I can't even count the number of lab-stage announcements that I have seen in HN. This will be of interest only when they can get it to scale
The announcement is suspicious. "100 hours" is not a meaningful number for a battery. The numbers you want to hear are KWH/Kg, KWH/m^3, max charge and discharge rates, number of charge/discharge cycles before storage drops off, efficiency, and cost/KWH.
"That project, announced in May last year, was originally due to be a 1MW/150MWh demonstration plant capable of outputting 1MW for 150 hours straight" hints that the discharge rate may be very low.
Previous work [2] indicates serious limits on charge/discharge cycles. Like 20-30 cycles.
[2] https://phys.org/news/2017-11-renaissance-iron-air-battery.h...
I'll make a note to check back in 20 years, then. :-)
I note they don't mention charge cycle durability - for grid, you'd want 10 000-ish cycles to 80% capacity. A bit of a red flag there.
Still, iron is cheap, plentiful, and environmentally benign. Even mining and refining iron ore isn't as bad as mining for many of the other chemistries, and we already mine a lot of iron, so there won't be shortages as it scales. I want this to work.
I note the web site[1] says its areal power intensity is about 3 MW per acre (presumably 300 - 450 MWh areal energy density, given the 100 - 150 hour claim), best case. They're not going to be installing this in Manhattan.
Battery prototypes are the moral equivalent of announcing you've found a new chemical that kills tumors - in vitro.
Back then the advice from others watching battery technology? If their tech is 10 years ahead of the current state of the art, then either it will take them 10 years to produce it or they will produce it sooner by sacrificing capacity for safety/manufacturing concerns.
Another important thing to remember in any discussion about power - there is almost always a substantial premium paid for portability. If someone is feeding you PR about a stationary power system and comparing it to any household-name portable system, then you should smile and nod and keep track of your wallet as you back away slowly. They are the snake in the grass your ancestors warned you about.
People are free to upvote whatever content they find valuable.
(And the headline + post title are not misleading clickbait, which is often the case with such stories.)
[ESS] already has a factory pumping out flow batteries just south of Portland, Oregon. Its core product is the Energy Warehouse, which fits a rotund tank and several stacks of battery materials into a shipping container, along with the necessary electronics... During a June visit, I saw a test line in which robots prepared cells to be glued together into the battery stack that the proprietary iron liquid flows through.
https://www.canarymedia.com/articles/ess-is-betting-the-worl...
This sounds a bit like what red blood cells are doing during breathing.
If they're planning a pilot plant to bring online in the next two years, then they have done their due diligence and believe the tech can work. It'll be interesting to watch.
It's an engineering problem that requires financing. After that, you still need to figure out how to do it profitably at scale.
The vast majority of "the science is done" breakthroughs fail to be commercialized because they're too expensive or time-consuming to figure out how to scale, or the unit economics just don't work out.
I've seen the inside of some magical labs...Their actual setup when you stared real close and knew enough of how they were doing things was anything but.
It's even more difficult than that. :-)
Ideas have negative value; lab prototypes also have negative value; pilot plants are worth a small fraction of what they cost - scrap value.
You get value from full scale production, and every bit of that value is paid for in the engineering required to get to full scale.
That project, announced in May last year, was originally due to be a 1MW/150MWh demonstration plant capable of outputting 1MW for 150 hours straight.
If the energy had gone up 300x from the originally announced pilot project the same way the power did, this would be a huge storage project boasting 45,000 MWh of storage capacity. It would surpass big pumped storage projects like the Bath County station (capacity: 24,000 MWh):
https://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Sta...
But this news article doesn't highlight any superlatives like that.
Reading between the lines, here's what I think has changed:
- The original announcement of a 1MW/150MWh project was an implicit admission that their battery could not charge or discharge quickly. It took nearly a week to fully charge/discharge. At the time they put a positive spin on it by emphasizing "long duration." That's not really an advantage, though. You can just discharge a high-rate-capable battery slower for long duration applications.
- Since this updated pilot project announcement touts more power and leaves any energy increase unspecified, I think that they found a way to increase the charge/discharge rate for their chemistry. That would be good because it would mean that the chemistry is suited for grid tied storage in general, more like lithium ion. If it can charge and discharge at high C-rates and it has lower lifetime cycle cost per megawatt hour than lithium ion batteries, it could be very successful.
