American Scientist

In a sense, it's very strange that modern society struggles so much to find a source of clean and abundant energy: After all, the power in the sunlight falling on the Earth exceeds civilization's needs by almost four orders of magnitude. If we harnessed even a tiny fraction of a percent of that total, we could dispense with the more messy and problematic forms of energy that we rely on so much today.

One of the more attractive ways to do so is to use photovoltaic cells, which convert light directly to electricity. These cells are safe, have no moving parts, operate at ambient temperature and last for decades. The obvious roadblock preventing their widespread application is expense: Electricity generated from photovoltaic cells is getting cheaper, but it still costs several times the going rate in most places.

It's tempting to imagine that one day soon prices will drop to the point that economic considerations alone allow photovoltaic panels rapidly to displace much of today's fossil-fuel-fired electric-power generation. But the true situation is more complicated. It turns out that the predominant type of solar panel being produced today cannot solve the world's energy problems anytime soon—simply because the device takes too much energy to manufacture. Fortunately, alternative strategies exist for making photovoltaic cells using much less energy, and one promising example is now beginning to be made in significant quantities.

The inherent problem with conventional photovoltaic cells is that they are composed of silicon. Although abundant in the form of silicon dioxide (say, from sand), the pure element requires considerable energy to extract. Analysts differ somewhat in their estimates, but the consensus is that it takes about three years for a conventional silicon photovoltaic panel and the equipment associated with it (the rigid frame used to mount it and the power-conditioning electronics that attach it to the grid) to produce the amount of electrical energy required to manufacture this equipment in the first place—assuming that it is set up in a reasonably sunny spot. These studies include, for example, a review of Tucson Electric Power's Springerville photovoltaic plant, which is one of the world's largest such installations, located in eastern Arizona. A careful life-cycle analysis of that plant determined the energy payback time to be 2.78 years.

At first blush, having an energy payback time of three years might not seem so bad. The problem comes when you consider how much the photovoltaic industry would need to grow in size to compete seriously with other forms of electric-power generation. Right now, the installed capacity of photovoltaic cells worldwide is about 6 gigawatts—a drop in the bucket compared with global electricity use, which demands terawatts of power.

The photovoltaic industry is, however, growing quite rapidly: Worldwide, the total installed capacity of photovoltaic panels increased by 36 percent in 2006. Alternative-energy advocates would, ideally, like to see such rates maintained so that photovoltaic cells could displace a large fraction of the fossil fuels being used to generate electricity. The rub is that with an energy payback time of three years, growing the industry at this pace requires more energy than all the existing photovoltaic cells produce. That is, even if you could somehow harvest all the energy produced from every last photovoltaic cell in one year, the total wouldn't be sufficient to produce the next year's crop of panels. As Lewis Carroll's Red Queen said in Through the Looking Glass, "Now here, it takes all the running you can do, to keep in the same place."

Andy Black, chief executive officer of OnGrid Solar Energy Systems, a San Jose company, pointed out this particular Red Queen effect in a presentation to the Solar World Congress in 2005. "It's sort of a mathematical oddity," says Black. "We've got this wonderful, clean industry that's actually using coal to power it." Were the growth rate more modest, of course, such photovoltaic systems would produce more energy than is being used to fuel their production. But Black says, "We're not going to make a difference unless we grow fast."

Another solar-energy advocate to express such concerns is Michael Graetzel, a professor of chemistry at the Ecole Polytechnique Fédérale de Lausanne in Switzerland. Two decades ago, Graetzel began work on a way to produce solar cells without silicon. He and his colleague Brian O'Regan published an influential paper in Nature showing how this could be done in 1991. The "Graetzel" or "dye-sensitized" solar cell uses a combination of titanium dioxide (a component found in many paints) and an organic dye molecule, often a compound containing ruthenium, which are together immersed in a liquid electrolyte. A. C. Veltkamp of ECN Solar Energy, an independent photovoltaic-research firm in the Netherlands, has estimated that such dye-sensitized cells installed in southern Europe would have an energy payback time of only a half-year or so.

So photovoltaic enthusiasts should be quite excited about the recent news that dye-sensitized cells are now going into large-scale production. G24 Innovations, a startup with headquarters in Cardiff, Wales, announced in October 2007 that it expects to be able to make enough dye-sensitized photovoltaic cells each year to provide 30 megawatts of peak generation capacity. For that, the company will be using a continuous process: Instead of coming off the assembly line in discrete, rigid units, the dye-sensitized cells are placed on half-mile-long rolls of flexible metal foil. The company's immediate market is in mobile applications, say, for recharging cellphones and the like.

Graetzel, who serves as a scientific advisor for G24 Innovations, points out that other companies, too, are working to commercialize dye-sensitized cells, based on a package of patents that the Ecole Polytechnique Fédérale de Lausanne controls. For a long while Graetzel and his academic colleagues used to arrange a licensee meeting every year (sometimes even more frequently) where representatives from these different companies shared ideas. With dye-sensitized solar cells standing on the threshold of large-scale commercialization, these companies are, naturally enough, no longer so inclined to talk with one another in this relaxed way. Their teams of technologists are presumably sprinting very hard now to beat out the competition. So it's good to know that, with this kind of solar cell at least, their energies won't be wasted just running in place.