Construction begins on world’s biggest liquid air battery | Hacker News Reader

Construction begins on world’s biggest liquid air battery

theguardian.com

96 pointszeristor98 comments

The technology sounds fantastic. The synergy with heat power plants is particularly cool (heh). The same goes for desalination: waste heat can be reduced by recycling heat between the processes, increasing the overall efficiency of the system.

Now for the rant: reputable news organizations need to stop enabling "creates x jobs" propaganda through repetition. It doesn't create 200 jobs, it briefly employs 200 to build the thing. Maybe 5-10 full-time jobs will be "created" to maintain a giant mechanical battery. This lie is particularly popular around pipeline projects. "You don't think we should spend billions of dollars on this pipeline? Do you want tens of thousands of imaginary people to lose their imaginary jobs?"

End rant.

I wonder in an era where clean energy costs fall more than 10x whether it is possible to channel surplus energy into desalination and pumping water into deserts near the coasts. Like a slow geo engineering project. Combined with future developments in vertical farming, it may make sustainable cities possible in places like Middle east, parts of South Asia and Africa

I recall reading about a proposed solution to global warming that included digging a channel from the Atlantic into the Sahara. Big sections are below sea level. So we'd create a new ocean/sea effectively, cool the local climate, and significantly reduce hurricanes on the East coat of America.

The Amazon rainforest relies on annual global sandstorms transporting nutrient-rich (phosphorus) sand from the Sahara to South America. If you flood significant parts of this desert, you'll disrupt this process.

A source: https://www.nasa.gov/content/goddard/nasa-satellite-reveals-...

I think the "creates 200 jobs" usage is completely fair here. The thing is due to be operational in 2022, so that's two years of work for 200 people, and the ongoing operation jobs will be "a few dozen", not 5-10. The article makes all this clear.

More broadly, infrastructure jobs are still jobs, even if they only last a few years instead of indefinitely. The average tenure of a software engineer at a big tech company is less than that; is it fair to say they don't have jobs? It's not like we're ever going to run out of projects to improve the world, so these 200 workers will hopefully finish this job and move onto the next "job-creating project".

==The average tenure of a software engineer at a big tech company is less than that; is it fair to say they don't have jobs?==

I don’t follow, after a software engineer leaves the big tech firm, the job still exists. They hire someone else to do it. In this instance the job doesn’t exist after 2 years. It’s also probably not 2 full years for those jobs as they are typically brought in with teams to perform their function (dig, foundation, build, framing, electrical/water/gas, finishing) then move on.

Software projects finish just like civil engineering ones do.

Hiring a software engineer as a full-time employee for a temporary software project, then firing them, is a poor use of company resources.

If you are talking about a contract software engineer, then you are correct that the worker rolls off. We typically don't see those considered "full-time jobs created" because they have a well-defined term and are never employees of the company.

Compressed air storage seems like a really appealing method, especially considering the low carbon footprint and massively longer lifetime than lithium batteries. This post covers it quite well https://solar.lowtechmagazine.com/2018/05/ditch-the-batterie... , I’d be really keen to see if compressed air can also be successful for independent households looking to go off grid or reduce dependence on the grid.

It has its own issues, not the least of which is safety. There is lots of energy in compressed air, that's the point. I would look heavily into the risks of structural fatigue. A lithium ion battery might catch fire, but an "air battery" can go boom.

Isn’t this the issue with any large battery? Water in large volumes can cause a lot of damage too. I suspect it’s hard to find a large battery that isn’t scary when mishandled.

Compressed air is much more dangerous than most other storage methods though. The failure mode tends to release stored energy very quickly.

Yes. Look for the vids of scuba tanks gone wrong. I'd rather drive around with a propane tank in my trunk than a 5000psi air tank.

Depends on the materials...

TBF I'm pretty comfortable with a proper steel tank.

But, nowadays I'd be afraid to use any of the fiber wrapped ones. I remember in the middle of the 2000s when a lot of new companies jumped into the Fiber-wrapped tank market... and shortly thereafter there were a notable number of recalls in/around the bonding between the fiber and metal layers of the tanks.