EDIT: new user "tiddelypom" below says that he works at Form Energy and that this article is incorrect about the project size:
Not sure where that article got the 300MW number, the GRE project is on track for the original size.
If that is the case, my remarks about the limitations of batteries with low C-rates still apply. But my speculation that the company has drastically improved the C-rate of its chemistry would be incorrect.
If the price is right, long duration is just fine. One week of storage to get through a spell of cloudy weather or poor wind conditions is quite useful. It doesn’t obviate the need for daily storage.
[1] https://newatlas.com/energy/bavarian-brewery-carbon-free-ren...
Coal energy density: 6.7 kWh/kg [2] Iron Ore energy density: 1.4 kWh/kg [1]
Total iron ore energy reserves: ~236.04 billion kWh [3] Total coal energy reserves: ~ 7,068 billion kWh [4]
So as far as energy goes we have 30x as much coal energy.
Now the rust can be renewed with energy ... but you need the energy to renew it in the first place.
[1] https://spectrum.ieee.org/energywise/energy/renewables/iron-... [2] https://en.wikipedia.org/wiki/Coal [3] https://www.nrcan.gc.ca/our-natural-resources/minerals-minin... [4] https://energyeducation.ca/encyclopedia/Coal_reserve
> but you need the energy to renew it in the first place.
So am I hearing you could create the iron-fuel someplace with excess energy and then ship this stored energy someplace that wants to burn it?
And while Iron loses out to energy density per kilogram, it wins on kilogram per cubic centimeter :D (your math is much appreciated btw)
What is left missing in the big picture for 100% clean, and renewable energy? As far as I can tell, Solar and Wind has already achieved cost/Wh lowered than all other form of energy and has a decent roadmap to further drive down the cost by another 50 - 60%.
Or are there some other puzzles we dont know? Or if Iron Battery works, and this will be "it" ?
- Steel production
- Ammonia production
- Organic chemicals production (polymers, lubricants, solvents...)
- Electrification of transport (world car and truck fleet is still 99% combustion powered, and these iron batteries aren't the right kind for mobile usage.)
- Synthetic fuel production for applications that can't use batteries, like rockets and trans-Atlantic passengrer flights.
- Cement production, which currently releases large amounts of CO2 from fossil combustion and from the chemical transformation of limestone.
Back to NiFe though, you can certainly get decades of use comfortably out of NiFe cells, but it's worth taking into account that they require a lot of maintenance, routinely topping up each cell with deionized water (there are automated systems you can buy or build yourself to manage this though), and every few years the KOH electrolyte inside them will need replacing every few years, plus you will want to keep your cells adequately ventilated as they vent hydrogen periodically.
I would definitely NOT recommend lead-acid, sure the upfront cost is lower than anything else by quite a ways, but they won't last long at all, and as a result I'd be surprised if you weren't paying a lot more in the long run.
Comparing for example a 100ah lifepo4 to 100ah lead acid, the useable capacity of the lithium will be close to 4x more. Using the full capacity of the lead acid will make them wear out quickly.
https://www.alexhsain.org/blog/ironair
This could be super useful for all sorts of backup situations, from homes to data centers to hospitals.
And once they are installed generally, they can be combined together as a virtual power plant and start partially paying for themselves.
The round trip efficiency of hydrogen is already something of a concern for grid backup power, as far as I know, ans that's only slightly less than 50%.
As storage costs decrease, those moments in time when there's oversupply of energy from solar and wind will become shorter and fewer. And (at least to me) it's really unclear what market designs will be viable if we were to have, say, 3x-5x oversupply of energy capacity so that renewables' minimums still meet our needs, and what new energy uses people will come up with when the can by intermittent energy for half a cent a kWh or something.
Unless there is a regulatory blip, then there is always a cost to energy. Stuff has to be maintained, connections repaired, monitored and upgraded.
The other thing to note is that its only "free" if you own both the infrastructure and the producer. If you are doing "arbitrage" then the efficiency is your profit margin.
