Liquid Air Energy Storage: A Power Grid Battery Using Regular Old Ambient Air

When you think of renewable energy, what comes to mind? We’d venture to guess that wind and solar are probably near the top of the list. And yes, wind and solar are great as long as the winds are favorable and the sun is shining. But what about all those short and bleak winter days? Rainy days? Night time?

Render of a Highview LAES plant. The air is cleaned, liquefied in the tower, and stored in the white tanks. The blue tanks hold waste cold which is reused in the liquefaction process. Image via Highview Power

Unfavorable conditions mean that storage is an important part of any viable solution that uses renewable energy. Either the energy itself has to be stored, or else the means to produce the energy on demand must be stored.

One possible answer has been right under our noses all along — air. Regular old ambient air can be cooled and compressed into a liquid, stored in tanks, and then reheated to its gaseous state to do work.

This technology is called Cryogenic Energy Storage (CES) or Liquid Air Energy storage (LAES). It’s a fairly new energy scheme that was first developed a decade ago by UK inventor Peter Dearman as a car engine. More recently, the technology has been re-imagined as power grid storage.

UK utility Highview Power have adopted the technology and are putting it to the test all over the world. They have just begun construction on the world’s largest liquid air battery plant, which will use off-peak energy to charge an ambient air liquifier, and then store the liquid air, re-gasifying it as needed to generate power via a turbine. The turbine will only be used to generate electricity during peak usage. By itself, the LAES process is not terribly efficient, but the system offsets this by capturing waste heat and cold from the process and reusing it. The biggest upside is that the only exhaust is plain, breathable air.

A high-level overview of the LAES process.
A high-level overview of the LAES process. Image via Highview Power

The so-called gigaplant is being built near Manchester, UK and is supposed to be complete by 2022. It will be able to power nearly 200,000 homes for five continuous hours, even if they all switch on their kettles as soon as the credits roll on Coronation Street. This plant will have a 250 MWh storage capacity, which is almost twice that of the Hornsdale Power Reserve — that’s the chemical battery that Tesla stood up in South Australia last year after a particularly chaotic storm took out the grid for nearly 2 million people.

LAES has a lot of positives compared to other renewable energy sources. Some existing green energy schemes have steep storage requirements — pumped hydroelectric power requires a mountain, for instance. Similarly, you can store energy in a pile of gravel, but you need a pretty big pile of gravel. Liquid air plants have a small footprint to begin with, but they’re also modular, so they can be stacked for greater output. They can also be used in conjunction with other components like peaking plants, which only run during periods of high demand.

In light of the recent heatwave in Siberia, it’s great to see large-scale renewable energy projects like this. Highview Power has other projects underway in the UK, Europe, and the US, so maybe we can take a tour someday. While you wait for that write-up, check out the animated walk-through of an LAES plant below.

82 thoughts on “Liquid Air Energy Storage: A Power Grid Battery Using Regular Old Ambient Air

      1. I am sorry, but the Carnot cycle isn’t really all that applicable when using air as a large spring.

        If we were building a sterling engine, yes, it is very applicable.
        If we build a refrigerator or heat pump, then yes, its applicable.

        But now we are working with what effectively is a very large rubber band.
        Storing energy in compressed air can be done with nearly 100% efficiency if done “correctly”.
        Though, in practice there is a few compromises one needs to make, since we generally need a solution that can handle a usable amount of power.

        In short, the Carnot cycle isn’t really a good description of what limits we have to our efficiency here, since we aren’t trying to bring forth work from a thermal gradient.

          1. You can have heat exchange during compression, cooling it down so that you don’t have to deal with the additional pressure created by elevated temperatures.

            Thereby it can be stored for a fairly indefinite period of time.

            During decompression, we also exchange heat to keep it from cooling down. Thereby getting out all the energy from it.

            Now, running the heat exchangers requires a bit of energy, but running a water pump isn’t all that intensive all things considered.

            And our heat storage can be as simple as a nearby lake, mountain, pile of sand, or just a fairly deep well using surrounding ground as the heat storage. (all these will have roughly the year round average temperature, as long as they aren’t too small.)

