22 ms·
24-Hour Solar Energy: Molten Salt Makes It Possible
- adrianN 7y agoThis is pretty old technology. Right now it's cheaper to build normal PV and rely on e.g. gas peakers. Eventually we'll want to deploy this, unless batteries become a lot cheaper.
- pjc50 7y agoOld? It's not really been properly widely deployed yet and is still in the falling cost phase. It is very much a Big Engineering thing though, unlike PV which can be deployed at the level of individual calculators.
- adrianN 7y agoNot old as the wheel of course, but the first commercial installation happened a decade ago. It's not like it's some kind of experimental stuff that was discovered only recently.
- pfdietz 7y agoA large part of the cost of concentrated solar thermal is from mature components like turbines and generators and heat exchangers. These are not going to get much cheaper, at least anytime soon.
- benj111 7y agoCoal/gas was cheaper than solar/wind (in certain areas) until it wasn't. We have to build these plants to get the experience to make them cheaper. And just comparing cost v fossil fuels is kind of missing the point. We have to move off carbon sooner rather than later. Fine, compare it the various battery technologies and nuclear, but comparing it to the thing we're moving away from, for good reason seems overly dismissive, and not particularly helpful.
- adrianN 7y agoI think you misread my comment. I'm comparing it to normal photovoltaics without overnight storage.
- benj111 7y agoI understood you to be comparing photovoltaics + night time gas v photovoltaics + molten salt night time storage? That seems the most reasonable scenario?
- adrianN 7y agoI'm kind of hoping that wind power covers most nighttime demand eventually. Right now we have so little renewable generation in most grids that just adding the cheapest option without caring about storage at all seems best.
- AstralStorm 7y agoWind is useful, but not really as base load. There are long lulls sometimes, more than a week. You could run gas peakers but then you haven't fixed anything... Fixing wind lulls requires continent-wide network of wind farms with corresponding grid. Computations suggest best case even 100 GW across Europe. That is nothing compared to demand. (Un)fortunately we have nuclear as a base load option. And fusion as research program. Potentially space solar including space mirrors too.
- adrianN 7y agoYou can use power-to-gas technology together with the existing infrastructure for strategic gas reserves to store enough renewable energy to run your gas plants during windstill winters. Maybe that's cheaper than nuclear power. It's probably easier to do politically.
- benj111 7y agoSo you're saying we should concentrate on (cheap) photovoltaics before moving onto expensive storage? Agreed although we do need to be thinking about storage, so we have technologies ready to go when the photovoltaics are built out. Speaking from a UK perspective, renewables now account for about a third of electricity production, so that's well on the way in certain scenarios to be thinking about storage. The UK is lucky in that most of this is wind which tends to coincide with peak demand, if it were solar generation, night time and evening could start being a problem.
- Jweb_Guru 7y agoOkay. Gas peakers are not a sustainable alternative to renewables, so we're going to need solutions that cost more.
- AstralStorm 7y agoOr you could deploy base load nuclear and use solar PV as peakers, with some load optimisation for big daytime spenders depending on insolation. Solar PV plus solar thermal, while seductive, competes for space and deployment.
- Jweb_Guru 7y agoNuclear is also much more expensive than solar, takes a long time to construct (so it's another "big infrastructure" project), and it's heavily regulated. Most countries demonstrably do not want to deal with it. Not to mention, nuclear isn't actually "sustainable" either, it just doesn't cause global warming.
- adrianN 7y agoI suggest we keep building normal PV and wind power until balancing the grid actually becomes a problem and then switch to more expensive technology. Gas peakers can actually be sustainable if you feed them with gas from power-to-gas plants btw. That's a convenient but expensive way for storing large amounts of power.
- Jweb_Guru 7y agoThat's not remotely practical and you know it. Let's not pretend that burning gas is a sustainable approach to power generation, regardless of how it's generated. I also don't really understand how you think we're going to suddenly flip a switch when "balancing the grid becomes a problem" if there are no alternative plant designs that people have actually constructed in large enough quantities to learn how to do it cheaply. In practice, if new nations onboarding with solar energy also add gas plants, we use more gas faster and contribute more to global warming. To me, it's pretty indefensible not to look at alternatives because they're more expensive.
- fphhotchips 7y agoSadly, the South Australian plant fell over financially: https://premier.sa.gov.au/news/statement-regarding-solarreserve https://premier.sa.gov.au/news/statement-regarding-solarrese... https://www.abc.net.au/news/2019-04-05/solar-thermal-plant-will-not-go-ahead-in-port-augusta/10973948 https://www.abc.net.au/news/2019-04-05/solar-thermal-plant-w... Given that the company website is refusing connections, I think it's likely dead, which is sad too.