And, mind you, this was all in the paintball industry, which was a field where many did not have an understanding of the dangers of a piece of fiberglass with 800-5000PSI pushing it somewhere.

Happened more than once:

phone rings

Boss working at field nearby: "Hey, someone might be coming across the street to try to fill a tank that we refused to fill because it was (damaged/out of hydro)"

Me: "Yeah, I turned them away 5 minutes ago."

Customer walks in

Customer: "Hey the field said they were out of air but otherwise they would fill this."

Joules are joules. A "grid scale" compressed air tank going pop is going to do a lot more damage in the immediate vicinity but won't do anything 50mi away. You can't say the same thing about a dam burst.

Just because it's dangerous in a way you're not used to doesn't mean it's more dangerous.

>Joules are joules.

When talking about battery failure how quickly the energy is released is important.

>A "grid scale" compressed air tank going pop is going to do a lot more damage in the immediate vicinity but won't do anything 50mi away. You can't say the same thing about a dam burst.

If your comparing it to an entire dam, then your looking at underground storage of compressed gas in caverns. For that I agree both are dangerous in different ways.

But the context I was talking about was compressed air tanks which are more comparable to chemical batteries and gravity storage. Catastrophic failure of compressed air tanks is generally going to be more destructive than many other similar alternatives.

>Just because it's dangerous in a way you're not used to doesn't mean it's more dangerous.

A large lithium battery is much less likely to level several buildings than an air tank storing a similar amount of energy. I'm not saying that the dangers of compressed air can't be engineered around, but by almost any reasonable, objective metric you could come up with, compressed air is more dangerous.

Given that anything that contains energy can go boom, I think of all of them this one doesn't sound too bad. Probably local casualties but no risk of cancer for the next century, or ground pollution for the next decades.

Literally anything you store energy in is going to have the same safety issue. You can muck around with space efficiency and peak wattage but at the end of the day you've got the same amount of joules that want out and if they get out in an uncontrolled manner it will be bad.

This plant is not using compressed air, it's using liquified air.

While it is appealing compared to many alternatives and is more portable for smaller settings, when used on an industrial scale like this giant battery, I wonder how it compares in overall utility to gravity batteries [0].

The nice thing about gravity batteries is that they can form a closed loop with water supply, too, pumping reclaimed water back up into reservoirs when there's excess power. I'd expect compressed air to have much higher energy density, but it would be interesting to see if anyone has the numbers.

[0] https://en.wikipedia.org/wiki/Gravity_battery

In my country, there are economic challenges: water isn't available everywhere, diverting water is not cheap, and it messes with water sharing agreements, which are a political issue in themselves.

As I once heard from a Wyoming rancher, “Whiskey’s for drinking, water’s for fighting.” Water rights are treacherous to navigate.

It’s gonna get worse, too, with upstream countries building dams at the expense of downstream countries.

Gravity batteries only need mass, it needn’t be water: https://youtu.be/gn5AM75AGvw

Pumped water batteries are geographically limited, since they can basically only be installed wherever you could build a dam anyways.

Yes, but you can be moving rocks up a hill on a train or cable car. Or concrete blocks up a crane. Geographic limitation solved.

Rocks and hills are equally geographically limited.

Cranes would work, but I think you’re ignoring the massive difference in mass between what the biggest cranes in the world can lift and even the smallest reservoir.

some people are trying this:

https://qz.com/1355672/stacking-concrete-blocks-is-a-surpris...

I believe there's a beta test facility under construction now.

Still pretty small compared to pumped storage - they give a value of 20 megawatt-hours, the Cruachan pumped storage scheme (which isn't particularly large) stores 7.1 gigawatt hours:

https://en.wikipedia.org/wiki/Cruachan_Power_Station

For comparison, the liquid air battery in this article is rated at 250mWh.

I assume you meant Mwh? :-)

Are they efficient ?