        1. The Carnot cycle certainly does come into play, along with many other source of losses.

          This isn’t simply compressing the air, which might have merely bad round-trip efficiency depending on how the energy from the heat transfer is recovered. Liquefying air involves compressing and cooling it in the gas phase down to the condensation point, then removing the heat of vaporization. Using the atmosphere means that you are dealing with all the components simultaneously, which is going to have horrible efficiency in part because they liquefy at different points.

          On the plus side, storage capacity is proportional to the size of the storage tanks. You can theorize an inexpensive giant ‘battery’ with a small plant and massive tanks.

          1. CAES or liquid air storage isn’t a carnot cycle, because the input and output happens at the same temperature (ideally). The efficiency can indeed approach 100% if you store the heat lost during compression and liquefaction, and use it to warm up the expanding gas.

            The main efficiency loss comes from the fact that you can’t store the heat. You have to warm up the expanding gas with something else, like a heat exchanger with the ambient air, so you’re storing the heat in the atmosphere – but you don’t get it back fast enough. That’s why the input is always hotter and the output is always colder than the ambient, and the difference is lost energy.

      1. Quite prepared to eat a hat if they’re more than 60% efficient. Whereas pumped hydro is often 80% efficient, and made better when it rains. Pumped hydro is also much, much bigger. Big is good.

        1. Pumped hyrdo is good when you have a location you can do it. But lots of the world is inconveniently flat, will have huge evaporation issues, is too damn dry in the first place so you’d have to pump sand etc etc..

          CAES type systems, and battery/capacitor systems can be put anywhere as they don’t need to worry about geography much at all.

          CAES systems certainly can reach their stated efficiency – if liquifiing ones can I am more doubtful, but if the charge/discharge cycle is frequent enough its plausible as little of the thermal work done in compression will be lost.

          1. yes Dude pumped hydro is always bigger plant – using enormous areas. In terms of energy density this could actually be significantly better than pumped hydro – as in hydro the energy is in the dropped distance but the drop is rarely that deep and the volume of water huge!

            Build a CAES to the scale of hydro (which doesn’t make sense as the biggest benefit of this type of tech is you can site it where you want the power not where the geography dictates) and that order of magnitude difference goes away

          2. Well, the question is, can you really build CAES or liquid air storage at a truly large scale? There are always marginal costs, so what does it cost to go from megawatts to gigawatts?

            Mind you, if you want to compare scales, we really need energy storage technologies in the Terawatt-hours. That’s a thousand times bigger than ANY storage technology we have at hand – except synthetic petrochemical fuels.

      2. For comparison the electrolysis-liquidH2-fuelcell(air) cycle is ~12% round trip efficiency. Though it both the liquifaction and fuel cell(oxygen) could be significantly better in a single site storage scenario

    1. No matter how inefficient it potentially is, it has to be better than curtailing renewable energy generation. If you can save the energy that would otherwise be lost to curtailment, then it’s actually more efficient than not having it.

      1. Thats what I was thinking, it really doesn’t matter how efficient it is (to a point) if the energy used would be wasted/not generated in the first place. I meam when they have to shutdown wind farms because there is to much sun AND wind, surely this is better than nothing?

        1. Wind and Solar power occasionally output *more* than the current demand. This excess power is effectively wasted, without storage at 100% loss. Capturing even 50% of that water power represents a real savings, if the power can be rapidly redeployed into the power grid.

          It’s a battery, without expensive mining and toxic byproducts.

  1. Got to wonder how efficient such a thing will be in practice – normal CAES can be about as efficient as batteries in energy in vs out, but with significantly larger footprint (it can also be worse depending on many factors – how great a pressure you pump to and how fast you run it out etc etc.)

    I always saw Liquefied as a bad idea for static installations because its pushing compressed air into the inefficient zone (and volume of storage doesn’t matter so much)- good for vehicles perhaps – fast recharge, no emission onboard, and energy dense enough to probably get sufficient range..But if you can make use of the heat differentials created from its operation perhaps the efficiencies will get high enough to be superior or at least worthwhile.

    1. I wondered if liquid air could be a good way to get energy back from marine windfarms. There might be efficiency in compressing the air with the mechanical energy of the turbine, rather than having to gear it and convert to electricity in each nacelle. The sea would make a good heat sink for the compressed air and perhaps the natural pressure at depth could assist with the liquefaction process (perhaps pipes and vessels would not need to be as strong because of the reduced pressure differential). Liquid air could perhaps be transferred back to shore in ships similar to the LNG vessels we have now or perhaps even towed in insulated barges or bladders. This might save on the grid costs of connecting to the windfarms.

