Answer Geek: Photovoltaics

-- Q U E S T I O N: You're on! Let's hear about the guts and volts of photovoltaic technology.

— Guy C.

Q U E S T I O N: Please tell us about photovoltaics. Could they be used as mass power generators? Love your columns. I learn so much.

— Julia S.

A N S W E R: Thanks, Julia. Love your question, too. And the compliments. Your question, too, Guy. "Guts and volts." That's pretty good.

Thanks to everyone else who wrote asking for about how photovoltaic cells, for that matter. For those of you who never quite got around to reading last week's column on solar energy, I mentioned in passing that the inner workings of the photovoltaic cell were interesting enough to merit a separate column. After receiving dozens of e-mails expressing interest, it looks like I'll have to make good on my threat.

So, just to bring everybody up to speed, a photovoltaic cell is a device that converts sunlight into electricity. It's a neat trick. And a useful one. You probably have one on your desk. Got a calculator? The kind that just needs a little light to work? The power to run that handy little device comes from a photovoltaic cell.

Although it may seem like cutting-edge technology, the fact that light could be harnessed to generate an electrical current is an old idea that was discovered in 1839 by a French physicist named Edmund Becquerel, who noticed that voltage was produced when light shone on a beaker full of conducting solution that held a pair of electrodes.

At First, Just a Little Juice

He called it the photovoltaic effect. Later in the 19th century, photovoltaic cells were built from selenium that were 1 to 2 percent efficient. In other words, those cells managed to convert about 2 percent of the available solar radiation into current. That's not a lot of juice, but it did have its practical benefit — even today, many light sensors in cameras use selenium photovoltaic cells.

It wasn't until the middle of the last century that a photovoltaic cell was produced that created enough current to actually serve as a power source. In 1954, engineers at Bell Laboratories built a photovoltaic cell from silicon that was 4 percent efficient, and they upped that number to 11 percent before long.

In 1958, a U.S. satellite was launched that included a radio powered by a photovoltaic cell, and photovoltaic cells have since played a major role in space exploration, providing power for all sorts of systems. For that matter, NASA has played a major role in the development of photovoltaic technology.

There was also a brief period during the mid-1970s when the U.S. Department of Energy poured money into solar energy and photovoltaic research. Remember the oil embargo?

Other than that, in the United States, at least, it is only in the last few years that photovoltaic solar power has been considered for anything but highly specialized niche applications, like space exploration.

How They Work

But enough of the history. You wanted to know how they work.

A typical photovoltaic cell is 4 inches square. The top layer is a glass cover with a non-reflective coating. Underneath that is something called a contact grid, which allows the current to flow to another contact layer at the back of the cell. In between these two layers are a pair of semiconductors, usually made of silicon and about 0.01 of an inch thick. It's in the two silicon semiconductor layers that the interesting stuff happens.

First let's start with the photovoltaic effect. When sunlight — which consists of photons — strikes a photovoltaic cell, one of three things can happen: The photons can be absorbed, reflected, or pass right through. When a photon is absorbed, it strikes a silicon atom and the photon's energy knocks an electron out of orbit, leaving a hole. If you get enough of these free electrons, and can get them to flow in the same direction, you've got yourself and an electrical current.

The trick is to get those electrons flowing. An electrical field is required. To create an electrical field, photovoltaic cells are made of two different types of semiconductors. One called n-type, the other, p-type.

In its pure form, the crystalline form of silicon used to make a photovoltaic cell actually wouldn't be terribly useful. The reason? The atomic structure of pure, crystalline silicon is very symmetrical and when photons strike, they don't knock many electrons free, and those electrons fall into holes left by other freed electrons.

Doping

The solution? Something called doping. During the manufacturing process, a very tiny amount of phosphorus, which has one more electron in its outer shell (it has five) than silicon (four), is added to the semiconductor. Suddenly, there are a lot more electrons that are free to be knocked about by photon energy. This semiconductor layer is called n-type because of all the negatively charged phosphorus electrons.

The p-type semiconductor layer is doped with boron, which has one electron less (three) in its outer shell than silicon. The result is a semiconductor with a lot of holes and a positive charge. Holes and free electrons want to pair up. And, get this, holes can move around. (Think of a hole as being something like a bubble.)

When you put the two types together, the free electrons and the holes scramble toward each other in a subatomic version of opposites attract. Eventually, a kind of equilibrium is reached at the junction between the two semiconductor slices. That equilibrium is an electrical field, a barrier that makes it impossible for anymore free electrons to flow directly to the p-side, or for any more holes to flow directly to the n-side.

Point of Equilibrium

Once you reach this point of equilibrium, you're ready to start making electricity. When photons strike the cell, free electrons on the n-side, which are unable to cross the junction to the p-side, instead head toward the contact grid sitting on top of the n-layer of silicon, which is wired to the contact layer on the bottom of the p-side.

Holes freed at the same time migrate toward the bottom contact layer where they wait for the incoming electrons. That flow of electrons, combined with the voltage created by the electrical field, yields power.

How much power? A module of about 40 standard-sized silicon produces about 50 watts. Enough to light a light bulb, give or take. So it takes a pretty massive array of photovoltaic cells before you get up to commercial power utility generating capacity. But they do exist, and rising energy prices, increasing efficiency of new cell designs that use different materials, and declining production costs make solar power look more and more like a useful alternative.

Todd Campbell is a writer and Internet consultant living in Seattle. The Answer Geek appears weekly, usually on Thursdays.