Massive underground air-battery project lands $1.76B DOE award

Compressed-air energy storage, a decades-old but rarely deployed technology that can store massive amounts of energy underground, could soon see a modern rebirth in California’s Central Valley.

On Thursday, the Biden administration offered a $1.76 billion conditional loan guarantee for GEM A-CAES, a wholly owned subsidiary of Canadian startup Hydrostor. That federal backing will help secure financing for the Willow Rock advanced compressed-air energy storage (A-CAES) project planned near the town of Rosamond in California’s eastern Kern County.

To finalize the loan guarantee from the Department of Energy’s Loan Programs Office (LPO), Hydrostor must satisfy technical, legal, environmental, and financial conditions — and retain the support of the incoming Trump administration. The project is also awaiting permitting review by the California Energy Commission.

But even this initial federal support is ​“a big deal” for what would be Hydrostor’s first North American project at commercial scale, Hydrostor CEO Curt VanWalleghem told Canary Media. The loan guarantee can back up to 80 percent of the estimated $1.5 billion in project construction costs, as well as covering interest during construction and providing a bridge loan for the federal investment tax credits the project will be entitled to when it’s completed.

“What LPO is saying is this is creditworthy technology that is ready for deployment,” VanWalleghem said.

If all goes to plan, Hydrostor would begin construction late this year, with the goal of starting operations in 2030. Much of the construction will happen aboveground, as Hydrostor installs the compressors that will use electricity to pressurize air, the turbines to turn that air back into electricity, and the tanks and reservoirs containing the water that’s vital to the company’s unique approach.

But the heart of the project will be a cavern, roughly the size of a football field in length and width and about 100 yards high, carved by miners out of the bedrock about 2,000 feet below the surface, VanWalleghem said.

Those caverns will be able to store up to 4,000 megawatt-hours of energy in the form of air compressed to high pressures using cheap excess renewable electricity. Hours, days, or weeks later, that air can be expelled to spin power-generating turbines to feed carbon-free power back to California’s grid, at a capacity of up to 500 megawatts for up to eight hours or longer.

Energy analysts and experts believe that long-duration energy storage (LDES) projects like this are crucial to removing fossil fuels from the grid. Today, lithium-ion batteries make up the lion’s share of new grid storage deployments. But power grids making the transition to renewable energy will eventually need longer-duration storage to fill the gaps during days or weeks of low wind and sun.

If built, Willow Rock would be one of the largest real-world examples of an LDES system — and one of the largest energy storage projects in the world, period. It would take the crown for biggest compressed-air energy storage (CAES) system on the planet, too, beating a 1,500 megawatt-hour CAES project that came online in China last year.

LDES technologies range from novel battery chemistries and thermal energy storage systems that can stretch durations into days at a time, to ​“seasonal” options like producing hydrogen with clean electricity and stashing it in underground caverns for use in generating power months later.

Right now, the most widely deployed LDES technologies are also the oldest. Those include pumped-hydropower reservoirs, which have been around since the 1930s and make up the vast majority of the world’s long-duration storage capacity, and CAES, which has been pursued in various formats, most of them unsuccessful to date.

In fact, beyond the newly built project in China, there are only two operational CAES projects in the world — a 290 MW plant in Huntorf, Germany, built in 1978, and a 110 MW plant in McIntosh, Alabama, built in 1991.

How adding water can solve CAES challenges 

CAES has been held back by its high upfront capital cost and its requirements for large and durable underground salt caverns to store compressed air. It’s also a relatively inefficient way to store energy. What’s more, CAES has a poor round-trip efficiency — the ratio of energy coming out of a storage system compared with the energy going into it — of less than 50 percent.

As befits its name, Hydrostor uses water to solve those problems. Every cavern the company excavates has two shafts connecting it to the surface — one for air and one for water. But while the air shaft is used only when air is injected or extracted, the water shaft is in continuous use to maintain a constant pressure within the cavern.

The water column descends about 2,000 feet from the reservoir above, takes a U-turn at the bottom like the piping under sinks, and then rises to flood the bottom of the cavern, stabilizing pressure within the cavern as air is pumped in and out. ​“It doesn’t matter how much air is in the cavern — it’s that weight of water that maintains the pressure,” VanWalleghem explained.