      1. iTM Power, who build electrolysers and Orsted are working on converting the electricity to Hydrogen in the offshore wind farm.

        The Dutch have stated, they will use redundant natural gas pipes to bring the hydrogen ashore.

        1. Thanks AnonW – seems like someone building a mechanical compressor wind turbine would be able to improve efficiency compared to that but perhaps they have a business model that takes advantage of the more developed market for hydrogen. Probably easier to strap an electrolysis plant to a conventional turbine than make a compressor turbine, sea-based air liquefaction plant and shore-side electrical conversion plant. I’d love to see it modelled or tried.

    2. Yes, compressed air as energy storage is really not a “dense” solution.

      But its very low cost compared to batteries. (since all one needs is a bit of high pressure rated pipe.)

      I do have to agree that liquefying the air seems a bit “overkill”, since it doesn’t really seem to add anything worth while as far as power storage on the electrical grid is concerned. A small form factor isn’t really needed.

      I myself think that the main area where CAES makes sense is if we build wind turbines with compressors in them, so that we don’t instantly get effected by a drop in the wind. (with sufficient amount of storage, it could last for a fair few hours if not days.)

      1. Not only is it overkill – but by having such massive thermal changes as well you are loosing energy that way – compress/decompress to the lower pressure levels and the energy lost is really minimal – going as far as liquification you have put in a lot of energy lost to the heat – for this system if its constantly in use so the heat/cold stores don’t radiate away much before its being run the other way and those stores are tapped again maybe the efficiency can be good. And worth the loss over other methods as the system has lower embodied energy than other systems (lower pressure CAES is cheap, but for the same storage is going to need alot more materials, and batteries are really complex energy hogs to make etc).

        Really won’t get good efficiency at all using it without that thermal store, or as is often done in current large scale CAES plants auxiliary heating by burning something.

        Not sure wind turbine CAES are workable.. A turbine when its actually working well produces a very large amount of power in a rather small area. Don’t know that you could come close to fitting a large enough tank to capture that sort of excess energy at each turbine – a more central system is probably more efficient.. But that would take some serious study to figure out -You have to consider that making a turbine compress air you can’t transmit that power efficiently – turn wind to electric and you can send it long distances with good efficiency. Turn wind to a different kinetic energy then empty that to make the electric just adds another conversion loss.

        To me CAES plants even a liquified one that doesn’t bother with thermal stores (or just was idle long enough it all radiated away) for more long term storage of the oversupply in large clusters makes most sense. Who cares if the energy stored in one of these only returns 20% (stupidly low estimate 40-50% is more reasonable) of the energy input if that input was days/weeks ago and it would have been wasted otherwise! Oversupply from the solar and indirect solar green supplies is better stored even at awful return rates than outright wasted.

        1. Yes, CAES is rather useful for long term energy storage. And primarily why I think the article shows an “overkill” solution.

          Though, in terms of energy storage in wind farms, I were more thinking that the energy is moved with a pipe. If all the power generated by the turbine is used to compress air or not is a different question. A hybrid solution is likely better. (Hybrid solutions taking the best of each thing is usually the better route in the end.)

          1. Trouble with hybrid solutions is you rarely get the best of each thing.. You hopefully get more of the good qualities of the parents than you loose..

            To me just use CAES systems in industrial parks and the ‘waste’ hot and cold suddenly need not be waste – so the system will get up to a theoretical 100% efficient all the energy that went it was usefully used or returned. And industries that use compressed air can be more efficient too as the large scale efficient pump and massive store will be way more efficient than the small shop compressors – and transmitting compressed air such short distances is easy.

      2. >A small form factor isn’t really needed.

        On the contrary. CAES is still difficult to place because on the larger scale it demands a volume the size of a cave. Liquid air can be built anywhere, and crucially closer to demand. Land prices are also a factor in energy storage costs.

        1. Look at the size of the current coal/gas powerplants the surge in green energy is closing down – not like they are small – finding space for this sort of thing is not hard.
          There is also the fact every large office complex/ tower etc can have its own UPS backup CAES as its so scaleable – give up a whole floor or cover the roof in the tank (and cover the tank in solar PV/water heater).

