BP Solar Sticks with Silicon

An industry giant says a tweak to silicon manufacturing could beat more exotic materials approaches over the next decade.

It's boom time for solar power, as a rising tide of startups tout various approaches--from organic thin films to concentrating light with holograms--for harvesting energy from the sun. But amid the flurry of nascent technologies, BP Solar, a 30-year-old subsidiary of oil giant BP, is betting that old-fashioned silicon still holds the most potential for cost-effective solar power in the next decade.

In its latest move, the company has developed a solar module--a collection of solar cells--using a new silicon-manufacturing approach that the company says drives down the cost of generating solar power. The new technology boosts power production 8 percent without a price increase, the company says. BP Solar will begin production of these modules by mid 2007.

Technology Review caught up with Lee Edwards, president and CEO of BP Solar, to ask about the new technology and other efforts at the company.

Technology Review: You say your new silicon prototype--which you call Mono²--increases efficiency of your solar cells without increasing the cost. What is Mono²?

Lee Edwards: Over the next 10 years, BP Solar believes that a silicon-based cell technology will continue to drive cost efficiency. This announcement of the Mono² approach to creating the silicon wafer was driven by an acknowledgement of the two different types of silicon available: monocrystalline and multicrystalline. Monocrystalline silicon makes for high efficiency, but it's relatively expensive, and the solar-power industry competes with the microprocessor industry for this type of silicon. Multicrystalline is cheaper, but it is lower quality. Mono², broadly speaking, gives the same electrical-efficiency benefits as monocrystalline wafers but uses a multicrystalline casting approach that is less expensive.

TR: How does it work?

LE: In the traditional multicrystalline manufacturing, you basically put a bunch of rocks in a ceramic crucible, heat it to 1,500 degrees C, let it sit there for a day, and cool it slowly. You get a block of silicon, but the crystal structure is random. Some people say it's more visually appealing because of the way light reflects off it, but each one of those grain boundaries creates a barrier to electron flow. The beauty of our technique is that we've found, in our protected intellectual property, a way to essentially get a single crystal using another approach. The details are proprietary, but it's a combination of metallurgy and the process that allows us to do it.

TR: What's the efficiency or cost-per-watt benefit?

LE: When we say "efficiency," there are two components to it. There are some who love to advertise their cell-conversion efficiency, so you'll see 19 or 20 percent efficiency quoted. That is the amount of sunlight that hits the surface that is converted to electricity. When we talk about efficiency within BP Solar, it is the dollar-per-watt cost to convert sunlight into electricity. The Mono2 module can produce 8 percent more power for the same price as a module made from multicrystalline silicon modules on the market today. This decreases the price per watt.

TR: What other ways are you trying to squeeze extra watts from silicon?

LE: We've got two approaches within BP Solar to do that. One involves grooving the wafer itself with precision lasers that increase the surface area, allowing more sunlight to be converted to electricity. And the other is printing material on the front and the back to create electrical contacts that pull out the electricity without blocking too much sunlight in the process.