Google’s latest renewable energy deal is all gas bags and hot air Caught in a constant race between its AI power needs and carbon emissions reduction pledges, Google's latest sustainability commitment sees it considering giant bags of carbon dioxide as a solution to dirty energy. Google and Italian startup Energy Dome recently announced a deal that sees Google not only deploying Energy Dome' …
How big and how expensive is the 20MW/200MWh "full scale" plant? For something "full scale" it's rather piffling, given datacentres are usually hundreds of megawatts these days
And I mean the real cost, not the subsidised cost.. No doubt they are farming carbon "use" credits.. (which really ought to be multiplied by the probability of it not leaking in the next 1000 years..)
It's far bigger than a battery of the same spec, far (i wager) costlier than a battery to build, and demonstrably far lower efficiency
Mind you, it's a reasonable fire suppressant. Maybe they should install some batteries inside their giant CO2 gas bag to puff up their performance figures
Is it easier to get CO2 than it is to get lithium? Is it easier to reuse the components of one of these gas bags than it is to reuse the components of a lithium battery? If you remove the constraints of it has to fit in your pocket or in a vehicle or even in a house, then all kinds of possibilities open up. The more interesting question to me is how does it compare to pumped storage, or a giant flywheel, or etc. Worth exploring I would say, and not as stupid as a lot of the ways that Google is currently spending money. A Marshall Plan level project to explore every possibility for mitigating catastrophic climate change should be barfing money into every possibility like this, as opposed to currently funding fossil fuel exploration and extraction and processing and transportation and "carbon capture" and "lowering emission intensity" which is all much more retarded than even the most retarded "offset trading" scam of which there are far too many examples. IMHO.
"pumped hydro is the only at scale solution"
For some values of "at scale". In order to achieve that you need suitable sites with an upper and lower reservoirs. Reservoirs involve drownin existing landscapes which isn't exactly environmentally firendly, nor is it firendly to those who lived there if the valleys were inhabited.
It's a useful technology but niche.
Couple of wrinkles that come immediately to mind:
- Anything that uses large volumes of CO2 as a working fluid will inevitably leak that CO2, only adding to anthropogenic climate change.
- There will be significant conversion losses across the cycle (might still be worthwhile, but will compare poorly with Lithium batteries).
- Green credentials are still predicated on the continued investment in renewable sources. So for Google to be credited for this they need to up their solar or wind investment further.
Other than that, inflate the gas bags, Smithers.
However, if the CO2 came from the exhaust of a coal/oil/gas power plant, that could be a massive reduction in emissions…
Which leads on to the big question: where is the CO2 in these gas bags going to come from? If it’s from the dedicated burning of fossil fuels - easy to get quantities and purity, then it is just another piece of greenwash.
You don't get pure CO2 in the exhaust gas from an efficient combustion - it will need more oxygen than the basic stoichiometric mix to ensure all the fuel burns and no partial combustion products (such carbon monoxide). And then you have to separate the CO2 from the nitrogen, oxygen, argon and water vapour.
These gas bags are not going to be filled by the exhaust from a nearby fossil fuel plant. You will want to have 100% pure CO2 or you will greatly complicate the design.
It's a great entry for the Halfbakery, it's a greenwashing distraction if you're one of the worlds leading tech companies.
The whole idea sounds very similar to Liquid Air Energy Storage which was first trialled as a way of smoothing peak demand way back in 1977.
https://www.sciencedirect.com/science/article/abs/pii/S1364032124007123
And since, with liquid air, you can just vent the turbine out to the atmosphere, the added cost of capturing and maintaining a giant bag of CO2 would seem to be exactly that - added cost.
However, I'm sure they know what they are doing. Icon: the engineering technical skills I don't have in order to understand fully. :-(
Well, the first person who knows they don't know, unlike all the pocket experts commenting here. Since these guys have had a plant running for years, there seems to be a lot of bumblebee effect going on.
Liquid air involves far greater temperature changes for liquifaction, making the liquifaction inefficient, the plant much more expensive per kWh and and the round trip energy loss greater. It has to be stored as a cryo fluid, though to be fair you can use unpressurised vessels. Since the whole generation thing happens as the liquid boils to gas, the energy is released at ~ -190C. Leaving you with gas at -190C. Unlike a liquid air plant you can't use that gas to cools the gas you are liquifying - because you are not liquifying gas at this time you are boilng it. So you either need some kind of giant -190C cold store and heat exchange mechanism, or you are venting that hard-won cold.