          Yeah CAES is big. But when people want power big is done when its approriate – and the facility to do CAES is itself stupendously cheap so it doesn’t matter if the land price was higher than other storage/generation options – you will still make heaps of return on the investment.

          1. They’re still small compared to the size of a proper CAES system, and you have to build these things in geologically stable locations. You can’t just dig a hole in some muddy riverbed and expect it to hold a cubic kilometer of compressed air.

          2. Also: the smaller your CAES system is, the less efficient it is, because it’s harder to maintain isothermal conditions with a small pressure vessel.

            A compressed air system the size of a car is hardly 10% efficient. The size of a locomotive, maybe 20%. A cavity the size of an entire apartment building might hold nearly isothermal conditions. The larger the capacity and the slower the discharge rate, the more efficient it is.

          3. That is all much more complex than just scale, and you don’t need a hole in the ground just arrays of pressure vessels – like this system here is stand alone, doesn’t have to be liquefied air you store in a tank.
            Engineering considerations in all things but used right a normal fire extinguisher sized pressure tank could be operated at high efficiency end and power something low draw for a good long while or set up for max peak power and get much less efficiency out but able to power a house for not very long at all. CAES systems can be set up everywhere in pretty much any scale you care to build them its such a simple technology – so distributed small/medium sized stores as grid ballencers and storage combined everywhere should work out far better than trying to stuff it all in one or two sites for a whole nation.

      3. Have you ever thought about using the waste heat in abandoned coal mines. In the UK, they are all filled with water approaching 15-20 degrees. London has just built a district heating system, that steals heat from the Underground in an abandoned station.

    3. As someone, who spent a couple of years doing the mathematics for a process engineer, I have a feeling, that this process could be open to a lot of powerful add-on improvements. When you’re designing a process plant that makes a product, you have to be careful, you don’t pollute the product with impurities. But here, your product is electricity, and so long as it’s got the right number of volts, amps and oscles, it doesn’t matter. Just look at some coal power stations converted to biomass, like Drax in England, they’ve got add-ons everywhere.

      We shouldn’t judge the efficiency of some of these energy storage systems for a year or so and we should constantly review all of them, as I’m certain some will improve efficiency in strange ways, soon after deployment.

  2. Compressed air energy storage is a very simple technology that can be used nearly anywhere in the world.
    It is cheaper than batteries, since all we really need is “a bit” of high pressure pipe. And the energy storage can also provide basic transportation as well.

    Though, one generally needs a fairly “slow”/purpose-built compressor/motor.

    Since if we rapidly compress a volume of x air into half the volume, then the temperature will also increase. And that increase in temperature leads to an increase in pressure above what we might at first expect. (Over time, the air will cool down and eventually we are left with twice the starting pressure.)

    If we on the other hand compress the air much slower, then the heat caused by compression will seep out into the surroundings, meaning that it won’t get as hot, and thereby we don’t waste energy fighting the additional pressure caused by the heat. (The heat is simply caused by the fact that a given volume of air has x amount of thermal energy in it, reducing the volume but keeping the thermal energy will directly result in a higher temperature due to the higher density in the thermal energy.)

    The same is true for decompression, but the temperature drops instead…

    This means that a shop compressor aiming at high flow isn’t going to be all that efficient as a compressor for energy storage.

    But, we could build compressors with inbuilt cooling fins (inside the cylinder itself) that exchange heat from the air much more rapidly (air itself is having fairly poor thermal conductivity), and thereby be able to run faster before reaching the same efficiency. (If one wants to comment that “one can’t place heat sink fins in a cylinder, since it will get crushed.” then please think. One set of fins on the piston head, that “meshes” with another set at the top of the cylinder, also we don’t really need extremely high compression ratios in our compression stages.)

    Another system for cooling down the air is to simply add more thermal mass to it, like spraying in a mist of water during compression. (obviously has its own downsides.)

    Or we could just run the compressor really really slowly… (But eventually our gears will eat up the efficiency instead, not to mention that electric motors are abhorrently inefficient at low speeds. So improving cooling is desired to be fair.)

