MOUNTAIN VIEW, Calif. In the face of uncertain markets and policies, solar panel makers are ramping up factories in hopes of driving costs down to a competitive level.
One market watcher said she will revise her forecasts for solar panels based on the financial crisis in the U.S. combined with uncertainty about renewal of federal tax credits here and a pullback of major solar plans in Spain.
Without incentives, we don't have a market, period," said Paula Mints, a principal analyst at Navigant Consulting (Palo Alto, Calif.), speaking at a solar event sponsored by the IEEE here.
Several speakers here also complained about a lack of federal funds for research in solar energy, something Tim Anderson, associate dean of research at the University of Florida, said he hopes will change soon.
"The reputation in the academic community is you don't want to go do research in photovoltaics because there's no money there," Anderson said. "But the promise of more money to come is out there, and people are adjusting to take advantage of this."
On the commercial front, a widening group of companies is expanding production of solar cells in a drive toward mainstream costs. "We're building the equivalent of a large scale nuclear power plant a year," said Richard Swanson, president of SunPower Corp. (San Jose), one of the top ten solar panel makers.
SunPower just completed a second plant capable of using 400,000 wafers per day and has plans for plants in Malaysia that will use as many as a million wafers a day. "We know how to [produce panels that cost] $1.50 per Watt by 2012, and we have detailed quarterly plans to get there," said Swanson.
Startup Signet Solar (Menlo Park, Calif.) shipped its first solar panels in October from a new factory in Germany that will be able to produce by early 2009 enough panels each year to generate up to 130 Megawatts. The company aims to set up another plant in India soon.
"I think you will see a trend toward large 10-20 Megawatt [solar installations] which will be very efficient," said Rajeeva Lahri, chief executive of Signet.
Applied Materials is supplying equipment to Signet and a handful of other panel makers now ramping up large scale operations, said Mark Pinto, Applied's chief technology officer and manager of its energy group.
"It took 20 years to get to the point where a solar factory could produce 10 Megawatts of capacity a year, but by 2010 we will see Gigawatt factories," Pinto predicted. "We are no longer a factor of ten away [from competing energy sources], we are within a factor of two or less today," he added.
Pinto talked about the outlook for Applied and the solar business in a video interview with EE Times.
Although production has been rising rapidly, panel prices have actually gone up in the last six years from historic lows in 2001-2002, said Mints of Navigant.
"This is a market that was unprofitable until 2004," she said. "You can't blame them for wanting to make profits. The industry was willing to sell at a loss to keep share," she added.
Meanwhile new and promising technologies are still emerging.
Russ Jones, business development manager for Boeing's Spectrolab, described his company's solar concentrators. They use 500x optics to create solar cells as small as one square centimeter. Lab versions of the gallium arsenide cells have hit record efficiency levels up to 40.7 percent, nearly twice the level of most commercial cells shipping today, he said.
"We believe 50 percent efficiency solar cells are within reach in the coming decade," Jones said. "We think we have an interesting new vector to help solar cells succeed."
James Gee, founder and chief scientists at Advent Solar (Albuquerque, N.M.) described a new architecture for cells and modules his company will produce starting next year based on research he conducted while at Sandia National Labs. The Ventura modules use a pre-patterned electrical circuit sheet on the back of the module rather that a wire grid on its front to decrease manufacturing time and increase cell efficiency.
"Free power has been thrown away due to non-optimal designs," said Gee. "It's time for a completely new process that gets around the current fundamental limitations," he said.