CO2 can be stored as liquid at normal temperatures, and the exit gas temperature is much higher.
Liquid air also involves the separation of oxygen and nitrogen, and if you vent to air, you will be doing that all the time. Liquid air plants have had massive explosions in the past: oxygen is not an especially safe liquid. Good luck running oxygen turbines.
But perhaps there is some economic model where the primary function is liquid oxygen production using renewable energy, and energy storage using the liquid nitrogen is a bycatch.
As an aside. the big failing of all long term energy storage ideas, is simply that the unit cost is the system cost / cycles.
The longer your storage period, the fewer cycle you get to divide the cost amongst. This dooms almost any long term storage idea, and conversely is why short term battery storage is economic despite the kwh cost. Even pumped hydro with daily cycling has struggled to make economic sense. Proposals for multi year storage like NZ's Lake Onslow, are an economic disaster zone.
To match the 200MWh capacity of the CO2 plant you would need a tower with a 90x90m base, the first 90m of height for the low storage tank, 10m more to raise the water an arbitrary 100m then another 90m for the upper tank. You would need much of that 10m so the base of the top tank can hold 720,000kg of water. You could decrease the area and mass by increasing the height but whatever height you choose 200MWh is a huge pumped water installation. It only makes sense when you have two natural lakes at different heights next to your data centre.
No. I am saying to deal with fluctuations in the price of electricity using pumped water storage would require putting a pair of water tanks each the size of a SpaceX gigabay on top of the data centre. The structure would be visible from miles away. People would object at the planning permission stage.
If they Choose France they can just balance a leftover Olympic swimming pool (from last year) on top of the Eiffel Tower, and voilà! ;)
OK here's one for the Google investment team:
1) Confiscate the metal boules from a million Frenchmen.
2) Forge said material into a circular ring with slightly less than the radius of the Eiffel Tower height.
3) Fashion some spokes and fix said ring at it's epicentre to the top (le troisieme etage, to be precise).
4) Attach to a motor/dynamo.
5) Spin it up with all that spare wind from Le Royaume-Uni or excess sunshine in Provence for that matter.
6) Draw it back as 'leccy when we have one of those rare dull, windless days.
The above ignores any interesting micro-magnetic phenomena that might arise from a load of forged, spinning balls in le 7e arrondissement.
Info is sparse though I also question their claims. As an example they claim the lifetime of the CO2 battery is 30+ years. And yet I know from personal experience no compressor lasts 30+ years, which is a key component. My home AC's are lucky to last 15 which is a much simpler compressor than that needed to chill/compress CO2 to a liquid.
Another thing I recall from a job where they had a giant N2 tank was periodically the thing would let off a huff that would scare the crap out of you. My guess was that as it heats up and not enough N2 gas was being used by the labs, the tank would blow off excess pressure since the tank was not cooled. They'd fill it up with fresh cold liquid N2 every couple of days. Not sure if the CO2 would have a similar requirement. If it did, then they'd either have to cool the liquid somehow after it was liquified or let some blow off back into the bag and reliquify wasting some power.
They say they have an operational unit, and I suspect "an" is the operative word, as in one. It might be interesting to see what kind of maintenance costs there are to keep that one running. Holes in the CO2 balloon needs patching, compressor needs a new valve assy, ... There is no mention of the "up" time on their web site.
Good idea indeed to check their website (linked under "announced" in TFA).
30 years here means (iiuc) the 200 MWh capacity can be maintained this long in such a system, where a battery (eg. in an EV) would have instead lost some good percentage of its capacity by that time. But yes, maintenance is needed, including replacing worn-out parts. Also, this has a 75% round-trip conversion efficiency which is not bad relative to 85% for lithium.
CO₂ goes liquid at -37°C which is much easier to get to than air with all its nitrogen (-212°C) which helps the process and should reduce the need for periodic "huffs".