    Storing the thermal energy we extract from the air is more or less important depending on where we run our setup. (If our ambient has a fairly constant year round temperature, then it isn’t majorly important.) Not that a drift in temperature by a few degrees has an all that major impact on storage efficiency.

    Another nice advantage of compressed air energy storage is that we could pipe out compressed air out to various industries, instead of them having less efficient compressors themselves. (I myself used to work in a factory being supplied by compressed air from an “off site” compressor facility, that also used compressors with rather good efficiency. Though, the factory had an air driven air compressor. Since the incoming air were around 40 bar, “a bit” higher than most air tools needs….)

    But in the end.
    Is compressed air energy storage the basket we can safely place all our eggs in?
    The answer is, no. Compressed air energy storage is just one of many promising technologies that has its set of areas where it makes a lot of sense.

    Advantages of compressed air storage is:
    1. Low cost per kWh
    2. Can be implemented nearly anywhere in the world.
    3. Low self discharge

    The disadvantage is that:
    1. Low efficiency at current compared to batteries. (I have seen some systems around 40-70% of the energy coming out compared to going in. But some more modern systems quote efficiencies towards 85%, so that is starting to be comparable to the lower end of batteries… But this is mainly due to point 2 bellow.)
    2. Its power in-/out-put is dependent on one’s compressors/motors. And this is where the real cost of this implementation exists. One can trade efficiency for power, and vice versa. Run the motors and compressors slower and they become more efficient, but if one need x amount of power, then slower isn’t always an option, unless one build more motors/compressors, but that costs more money…

    Though, when it comes to wind power, then the biggest advantage of compressed air energy storage is that a wind turbine could have the compressor in it. (Removing two whole conversions from our chain. Wind -> mechanical -> electrical -> mechanical -> compression. Could just be: Wind -> mechanical -> compression. (removes the efficiency of the generator and motor. If both are 95% efficient, then its still a loss of 9.75% of the energy.))

      1. Using water as a heat sink has some practical advantages, but generally we can use any thermal storage medium, be it a mountain, a pile of sand, or a lake, or even a hole drilled into the ground.

        But generally we want to use the same heat sink for both compression and decompression. (Or we could use a cooler one for compression and a warmer one for decompression, that will lead to a slightly larger power output.)

        But the main advantage of having a compressor in a wind turbine is to get rid of the additional conversions.
        Also, a lot of wind farms use underground AC transmission lines, and these are rather inefficient due to capacitive couplings between the conductors.

        Also, when it comes to trying to store compressed air in the wind turbines and move them with ships, this will likely be rather inefficient. (the amounts of power a wind farm generates will fill up a ship in very short order. A pipe is logistically easier.)

        How the friction losses of the pipe compares to resistiv losses in a cable is another good question. (But a nylon covered, aluminium coated, fiber glass pipe (if looking at the layers from the inside out.) of a large diameter should be both exceptionally gas tight, and have very low friction and likely be cheaper than a cable with similar resistance.)

        Otherwise yes, putting a compressor in wind turbines makes a lot of sense. (and why I wrote two comments about it before even seeing yours.)

        Why wind turbines doesn’t currently do this is likely due to the additional cost of having both a compressor and an air motor. (since the air motor would likely be fairly expensive all things considered, and large…) But I have seen some companies/researchers explore the concept, mainly since energy storage is frankly needed for wind/solar to be practical as a main source of energy. (wind power has historically just been used as a supplement, mainly by land owners seeing it as a secondary form of income. (unless we go back to the time when wind mills were everywhere.))

        1. I recognise that the liquefaction might involve some losses but it may be worth it for storage and transport. I was thinking that you’d probably use the sea at the shore as the heat sink for conversion to electricity, so it would be complementary again.

    1. Use ocean-based wind turbines to drive a liquid-piston compressor, that stores air in under-sea bags, and also gradually pull them lower in the water to help with compression.

    2. Good summary, and a key point that we actually need and want diversified energy storage. They all have pros and cons plus some have strange side advantages.

      I used to work for the Dearman Engine Company which was a Spinout from Highview nearly 10 years ago now (they’re still going as Clean Cold Power). Actually, I think the Dearman engine concept is what kick started Highview – look up Peter Dearman.

