Storing Solar Power Efficiently
Thermal-power plants that store heat for cloudy days could solve some of the problems with solar power.
Solar proponents love to boast that just a few hundred square kilometers’ worth of photovoltaic solar panels installed in Southwestern deserts could power the United States. Their schemes come with a caveat, of course: without backup power plants or expensive investments in giant batteries, flywheels, or other energy-storage systems, this solar-power supply would fluctuate wildly with each passing cloud (not to mention with the sun’s daily rise and fall and seasonal ebbs and flows). Solar-power startup Ausra, based in Palo Alto, thinks it has the solution: solar-thermal-power plants that turn sunlight into steam and efficiently store heat for cloudy days.
“Fossil-fuel proponents often say that solar can’t do the job, that solar can’t run at night, solar can’t run the economy,” says David Mills, Ausra’s founder and chairman. “That’s true if you don’t have storage.” He says that solar-thermal plants are the solution because storing heat is much easier than storing electricity. Mills estimates that, thanks to that advantage, solar-thermal plants capable of storing 16 hours’ worth of heat could provide more than 90 percent of current U.S. power demand at prices competitive with coal and natural gas. “There’s almost no limit to how much you can put into the grid,” he says.
Major utilities are buying the idea. In July, the Pacific Gas and Electric Company (PG&E) signed a 25-year deal with Ausra competitor Solel Solar Systems of Beit Shemesh, Israel, to buy power from a 553-megawatt solar-thermal plant that Solel is developing in California’s Mojave Desert. The plant will supply 400,000 homes in northern and central California when it is completed in 2011. Florida Power & Light, meanwhile, hired Solel to upgrade the 1980s-era solar-thermal plants it operates in the Mojave.
Ausra, meanwhile, is negotiating with PG&E to supply power from a 175-megawatt plant that it plans to build in California, for which it secured $40 million in venture financing this month.
What distinguishes Ausra’s design is its relative simplicity. In conventional solar-thermal plants such as Solel’s, a long trough of parabolic mirrors focuses sunlight on a tube filled with a heat-transfer fluid, often some sort of oil or brine. The fluid, in turn, produces steam to drive a turbine and produce electricity. Ausra’s solar collectors employ mass-produced and thus cheaper flat mirrors, and they focus light onto tubes filled with water, thus directly producing steam. Ausra’s collectors produce less power, but that power costs less to produce.
One megawatt’s worth of Ausra’s solar collectors has been producing steam in New South Wales, Australia, since 2004; the steam is fed into the turbines of a primarily coal-fired power plant. The final piece of the system–a proprietary heat-energy-storage system–should be ready by 2009.
Mills will not say what material his company’s system will heat, although several recent solar-thermal plants by Ausra competitors–including one in Nevada that started up this summer and two under construction in Spain near Granada–plan to use molten-salt storage. Molten salts are inexpensive salt solutions that absorb considerable energy when they melt and give up that energy when they freeze.
What Mills can say for certain is that Ausra’s storage system will lower its power-generation costs. That is a surprising statement since energy storage can as much as double the cost of electricity from photovoltaics or wind turbines.
Heat storage is more efficient than electricity storage: just 2 to 7 percent of the energy is lost when heat is banked in a storage system, compared with losses of at least 15 percent when energy is stored in a battery. More important, says Mills, is the fact that storage enables thermal plants to use cheaper turbines.
The bottom line is that Mills vows that adding storage plus savings from economies of scale and lower cost of capital (as banks become familiar with solar-thermal technology) will cut Ausra’s current 10 to 11 cents per kilowatt-hour cost of power in half. By 2010, he expects solar thermal to provide California with baseline power cheaper than natural gas, currently set by the state at 9.2 cents per kilowatt-hour.
Why has solar-thermal power received little attention from the energy-storage community despite such promise? John Boyes, manager of the Energy Storage & Distributed Energy Resources at Sandia National Laboratories, in Albuquerque, NM, says that solar thermal is viable but inflexible compared with other means of storing energy, such as, say, coupling wind farms to large batteries, flywheels, and supercapacitors that can be placed almost anywhere on a power grid. “You can store energy anywhere you have electricity and a little bit of floor space,” says Boyes.
The footprint of Ausra’s planned 175-megawatt plant will be, in contrast, about one square mile.
Become an MIT Technology Review Insider for in-depth analysis and unparalleled perspective.Subscribe today