Via Climate Progress: MIT Creating 24-Hour Solar Power on the Cheap?

Researchers at MIT are designing a new method of building concentrating solar power plants with thermal storage that they say could lower the cost of energy by 50% compared with existing technologies.
Last month, a 19.9 MW power-tower concentrating solar power plant in Spain became the first to generate electricity for 24 hours using molten-salt storage. But the cost of building that demonstration plant is higher than most CSP technologies – around $18 per watt, putting the cost of electricity somewhere around 30 U.S. cents per kilowatt-hour.

The company developing the plant, Torresol, wasn’t building it to prove the design could be the cheapest. It was a demonstration plant to prove molten storage technology and allow the company to scale up a much larger plant. But it also showed that there’s still work to be done in order to bring down costs of concentrating solar power designs.

MIT Mechanical Engineering Professor Alexander Slocum – along with a group of other researchers – says he’s designed a new type of tank for molten salt storage that could reduce equipment needs, increase durability and ultimately reduce the cost of electricity being generated by a plant.

Rather than use a complicated plumbing infrastructure to heat and pump the molten salt for storage, Slocum’s design puts the salt storage and water heating in a single tank mounted on the ground, rather than on a tower far above the field of mirrors. Under the new design, the mirrors are actually mounted on a hillside above the storage tank and reflect sunlight down into a small opening in the top.

The system could be “cheap, with a minimum number of parts,” says Slocum, the Pappalardo Professor of Mechanical Engineering at MIT and lead author of the paper. Reflecting the system’s 24/7 power capability, it is called CSPonD (for Concentrated Solar Power on Demand).
The new system could also be more durable than existing CSP systems whose heat-absorbing receivers cool down at night or on cloudy days. “It’s the swings in temperature that cause [metal] fatigue and failure,” Slocum says. The traditional way to address temperature swings, he says: “You have to way oversize” the system’s components. “That adds cost and reduces efficiency.”

As this technology is still in the research phase, the actual cost projections for such a plant can’t be precisely mapped. But the team says electricity could be as low as 7 cents per kilowatt-hour and as high as 33 cents per kilowatt-hour. Those cost ranges would be determined by many factors, including turbine capacity, size of the mirror field, quality of solar resources and a reduction in equipment needs.

If the design works as the researchers suggest, it could hold a lot of potential. Because CSP plants have a higher cost per watt — and thus a higher cost of electricity — the use of storage like molten salt will be key to the success of the industry, GTM Research Senior Analyst Brett Prior tells Climate Progress:

“If you want to make real progress with deploying these technologies, you need to have the same dispatchable characteristics as natural gas — you need firm power. Companies need to market their technology as something different than PV, which is a peaking resource, but can be more attractive on a cost basis right now. That’s where these CSP technologies can have a real advantage.”

Prior says that Solar Reserve, a U.S.-based company developing power tower technology with molten salt storage in California and Nevada, has a projected cost of energy at 11 cents per kilowatt-hour. That’s just below what large-scale thin-film PV projects are producing today.

So in theory, the MIT design could have major cost advantages if it works at scale. But as history shows, actually getting to that point isn’t easy. Many of the CSP designs that are being deployed on the demonstration and commercial scale have taken many years to get to this point.

However, MIT’s Slocum says that the plant would be built using existing technologies — it would just require a new way of thinking about how to construct the project.

Most of the individual elements of the proposed system — with the exception of mirror arrays positioned on hillsides — have been suggested or tested before, Slocum says. What this team has done is essentially an “assemblage and simplification of known elements,” Slocum says. “We did not have to invent any new physics, and we’re not using anything that’s not already proven” in other applications.

If this actually works, we may yet see more solar power after twilight.