Also, working for 150 hours is no great feat. Just means they undersized the inverter and power electronics. If you did that with lithium ion, you’d also have much lower cost per kWh (for small systems with only an hour or two, the inverter could be half the cost or more), so going to 150 hours would also reduce costs by a lot as you’d be closer to the raw cell price.
I suspect that unless there’s a chemical reason why they have to discharge over many days, any real system is likely to be more like 12-24 hours as you’d be wasting the usefulness of the battery by picking an impractically small inverter.
Here’s hoping it’s actually a huge cell cost reduction while keeping decent round trip efficiency and cycle life.
$6/kWh raw materials, or $20/kWh for full system
(I work at Form fyi)
Are there any publicly available documents about the chemistry, or how the product compares with e.g. the ESS Inc. iron flow battery?
BTW, hydrogen-air battery = hydrogen fuel cell, if you didn't realize that.
Basically, if you're cycling regularly (e.g. smoothing solar and wind over a day) then Lithium is already pretty good and you can expect it to get bettwr as it scales out to the entire automotive industry and indeed those car batteries will be fed by the grid and can also act as short term storage and demand management.
If you cycle less regularly though, storing power for weeks or more then you need something much cheaper than lithium can ever be but if you're cheap enough you can sacrifice some conversion efficiency and still be useful in a 100% renewable grid.
This is where flow batteries are targetting, you can have a small/cheap "converter" but store the energy in tanks longer term.
But as you say, thats also basically what you can do with hydrogen/ammmonia. And as an added bonus you can buy sell hydrogen/ammonia on the open market as it's used for other purposes, which lets you insure against under/over production and take advantage of economies of scale on the converter and storage parts.
As a final bonus, during the transition you can add a percentage of hydrogen to existing gas turbines to reduce their carbon intensity and GE and other sell turbines that are built to run on gas, hydrogen/gas mixes and also pure hydrogen. This gives an easy ramp up as a carbon price and/or minimum targets can kickstart the green hydrogen industry without any particular customer needing to bear 100% of the cost.
Methane from waste can also be used as a source of hydrogen, making it carbon negative, with a promising tech looking to generate solid carbon in the form of graphite. But even if you released the carbon I to the air it's better than releasing the methane.
Low-cost methods for its safe storage and handling are well established, despite the need of being careful to avoid leaks.
Also the fuel cells using ammonia are not too different in performances compared to those using dihydrogen.
Hydrogen might be a possible choice when high energy per mass or power per mass is desired, but it is a very bad choice for the purpose of this new iron-air battery, i.e. stationary storage with very high energy capacity and very low cost.
Iron-air might indeed be the best choice for medium-time energy storage, with low cost and good full-cycle efficiency.
For very long energy storage times, e.g. years, synthetic hydrocarbons would be preferable to hydrogen, due to much easier storage and handling.
Since water is significantly more available than just about any other material, hydrogen-air cells should be the ideal battery for anything that isn't volume limited. Which is frankly a lot of cases. Unless there's some specific need for an iron-air battery where hydrogen-air can't be used, it's hard to conceive of a situation where we wouldn't use hydrogen-air.
Synthetic hydrocarbons are basically extensions of hydrogen electrochemistry. You are just adding carbon to the hydrogen made with the electrolysis step of a hydrogen-air battery. It's even possible to make a hydrocarbon-air cell such as direct-alcohol fuel cells or solid oxide fuel cells.
There's some tantalising research from a few years ago that suggests it could get way better energy density, but the same comes up frequently with battery technology.
Or as Greg Lydkovsky, global head of R&D at steel giant ArcelorMittal — Form Energy’s latest investor — put it, the technology “holds exciting potential to overcome the intermittent supply of renewable energy”.
Form Energy president and chief operating officer Ted Wiley said: “We’ve completed the science, what’s left to do is scale up from lab-scale prototypes to grid-scale power plants.
“[At full production], the modules will produce electricity for one-tenth the cost of any technology available today for grid storage.”
The battery is said to work through “reversible oxidation of iron”. In discharge mode, thousands of tiny iron pellets are exposed to the air, which makes them rust (ie, the iron turning to iron oxide). When the system is charged with an electric current, the oxygen in the rust is removed, and it reverts back to iron.
Wiley said that a 300MW “pilot” project for Minnesota-based Great River Energy will be commissioned in 2023.