Pumped hydro hits about 80% efficiency. So if you’ve got the right geography for it, it’s one of the best choices available.

Those methods are actually quite similar, they're all ultimately related (limited by) tensile strength of materials. So costs and specific energy densities (per construction material unit) are similar. An important figure is cost/tensile strenght I guess.

(Compressed air: limited by container tensile strength; gravity batteries: limited by strength of cables)

I suspect even the constants involved are the same (given the materials are almost uniformly under nominal load), although I don't have time to investigate right now (a good curiosity research topic!).

They are storing liquid air at low pressure, so container tensile strength is not a limiting factor at all.

I believe these ones are not limited by tensile strength: https://interestingengineering.com/concrete-gravity-trains-m.... They are a bit limited by available geography, but far less than say pumped hydro.

Indeed! In this case it's the integrity and friction of the soil (of the hill) that's keeping the potential energy contained, and this hill soil is "free".

look into energy stored on energy invested

https://en.wikipedia.org/wiki/Energy_return_on_investment#ES...

according to the wiki, pumped hydroelectric storage has a ratio of 704 where as compressed air energy storage has a ratio of 792.

Several startups have tried and failed. Most notable was Lightsail which raised $80M from Gates, Thiel, etc.

IIRC LightSail was founded by our own very brilliant Dani Fong.

Here’s an article with some of the technical details of chilling the air to liquify it:

https://www.bbc.com/news/science-environment-19785689

“IMechE says this process is only 25% efficient but it is massively improved by co-siting the cryo-generator next to an industrial plant or power station producing low-grade heat.“

“More energy is saved by taking the waste cool air when the air has finished chilling, and passing it through three tanks containing gravel“ as a thermal store.

“their kits could be up to 70% efficient, and IMechE agrees this figure is realistic.”

Sorry for the rant but this is just so poorly designed that I don't see any relevant content when I open the link without scrolling right away:

With adblocker: https://imgur.com/a/fWuoJEi

Without adblocker: https://imgur.com/a/cvmu3dm

Plus ublock origin says it blocked 30% of requests. I'm all up for paying for good journalism but part of me wonders their paid customers aren't getting a great experience either.

I've found the Fanboy Annoyance List to be pretty effective vs the sort of GDPR cookie acceptance things you see in your blocked version: https://easylist.to/easylist/fanboy-annoyance.txt

I mistakenly interpreted the headline as a battery with some kind of liquid electrolyte and outside air as a reactant. Kind of like a https://en.wikipedia.org/wiki/Zinc%E2%80%93air_battery . But nope, it's just the liquefaction of air, no chemical reactions involved.

Yes. The term "battery" is being thrown around too much. Are hydroelectric dams to be called "rain batteries"?

Yep. A lot of people call pumped storage "water batteries" or the like[0]. It baffles me as well, but now that language is evolving to redefine a battery as pretty much anything that generates electrical power than can be recharged with electrical power, I guess that's the world we're walking into and I should get used to that.

[0] https://www.hydropower.org/sites/default/files/publications-...

Have you seen where "battery" comes from? Electrical batteries are called that since they were made of sets of electrical cells working together, analogously to a battery of artillery.

Yes I'm aware of that - good point. I could have included that in my post above to highlight that this word itself has gone through plenty of language change already, from meaning "a number of pieces of artillery used together" to "electrochemical energy storage" to "energy storage." While it keeps most of those original definitions as well - including "to strike" etc. The only definition I don't think has really survived to the current day that I see when looking at its etymology is "metal articles wrought by hammering." Wonderful word.

Government cheese can be thought of as a milk battery.

all food is batteries

Likewise, I was expecting some kind of redox flow battery. Compressed air is heinously inefficient due to thermal losses (and pumping losses to a lesser extent) but I guess if the energy is literally free (eg. solar or wind overproduction) then that matters less than the cost per kWh stored.

