Last week CSP Today spoke to Craig Turchi about the DOE and SunShot initiative in the USA. This week Susan Kraemer delves deeper into the potential of s-C02 in reducing CSP costs.
This year, the DOE’s SunShot Initiative is investing $56 million into technologies that promise to truly revolutionize CSP. The goal is to reduce the wholesale rate for CSP to 6 cents a kilowatt-hour by 2020 (the same target it is working on for PV) in order for solar technologies to competitively supply 15 to 18% of America's electricity generation by 2030.
“That’s going to require a very significant transformation of the technology to wring that much cost out of it,” said CSP Alliance Founder Tex Wilkins, who has spent 30 years at the U.S. Department of Energy, with the last 11 leading its CSP program.
“Industry usually does best on evolutionary improvements,” Wilkins points out, “but the DOE looks for technical break throughs that bring about revolutionary cost reduction.”
The largest of the grants, $8 million, is going to the National Renewable Energy Laboratory (NREL) to build and test a commercial-design turbine with supercritical carbon dioxide - s-CO2 - as the working fluid within a closed-loop Brayton power cycle.
NREL CSP project leader, Craig Turchi, says the 10 MW project will be the largest s-CO2 turbine ever built. It comes after a thorough evaluation of the optimal temperature for running solar power towers conducted by NREL, Sandia and the DOE. “Based on our assessment of optical losses, thermal losses, material issues, and power cycle efficiency, we believe s-CO2 power cycles running at 600-700°C represent an excellent case for improved CSP performance,” he says. Heating and compressing the carbon dioxide to create s-CO2 will be done within the system.
“You can’t simply buy supercritical CO2” says Wilkins. Rather, “You buy the CO2 and then have to include that pressurizing and heating as part of the CSP plant that has to be developed. It hasn’t been done commercially at the scale solar would require.”
Supercritical carbon dioxide expands to fill its container like a gas but with a density like a liquid. These supercritical properties at above 500 °C and 20 MPa enable very high thermal efficiencies, approaching 45%, potentially increasing the electrical power produced by 40% or more.
‘I’m not surprised about the s-CO2 award’ agrees Wilkins. ‘The DOE is focusing on that power cycle because of its high efficiency and potential for low cost’.
“Brayton-cycle systems using s-CO2 have smaller weight and volume, lower thermal mass, and less complex power blocks versus Rankine cycles due to the higher density of the fluid and simpler cycle design”, wrote Turchi in his 2011 paper; Advanced Supercritical Carbon Dioxide Power Cycle Configurations for Use in Concentrating Solar Power Systems. Furthermore, “The lower thermal mass makes startup and load change faster for frequent start-up/shut down operations and load adaption than a HTF/steam based system.”
This is particularly relevant as CSP plants are currently too slow to act as fast-acting peaker plants. According to Bill Gould, CTO of SolarReserve, they are limited by the steam cycle, and are only able to ramp up at about 10% a minute. Gas turbines are faster.
As we add more renewable energy, a fast startup is increasingly necessary. The Obama administration’s FERC Chairman, Jon Wellinghoff, proposed last year that utilities should pay more for peaking resources able to ramp up fast.
The NREL demonstration builds on five years of DOE-funded research and development at Sandia Labs conducted on a much smaller scale - under 1 MW - on advanced Brayton cycle loops, which work like a jet engine running on hot liquid.
Turchi says that the project is a team effort with researchers in other energy disciplines who are also interested in this power cycle. “We are leveraging prior investments made by the nuclear energy program at Sandia National Lab where the test will be located,” he explains, adding that the NREL project focuses on operating conditions specific to CSP and builds on their work in s-CO2. “We believe the end result of this project will be a tested commercial turbine design and the understanding of how to control a s-CO2 power cycle running at conditions relevant to CSP, in particular one that can use dry cooling.”
“One of the ways to lower costs is to raise the temperature of the operating system, because the higher the temperature, the more efficient the system is” says Wilkins. “The correlation is the higher the efficiency the lower the cost. And with the higher temperatures the cost of storage then also becomes a lot cheaper.” To save money, the demonstration will simply use natural gas to run the test loop, since no solar source of the required size and temperature is currently available for the test.
NREL anticipates two applications to CSP. Firstly, the Brayton closed loop power cycle would be coupled to a molten-salt HTF solar plant with traditional receiver and thermal storage systems (s-CO2 is not suitable as storage media, because of its high pressure.) However, to realize the full potential of the s-CO2 power block, higher temperature salts are needed. These are currently being developed in other DOE projects, for example with Halotechnics which is using molten glass.
“In the second application the s-CO2 would also be used as the HTF, analogous to a direct steam generation design” says Turchi. “We believe such a configuration makes sense for small (10-20 MW) power towers using phase-change thermal storage. We are currently running an analysis of this design.”
At 10 MW, the scale is large enough that the next step can be industry-deployment of the power system on a CSP plant. While the DOE is supplying $8 million, industry partners are putting up half the money for this test - for a total cost of $16 million.
“That, in itself, indicates the excellent potential that is foreseen for this technology” concludes Turchi.
To comment on this article write to Susan Kraemer
Or contact the editor, Jennifer Muirhead