BERKELEY, CA. -- Common manufacturing methods produce solar cells with an efficiency of 12 to 15 percent in converting sunlight to electricity; to make a profit, 14 percent is the bare minimum. In work done at the Ernest Orlando Lawrence Berkeley National Laboratory, scientist Scott McHugo has discovered important clues to the poor performance of solar cells manufactured from polycrystalline silicon.
The solar-cell market is potentially vast; because there's no need to build transmission lines or truck in generator fuel, lightweight solar panels are ideal for bringing electrical power to remote locations in developing nations. Industrial nations faced with diminishing resources also have active programs aimed at producing better, cheaper solar cells.
"In a solar cell there's a junction between p-type silicon and an n-type layer such as diffused-in phosphorus. When sunlight is absorbed, it frees electrons which start migrating in a random-walk fashion toward that junction," explains McHugo; now with Berkeley Lab's Accelerator and Fusion Research Division, McHugo became interested in solar cells when he was doing graduate work in materials science at UC Berkeley. "If the electrons make it to the junction, they contribute to the cell's output of electric current. Often, however, before they reach the junction they recombine at specific sites in the crystal," and thus can't contribute to current output.
McHugo looked at a map of a silicon wafer in which sites of high recombination showed up as dark regions. Researchers before him had shown that these occurred not primarily at grain boundaries in the polycrystalline material, as might be expected, but more often at dislocations in the crystal -- yet the dislocations themselves were not the problem. Using a unique heat treatment, McHugo performed electrical measurements to investigate the material at the dislocations; he was the first to show that they were "decorated" with iron.
"When I came to Berkeley Lab as a postdoc, I was able to employ a technique using x-rays at the Advanced Light Source (ALS) which is orders of magnitude better than what can be done with standard techniques that use an electron beam," says McHugo, who worked with the x-ray fluorescence microprobe beamline built and operated at the ALS by the Center for X-Ray Optics, part of the Lab's Materials Science Division. The one-micron spot of hard x-rays produced by the beamline allowed McHugo to align the resulting x-ray fluorescence spectra with maps of the defects made with a scanning electron microscope, comparing defects and impurities directly. "That's when I found that not only iron but copper and nickel were also concentrated in these high-recombination sites."
Metal from valves, couplings, and other machinery can contaminate solar cells as they are grown from molten silicon, cut into wafers, and finished by adding dopants and attaching contacts. In an industry with a narrow profit margin, where cheap polycrystalline silicon must be used instead of easy-to-purify but far more costly single-crystal silicon, rigorous cleanliness at every step of the way would be too expensive.
However, when it comes to purifying solar cells, cleanliness is not the only variable. Doping with phosphorus, as well as sintering aluminum contacts onto the wafers (heating them almost to melting), both actually help in "gettering" the silicon -- getting out the contaminants chemically. By adjusting time and temperature, these standard processes could be optimized to do a better job. McHugo has shown that briefly annealing the finished solar cell at high temperatures is enough to remove copper and nickel precipitates of moderate size, although dissolved copper and nickel or very small precipitates of these metals may remain.
McHugo is currently investigating what techniques are necessary to remove stubborn iron impurities from their hiding places in crystal defects. "We're looking at a two-step process," he says, "first subjecting the wafer to very high temperatures and then lowering the temperature to finish the proper processing of the solar cell."
"If a dirty manufacturing run produces solar cells of 12 percent efficiency, and a manufacturer can make money at 15 percent, think how profitable cells of 18 percent would be," says McHugo, who has collaborated with American and Japanese manufacturers and is now working with a consortium of government, university, and industry researchers in Germany. "Investigators have already achieved 18 percent in the lab with small samples; the challenge is to do it on the production line with full-sized solar cell wafers. It's a goal we're close to reaching."
Some of McHugo's findings were presented at the Materials Research Society meeting held last spring in San Francisco and will appear in the October, 1997 issue of Applied Physics Letters.
Berkeley Lab is a U.S. Department of Energy national laboratory located in Berkeley, California. It conducts unclassified scientific research and is managed by the University of California.