Wondered what they were doing with the thermal energy from the process. The manufacturer has a video here that helps explain the technology a little bit: https://www.youtube.com/watch?v=kDvlh_aG7iA

The "air" gets stored in liquid form... but what I don't understand is what parts of the air? The gasses in air mix as gasses, no so much when liquid. Will the tanks include a layer of liquid oxygen? What prevents those layers vaporize selectively as the tank temperature is raised again? That would seem very dangerous. Or am I misreading this and they are only liquefying nitrogen and aren't playing around with liquid O2?

This article says they remove CO2 and water because they freeze solid, and it implies oxygen remains in the mix: https://www.bbc.com/news/science-environment-19785689

I am guessing they pump the liquid air and then “boil” it, so there is no risk of selective vapourisation as they are not boiling the bulk liquid within the tank.

The "concrete battery". Or, because it runs on gravity, the "earth-powered"... relativity-powered "Hawking battery".

https://newatlas.com/energy-vault-concrete-tower-battery/571...

(It stores energy by stacking rocks.)

If they did something with that CO2, the whole setup could be carbon negative.

Dry ice is actually a precious commodity among algae growers - another alternative energy source - algae diesel. Or any frozen food dealers - and sometimes they run into shortages: https://www.marken.com/alerts/dry-ice-shortage/

Not just alge but ag in general. CO2 makes plants grow fast, farmers would buy it for greenhouses if it was produced on a large scale, although finding a way to stop it going back into the atmosphere would be better.

They aren't going to be handling large enough volumes of air to make this worthwhile. Co2 is such a small part of air that direct air capture of it requires massive fan farms.

Why wouldn't they stay mixed as a liquid? They're highly miscible, if you leave liquid N2 out it can actually enrich with oxygen with time as oxygen dissolves into it. You need to distill them to separate oxygen and nitrogen cryogenically. That happens in the boiling off part, but as another commenter mentioned you could pump out what you want to boil to avoid boiling the bulk liquid and causing the remaining liquid to be enriched in oxygen.

It doesn't sound like they are doing much separation though maybe they should for a secondary income stream. I wonder how the scale of this compares to traditional cryogenic ASUs.

> will use spare green energy to compress air into a liquid and store it.

Is that really right? It's not just green energy sources that benefit from energy-storage. Fossil-fuel plants and nuclear plants prefer to run at a constant output, but demand varies depending on the time of day.

See the 'Duck curve' https://en.wikipedia.org/wiki/Duck_curve

This seems like useless pedantry on the word "green", no? The electric field is not coloured, you can't really attribute grid power to a specific source sensibly.

Coal is almost entirely gone from the UK. Nuclear does indeed prefer to run at constant output, but almost all the remaining fossil generation in the UK is CCGT which can ramp up and down fairly quickly.

Cool. Technology like liquid air, compressed air, electrolysed hydrogen etc. has a very different use case to lithium ion batteries.

Batteries are good for short cycle times, if you need to balance your grid and even to some extent if you need to correct for diurnal patterns, you want batteries. The high capital cost is compensated for by the high round-trip efficiency.

For seasonal storage or even inter-year storage, batteries do not work. You spend a fortune on them and then cycle the system once a year or less.

The advantage of these other systems is you can decouple three elements:

The conversion in

The storage

The conversion out

and match the capacity of each to what you need. For long term storage of hydrogen for instance, you want relatively small electrolysers with fast response times so that you can use excess electricity during sunny windy periods, a massive geological storage that can hold a year+ of production, and modestly sized fuel cells to turn it back into electricity in the winter (in Europe anyway) with good heat recovery.

In the case of this technology, you can again size the charging, discharging, and storage separately. The scaling for tanks is much cheaper than batteries.

Flow batteries use a similar idea by separating out the charging/discharging membrane and the electrolyte tanks but I don't think anyone has gotten satisfactory performance from them yet.

You can't build a lithium ion battery where you specify the storage capacity separately from the charge and discharge rates with anything like that kind of freedom. You can adjust the battery chemistry of course but nothing like the same.

This is why pumped hydro using existing dams is so nice. I think constructing enough storage to handle months of lower solar output will be difficult.

