Linking oversized large scale PV with molten salt storage tanks is claimed to be a workable technical solution for regions with high energy consumption, according to recent research from Israeli and French scientists.
In the study Providing large-scale electricity demand with photovoltaics and molten-salt storage, published in Renewable and Sustainable Energy Reviews, the researchers presented a model to integrate solar power generation from utility scale facilities with high-temperature molten-salt storage in regions with low direct solar beam radiation and high levels of global solar radiation.
The PV-plus-thermal-storage (PV-TS) solution proposed by the academics, which is claimed to be “ready for immediate implementation,” due to the “unusually favorable economics” provided by PV technology, represents an alternative to CSP molten salt towers in areas where CSP technology is not considered viable because concentrators cannot exploit diffuse solar radiation. “Rather than aiming for precise answers for specific locations, the objective here is to see if the magnitudes for PV system size, storage capacity, grid penetration levels and cost estimates are feasible,” the academics explained.
In the PV-TS unit, a significant part of the generated solar power would be used to resistively heat molten-salt thermal storage to temperatures over 565 degrees Celsius, and the stored thermal energy would be in turn used to drive high-efficiency superheated steam turbines for power generation.
The simulations conducted by the Israeli-French group showed that in certain areas PV can see its grid penetration rate increase from around 30%, when no thermal storage is used, to around 80% with just 12 hours of thermal storage. “Furthermore, with only a 25% increase in solar input, 90% grid penetration can be attained,” the group further explained. “For the higher-insolation locations, where proportionality can be maintained up to approximately 90%, an extra 25% of solar input can raise grid penetration to about 95%.”
In this kind of project, the PV plant should not be sized as a common facility that needs to meet a particular peak daytime demand and most of the generated electricity should be used for heat storage in the molten salt tanks. “And that stored heat would satisfy the power demand not only at night, but also during daytime periods of sub-peak insolation,” the researchers specified, adding that an average land area of approximately 0.64km2 would be necessary for a terawatt-hour of annual power generation.
According to them, the “unconventional” solution proposed in the study may also be integrated with rooftop PV arrays and large steam turbines already operating in fossil fuel and nuclear power plants that are being decommissioned in several countries.
The findings of the research relate only to the U.S. territory but they could be extended to all regions with similar climatic conditions and utility demand profiles. “For regions with average insolation that is higher than the U.S. – some of which also happen to have electricity demand profiles that are better correlated with solar availability – the PV and storage requirements per terrawatt-hour of electricity consumption will be lower,” the paper notes. “Furthermore, the transition to all-electric vehicles may increase the fraction of electricity demand during daytime hours, when much of the battery charging will be performed.”
Russia, the former Soviet republics, Japan, northern Asia, and mid-to-northern Europe are pointed out as the most suitable areas, along with the United States, for the deployment of PV-TS projects.
The research team is composed of scientists from Israel's Ben-Gurion University of the Negev, France's Aix-Marseille University and Promes, which is the French national R&D laboratory on solar concentrating systems.
*The article was updated on January 26 to add an additional paragraph on the PV-TS system configuration.
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: editors@pv-magazine.com.
So if it’s not CSP, does that mean they’re using electric-powered heating elements to warm the molten salt?
PV power would resistively heat molten-salt thermal storage, according to the research
For some areas such as Scotland and Scandinavia, Wind / thermal storage would I think make more sense, especially if the thermal storage can feed district heating as well as steam turbines.
I won’t get it. Knowing latest pv efficiency only reach about 20+% for system size. And CSP is well above such a rate, why don’t simply use it instead of pv to heat molten salt?
All the mirrors and specialized tanks are more expensive than just ground tanks. PV isn’t much more expensive than the mirrors either.
When it’s overcast, there’s not enough of the focused beam to superheat the salt, whereas there’s still electricity available from the PV to power an electrode.
Another problem is dealing with rapid temp changes from low clouds. Common materials would instantly explode going from say 0° C to the 500 or 600° or so temps, and back again, all day long on certain cloudy days. So they have to always move the mirrors (I believe) to gradually warm up and cool down, wasting even more time “between clouds”.
Another problem is the inefficiency, however, same with the PV sourced “heat”. PV and batteries will probably win in the end because of Tesla style industrialism (and Chinese cell makers like CATL, too).
Oh, and there’s the enviro issue of zapping birds (however, I would expect bird herding drones to deal with that).
They need to put water by large PV farms, too, as they also create mirages (birds presumably die from exhaustion going to the fake lakes).
Oh, and there’s another thing. The stats (mirrors) can’t be as close to each other as the PV, causing even more land requirements. Towards the outer edges of the stat field, the tower appears really low in the sky, thus forcing the mirrors to be much further apart. PV only has this problem early morning and evenings.
So, the ⅓ efficiency of its steam compared to batteries, and the larger land requirements for a stat field might be even less than a third as efficient than batteries and PV panels.
Also, I would think less bulldozing needed for panels compared to stats, not sure though. Panels don’t have to be precise and thus should be able to be placed on framework that doesn’t require the complete destruction of the land…
The inefficiency is stated in the title. Batteries, especially the coming lifepo4 (lithium iron phosphate or LFP) though not as energy dense as Tesla style NCM, are a perfect solution for “oversized solar” because they are close to 3x more efficient than steam turbines. Thus 3x less of the planet needed to be covered by solar.
I don’t know if the salt tanks would last as long as the batteries, either, which should last for about 5,000 cycles.
Granted, salt should be at least 3x cheaper, though, and thus be a longer term solution.
I wonder what the “self discharge rate” is from the extra energy needed to keep it hot when the sun is down, like how good is the insulation around the tanks that prevent radiative and convective heat loss?
CATL is gearing up with Tesla to literally industrialize them.
Meanwhile, Tesla is gearing up to mass produce cell assembly lines, as well! Not sure if their silicon loaded high nickel content 4680 format could also be used for the much cheaper, more plentiful material needed for LFP cells.
Nevertheless, it’s best to have multiple solutions!
How does this compare to PV + batteries?
The terminology ” regions with low direct solar beam radiation and high levels of global solar radiation” is unnecessarily pompous. It means “places where light clouds frequently diffuse the sunlight”. That makes it impractical to focus the light onto a small target, but PV panels can still work fine. In fact, I have seen better production from my panels under very light clouds.