When it comes to batteries, we’re used to thinking small. Small phones, light laptops, compact electric vehicles. But what the world needs now is batteries that are big.

Big enough to smooth out the unavoidable peaks and troughs of solar and wind farms, and big enough to let us shutter fossil fuel power stations for good.

That means we need to store huge amounts of energy across the globe. The United States alone will need to add hundreds of gigawatts of storage by 2050 to meet its clean energy goals, the Department of Energy calculates.

Pumped storage hydroelectricity, where water is pumped uphill behind a dam, is a proven technology but requires mountainous terrain. Banks of lithium-ion batteries, scaled up from those found in most of today’s gadgets and EVs, can also help. If backup power is needed for more than a few hours, however, the cost of lithium–ion storage rockets. Added to that, almost all the proposed new lithium-ion capacity in the next decade is destined to decarbonize road transport.

But other ways to store energy reliably and durably at scale are emerging, leveraging natural elements and processes. Start-ups are battling to turn the world around us into a battery, using technologies that seem ripped from the pages of science fiction. Here are some of the wildest new ideas for the big batteries of the future.

Kern County, California is best known for its cattle ranches and oil fields. But it may soon be home to an energy storage landmark: One of the world’s largest advanced compressed air energy storage, or A-CAES, systems. 

To build it, Canadian company Hydrostor proposes drilling shafts nearly 2000 feet deep into the earth, and excavating two football field-sized subterranean caverns. On sunny and windy days, the company will use surplus renewable energy to run air compressors that heat and squeeze air into a smaller volume. After extracting the heat, they’ll force the cool, compressed air down the shafts, displacing a reservoir of water. When energy is needed, the operators will release and heat the compressed air. As the air rushes out, it will spin electricity-generating turbines. 

The proposed facility will deliver around 500 megawatts of electricity per hour for eight hours, before it needs to “recharge” by refilling the caverns with more compressed air, Hydrostor says. First, however, the company will need to address concerns about the project – environmental groups claim Hydrostor hasn’t adequately assessed how it will impact threatened species on the surface, like the western Joshua tree. 

CAES facilities (without the “advanced” moniker) have been operating in Alabama and Germany since the 1970s. But these differ from Hydrostor’s technology in several important ways. First, previous plants have used naturally existing underground salt caverns to store the air, and burned natural gas to heat the air as it escapes. Hydrostor’s system doesn’t require pre-existing caverns, and avoids using fossil fuels by capturing and storing waste heat from the compression process, then using it to reheat the air as it is released. 

Hydrostor’s approach is just one way to store energy by squeezing a gas. Italian start-up Energy Dome is forcing CO2 into above-ground vaults, and a U.S. company called Breeze cleverly plans to store compressed air in the vast network of empty gas and oil pipelines currently lying idle across America. 

Because they use a substrate that’s free – air – A-CAES systems are among the cheapest technologies being developed, according to Vince Sprenkle, project manager of the Office of Electricity Energy Storage Program at the Pacific Northwest National Lab in Richland, Washington. “It’s been really hard to beat compressed air energy storage,” he says. In a PNNL analysis of competing technologies it had one of the lowest costs on a levelized basis. That is to say, taking into account the longevity of the system and the costs of operating and maintaining it, as well as the upfront price to build it.  

Capacity: Hundreds of megawatts
Duration: 8-10 hours
Lifespan: 60+ years
Pros: Uses a fraction of the land required for pumped hydro, can be located almost anywhere
Cons: Not suitable for small capacity, short duration projects

When the Swiss company Energy Vault first unveiled its six-armed prototype for a “gravity battery” in 2020, it raised some eyebrows. The concept was appealingly simple: Use cranes powered by surplus solar or wind energy to lift 35-ton bricks high into the air on cables and stack them in a cylinder. When energy is needed, lower the blocks back down. 

The starfish-like system would use regenerative brakes, like the ones used in electrical vehicles, to slow the bricks’ descent and convert that energy into electricity. 

The system appears to have several clear advantages over batteries. First, energy stored by elevating heavy objects doesn’t degrade over time as the chemical energy in batteries does. A gravity storage facility is also likely to last longer—around 50 years, compared to the roughly 20-year lifespan of a battery. 

Energy Vault’s pilot project got positive press, but there were nagging questions about its practicality. Would high winds make it difficult to stack the blocks? Could one defective block or cable disable the whole tower? By 2021, the company had redesigned its system into a giant, enclosed steel matrix in which the bricks get shuttled up and down like elevator cars. 

Its first commercial system, recently built in China, dispatches 25 megawatts of electricity for four hours at a time—about the same duration as lithium-ion batteries. But Energy Vault says future facilities will deliver hundreds of megawatts for around 12 hours. At that scale, the cost could compete with lithium-ion battery plants.

A gravity storage facility would need a storage duration of 100 hours to reduce its costs to half that of a lithium-ion plant, the DOE calculates. To hit that target, other companies are using empty skyscrapers and abandoned mine shafts to build conceptually similar systems, potentially lowering upfront building costs and using still less land. Still others are using slopes instead of vertical shafts—“basically ski-lifts with rocks,” says Alexander Morris, general manager of California Community Power, a consortium that negotiates affordable renewable energy contracts for cities. One railroad-inspired system built on a 20-acre gravel mine in Nevada, for example, will roll more than 200 cars packed with 75,000 tons of material over steep tracks. 