Phase locking human activity, including economic to the energy cycle would obviate the need for many technical solutions. We should just hibernate.

Pumped hydro using existing dams is brilliant and should be maximally exploited but we already have dams on all the really good sites. We can improve by adding the pump-up stage to many of them but there is simply no scope for massive expansion in total reservoir / energy storage capacity.

I was looking into this briefly when I was investigating molten salt batteries. Can anyone who has expertise in this field tell me a bit about why molten salt batteries aren't more common? From what I can tell, the materials used are more renewable than Lithium Ion batteries, and with comparable (potentially higher) density. I can see why molten salt isn't as common in the more portable form factor, but it seems like it would be a great fit for large energy reservoir/grid type applications.

I believe the main use of molten salt batteries is for high heat applications e.g. in solar collectors

The heat is essentially "free" which makes the molten salt heat retention properties perfect for storing energy overnight.

Stationary batteries in general aren't common. We don't have enough transient power generation for them to be very lucrative. (What, is very obviously going to change in the near future.)

Added to that, for molten salt batteries, there are old designs with bad cost and density properties and new ones with dangerous materials and patent minefields. I don't think anybody was ever able to sell a Na-S design without being destroyed in some court.

I wonder what is the round-trip efficiency for this cycle. I read that you can improve it by matching with large scale but low-grade heat source.

The article says 25%.

That’s a pretty tough sell versus pumped hydro which is generally 70%+

The manufacturer claims 60-75%, with a lifespan of 30-40 years https://highviewpower.com/technology/

That's not increasing the efficiency of the storage, though. It's just using otherwise wasted heat from a different process (which could probably be otherwise used for cogeneration or something).

In hot countries it might be efficient if the chilled exhaust is used for air conditioning or something, I guess.

Per https://kth.diva-portal.org/smash/get/diva2:1216213/FULLTEXT... , the theoretical maximum seems to be ~70%. Which is not bad but not great either if you compare it to Pumped hydro storage.

The power cycle used here is quite an interesting one.

The heat from the compression cycle and cold from the expansion cycle are stored separately and re-used which substantially increases storage efficiency.

Waste heat from co-located industrial plants is used to further boost expansion performance. A lot of "low grade" industrial heat is available but when you're re-heating from cryogenic temperatures, 100 C is actually high grade heat.

I suspect that the time over which it's efficient to store is limited by losses from the temperature stores.

> Waste heat from co-located industrial plants is used to further boost expansion performance.

Let's not greenwash this. The "waste heat" is fossil fueled.

That's not "greenwashing" because it's obvious. Thermal input in industrial processes is essentially all fossil fuel, high temperature input is 100% fossil fuel. To the extent that we need these processes, we may as well get as much bang for our CO2e buck as we can. Even if and when these processes can be driven using hydrogen combustion or other heat sources, there will still be low grade waste heat to use.

It's greenwashing because it's not obvious, and the wording is deliberately deceptive. An accurate description would give readers the relevant facts, not assume they can figure out the hidden catch.

The question is, would it be possible to use large scale liquid air storage to speed up removing CO2 from atmosphere?

I have no idea if removing CO2 is easier in liquid form but if it is feasible then ff we already had huge infrastructure to store energy that requires pumping astronomical amounts of air back and forth, it might make sense make sense to add a bit to it to remove CO2 while in liquid form.

Hmm. This actually has legs - CO2 can't be present in liquid air, because it freezes at only about -80 rather than the -200ish for air. So at some point CO2 "snow" will collect in the system. Unfortunately it's ~0.04% of the air, so this will take a long time to collect much. But if it's removed anyway, sure. Then you have to work out where to put it.

The energy required to liquefy air is probably too large. Say we have 400pm CO2 in air - that's ~630mg of CO2 per kilogram of air.

Given that the specific heat of air sits at around 1kJ/kg[0], you'd have to spend over 125kJ of heat energy to get just one gram of CO2 - that's 35Wh.

