Iron-Air Batteries: The Grid Battery Made of Rust

Google is spending about a billion dollars on the “largest battery ever announced.” It runs on rust, and it will hand back barely half the electricity you feed into it. That isn’t a defect. It’s the whole point.

What’s actually true: Iron-air batteries store energy by rusting iron on purpose and reversing the reaction to charge back up. They’re cheap, hard to set on fire, and can discharge for 100 hours straight — exactly the multi-day gap a wind-and-solar grid can’t cover affordably any other way. What’s misleading or unregulated: The record-setting “gigawatt-hour” headlines invite you to compare these to the lithium battery in your phone or car. They’re a different class of machine — slower, far bulkier, and only about 40 to 60 percent round-trip efficient — and most of the giant projects are contracts with in-service dates of 2027 to 2029, not power flowing today.

Every so often a battery story shows up wearing a headline that almost begs to be misread, and this is one of them. In February 2026, the utility Xcel Energy and Google announced a definitive agreement to build a 300-megawatt, 30-gigawatt-hour battery in Pine Island, Minnesota — what Xcel called the largest battery project by energy capacity announced anywhere in the world, reportedly a roughly one-billion-dollar commitment. Weeks later came the first project outside the United States, in Ireland, and a 12-gigawatt-hour supply deal with the AI data-center developer Crusoe. The instinct on reading any of this is to picture a colossal version of the battery in your pocket. That instinct is wrong.

What Google is buying is not scaled-up lithium; it’s a machine built on reversible rusting. The chemistry is close to something you already understand. Iron left out in the weather turns to rust. An iron-air battery does that on command and then undoes it. On discharge, metallic iron reacts with oxygen from the air and water from the electrolyte to form iron hydroxide — rust — and that reaction releases electrons into the grid. To recharge, you push electricity back in, which strips the oxygen away and returns the rust to metallic iron, venting oxygen as it goes. Form Energy, the company commercializing the technology, describes it as a battery that breathes: inhaling oxygen to discharge, exhaling it to charge. Lithium-ion works nothing like this — there, lithium ions simply shuttle back and forth between two electrodes and very little is lost in the trip.

And that difference is where the honest catch lives. An iron-air cell gives back only about 40 to 60 percent of the energy you put into it; lithium-ion returns well over 90 percent. Put 100 kilowatt-hours in, get roughly 50 to 60 back. Stated flatly like that, it sounds like a broken product. It isn’t — once you know what the thing is actually for.

The trick is to stop reading “efficiency” as the headline number and start reading “duration” and “cost.” Iron is one of the cheapest materials on Earth — on the order of ten cents a pound — abundant, non-toxic, and hard to make catch fire. You charge an iron-air battery with surplus wind and solar that would otherwise be curtailed and thrown away, so the energy you “waste” to inefficiency was nearly free to begin with. In exchange, you get something lithium can’t deliver at a sane price: 100 hours of continuous discharge. Days, not hours.

So when a press release says “world’s largest battery,” two questions tell you more than the gigawatt-hour figure ever will: how many hours can it run, and how much energy does it lose on the round trip? Thirty gigawatt-hours of iron-air and thirty gigawatt-hours of lithium are not interchangeable. Lithium wins the daily four-hour shuffle — soak up the midday solar peak, release it at dinnertime, do it again tomorrow. Iron-air wins the four-day windless cold snap that European grid operators grimly call a “Dunkelflaute,” when the sun and wind both go quiet and a short-duration battery empties in an afternoon. Think of lithium as a sprinter and iron-air as a very cheap, very patient ox. You would never put the ox in your car. That’s not an insult to the ox.

None of this is new chemistry — what’s new is the question we’re asking of it. Metal-air batteries were first sketched out in the 1870s, and zinc-air cells were running hearing aids by the 1930s. Iron-air had its first serious moment during the 1970s oil crisis, when Westinghouse, Sweden’s National Development Corporation, Siemens, Matsushita, and NASA all chased iron-air for electric cars. It failed at that job. As a peer-reviewed review of the technology lays out, the batteries were too heavy, too slow to charge, and lost too much energy to be a practical vehicle battery, and the effort was largely shelved by the 1980s. The very traits that killed iron-air in a car — bulk, sluggishness, low efficiency — simply don’t matter when the battery sits in a field beside a substation and is never asked to move. The renaissance came when engineers stopped asking iron-air to power a car and started asking it to back up a grid.