- iagovar 7y agoAcciona also sold its assets in Australia to some other companies. It seems that CSP is no easy business. There are still a few in Spain though.
- taneq 7y agoCSP was very competitive 10 years ago before the bottom fell out of the solar panel market. Solar panels and batteries are now so ridiculously cheap that it's kind of silly not to just use battery backed PV solar.
- Gibbon1 7y agoI went and did a simple cost analysis and CSP is a dead end. A selling point is that it can store energy as molten salt. But when you run the numbers it's cheaper to heat the salt with PV solar energy. The problem is PV silicon wafers have gotten so cheap that rest of the systems costs are dominating. The difference in costs between a mirror and a PV panel isn't large. Then add the mechanicals to focus the mirror and the mirror costs more.
- pfdietz 7y agoConcentrated solar (thermal or PV) has not kept up with ordinary unconcentrated PV. And it doesn't work with diffused sunlight, which the latter collects just fine.
- mcv 7y agoWould it make sense to use ordinary PV cells and use that energy to heat molten salt to use during the night? Or are there too many efficiency losses there?
- ZeroGravitas 7y agoThere's a variety of storage ideas that just use any excess electricity e.g. Google X's Malta https://www.maltainc.com/our-solution https://www.maltainc.com/our-solution There's various prototypes in testing but at least for short term storage I think lithium batteries continual price drop is leading the pack.
- entropicdrifter 7y agoUnfortunately lithium is becoming increasingly difficult to obtain at scale. If graphite becomes commercially viable, that might be a good step up since Carbon is freaking everywhere.
- pfdietz 7y agoLithium reserves are more than 100x annual global production. Estimate lithium resources are something like 600x annual global production. There is no shortage of lithium, although there are temporary bottlenecks as capacity ramps up.
- xorfish 7y agoYou could use a molten salt nuclear reactor to heat the molten salt and use the heat when you need it. Terrestrial Energy will produce such plants commercially in the next 10 years.
- 7y ago
- reallydontask 7y agoIt would be pretty ironic if we got to the point of too cheap to meter with renewable based electricity. I realise that this is unlikely to happen but given the way prices are falling ...
- flgb 7y agoThe price of energy is definitely trending to free in Australia due to renewables —- just not for every hour of the day, yet. It’s not that electricity will be free when we get there, it’s just that consumers will stop paying for volume (how much energy did you use) and start paying for capacity / maximum demand (how much generation and network infrastructure was required to serve you).
- gwbas1c 7y ago> too cheap to meter I like to joke that electricity is already too cheap to meter; just look at all the free places to plug in a phone or laptop! IMO: Energy will never be "too cheap to meter," because someone will always come up with a way to consume lots of cheap energy.
- anon1253 7y agoThese types of systems are generally not competitive with PV for electricity, since they require a lot of concentrated solar energy thus limiting the options for deployment. However, they are great for generating heat. And with heat you can trivially desalinate water. Since trees can be a nice carbon sink, I've often wondered if this type of scheme might make sense: - Plant heliostat molten salt reactors in desert coastal areas (the concentrated solar kind) - Use generated heat to desalinate water (e.g. Multi-stage flash distillation) - Use desalinated water to provide irrigation - Plant (potentially GMO'ed) trees and other bootstrapping organisms to "green" the desert - Rinse repeat. The net effect is that you can potentially turn large swaths of desert into forests, with minimal external energy requirements (and potentially build an agricultural community around them, more or less self contained in terms of water/energy)
- flgb 7y agoSomething like this? https://www.sundropfarms.com/facilities/ https://www.sundropfarms.com/facilities/
- anon1253 7y agoYep! but then with fewer greenhouses so it has more of a https://en.wikipedia.org/wiki/Desert_greening https://en.wikipedia.org/wiki/Desert_greening effect
- dreamcompiler 7y agoCoastal areas (even desert ones) are often plagued by clouds, and clouds are a big problem for concentrated solar. This doesn't make your idea bad but it's an obstacle. Edit: Now that I think more about it, who cares if the heat is less reliable if all you're doing is distilling water? I need to take off my electric generation hat sometimes.
- anon1253 7y agoYeah it does limit it a bit, and if the goal is really to plant trees, then eventually it will be somewhat self defeating as forests cool the air and will generate clouds. On the other hand, desalination doesn't require extremely high temperatures, so it also kinda depends on the heat capacity of the molten salt. It probably also doesn't really have to be 24/7/365. But that's I guess a numbers game, so obviously it has to be modeled first.
- ReptileMan 7y agoAny reason we cannot melt the salt the conventional way while using excess electricity from renewables?
