Potentially cheap 24-hour storage is renewables' missing piece. A CO2 dome in Sardinia is the first commercial bet to crack it.

Wind and solar are the cheapest, fastest things to put on a grid. They are also intermittent, and a grid running on them needs a way to ride through 8-to-24-hour lulls. Lithium-ion batteries handle the first four hours; beyond that, the economics get ugly. A 200 MWh plant switched on in Sardinia in 2025 suggests the long-duration gap is finally addressable, and the answer is a white dome the size of seven soccer fields holding about 2,000 metric tons of carbon dioxide.

The plant belongs to Energy Dome, a Milan-based storage startup founded in 2020. Its "CO2 battery" is a closed-loop thermomechanical system. When grid power is cheap or abundant, electric compressors squeeze CO2 gas into a liquid and pump it into carbon-steel tanks. The compression throws off heat, which the system captures in a proprietary thermal-storage material. On discharge, the liquid is released from the tanks, warmed with the stored heat, flashed back into a gas, and pushed through a turbine to make electricity. The CO2 then flows back into the dome to start the cycle over.

The mechanism is not new in physics, but the category has been stuck on a siting problem. Legacy compressed-air energy storage needs salt caverns, depleted gas fields, or hard-rock aquifers to hold high-pressure air, which limits where it can be built. Energy Dome's process runs on liquefied CO2 at much lower pressures, stored in standard steel tanks, on flat ground. The dome itself is mostly a low-pressure gas holder, the same job as a balloon. That single change is what makes the technology portable to almost any grid.

The economics get interesting past the four-hour mark. Lithium-ion is the default for short-duration storage: cheap modules, mature supply chain, well-understood fire and degradation behavior. But an eight-hour lithium system at the same power output roughly doubles the cost footprint, because the buyer is paying for energy capacity, not power capacity, and cells are the expensive part. Long-duration storage is the category trying to break that curve with thermomechanical, flow, thermal, or hydrogen approaches. CO2 is one of several bets in that race.

Traction around Energy Dome is now real, if early. The Sardinia plant, the first commercial deployment, can deliver 200 MWh — enough to power about 18,000 Italian homes for 10 hours. The company has a 30 GWh pipeline across five continents, with a Google-backed project at Salt River Project in Arizona and a storage offtake agreement with ENGIE on file. The European Investment Bank has listed the project in its CATALYST pipeline, and EPRI has published a feasibility study on the technology. Latitude Media has tracked the business model, which positions the company as an energy-storage-as-a-service provider rather than a hardware vendor.

A 200 MWh site is one data point against 30 GWh of announced projects, so the real test is the gap between the two. Lithium-ion still rules anywhere storage duration is four hours or less, and nobody is going to retire peaker plants for a first commercial plant of any size. The Sardinia plant is meaningful because it proves the thermomechanical CO2 cycle can be operated as a grid asset, with steel tanks and a turbine, on flat ground.

What to watch next: round-trip efficiency on a sustained basis, capacity factor at the Salt River Project site, and whether ENGIE's offtake agreement turns into a second commissioned plant. If those land in 2026 and 2027, the grid's long-duration ceiling moves with them.