      Dearman were using liquid air or liquid nitrogen in small/medium scale engines for applications requiring cold and power e.g. Data centres, busses in hot environments, and our key application, transport refrigeration. The Dearman Engine expanded the compressed nitrogen in a piston engine with a direct contact heat exchange fluid in order to achieve closer to isothermal expansion and therefore improve the expansion efficiency.

      Liquid nitrogen provides a great energy vector for transport refrigeration. I developed all our engine test cells and control systems, and we had a number of field trial vehicles successfully running around daily doing food deliveries.

      There were lots of additional advantages which really interested our customers: zero emissions, very low noise, low operating costs, very fast ‘pull down’, backup cold solution.

      We were using the cold in the compartment, but also cooling down the condenser of a vapour cycle to improve its coefficient of performance.

      Sounds like you know a bit about the processes so you might already be aware of some of this!

      1. There are significant refrigerated storage facilities at many ports where food is imported/exported. Might be a good place for liquid nitrogen to be converted to gas and electricity. Could be a good fit if we started liquifying the air with marine wind turbines and bringing it ashore in tankers (or pipes if well insulated?). From what you know of these systems, might that be feasible/economic Scott S?

  3. Great video, I learned a lot about a very interesting and important topic. I do have one major criticism of the production, however.

    The background music so loud that it drowns out what the speaker is saying. This is a very real problem for many people, especially those sensorineural hearing loss, the kind you get from listening to music too loud, or from just aging. If the background music is knocked down by 10 dB or so comprehension will improve markedly and people everyone will get a lot more out of the video. Masking of high-pitched voices by competing sounds of similarly pitched and lower pitched sounds is a very real and well understood phenomenon that every first year audiology undergraduate studies. Please consider re-recording the music on this video at a more tolerable level. Thank you.

  4. What this needs is deep equatorial seas to be efficient. Sink heat into cold deep waters, source heat from solar. Yes it could work well other places but we will have to build infrastructure to transfer thermal gradient, where as now we just waste it to the atmosphere.

    1. Don’t like the idea of deliberately dumping heat/cold into deep water – we really don’t understand that ecosystem enough and could bugger up ocean currents too, yet we know the sea is crucial to our survival. Though that all depends on the scale – if you are talking 1W per 100000 litres its so low the water will conduct it away, radiate it into the atmosphere doesn’t upset the local balance meaningfully…

      Instead of wasting it as these systems can be sited on industrial parks full of business that need either heat or cold use the ‘waste’ locally and suddenly the system could even become 100% efficient – all the energy it takes in is usefully used or returned.

      1. Heck many business need compressed air too – much more efficent to run a short pipe to them than convert compressed air back to electric, transmit it to them, have them run a probably much less efficient compressor to turn it back into compressed air… All those conversion losses!

        1. Some larger factories and industrial areas have compressed air as a service, just like electricity, gas and water.

          Apparently, when a factory starts needing more than a few kW of continuous air, then efficiency starts to be a lit bit more noticeable to one’s wallet. (if one has a compressor that is only 40% efficient and one use 10kW of air continuously, then one is paying a lot for it…)

          Now most smaller places doesn’t use enough air for efficiency to be important, but rather look at max continuous flow, since waiting for the compressor to build up pressure is an annoyance.

          And this is another advantage of an air grid, a shop can suddenly have more flow than they could ever need….

          Now a compressed air grid is likely only logical in industrial areas/districts, and the incoming pressure would be higher than the normal 8-12 bar. A factory I worked at used 40 bar for air distribution, then they ran air driven air compressors to convert that 40 bar into a more usable pressure. (The efficiency of the air driven air compressor must have been rather good.)

          1. I can see why they wouldn’t Dude – when regulators fail they tend to fail open, which would wreck via overpressure something down stream.

            Of course once you take some of that energy out and use it for something then the regulator becomes more likely – a few BAR over probably doesn’t cause any harm in the short time it takes to shut down and fix but 4x or greater the pressure you are looking at ruptured something…

          2. Pressure regulators would get covered in plenty of ice if dropping 40 bar down to 10 bar for supplying a factory with literally hundreds of air tools used fairly continuously on a manufacturing line. (not to mention all the other air powered lifting equipment, and venturi driven vacuum systems too..)

            So at least a couple of thousand liters a second.
            The 40 bar pipe weren’t small btw. (and there were more than one entering the building.)

