Solar Needs Brain Power

"If we can get the brain trust of the semiconductor industry behind solar, we can bring down costs and become mainstream," he said.

The global solar industry spent some $2.8 billion on plants and equipment in 2006, according to Solarbuzz's 2007 Marketbuzz report, as it added 548 MW, for a 33 percent jump in capacity, to a worldwide total of 2204 MW. Solar sales reached $10.6 billion last year and will likely grow to somewhere between $18 and $31 billion by 2011, for a 170 to 290 percent jump.

"Much of this forecasted growth depends on lower costs, and that depends on such things as deposition tools with better yields and higher throughput, economical slicing and handling tools for thinner wafers, better coating materials and lower cost polysilicon," Resch said in a recent statement.

He elaborated on these opportunities in "Solar Energy: The Next Great Growth Opportunity for the Semiconductor Industry," his keynote speech at Semicon West 2007, which was held this week at the Moscone Center in San Francisco.

"The single biggest factor to bring down solar cost per watt is moving to larger substrates, on tools whose development was paid for by the folks who make television sets," said Charlie Gay, corporate vice president and general manager of Applied Materials Inc.'s solar business group.

Gay presented "Issues of Going to Gigawatt-scale Solar Manufacturing" at renewable energy sessions during Semicon's Emerging Technologies & Markets TechXPOT on Thursday.

Applied Materials has said the large substrates could potentially bring thin-film solar costs down by 25 percent. But Gay said he sees a continuum of further cost reductions to come from other suppliers scaling up near big-thin film solar plants, from the makers of clear solar soda lime glass to gas suppliers providing the silane and hydrogen precursors, to those making the environmental control and packaging equipment.

"It's all about economies of scale," he said. "There's a wider range of opportunities supplying the solar industry with the same equipment and materials as the semiconductor industry, because it's earlier on in the maturation of the industry. It's like the integrated circuit industry years ago, when the makers were doing a lot of their own equipment design. But now the industry is growing so fast there's no longer time to do it all in-house."

New materials are another area for reducing costs, since materials account for 60 percent of the typical cost of a conventional wafer-based solar cell, said Gaetan Borgers, Dow Corning's global solar industry director.

Now that solar is becoming a significant market, suppliers are starting to invest in developing products tailored to its specific needs, he said, "so there was no focus on the market. You had nine-nines purity silicon for the semiconductor industry, and two-nines for the chemical industry. Now there is a big new market emerging that needs something in between."

Borgers said Corning is stepping into that gap with a lower-cost, lower-purity solar silicon feedstock made with a metallurgical process that can be blended into high-purity polysilicon without impacting cell performance. He said it also plans to introduce, possibly in 2008, a lower-cost encapsulant of silicone that may allow higher efficiency cell performance and that may last longer.

Other companies applying new kinds of nanoscale manufacturing technologies to make solar cells more affordable: Miasolé of Santa Clara, Calif., said it will start volume production of its sputtered thin-film copper indium gallium selenide (CIGS) on flexible stainless-steel foil sometime later this year.

President and CEO Dave Pearce said the company is "running the first vacuum system every day and now bringing up a second roll-to-roll coater" but that he is waiting for higher efficiencies before starting commercial production. Although the company's R&D tool reportedly consistently makes cells of 8 to 10 percent efficiency on 5 ft 2 in substrates, the production roll coaters are still running mostly in the 4 to 6 percent range.

"But that's about a point higher than last month," Pearce said. "I'm comfortable we can get 8 to 10 percent this year. I think we'll get there this quarter."

The atomic-level control of the tricky deposition of CIGS film is key to efficient performance, he said. Miasolé's approach relies on much of its control of the thin-film sputter target materials, then uses multiple rotating targets to control the gradient of composition through the CIGS film at the nanoscale.

Pearce said the low-cost production tools and high-speed process—about an hour from start to the final testing of the cells—should allow the nanoscale CIGS film to bring solar costs down dramatically.

Other companies see opportunity in supplying their nanotechnology products to solar cell makers. Eikos Inc. of Franklin, Massachusetts, reports good results using its carbon nanotube (CNT) film as a transparent conductor for organic solar cells, since the naturally hole-conducting CNT films are the p-type transparent electrode the cells require. The binder-infused CNT film is applied at ambient temperatures, so it is well-suited to the heat-sensitive organics and is flexible enough for use on its targeted applications such as cloth and tents.

NanoGram Corp. is also turning its nanoparticle deposition technology to the development of solar cells. The Milpitas, California, company creates nanoparticles by laser pyrolysis, using a laser to provide uniform initiation of the chemical reaction to condense compounds from the gas phase into nanoparticle form.

Particles produced in this continuous process are deposited directly onto a substrate, at thicknesses up to 30 µms in one pass, at throughput potentially high enough to be practical for production of thick-film coatings. The company is working on depositing crystalline silicon thick film for solar cells—potentially avoiding the contamination from the substrate usually generated at the high temperatures needed to crystallize silicon—to allow higher efficiencies than amorphous thin film without the cost of silicon wafers.

This article was adapted from the original July 17th article on Photonics.com and was republished with permission from Photonics.com.