Solar energy offers a vast supply of power, but harnessing it is a challenge

By Brian Palmer
Tuesday, June 22, 2010; HE03

We have a solar-based economy, whether or not we realize it. Ninety-four percent of the world's energy comes from the sun, even energy that doesn't at first glance seem solar. Coal, oil and natural gas are mostly the products of ancient plants that grew with the sun's help. The sun drives hydroelectric power by evaporating low-lying water, then dumping it at higher altitudes. Windmills turn because the sun warms the planet's air unevenly.

Fortunately, there's plenty of sun to go around. Our local star is continuously transmitting 180 quadrillion watts of energy to the Earth, 14,000 times our requirements for generating power. So the question isn't where to get our energy, but how to capture it.

Solar cells, also known as photovoltaic cells, are our most identifiable effort to convert the sun's energy into electricity. They depend on a phenomenon known as the photovoltaic effect, discovered in 1839 by a French teenager. Alexandre Edmond Becquerel, then 19, placed two metal plates in a salt solution and generated an electric current by simply placing his rig in the sun.

Sixty-six years later, Albert Einstein demonstrated the physics behind Becquerel's electric soup, that it worked because sunlight provided enough energy to move some electrons through the solution, creating a current. Einstein won the 1921 Nobel Prize for that explanation.

Ever since, engineers have been working to make the conversion of sunlight to usable energy more efficient.

Today's commercial solar cells consist of a layer of silicon mixed with boron, which faces the sun, stacked on top of a layer of silicon mixed with phosphorous. The silicon-phosphorous layer is known as the negative, or n-type layer, because it has lots of spare electrons. The silicon-boron molecules, in contrast, have a gaping hole in their electron layer, yearning to be filled.

Sounds like a perfect match, right? The free electrons on the bottom could just flow to the empty spaces on the top, and all the molecules would live happily ever after. But we can't let that happen. We need to keep some of those holes empty, or the electrons won't have anywhere to go and they'll stop flowing. The imbalance between the two layers creates the voltage we need to power our smoke detectors, MP3 players and hair dryers.

Fortunately, something special happens where the two layers meet that prevents too many electrons from escaping the crowded n-type layer. Have you ever inflated a pool float? The rubber valve lets you blow air in, but it won't let air escape, no matter how crowded the air molecules get inside the float. The same sort of thing happens in a solar cell. The initial exchange of electrons creates a charged electric field between the layers, in an area called the P-N junction. Like a rubber inflation valve, the electric field is a one-way street. Electrons can cross from the spacious boron-silicon (or p-type) layer to the crowded n-type layer, but not the other way. So those electrons remain stuffed in the n-type layer.

Just as you can pump more air into the pool float, the sun's energy forces more electrons into the already crowded n-type layer. The electrons find this overcrowding intolerable and look for a way out. Engineers have been kind enough to provide it, but it will cost them. An electrical wire located on the far end of the n-type layer gathers the suffocating electrons and shepherds them back across. Along the way, we take a little bit of energy from the grateful electrons to power our gizmos.

Simple enough. So why aren't we ready to kick the fossil fuel habit and go 100 percent photovoltaic? Building an efficient solar cell presents a number of challenges, and we're still working to improve photovoltaic efficiency. Most commercially available systems convert only about 15 percent of the sun's energy into electricity.

Beyond that, timing and costs are problems. We have to store surplus energy collected during the day or we won't be able to watch late-night television. The problem is that high-capacity batteries are really expensive. "The battery adds about $1 per kilowatt-hour to the cost of solar energy," according to solar pioneer Allen Barnett at the University of Delaware. Americans pay only about 12 cents per kilowatt-hour for electricity bought from the utility company. So if you put solar panels on your house and can't sell your surplus energy back to the utility company, solar energy can be cost-prohibitive.

Despite these challenges, there is great promise in solar laboratories. Researchers are producing far more efficient cells than you can buy on the market. Barnett's laboratory just produced a prototype that converted more than 36 percent of its solar energy into electricity, a record for a photovoltaic cell that is usable by a consumer.

Some researchers believe cost, not efficiency, is the real barrier to a solar-powered America. "Silicon has to be processed to a high level of purity. It's an expensive process," says Alan Sellinger, who directs the Center for Advanced Molecular Photovoltaics at Stanford University. In 2000, three scientists won the Nobel Prize in Chemistry for showing that plastic can conduct electricity under certain circumstances. Sellinger's lab is exploiting that discovery to make what he hopes will be inexpensive solar cells.

"We're talking about photovoltaic cells that can be printed onto any backing using an ink-jet printer," says Sellinger, who also imagines other advances: Stained-glass windows will collect a wide range of wavelengths of light. The shaded blue stripe across of the top of your windshield will generate electricity. Soldiers' electronics will be driven by solar-collecting clothing.

Many solar scientists also think silicon is yesterday's news. Sandra Rosenthal, a professor of chemistry at Vanderbilt University, works with ultra-thin layers of cadmium selenide. She has shown that, when you reduce the material to a thickness of only a couple of nanometers -- that's about 200 atoms -- you can capture a much wider spectrum of the sun's energy, which has the potential to create dramatically more efficient solar cells.

Despite the fast-moving research, Barnett believes there's no point in waiting: "It makes sense to buy right now. The current technology is cost-effective, especially with the tax incentives available in the Washington area." So if you've been thinking about making the shift, go buy your roof panels. The solar windows, car and clothes are on their way.

Palmer, a freelance writer living in New York, is a regular contributor to's Explainer column.

© 2010 The Washington Post Company