          3. Not to mention the rather huge loss in overall system efficiency if we used a pressure regulator to drop the pressure by around 75% without getting any useful work out of it.

            Then smaller local compressors would make more sense than a large off site compressor facility. (And the reason for the larger facility were primarily for increased power efficiency.)

          4. >Pressure regulators would get covered in plenty of ice

            So would an air turbine. The trick is to add a radiator and blow enough ambient air through it to keep it near the ambient temperature.

          5. >Not to mention the rather huge loss in overall system efficiency

            Pressure regulators don’t work like linear voltage regulators. The pressure is exchanged into volume – energy is conserved as long as you maintain isothermal conditions, which is accomplished by adding a big radiator after the regulator to warm up the lower pressure air back to ambient temperature.

          6. The Air powered Air pump won’t get covered in ice easily Dude – its very nature is pulling in ambient air and making it smaller and hotter. Its already got by its function the effective heat regulation locally. And its easy stage them up, you will get some additional friction losses, but those aren’t going to be meaningful.

            Trying to use a radiator to radiate away KW of cool every second just isn’t feasible the radiators would need to be far too large and actively blown or heated to manage that level of heat change at such high flow rates – which is incidentally why lots of CAES plants burn crap to get better efficiency back out.

  5. It seems like it wouldn’t be that hard to do fractional distillation as well and strip out useful gasses. I mean, air separation plants already exist, and it’s only a tiny loss of stored energy to retain anything other than oxygen and nitrogen.

    1. Seems like an opportunity to separate out CO2 for sequestration as well, maybe cash in on a carbon tax credit of some sort. My understanding is CO2 scrubbers usually need high concentration.

      1. The CO2 pretty much has to be separated out. Nitrogen and oxygen are liquid at equivalent temperatures, but CO2’s frozen by that point. I’m a bit surprised they don’t mention that, although it’d be a marketing point more than anything else – volume-wise it’s completely negligible. Air separation plants filter out CO2, for instance, but they never try to talk about that as a benefit.

        The argument everyone’s making here is “the efficiency sucks,” but that somewhat misses the point – it’s a battery. The efficiency doesn’t matter *that* much: air separation plants waste *all* the energy that goes into making the gasses, so in effect if you combine the two situations you might win out overall depending on scaling benefits. In fact if you do fractional distillation and split out the nitrogen/oxygen, you effectively have “infinite storage” since you can almost certainly sell off the liquified gasses for profit anyway.

        I’m guessing there’s some sleaze in the way this is done that makes this impractical, though.

      2. Exactly! You could use the waste heat to keep a liquid salt electrochemical cell and the excess energy to transform c02 into graphite and thus sequestrate the carbon from the atmosphere.

    1. The ability to go “boom” is limited by the heat transfer rate. If you open a vat of liquid nitrogen, it won’t instantly expand by a factor of 700 because it would take a lot of heat to do so. That’s basically the reason why you can have liquid nitrogen in an open cup in your hand anyways.

      1. To elaborate: how the liquid air system works, they put the liquid air through a radiator and blow warm ambient air through it, which causes the liquefied air to boil out inside the tubes. The pressure of the boiling air is used to drive a turbine to make electricity. When you stop the fans, the radiator goes really cold and the boiling stops.

      2. I meant “boom” more in an oxidation sense rather than a pressure differential equalizing. I’ve seen the popsicle stick of science “instant oxidation” experiments done with 100% O2 but I can’t imagine liquefied atmosphere at 20% O2 is much kinder to cellulose.

        1. Well, yeah, if you literally pour liquid air into a fire then it’s going to go boom. Just dropping a popsicle stick in isn’t going to cause an explosion though – just some bubbling as the stick freezes. One corner of the fire triangle is missing: heat.

          1. H20 is an ideal explosive mix of liquid air, but a mist-burst of it into a car fire, snuffs the fire. New-ish portable fire-fighting device… so, please elaborate.

  6. I wonder if could you create a gas mixture of several dozen different gas molecules with a large range of different latent heat of vaporization/condensation temperatures. Say 2 to 20 degrees apart, to store more energy in a smaller volume. It would need to be a closed system, which would cause lots of headaches, and selecting gasses that did not react with each other would also be a difficult problem. And that is even before you start planning for all the of failure modes.

