sachs Explains. “Solar panels are blue—which immediately tells you that they are reflecting blue light and not making use of it. What we are doing is building a structure to trap that light in the wafer, so it has a chance to be absorbed and produce electricity. Metallization has to do with the fingers that collect the current from the wafers. The standard way of screen-printing gives you finger 125 microns wide, but in our lab prototype we can make fingers 25 microns wide, which reduces the amount of the cell that’s shaded, immediately giving you more electricity, and also allowing you to put the lines closer together, which leads to other benefits.”

In other words, Sachs is applying some basic mechanical-engineering smarts to the problem. “We’re not trying to invent a new photovoltaic absorber material,” he says. “That’s a critical distinction because it means you can move fast.”

In fact, 1366 hopes to show within 18 months that it can churn out 6-inch-square PV cells fast enough to build a 3-megawatt solar plant, as the NREL award requires. “That forces you to automate critical parts of the process,” says van Mierlo. “You can have an innovative cell architecture and produce it in the lab with tiny cells that are 1-by-1 inch, and show that the physics works. But by the time you do a 3-megawatt plant, you have to have truly solved all the scaling problems and have a handle on the true economics. That’s what the DOE program is about.”

Ultimately, 1366 plans to build and sell a few key pieces of equipment that would be added to standard photovoltaic production lines. But it won’t get the whole $3 million from DOE unless it can show that the scale-up effort is succeeding. “There are approximately 20 to 30 payments, and we have to show NREL our technical results at each step and prove that we did what we said we would do to get the next payment,” says van Mierlo. “And we also have to match some of the public money with private money. It’s a very well-organized program with a lot of accountability. There is no free lunch. In fact, you’d wish to have a similar program for Wall Street.”

Wade Roush is a contributing editor at Xconomy. Follow @wroush

By posting a comment, you agree to our terms and conditions.