Capacity: Hundreds of megawatts
Duration: 4+ hours
Lifespan: 45-50 years
Pros: Can use existing infrastructure and cheap materials
Cons: Not suitable for small capacity, short duration projects

According to legend, ancient Greek soldiers used bronze shields to focus the sun’s rays on invading Roman ships, setting them on fire. Today, engineers are using a similar concept to store heat energy from the sun. 

One common approach uses circles of concentric mirrors called heliostats to reflect sunlight toward a central tower, topped with a reservoir of molten salt. The beams heat the liquid to more than 500 degrees Celsius, which is then stored in an insulated tank, typically for several hours, although some systems last for days or more.  At night, the molten salt can be used to boil water and turn steam turbines to generate electricity. The technology, called concentrating solar power, has already been deployed in dozens of sun-blasted locations around the world. The largest facility to date is under construction in the desert near Dubai and is expected to be able to dispatch 700 megawatts of electricity for up to 15 hours.

Although the levelized cost of electricity from these farms is on roughly par with other options, their massive up-front construction costs and large environmental footprints have inspired some companies to explore smaller-scale solutions. 

One Bay Area start-up called Antora is using surplus renewable energy to heat blocks of graphite until they glow white-hot, then storing them in heavily insulated boxes. When needed, a window in the box directs a beam of thermal energy at a modified solar panel that converts heat (rather than light) into electricity. Antora can also use the stored thermal energy—reaching up to 1500 degrees Celsius—for industrial applications like making cement.

One advantage of thermal storage is that many common materials can be used to store heat, including metals, rocks, and water, says Morris. In Berlin, for example, engineers have built a massive Thermos-like tank that will keep over 14 million gallons of water hot for 15 hours, fulfilling much of the city’s demand for hot water. Engineers say there’s no reason why we couldn’t even copy the same idea in our own homes, using ultra-insulated hot water heaters. 

Capacity: Hundreds of megawatts
Duration: Up to a week
Lifespan: 30-35 years
Pros: Good for decarbonizing industrial processes
Cons: Potential for toxic byproducts and heat pollution

Heat isn’t the only way to store energy—freezing can also work, in sunny climes. In Jerusalem, a company called Nostromo is installing what it calls IceBricks on the rooftops of shopping malls and other large buildings. The bulky rectangular containers are filled with water mixed with a chemical that lowers its freezing point. When renewable energy is cheap, IceBricks use it to freeze the solution. When temperatures rise, the system circulates warm air from inside the building across the melting ice, cooling it and reducing the energy needed to keep buildings at a comfortable temperature. 

“Huge amounts of energy go toward solving the problems of temperature,” says Morris—from cooling data centers to refrigerating food to ensuring people have access to air-conditioning during deadly heat waves. Nostromo’s IceBrick and other ice storage technologies can’t put electricity back onto the grid during a shortage. But they could play a key role in reducing peak energy use during heatwaves and avoiding dangerous blackouts. As heatwaves get more frequent and extreme, some of us may soon be recharging our own ice batteries.  

Capacity: Roughly 2.5 megawatts
Duration: 4 hours
Lifespan: 20 years
Pros: Lowers cost of air conditioning and refrigeration by reducing electricity use
Cons: Not a true “battery” because it isn’t collecting electrons from the grid or returning them when needed

Day after day, a cargo crane on Kodiak Island, Alaska, lifts and lowers metal shipping containers on and off a dock in the bay. It might consume two megawatts of electricity one moment, then nothing at all a minute later. Such an extreme fluctuation in power consumption would normally put a sizable strain on the island’s delicately balanced electrical grid, which runs almost entirely on renewables. But this crane doesn’t need to, because it gets all the power it needs from two spinning flywheels. 

Although they’re ancient as pottery wheels, flywheels have recently been revitalized to store renewable energy in applications ranging from capturing energy in braking commuter trains, to providing fast bursts of power to launch aircraft. 

There’s a perception that flywheels are mostly useful for these short-term energy needs, but that’s not the case, argues Seth Sanders, an electrical engineer at UC Berkeley who co-founded a flywheel company called Amber Kinetics. 

Amber’s flywheels are smooth, 2,000-pound steel discs stored in vacuum-sealed chambers. A magnet slightly levitates them off their bearings to keep friction from slowing them as they spin. Surplus energy from wind or solar powers an electric motor that starts them spinning, reaching up to 9000 rpm as they convert that electricity into kinetic energy. Once moving, it only takes a small amount of energy to keep the wheels going—about the same as a household light bulb, according to the company. When energy is needed, the system changes its configuration: Now the motor acts as a generator, turning mechanical energy back into electricity for use on the grid. 

Each individual flywheel can discharge 32 kWh –  8 kilowatts of energy per hour for 4 hours before it must recharge. That’s a relatively small amount of energy, with the same limited duration as a lithium-ion battery. But unlike a battery, the flywheels never lose their charging capacity. Their rugged design makes them ideal for time-shifting renewable energy in remote and rural places with harsh weather, the company says. So far, Amber’s flywheels have been installed to stabilize electrical grids in Hawaii, Japan, Taiwan, the Philippines, rural Australia, and Tibet.

Capacity: Currently 8 kilowatts (0.008 megawatts) per wheel
Duration: 4 hours
Lifespan: 30 years
Pros: Durable, rugged, portable
Cons: Tiny capacity; remains to be seen if it can reach megawatt scale

Emily Underwood is a science journalist at Knowable Magazine. She lives in Coloma, California.

_________________
Noun Project Icons
Squeeze: Magicon
Ice: Natthapong Mueangmoon
Spin  Andrejs Kirma
Gravity Lastspark