In Tower CSP, thermal storage had issues at first
A thermal storage tank failure ended Concentrated Solar Power (CSP) development in the US. At the world’s first utility-scale Tower CSP project with storage, the molten salt thermal energy storage tank sprang a leak.
SolarReserve’s Crescent Dunes 110 MW project was the only Tower CSP with thermal energy storage among the first five commercial CSP projects deployed in the US.
[In Tower CSP, an array of heliostats is focused on a receiver atop a tower, generating much higher temperatures than the initial form of CSP; parabolic Trough. While Trough operates at a relatively low temperature of 400°C, Tower’s higher temperatures enable more efficient operation at closer to 600°C.]
Since its tank leak nearly ten years ago, SolarReserve, the developer of Crescent Dunes, could not land its global pipeline of permitted projects.
ACWA Power is now completing SolarReserve’s Redstone project in South Africa, Grupo Cerro is bidding SolarReserve tower projects in Chile; Likana and Copiapo and Vast Solar in Australia is currently building its own unique CSP plant at the Port Augusta site where SolarReserve was not able to complete financing on its Aurora Tower project.
And no new tower CSP with thermal energy storage was commercially bid in the US.
Sharing knowledge helps all CSP
Knowing what caused this tank failure is crucial, Vast Solar CTO Kurt Drewes told SolarPACES in a call from Australia. He believes that to advance CSP, the industry must share solutions in commercial plants.
“Few technical problems cannot be solved with simple communication. They’re social problems,” he commented. “If we all share our problems, ideas, and lessons learned, we can solve these. But it’s been very hard because everybody’s behind commercial curtains.”
In a call from Australia, he pointed out that sharing commercial knowledge on thermal energy storage is not just important for the future of Tower CSP but also impacts the development of other technologies. These include standalone electric-to-thermal grid storage, solar heat for industrial processes, and hot solar thermochemistry for producing hydrogen and hydrocarbons like jet fuel.
“This is because the core value proposition in CSP is the ability to store thermal energy,” he explained. “That’s fundamental, and that benefit is seen in all kinds of other system configurations. And so it’s in everybody’s interest that we get this right.”
The problem is not the higher temperatures of Tower CSP
After studying the issue, engineers at Vast Solar concluded that the problem was not so much the higher temperature of Tower CSP but rather a greater temperature differential and higher friction that occurs within the hot tank.
“It took a while for us to understand that the buckling failure is more of a function of temperature distribution than the actual temperature,” Drewes said.
“Because the primary data was that the thermal storage in parabolic trough plants is fine, but not in Tower at higher temperatures. And the first obvious answer is that the only thing that’s different is the temperature. But we have shown that it’s a temperature distribution that changes with respect to space and, importantly, time.”
“These are very large storage tanks; with very large diameters and, in proportion, very thin floors. The ratio between these two is extremely low with these very big tanks. Certain parts of the tanks are hotter and colder; process changes can cause that. So it’s really a temperature distribution problem. ”
Vast Solar’s solution for buckling in thermal storage tanks
Vast Solar is developing an innovative multi-tower form of Tower CSP that uses liquid sodium for heat transfer throughout its solar field. Still, the heat is then stored in molten salts, a combination of sodium nitrate and potassium nitrite. With less than 1% of heat loss daily, in liquid form, these molten salts are well suited to hold and store thermal energy.
Their innovative use of sodium for heat transfer is interesting because sodium is capable of even higher temperatures, up to 800 C, ensuring that delivery to the thermal storage tank will always be higher than needed.
He noted that the issue of too great a temperature differential can arise not just during commissioning when the first liquid is poured but also in daily operation.
“The core problem here is that we’ve got compressive forces that result as a function of temperature distributions and the friction between the floor and the foundation,” he said. “So even in commissioning, we need procedures that minimize or eliminate these. We have been working with our partners CyD and others for many years on this.”