Conventional photovoltaic (PV) panels made from silicon to provide electricity to office buildings and homes are still too expensive. Unless they are heavily subsidized, it rarely makes sense to install them where electricity is available from the grid. Taking a new approach to solar conversion, using advanced materials and solar-concentrator technology, a group based at Rensselaer Polytechnic Institute (RPI) in Troy, N.Y., is developing a system that promises to be cheaper and smarter.

Solar-concentrator technology relies on optical methods to focus light on highly efficient photovoltaic materials. A novel way of using such concentrators has been hatched by an interdisciplinary group that includes architects, materials scientists, and electrical and mechanical engineers at Materialab, a research firm that grew out of RPI. The key element in their design is a concentrator with a Fresnel lens, whose concentric grooves focus light on a postage stamp­size cell made of gallium arsenide. The lens forms the flat base of a plastic pyramid, 25 centimeters on a side; the photovoltaic material, made by Spectrolab Inc., in Sylmar, Calif., is at the apex of the pyramid. Developed mainly for space applications, Spectrolab's gallium arsenide multijunction cells have layers of subtly varying PV materials that convert different wavelengths of light into electricity.

In the full system, an array of the modules is hung on wires between glass panels in an office building's facade [see artist's conception, "Let There Be Light"]. A computerized tracking system adjusts the orientation of the modules, so that the flat lenses always face the sun directly, for maximum photon collection. The modules convert about 30 percent of the sun's energy falling on them into electricity—about twice as much as flat silicon solar cells manage. But that's only part of the efficiency story. Heat collected at the tip of the pyramid is to be absorbed by water and transferred into the building through clear plastic tubes, to provide energy for heating and cooling systems. At the same time, the translucent modules allow diffuse light to pass into the building, reducing the need for artificial lighting [see diagram, "Pyramid Scheme"].

In effect, what the team has come up with is a smart venetian blind, says Anna Dyson, a professor of architecture at RPI and the founder and director of Materialab. "Instead of closing your blinds to reflect [intense] sunlight away and then using artificial lights," she says, "this system uses that [sunlight], reduces the current load on the building, and spreads light into the interior space, reducing the consumption of fluorescent lights."

The overall system is far less expensive than present solar panels in terms of dollars per watt, Dyson claims, although she won't give an exact number because the group is still negotiating prices with manufacturers. Some costs are lower, even though the gallium arsenide multijunction cells are costly, because the total area of PV material is two orders of magnitude smaller than in a standard system. The concentrators' tracking mechanism and the more complicated installation represent added expenses, compared with regular solar panels. Despite that, Dyson says the system will pay for itself in a quarter or a fifth of the time it takes for silicon solar panels to do so.

Dyson was inspired in part by the notion that architects need to be more than just consumers of materials and components delivered by others. She says architects should be more involved in the development of the building blocks themselves.

"I realized that in order to produce enough energy to actually change the energy-consumption profile of a building, photovoltaic systems had to move well beyond what was available in the marketplace," she says.

The RPI concentrator technology could be a breakthrough in solar facades, says Brian Dougherty, a mechanical engineer who works on building-integrated PV systems at the National Institute of Standards and Technology, in Gaithersburg, Md. But for that to happen, Materialab will have to deliver the lower costs it promises. "Right now, standard applied systems cost a few dollars a watt," Dougherty says. "If they can bring the price of the overall system down to a dollar a watt, that would be a significant accomplishment."

—Prachi Patel-Predd