    And could such a system, if it worked, provide any useful advantage over the CES/LAES. Like would a higher density of energy storage be worth the additional complexity.

    1. I can’t see any way you’d get meaningful improvements in energy density or efficiency by a mix of gas. I’m sure there will be some gasses you can compress more effectively or get energy out easier – for one thing if you just use a single gas you can optimise your pump/turbines for that gas, and be sure of longjevity by selecting non-reactive materials for that gas.

      But I don’t see it being enough to be worth the effort – especially where this system can and should as a by-product of working be set up to extract say Argon and Carbondioxide for all those industies that need them. Maybe I am missing something but I don’t see any loss and infact big gains in overall efficiencies doing that – not burning power somewhere else to make liquid air, and only loosing a small volume of the air you liquified – don’t expect it would be an always on operation though – when highest power draw so highest flow is required its all just going back out . Heck just catching that carbon and releasing it in greenhouses would effectively pull the concentrations in the air down as a useful byproduct of storing electric.

  7. Dig a hole in the ground put a winch and generator with a concrete block. Or use your building as as the wait like the floating homes up and down. Simpler and more efficient

  8. The cheapest way to build CAES is in a natural cavern or a mine. It really doesn’t make any sense on a small scale because the unit cost is so high and the efficiency is so low.

    1. Bigger volume for the same ballpark of efficiency as batteries. But its much much cheaper, both to build and maintain, so over its lifetime will actually work out considerably more efficient.
      Pretty much the best argument, though there’s also the fact it doesn’t need exotic rare materials the way batteries do. I know I don’t want a dive tank on my back to power my personal electronics thanks – I’d rather those materials went to batteries for devices that need the energy density.

  9. This seems like an idea taken from Steamboy (Anime, 2004). I still think that storing heat and trying to make the heat-to-electricity conversion more efficient would be a better path (developing a meta-material for instance). And/Or maybe a hybrid, after all, when air is decompressed it cools its surroundings, making a higher difference between the lowest and highest temperatures…

  10. An important consideration is the absence of water used to generate steam.

    This is a hidden cost in any other gas turbine driven conversion to electrical current.
    Pursuit of perfection is the enemy of progress.

    Plenty of positives abound in a scalable approach that doesn’t require bespoke solutions.

  11. They’ll want to build a plant like this in a place with very, very dry air. At 50% humidity at 100 F, approximately 28% of the energy used to liquefy the air will be expended just removing the water from the air. I doubt that they will incorporate a boiler or flash vaporizer for the water in the expansion cycle, so best not to throw that energy away in the first place.

    There are some other interesting problems to solve: A leak or failure of the storage tank will result in an uncontrolled boiloff, which could result in fractionation, which will eventually provide a very oxygen rich environment that will for sure cause a fire that will burn anything.

    The process is very much Carnot cycle limited, which is why there is the need to have the heat reservoirs, but like the video says, any low-quality waste heat source that you have lying around can be used during the expansion cycle.

  12. Greetings from Manchester. Haven’t seen it yet, but this is quite exciting. To those who think it’s inefficient, of course it is. It just has to be efficient enough to capture enough of the excess power to get through the day/night. Compared to hydrogen electrolysis, which is being trialed in Orkney, 60% is a step up.

    I do wonder what the heat storage is going to be made from (cryo is a lot simpler to imagine) and what the working liquids will be.

    Interestingly, a quick search points out that the chief cost of industrial compressed-air plant isn’t the equipment, it’s the electricity used to compress the air – 72-73% of the cost. If they can use the heat/cold storage to reduce that, and be paid for outputting an appreciable fraction at peaker plant rates, then even if they buy the electricity on the open market (even better, if they have a purchase agreement to take excess energy from a wind farm), it could be a very good business for them: the Horndale battery made a million Australian dollars in two days. If you can emulate 60% of that performance with simple plant equipment, you do it.

    Hell, even if the power storage doesn’t work out, the technology to make compressed air more efficiently would be a game-changer.

  13. Couldn’t you just use a bunch of small tanks at 5 bar each to solve all the problems y’all are debating? Would be cheaper than a big tank and you could do thing’s like just use the first tank then the next when one gets low or use all at the same time.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.