Less than a year from now, a giant space that was once a distribution center for Netflix in Sunnyvale, Calif., could be filled with solar machines that will crank out solar cells from Kleiner-backed Alta Devices. Alta’s CEO, Christopher Norris, recently gave us an exclusive tour of the new, 100,000-square-foot space that is supposed to house the company’s pilot production line to make solar cells from gallium-arsenide in 2013.
The gallium-arsenide compound is more efficient at converting sunlight into electricity than the more common semiconductors found in today’s solar panels, such as silicon, cadmium-telluride or copper-indium-gallium-arsenide (CIGS). But gallium-arsenide also is much more expensive and commonly used sparingly as one of several ingredients for solar cells.
Alta’s simulation and field data show that its solar cells can produce two to three times more electricity over a one-year period than those available on the market today, Norris said. The company broke the world record for gallium-arsenide solar cells last year with a 28.2 percent cell last year.
Experimental cells like that are typically made in labs and demonstrate what a company might be able to achieve after it starts to mass-produce them. Whether a company will ever do so will depend on factors such as its ability to raise money and operate factory equipment efficiently, and these hurdles have tripped up many venture-backed CIGS startups.
Earlier this month, Alta announced it was able to produce a solar panel with 23.5 percent efficiency. When solar cells are selected and assembled into a panel, there tends to be a small loss of efficiency. That’s why cell efficiency figures tend to be higher than panel efficiencies. The most efficient solar panels sold today are the 20 percent panels launched by SunPower last year.
While gallium-arsenide cells are more efficient and should fetch higher prices, they won’t attract many buyers if they are too expensive and require project developers to make a substantial upfront investment.
Alta’s secret sauce
Alta’s technology for cutting production costs focuses on creating ultra thin, 1-micron gallium-arsenide solar cells that are grown on re-usable gallium-arsenide wafers, Norris said. Using the same compound for the cell and wafer ensures that the crystalline structure of the two match, which minimizes defects and leads to more efficient cells than using materials with mismatched crystalline structures.
To slice off that top layer without damaging it, Alta creates an intermediate layer between the two, using aluminum arsenide. Aluminum arsenide has virtually the same crystalline structure as gallium-arsenide, making it a good substitute. Alta harvests that top layer to make solar cells by etching away the intermediate, aluminum-arsenide layer, Norris said. After that, the wafer at the bottom is re-used to grow more cells. The wafers, at $100 each, are good to be used repeatedly as long as they don’t break.
Many research labs know how to grow and slice off the top layer in tiny pieces. “The trick is to do it on large pieces and make it possible to do it at scale,” Norris said of Alta’s challenge to commercialize the technology.
Alta wants to make cells and sell them to customers who will use them for making either conventional solar panels or other devices such as roofing materials. In fact, Alta is working with one of its investors, Dow Chemical, on a federally-funded project to embed solar cells in building materials. Dow already is selling solar shingles containing CIGS solar cells.
The startup has raised about $120 million since its inception in 2007, and it’s building the pilot line that will be able to produce a few megawatts of solar cells per year, Norris said. With an undisclosed equipment supplier, Alta has designed three key pieces of equipment for slicing off the top gallium-arsenide layer for making solar cells. The rest of the factory tools will be more or less off-the-shelf. Commercial shipment should take place by the end of 2013.
Check out photos from our factory tour: