Ambri Utility Scale Liquid Metal Batteries Arriving in 2020

by

Ambri will be commercializing a utility-scale liquid metal battery within three years. In 2020, they want 10-foot cube commercial units in proof-of-concept field deployments.

Ambri’s liquid metal battery was initially based on magnesium and antimony as the negative and positive electrodes and a low-cost molten salt electrolyte. Ambri has transitioned to using a higher voltage and lower cost chemistry.

They halted commercialization because of a problem with the seals on its high-temperature battery cells. They restarted with the new, lower-cost chemistry.

Utility battery storages projects are currently scaled for the 4-hour duration. Ambri is targeting 4 to 12 hours of duration.

The base unit for Ambri’s system is a fully-sealed liquid metal battery cell. Ambri’s cells are connected in a thermal enclosure to form an Ambri Core.

55 thoughts on “Ambri Utility Scale Liquid Metal Batteries Arriving in 2020”

  3. There is no cost of CO2. There might be if we were going to use primarily fossil fuels for another 70 years, but the reality is that we will have effectively eliminated fossil fuels used for energy by about 2050 for the same reasons we stopped buying typewriters and steam engines. Greed. Nobody buys inferior products that cost more except for maybe a few oddballs.

  4. Or as long as the idea holds true promise of profit. Greed drives our economy. Think about it – although they are cautiously advertising these as 20 year lifetime batteries, as long as the seals hold they should last forever. What is the cost per cycle on a forever battery? $0.

  6. That was years ago, and the result was that mining companies uncovered more sources. Vanadium is not truly rare, and in the perverse manner of so many other resources mankind uses demand actually increases supply and reduces cost.

  7. Actually, 24-M started out in pursuit of a flow battery. After several years working under the assumption that it would be cheaper, one of their scientists actually figured out a way to predict HOW MUCH bigger it would have to be to save money, including factors like tanks and pumps. That was when they switched to semi solid.

    Batteries are actually simpler than flow batteries, having no moving parts and easily shipped in discrete units.

  8. Eight years would not be unusual to commercialize a product, what is more concerning is that they already sold products and had to shut down production a couple of years ago for technical reasons (leaking seals) and have not yet restarted manufacture.

  10. 2020 running about $25/kg. Presuming they use 1,000 lbs for one if their cubical 500kWh units, that’s $50/kWh just for the vanadium alone. That still may be worth it based on the claim that the battery is not expected to experience material degradation over 20 years of operation.

  13. I 100% agree. Nuclear can be made super safe and super clean. To bad the navy was the prime mover on this tech when it first started or else we would have been so much better of now.

  14. Except for the fact that its highly hazardous and any large quantities is highly regulated as well for good reason. It’s explosive and combustible at certain concentrations. Not to mention humans and ammonia don’t mix well.

    Don’t get me wrong, I like ammonia as long as it stays in the pipe. But you might think differently about it if you see 100 gallons of it dumped on to the ground.

  16. At $50/kW-hr, that is $600/kW for 12 hours. For a 20 amp circuit, that would be ~$12,000. Since the household would be sleeping for about 8 of those 12 hours, that might be enough battery for up to 60 amps of service. Considering that these batteries deteriorate minimally you can amortize the cost over at least 10 years. If you were going to buy a tesla, I could see using this technology for residential (or better community, if you can get that night owl down the street to quit hogging all the battery at night.)

  17. 4 hours is good for getting coal plants started, but you need a minimum of 12 hours for solar and 72 hours for wind. You really only need minutes for natural gas to start. So batteries are needed with either minutes of capacity or 12+ hours of capacity.

  18. Remember… its vitally important for any energy that competes with fossil fuel to count all possible costs… and just as important that we all put on blinders to the external costs of further CO2 release. How else can the current owners get a decent ROI?

  20. Gee Whiz, before kvetching watch the videos on it. It starts by point out that they started the project looking at the cheapest and commonest elements in the periodic table… Looking at your comment, you sort of looks like a fossil fuel troll – “Oh! It can’t work economically”. Well, it does. It works good and will get a lot better.

  21. Problem here is that they target grid systems. I really don’t see why. The basic technology seems very good for stationairy residential usage. An “Ambri-core” (about the size of a wasing machine) seems to almost fullfill the needs here.

    Selling and upscaling would be way easier for these type of complete of-the-shelf systems would b e way easier too.

  22. No moving parts, low cost using plentiful materials, good density, very low maintenance, safe and recyclable.  They can also be used in a much smaller form factor, smaller than a breadbox.

  23. We need a good beer battery. If it leaks, we just have a party but they need to include a refrigeration unit to make sure it’s cold.

  24. There is a section in there about batteries and how utility scale storage is about to change everything.

  27. I think he has trouble sleeping, and reads a phenomenal amount. He got slung off his previous blog venue, Daily Kos, for abusing antinukes and other faith-based types too harshly.
    ‘Magic alloys’ are what has made the cheap jet transport of today possible – the refractory alloys in turbine blades need precise mixtures of very rare alloys to operate at those temperatures. Quoting myself from last week, at the end of WW2 jet aircraft and nuclear reactors were both brand new and exclusively military. Now every country has an airline, and everyone uses them to get anywhere, but nuclear power, with all due respect to your profession, is still a tiny niche product on the energy scene. Water cooled reactors are only a fraction of the range of possible configurations.
    Nnadir did a discussion podcast with Atomic Rod Adams a while back. You might have the background to do something similar, if there’s some particular points you want to make.

  28. ‘When I contemplate potential LAMPRE based reactor designs, I do so in imagining a “breed and burn system, reactors designed to run without refueling for significant fractions of a century.’ Nnadir discusses using technetium as an alloying agent to render tungsten more easily machined – tungsten and tantalum were the only metals capable of holding liquid Pu/Fe.
    In other articles I haven’t run into yet, he proposes a uranium/plutonium fuel cycle, with no need for any isotopic enrichment, and liquid-phase separation of fission products. https://www.democraticunderground.com/1127121237

  29. His article about liquid plutonium reactors was more recent, he just hints at it in the one I linked. If I run across it I’ll put it here. He’s a chemist in New Jersey, not too far from you I think. Blogs on Democratic Underground.

  30. I wasn’t worried about the flammable or carcinogenic nature of ammonia, accidental large scale release of the stuff terrifies me.

  31. nice article. He discussed iron/Pu alloy having a low melting point, but he didn’t take it further to state that this would be a useful material. I think you are putting words in his mouth WRT using it in a reactor.

    This guy still blogging? I get quite a few hits on google for his username.

    Interesting analogy about running an aluminum smelter in reverse. I understand that one form of aluminum smelters uses graphite electrodes to pass DC through molten clay. So this battery is in a way related to that? Ok.

  33. At a regional electricity supplier scale then sure, pumped hydro may be a valid option (depending on geography).

    But at the scale of a factory or something there isn’t a way to incorporate pumped hydro. There is definitely an option to put 20 containers in the back storage area. And if that takes you from being on a 5 MW peak supply account to a 1 MW peak supply account, you could save a lot of money.

  34. What happened to Hyman Rickover? I grew up in Groton, CT. He was well known there, nuclear subs and all.
    Supposedly the Ambri thing stays hot just by charging and discharging. Though it seems that this high temperature aspect is its down side.

  35. Vanadium sounds expensive. And your Chinese project is making it more so. Google “China Rattles Vanadium Supply Chain”

  37. You know the stuff in your gas tank is highly flammable and carcinogenic, right?

    And the stuff most of us heat our homes with is explosive.

    Ammonia is extremely widely used and transported chemical with known safety precautions.

    That said, I tend to agree that public acceptance could be challenging for some applications. Ammonia for shipping, locomotives, and agriculture energy make sense because they are all applications that already transport, store, and use ammonia. They already understand and manage the safety risks.

  38. Companies are in business to sell at the highest possible price the market will bear. They said it should be 1/3 the cost of LI Batteries, but I doubt you will be able to afford one for your home solar array.

  39. I will support your idea, if and only if, you can guarantee no amount of ammonia over 2 liters will ever be stored within 5 km of any place I’m likely to be at any point in my life.

  40. I really thought this company was long ago defunct.
    I guess you’re never really out of the game as long as the VC cash keeps flowing.

  41. I like hydrogen/ammonia.

    Make hydrogen, put it in a salt cavern or steel tank. With reversible solid oxide fuel cells 70% round trip efficiency is possible.

    If you want to store the hydrogen for seasons or move it somewhere else, bond it to nitrogen from the atmosphere to make ammonia. Ammonia is delivered anywhere in the world at reasonable cost today.

    Ammonia can be used directly as a liquid fuel in reciprocating engines, combustion turbines, etc. Or you can reform it back into hydrogen for PEM fuel cell vehicles.

    Ammonia is easy enough to store that it is a feasible energy carrier for shipping, locomotives, large trucks, etc electrification.

  42. Wow, it only has to sit there at 600C in stand-by mode. So convenient. Nothing says simple like ‘molten’.

  43. Generally you add tankage to increase capacity but to increase output you need to have a larger area where the electrolytes contact.

    This means that it is much easier to add more capacity (add a second tank) than it is to double the power output.

  45. If you don’t need instant on then you don’t need batteries. Pumped hydro scales to GWhr and costs one tenth what lithium ion costs for equal units of storage.

  48. If they do not openly state the costs involved, suspect the cost is huge. All in an attempt to compensate for intermittently unreliable energy sources like industrial wind and solar. Add all the costs together and calculate the discrepancies between nameplate capacity and actual usable power, suddenly this charade is a klutzy solution looking for an actual problem.

  50. Question is cost per KWH. This should open up much more solar and wind backup. But still not enough to fill the gaps when there are several days without wind or sun. Still need a base load of nuclear.

  51. Only two things matter. Are the batteries reliable and how much do they cost per $MW/hr. If the cost is low enough then large customers paying bills based on peak demand could buy them to shave their peak demands.

  52. I downloaded their brochure the only numbers they list are in the marketing blurb:

    “Ambri’s systems are modular – units can be deployed as small as 200 kilowatthours and as large as hundreds of megawatt-hours. ”

    So I assume that a single Ambri core is rated at 200 kilowatt hours. Pile them up until you get the capacity you crave.

  53. That is really impressive! Just think how many of those two dimensional batteries you can stack on top of each other!

  54. I’ve had to RFA. Burried quite deep it says that a 10ft*8ft unit is 500 kWh, at 80% efficiency. Thats quite respectable.

Comments are closed.