But yes, they're looking at 5 hectares for 200 MWh which seems a bit large (500m x 100m, or 1/3 mile x 300 ft), though several of those could certainly fit around a datacenter the size of Manhattan Island.
Overall I think it's a quite decent concept until we get better at readily reclaimed transient energy storage (a must at this juncture!), imho.
How often does the balloon need to be replaced?
Seems like Solar Radiation (well, unless they put the balloon underground, or, maybe, have a shelf of photo-voltaics covering the top and sides) would "sun rot" the material quicker than that.
Also, how carbon neutral is the process for making the balloons?
And, since curiosity is a thing, when the balloon needs patching, do they patch it from the inside (people wearing breathing devices to survive the CO2), or patch from the outside (which probably requires being suspended from a crane, unless they pump the balloon down, first)? Great documentary material.
I'd love to find one, particularly if it's flexible (like the bags would need). Plastics get extremely brittle, and organic materials rot. Stainless steel might survive 15 years of sun, but isn't flexible enough.
I'm particularly interested in rope or string. Like a hammock that won't dry-rot away if left outside for 2 years.
They would simply put up with the leaks until replacing the CO2 got too expensive. Then they would deflate the balloon and patch it while it is flat on the ground. There's really no other practical way to deal with it that would be safe from a workforce perspective.
I suspect however that the concept will be abandoned long before the balloons in the demo systems decay enough to be an issue. The balloons will then be cut up and hauled away to an incinerator.
https://en.wikipedia.org/wiki/Energy_dome
The density of liquid CO2 seems to be roughly that of water ~1 kg/L ~ 23 mol/L of which 1L as a gas at STP would become roughly 500L or 0.5 m3.
I don't know how much liquid CO2 is required to generated 200 MWh but fairly obvious that some pretty big gas bags covering hectares would be required. Just moving the gas around and back to the liquifiers is probably a major engineering problem. If the liquifiers have any possibility of producing liquid oxygen there is even more fun.
These days you could probably pitch a huge hangar housing a trillion clockwork spring mechanisms to store energy and not be carted of to the funny farm.
> I don't know how much liquid CO2 is required to generated 200 MWh but fairly obvious that some pretty big gas bags covering hectares would be required. Just moving the gas around and back to the liquifiers is probably a major engineering problem. If the liquifiers have any possibility of producing liquid oxygen there is even more fun.
Energy Dome say 5 hectares (12 acres) for a 200 MWh plant.
LOX is unlikely to be a problem, given that liquid carbon dioxide doesn't exist below -60 C, and LOX boils around -180 C at 1 atm -- it shouldn't be the hazard it poses when handling LN2 (-196 C), for instance. (Icon: eye protection when handling cryogenic liquids.)
Funny how, with any company that is valued in the billions, everything suddelny becomes "strategic".
There is no "give it a shot" investment. There's no "We're going to try this" investment.
No. It's all done with military precision - even though the military has the saying "no plan survives first contact with the enemy".
But it makes the suits (who've never seen a battlefield outside of films) feel more important, as if they've carefully considered all the variables and come to the best conclusion.
Yeah, sure.
I'd want to see the safety precautions first. There's a reason all these CO2 storage proposals are using undersea reservoirs.
Rapid unscheduled releases of CO2 will flow downhill and pool. If that's over your house, you'll suffocate.
Google Lake Nyos disaster to see the geological scale demonstration.
For a smaller-scale demo you can try at home:
Form an index card into a V-shape.
Light a candle.
Mix baking soda and vinegar in a cup, keeping the cup less than half-full at any time. Let the reaction settle out a bit.
Hold one end of the index card over the candle, keeping the other end slightly higher. "Pour" the gas in the cup onto that upper end of the index card.
The candle will go out from the CO2 pouring over it.
Unless the trees are then used for other purposes. My great-grandfather's dining room set, still in active use, contains a fair bit of carbon. Likewise the wooden framing of my house.
All things eventually die and decay. So I suppose it depends on what you mean by "medium-duration"!
So, they want to re-vaporize the CO2 (no idea how to do subscript on here) and drive a turbine. Where does the heat to change liquid CO2 to gas come from? There's also the question of what do they do with the heat from liquefying it - though if they have some kind of storage that could be the answer to the first question.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
