The biggest advantage, van Hest says, is that ink-jet printing, unlike screen printing, does not involve applying pressure to the silicon wafer. That makes it possible to use far thinner wafers, he says. In conventional solar cells, the wafers are about 200 micrometers thick. "If you go any thinner than that, most of them will break" during manufacturing, van Hest says. "If you go to a noncontact method, you don't have to worry about that. You can use cells as thin as a hundred micrometers, or even thinner. That means you can save 50 percent on the cost of silicon." NREL decided to use iTi Solar's printers, as opposed to those from other manufacturers, for two main reasons. The first is the accuracy of the system: the printers can apply ink with an accuracy of within 1 micrometer, as opposed to 10 to 15 micrometers with other systems, van Hest says. It's also easy to adapt to working with different inks and different solar-cell technologies. The system could, for example, be used to manufacture thin-film solar cells made of semiconductors such as copper indium gallium (di)selenide (CIGS), employing inks that NREL has developed.

Ink-jet printing has been considered before for solar-cell manufacturing. "There initially was a lot of buzz about it--two years ago, everyone was talking about it," says Bruce Morgan, iTi Solar's CEO. "Then as the practical results were seen, people got discouraged." He says the problem was that the printheads didn't have high enough resolution, or companies didn't understand the distortions in the materials that were being printed on. The initial results of the iTi Solar system suggest that it has solved these problems. "We've seen what it can do, and so far, it's very impressive," van Hest says.

The NREL system is designed to produce only prototype cells. Van Hest says that the technology can easily be scaled up, such as by increasing the number of printheads, to make many cells at once for commercial production.