That reframing is essentially the founding story of Form Energy. The company was founded in 2017 out of the merger of two startups — one an MIT spinout, the other led by former Tesla energy executive Mateo Jaramillo — with early seed money and lab space from MIT’s venture firm, The Engine. Its co-founders include the MIT battery scientist Yet-Ming Chiang. After running the numbers on what a renewable grid actually needs, the team deliberately chose 100-hour iron-air as its first product. Today those batteries are being built at “Form Factory 1” in Weirton, West Virginia, on the site of an early-1900s steel mill on the Ohio River — backed by up to 150 million dollars from the U.S. Department of Energy plus federal manufacturing incentives, and aiming for 500 megawatts of annual production capacity by 2028. Form’s first commercial pilot, a small system with the Minnesota utility Great River Energy, has been operating since late 2025.

The manufacturing reality carries its own asterisks, and the company is fairly open about them. Form’s own cost target — roughly 20 dollars per kilowatt-hour of storage at the system level, which it pitches as close to a tenth of lithium’s cost per unit of capacity — is a company projection tied to full-scale serial production, not a bill anyone has yet paid at scale. The batteries are also land-hungry: industry analyses put an iron-air installation at something like five to ten times the footprint of an equivalent lithium system per unit of energy, which is fine next to a rural substation and a non-starter in a dense city. On the official side, the efficiency figures aren’t marketing spin — peer-reviewed work pegs round-trip efficiency near 50 percent, with well-documented losses to a side reaction that generates hydrogen — and the projects lean on Department of Energy funding and standard grid-battery safety certification to get built.

The marketing, predictably, runs hotter than the engineering. “Rust-powered,” “world’s largest,” “the next storage revolution” — and, from the buyers, the promise of a “24/7 carbon-free” data center. Some of that is fair; some of it gets ahead of the meter. The useful tell is that the people closest to the technology don’t hide the tradeoff. Form’s own spokespeople and independent analysts will tell you plainly that iron-air is less efficient than lithium and is meant for a different job, arguing that cheap iron and near-free surplus power make the losses worth eating. That candor is a good sign: it’s usually the products whose builders won’t name the tradeoff that you should worry about.

So here’s the honest bottom line. The chemistry is real, the deals are real, and the fit between a rusting-iron battery and a renewable grid is genuine rather than hype. But read the fine print on those record-breaking announcements and you’ll see in-service dates of 2027 and 2029 — these are commitments and factory orders, not plants humming today. Because iron-air loses roughly half its energy on the round trip, you have to overbuild wind and solar to feed it, which means a “24/7 carbon-free” data center is a design goal being financed now, not a switch someone has already flipped.

What’s still unknown is worth holding onto. Whether 100-hour systems perform and hold up as promised across many years at true grid scale is something only time and the young pilot projects can answer. Whether Form’s roughly-20-dollar cost target survives contact with serial production is an open question. And whether the land and efficiency tradeoffs pencil out everywhere, or only in wind-rich, land-rich regions, will decide how far this technology actually spreads. The single most useful thing to remember cuts against the headline that started all of this: iron-air isn’t a bigger lithium battery, it’s a different tool — and the gigawatt-hour number that earns the headline is the one that tells you the least about what it does.

Citations

1. World’s ‘largest’ grid battery part of Google–Xcel Energy agreement. Utility Dive, 2026. utilitydive.com
2. Minnesota Google Data Center Agreement Includes 300-MW Iron Air Battery System. American Public Power Association, 2026. publicpower.org
3. Form Energy and FuturEnergy Ireland Announce Agreement To Deploy First Iron-Air Battery Storage Project In Ireland. Form Energy, March 17, 2026. formenergy.com
4. Form Energy signs 12GWh agreement to supply multi-day iron-air batteries to new US AI data centres. Energy-Storage.News, March 2026. energy-storage.news
5. Google paid startup Form Energy about 1 billion dollars for its massive 100-hour battery. TechCrunch, February 26, 2026. techcrunch.com
6. McKerracher, R. et al. A Review of the Iron–Air Secondary Battery for Energy Storage. ChemPlusChem, 2015. doi:10.1002/cplu.201402238. doi.org
7. Iron–air battery development program, interim report 1976–1977 (Westinghouse Electric Corp.). U.S. Department of Energy / OSTI. osti.gov
8. About — company history and co-founders. Form Energy. formenergy.com
9. Rust-powered battery to deliver 100-hour backup in California (reversible-rusting explanation, Form Energy). Interesting Engineering, 2025. interestingengineering.com
10. Gigantic Form Energy battery to power Google data center in Minnesota. Canary Media, February 24, 2026. canarymedia.com

Disclaimer: This content includes personal opinions and interpretations based on available sources and should not replace medical advice. This content includes interpretation of available research and should not replace medical advice. Although the data found in this blog and infographic has been produced and processed from sources believed to be reliable, no warranty expressed or implied can be made regarding the accuracy, completeness, legality or reliability of any such information. This disclaimer applies to any uses of the information whether isolated or aggregate uses thereof.