That offers two key benefits, he said. First, unlike traditional CAES, which has to contend with changing pressures while releasing air, Hydrostor’s method maintains steady output pressures as air is expelled from the cavern. That helps turbines run more efficiently.

Second, the high air pressures of traditional CAES projects require the use of impermeable salt caverns, which are relatively rare geologic formations and in high demand as sites to store oil and gas. By using water as a sort of pressure absorber, Hydrostor’s caverns, in contrast, can be excavated anywhere ​“the rock is hard enough to hold itself open,” he said. About 80 percent of the geology of the U.S. qualifies, according to the DOE.

Hydrostor’s other water-based innovation deals with temperature. Gas gets hotter as it’s compressed and colder as it expands. Traditional CAES systems deal with that cold-air outflow by burning fuel to heat it. Otherwise, it would be cold enough to ​“turn the turbine into an icicle,” VanWalleghem said.

Hydrostor avoids burning fuel — and emitting carbon and air pollutants — by capturing and reusing thermal energy generated during the air-compression process in pressurized water storage tanks that reach about 200 degrees Celsius, he said. The air being released from the caverns is reheated by running it through the same thermal exchange system that heats the water during compression, only in reverse.

Competing against batteries to fill a future need 

These innovations — the ​“advanced” part of its A-CAES designation — allow Hydrostor to achieve a round-trip efficiency of about 65 percent, he said. That’s been proved out in the company’s first 10 megawatt-hour project in Ontario, Canada, which has been running since 2020 and actively bids its energy storage capacity into energy markets.

That’s a lot better round-trip efficiency than traditional CAES. But it’s well below the round-trip efficiencies of about 80 percent achieved by pumped-hydro projects, the other large-scale long-duration contender. On the other hand, Hydrostor doesn’t need to find places to build dams and reservoirs, VanWalleghem said.

It’s also well below typical lithium-ion battery round-trip efficiencies in the high-80- to low-90-percent range. Lithium-ion batteries also have the advantages of being mass manufactured, which has driven down costs dramatically — and, of course, they can be deployed in various formats on open ground, without the need for enormous caverns or dams and reservoirs.

Those advantages have put lithium-ion batteries in the pole position for storing energy on power grids over the past decade. But states with aggressive clean energy mandates like California and New York are hunting for technologies that can last for longer than the typical four-hour duration of today’s lithium-ion battery projects, which is enough to shift solar power from afternoons to evenings, as California is doing today, but not to cover longer and less predictable gaps in electricity supply.

In California, for example, utilities and community power organizations are under a mandate to secure a collective 2,000 MW of LDES to meet the state’s 2035 clean electricity targets. But the first procurements under that mandate — defined as projects that can store at least eight hours of energy — selected a project using lithium-ion batteries to achieve the target, indicating the challenge that alternative LDES technologies will face in competing against a well-tested and ever-cheaper technology.

Last year, the California Public Utilities Commission upped its LDES goals to target at least 1,000 megawatts of storage with at least 12 hours of duration and another 1,000 megawatts of ​“multi-day” storage. Hydrostor has already signed a $775 million contract with a California power provider, Central Coast Community Energy, to supply 200 megawatts of eight-hour-duration energy storage supply over 25 years, and VanWalleghem said that the company is in discussion with other would-be partners seeking various durations of its remaining storage capacity. These agreements will help anchor the Willow Rock project.

Hydrostor has raised $250 million from Goldman Sachs Asset Management and $25 million from the Canada Pension Plan Investment Board. It’s planning to start construction this year on its second large-scale project in Australia and has about 10 more large-scale projects under development in North America and Australia, he said.

“We’re all domestic labor and domestic content, with no supply chain limitations,” VanWalleghem said of Hydrostor’s competitive stance against lithium-ion and other LDES technologies. ​“We also last 50 years with no cost degradation,” unlike lithium-ion batteries, which tend to require replacement after 10 to 15 years.

And, of course, the longer the duration of energy storage required, the less feasible it will be to simply add more lithium-ion batteries to try to achieve it, he said. ​“At some point there’s a pretty big crossover for cost advantage.”