- asdfman123 7y agoYou mean melting the salt using electricity with excess capacity? I'm not an expert but I'm guessing we don't do this because the process of converting heat back to electricity is pretty inefficient. I think I saw someone say you only capture 1/3rd of the energy when you do that. However, if you're already starting with heat and need to convert it to electricity, you can just store it for later conversion.
- beat 7y agoEngineering-wise, no. It's just a question of conversion efficiency and cost. Use electricity to heat up a thermal store (molten salt), pull power from the thermal store via generator. But how much does it cost, relative to any number of other storage mechanisms that do basically the same thing - but particularly batteries?
- mikece 7y agoThe article mentions that the solar plant produces 110MW, which I presume is the thermal energy captured. On average converting thermal to electric is about 1/3 efficient so this two-mile-wide field of mirrors, all of which require electric motors, maintenance, and regular cleaning for actual operation, will produce maybe 40MW of electricity? How many megawatts would be generated if all of those mirrors had PV panels installed instead?
- Beltiras 7y agoI would think produces in this sense is the electrical output of the facility.
- mixmax 7y agoIn the daytime - probably more. In the nighttime - zero. that is one of the primary points of the system; You can generate molten salt when the sun is out, and genereate the electricity via a steam turbine whenever you want. This gives you the possibility to use this technology as a buffer.
- entropicdrifter 7y agoPerhaps a hybrid system would be the best compromise: use PV panels to capture electricity, but keep some salt molten using a combination of electrical heating and solar energy redirected off the panels' protective coatings, then at night you can supplement your batteries with the far longer lasting molten salt.
- strainer 7y ago> 110MW, which I presume is the thermal energy captured That would be deceptive advertising. 110 MW is the "nameplate capacity" which is basically its maximum sustainable generation capability. Its storage capacity is specified straight-forwardly as 1,100 MWh (electric) [1] https://en.wikipedia.org/wiki/Crescent_Dunes_Solar_Energy_Project https://en.wikipedia.org/wiki/Crescent_Dunes_Solar_Energy_Pr...
- acidburnNSA 7y agoPeripherally related: molten salt thermal energy storage is also something the nuclear industry has been considering using to help improve load-following economics of nuclear in a heavily variable renewable world. The idea is the nuke would run 24/7 at constant power but the load of the plant would be swapped from the thermal storage tank (in the day while solar is providing all the electricity) to the grid. Meanwhile, the load of the solar thermal system would be turned off during the day and turn on at night. This effectively doubles the power capacity of the nuclear plant to help fill in the duck curve as the sun sets. Doesn't help with seasonal renewable variability, but for day-to-day, it's pretty slick. The cool thing over TFA is that now it can work in many more geographic areas (e.g. where solar thermal doesn't make as much sense due to low intensity sunlight and there's more solar PV). Conventional nukes aren't great at going to these temperatures so this is mostly a thing "advanced" reactor people talk about. e.g.: https://inldigitallibrary.inl.gov/sites/sti/sti/6339782.pdf https://inldigitallibrary.inl.gov/sites/sti/sti/6339782.pdf Of course then there are molten salt cooled/fueled reactors, which is different from energy storage, making this topic even more confusing.
- ambicapter 7y ago> Conventional nukes aren't great at going to these temperatures so this is mostly a thing "advanced" reactor people talk about. Are the temperatures too high or too low?
- Analemma_ 7y agoToo low - water-cooled reactors are designed to bring the water in the secondary cooling system to just above the boiling point, because anything beyond that is wasted energy (the temperature of the steam doesn't really matter as far as driving turbines). So retrofitting a molten-salt system onto these reactors isn't very efficient and you would ideally want a new reactor design intended for that from the beginning.
- jandrese 7y agoIt seems like a serious materials challenge problem to come up with a molten salt that has the right nuclear characteristics (neutron moderating, etc...) and is not horribly corrosive/flammable/etc... Especially if your usage model is to let the salts heat up to outrageously high temperatures during the day and then draw them down over night.
- hawkesnest 7y agoMissing [2018] tag
- LoveDeathRobots 7y ago"The Crescent Dunes Solar Energy Project .. substantially missed its intended power production over its four-year lifespan by only achieving about 20% of its capacity on an annual basis" https://en.wikipedia.org/wiki/Crescent_Dunes_Solar_Energy_Project https://en.wikipedia.org/wiki/Crescent_Dunes_Solar_Energy_Pr...
- vondur 7y agoThere is one of these on the Nevada-California border outside of Las Vegas. I know they have had issues with the plant, such as the valley it's located gets foggy, which lowers the output. What's bad is the concentrated sun rays have been killing birds from what I've heard. Not sure of any easy way to dissuade birds from coming near it. It does look really cool off Interstate 15.
- effnorwood 7y agoIt doesn’t