Solar cells are improving

Every minute, the sun bombards Earth with enough energy to supply its power needs for a year. Yet only two one-hundredths of a percent of all the electricity fed into the U.S. grid originates from sunlight.

By Matthew Fordahl
AP technology writer Deseret News

The world still largely relies on diminishing supplies of environmentally unfriendly and politically destabilizing fossil fuels. Despite decades of research, it’s still cheaper to burn coal than get power from the sun.

But photovoltaic technology is improving efficiency and lowering costs for solar power, and experts believe the development will in the next few years drive solar adoption far faster than any government incentives or environmental concerns.

"Today, if solar energy were available in a quantity and at a cost comparable to fossil fuel, it would be a revolutionary change," said Stephen Empedocles, business development director at Nanosys Inc. The Palo Alto-based startup is working on photovoltaic cells so small and cheap that they can be sprayed or even painted onto surfaces.

Solar power research is proceeding on two fronts: Making cheaper versions of crystalline silicon cells that comprise 80 percent of the solar market, and creating less expensive photovoltaic technologies with the reliability and efficiency of crystalline silicon.

All the research is built on semiconductors, which absorb the sun’s photons. Electrons are bumped along a predictable path by those photons until the current flows into something useful, such as an appliance.

This photovoltaic effect, first noticed in 1839, is the same way the flow of current is manipulated in computer logic and memory chips.

In fact, money-saving advances in the chip industry, such as increasing the size of silicon wafers and perfecting how they are sawed from ingots, are now being exploited in photovoltaics, said Dan Shugar, president of PowerLight Solar Electric Systems.

Since the 1950s, the uses for solar power have exploded. It’s no longer limited to such specialty uses as satellites, remote cabins and highway call boxes. "Now it’s cost effective against grid power for homes and businesses in states that have the incentives," Shugar said.

Yet manufacturing costs remain relatively high, making solar power as much as five times more expensive than energy from fossil fuels — before tax credits and other incentives that states such as California offer.

To lower costs, companies and researchers are trying different types of semiconducting materials, including amorphous silicon and gallium arsenide. These materials are also more flexible than crystalline silicon. Rather than being grown as crystals in high-vacuum chambers, they can be deposited in flexible thin layers, making bulk production cheaper.

But there’s a catch: Such photovoltaics harness about 8 percent of the total energy in sunlight, compared with about 15 percent for crystalline silicon.

"You basically get half as much electricity from the same amount of sunlight," said Thomas Surek, program manager at the Department of Energy’s National Renewable Energy Laboratory in Golden, Colo.

Ways to boost efficiency include layering photovoltaic cells so they capture a wider portion of the sun’s range of energy. Another involves using lenses and reflectors to concentrate more sunlight on each cell.

These techniques do raise the overall cost, but the Boeing Co. subsidiary Spectrolab, which makes solar power systems for satellites, has used them to reach a record efficiency of more than 34 percent — in the lab. Researchers found sunlight can be concentrated as much as 500 times.

The most futuristic approach involves arranging nanosize semiconductors in a matrix of plastic-like materials that are expected to be much less expensive to produce.

Nanosys is working on nanorods that are just 7 nanometers by 60 nanometers in a polymer. Because of their size — a nanometer is about 10,000 times narrower than a human hair — nanorods are arranged by chemical reactions.

"That is an interesting technology to watch but today it is not a viable technology," said Nasser Karam, director of advanced programs and optoelectronics for Spectrolab. "I would suspect with a lot of government funding, the feasibility will probably show in three to five years."

Though some are skeptical about the future technologies, companies that sell today’s solar-power systems note that traditional photovoltaics are dropping in price and eventually pay for themselves even at today’s costs. PowerLight even touts some of its systems as added protection for roofs.

"Those other technologies are really interesting and I hope they succeed, but we don’t need those to succeed for us to succeed," Shugar said. "We’re not betting our future on one technology horse."

http://deseretnews.com/dn/view/0,1249,480035308,00.html