—By David Pacchioli
Solar power is clean, renewable, and evenly distributed around the globe. So why is it still the energy of the future? One reason is the cost of silicon, the basis for most solar-cell technologies.
A solar cell works by converting sunlight to electricity. When sunlight strikes the cell's semiconductor (typically silicon) surface, some of that light is absorbed, knocking loose electrons from the silicon atoms, allowing them to flow freely within the solid's molecular matrix. The built-in presence of an electrical field draws these free electrons in the same direction, creating a current which exits the cell through a conducting wire.
The best commercially available silicon cells convert sunlight at a rate of 14-16 percent. "But silicon takes a great deal of energy to produce," says Craig Grimes, "and we get that energy by burning coal. If you look at the economics of it, solar cells have a hard time catching up."
That's why in recent years researchers including Grimes, an engineering professor at Penn State, have been looking for cheaper solar-cell materials. One that shows promise is titanium oxide.
Titanium is a metal, plentiful in the Earth's crust. Exposed to oxygen, it becomes titanium oxide, commonly seen as a white powder that's used to make white paint and sunscreen. "It's real cheap stuff," Grimes says. It is also a semiconductor.
For solar cells, "The conventional approach has been to use a paste made up of nanoparticles of titanium oxide, something you squeegee onto a glass substrate as a thin film," Grimes says. The nanoparticles create increased surface area, which improves the light absorption. But the arrangement of those particles is random, which means too many electrons get lost on the way to the cell's conducting wire.
"An electron has to go from particle to particle to particle," Grimes explains. "You can see how many hops it has to make, depending on the thickness of the film—it's many thousands. And every time that electron makes a hop, it has a chance to recombine, generating heat instead of useful work."
Scanning electron micrograph (SEM) image of a titanium oxide nanotube array produced in Craig Grimes's lab. Courtesy Craig Grimes
In 2001, Grimes happened on a different approach, a simple chemical etching process that turns a flat titanium surface into a densely packed forest of tiny metal oxide tubes, each about 46 nanometers across. This highly ordered arrangement, it turns out, brings with it some unique sensing and charge-transfer properties. In addition to being able to pick up infinitessimal traces of gases, the nanotubes are great for transporting electrons.
To make a solar cell, Grimes and his group take a piece of conductive glass and sputter on a layer of titanium, then dip the structure into an acid bath charged with a mild electric current. The combination of acid and oxygen eats away at the metal, leaving the neat array of nanotubes. "They're so uniform it's almost scary," Grimes says.
Next the tubes are heated in oxygen until they crystallize and become transparent. Coated with a light-absorbing dye, the see-through array becomes the cell's negative electrode. When sunlight shines through the glass, the energy falls on the dye molecules and electrons are knocked free. Before they have a chance to recombine, the tube structure of the titanium oxide ferries them directly to the conducting wire.
"It's the perfect material architecture," Grimes notes. "You have these pipes which electrons just love to travel up and down. They're effectively electron superhighways."
In a paper published in Nano Letters in January 2006, Grimes and company reported a three percent photoconversion rate using nanotube arrays 360 nanometers tall, the tallest they had "grown" at that time. That initial result was enough to create a stir in solar energy circles. "The conversion efficiency is proportional to the length of the tubes," Grimes explains. The taller the tube, the greater the amount of light absorbed and the number of electrons transported. Achieving three percent efficiency with such short tubes suggests the technology has real potential.
In fact, Grimes recently reported an increase to 7.2 percent efficiency; but further increases will require taller tubes, which in turn means starting with a thicker deposit of titanium. That's the current hold-up, he says. Although he and his team have produced uniform arrays over six micrometers (that's 6,000 nanometers) tall from titanium foil—"which should suffice for a new world's record in efficiency," he says—getting them to adhere to the glass substrate of a solar cell has proved a bit tricky.
"It's been frustrating to be stuck on something so seemingly simple as adhesion, but I'm confident we'll figure it out," Grimes says. "When we do, I think 18 percent efficiency is within our reach. And that's with a relatively easy fabrication system that is commercially viable."
Whether that would be enough to win the race for a cheaper alternative to silicon is not clear, he acknowledges. "If there's one thing I've learned working with new solar cell technologies it's that there are an awful lot details to be worked out.
"Ultimately it will be the market that sorts them out."" RPS
Craig A. Grimes, Ph.D., is professor of electrical engineering and materials science and engineering in the College of Engineering, cag14@psu.edu. His work on solar cells is funded by the National Science Foundation, the Department of Energy, and the Pennsylvania Energy Development Authority.