If you used wind power(14g CO2/kWh) and a heat pump with a coefficient of performance of 1 (not sure how possible at these gradients) you'd "emit" 0.5g of CO2 in the process.

[0] https://www.ohio.edu/mechanical/thermo/property_tables/air/a...

Well, you missed the fact that the plan is already to liquefy the air. So the air will already be in liquid form.

Also I don't think they are going to cool it but rather compress it. Cooling would require constant expenditure of energy to keep it cool whereas you can compress it and leave it in the vessel at room temperature. It takes 33 atmospheres to liquefy nitrogen and 50 for oxygen.

>The new liquid air battery, being developed by Highview Power, is due to be operational in 2022 and will be able to power up to 200,000 homes for five hours, and store power for many weeks.

Is it just they're expecting many leaks in the system? I don't see why they couldn't just store things for years basically.

Liquid air is cryogenic - they'll have to either bleed off or spend energy to keep it cooled. The "many weeks" measurements is probably some measurement of bleed off rate assuming no top offs (but I'm just guessing - the company website https://www.highviewpower.com/technology/ doesn't seem to call it out clearly).

If the tanks are strong enough, is cooling required? My understanding is that you'd only need to cool the liquid if the structure holding it wasn't sufficiently strong enough to handle the pressures.

Approximating air as nitrogen... nitrogen cannot exist as liquid phase at room temperature - it can exist as a supercritical fluid though at 5MPa (so 50 atm). At that point it has a density of roughly 56 kg/m^3. At ~1 atm, nitrogen becomes liquid at -194C and has a density of roughly 800 kg/m^3. So your mass density of the system is like 14 times better with liquid nitrogen, and the pressure loading requirements are stupidly lower.

Making the tanks strong enough to handle the pressure if the liquid turns to gas is very expensive. It is much cheaper to make a tank that just holds a liquid.

It's not cost effective to store for a very long time. If you stored for a year, and the tanks last 30 years, you only get 30 tanks worth of electricity for the price of the tank. If you store for a few hours, you get 30*365 tanks worth of electricity for the price of the tank, but you need to factor in the cost of the compressor and generator. Presumably there's some balance when you're getting best value out of the compressor and generator, but don't need an excessive number of tanks.

When you compress air it produces quite a bit of heat that is usually cooled off... anyone that has an air compressor can attest.

Help me understand the thermodynamics, are they storing the heat too?

What kind of pressure is required to do this?

This cannot possibly be more efficient than lithium batteries...

Maybe cost efficient?

> This cannot possibly be more efficient than lithium batteries...

It really depends on what you perceive to be efficiency.

Lifepo4 has 99% coulombic efficiency

There's probably a much greater amount of energy lost from conversion to liquid form and stored and converted back into usable electricity from steam generators. I believe the article mentions it but it likely has better long term storage than batteries as well. The equipment required for the 'air' battery would have a longer lifespan and be cheaper to repair then a massive battery bank. On top of that batteries aren't exactly the most 'green' thing to produce, so from a carbon footprint perspective, I imagine the air battery is much more efficient.

It's not of course but for long term storage the capital costs of the storage are more important than the the round trip efficiency.

At first, I thought this was interesting. And then I read the part about the battery only lasting for a few weeks, before it must be recharged. This sounds terribly inefficient.

I recall reading about a spring type of battery made of diamond nano fibers a few weeks ago. They were saying this was even more efficient than lithium ions, since it was a physical process of releasing its energy, instead of a chemical process. It behaves like a child’s wind up toy, but at the molecular level.

Would it be better instead to have a physical battery, instead of a chemical battery?

Perhaps something like a big wind up spring can be built. Then it can hold its charge indefinitely. And you stack thousands of these together, and they can just sit there for a long time, until they are needed.

Or even that idea to use large cement blocks stacked into skyscrapers. Each block holds potential energy, and when released, it can glide down a magnetic track and generate electrical energy. This is also a concept of a physical battery.

Even a hydroelectric dam is physical energy. You’re moving the water uphill.

Reply on news.ycombinator.com