Material Insights · Understanding the Material World

Carnot batteries: Engineering and performance

Why efficiency hinges on the quality of the turbines

What this tab covers Engineering and operating performance: round-trip efficiency, whether power and energy can be scaled separately, footprint and siting, and safety.

Carnot batteries can be sized freely in MW and MWh and sited almost anywhere, including at retiring coal plants. Their weak point is efficiency, which depends heavily on turbomachinery quality.

Round-trip efficiency

Mixed

A review of studies puts Brayton PTES at 52% to 70% and Rankine designs at 45% to 65%. NLR models 66% for a silica-particle design against 57% for molten salt with methanol, and 54% for its reference nitrogen cycle. Measured results are few: Isentropic reached 77% when its 150 kW demonstrator ran at about 10 kW, SPIC claims over 65% for its 1 MW pilot in Beijing, and Siemens Gamesa's simpler heater-and-steam-turbine rock store managed about 25%. Every point of compressor and turbine efficiency matters: in NLR's heat-and-power variants, round-trip efficiency falls to about 35% to 40% if compressor and turbine (polytropic) efficiency drops from 90% to 85%.12349

Brayton / Rankine PTES, reviewed 52-70% / 45-65%1 NLR silica vs salt PTES (modelled) 66% vs 57%3 Isentropic test (about 10 kW on a 150 kW unit) 77%2 SPIC 1 MW / 4 MWh pilot (company claim) over 65%9 Siemens Gamesa rock store, electricity out about 25%2

Power and energy decoupling

Favourable

Yes. Turbomachinery sets MW and tanks of salt, sand or rock set MWh. NLR's silica design falls from USD 245 per kWh of electricity at 10 hours to USD 38 at 100 hours, as fixed machinery is spread over more storage. For Isentropic's design the storage part was estimated at only EUR 13 to 21 per kWh.23

Silica PTES capital cost, 10 h vs 100 h USD 245 vs 38 per kWh3 Isentropic, energy and power parts EUR 13-21 per kWh; EUR 350-797 per kW2

Footprint and siting flexibility

Favourable

Stores need no special geology or water, unlike pumped hydro or cavern air storage, but they are not dense: 10 to 30 kWh per m3 for Brayton systems. A natural fit is a retiring coal plant, which already has a grid connection, turbine hall and steam cycle; in Chile, 2019 agreements committed eight coal units (1,047 MW) to retire within five years.1

Energy density, Brayton PTES 10-30 kWh per m31 Chile coal units due to retire 8 units, 1,047 MW (2019 agreements)1

Safety and thermal runaway risk

Favourable

There is no electrochemical runaway. Risks are those of a thermal power station: pressurised gas machinery, hot salt that can leak and freeze, and in some designs flammable cold-store liquids such as methanol or hexane. Pressure-equipment, turbine and process-safety rules apply instead of battery codes.24

Sources

  1. Carnot Battery development: A review on system performance, applications and commercial state-of-the-art (Vecchi et al., Journal of Energy Storage 55, 105782) University of Birmingham repository, 2022
  2. Review of Carnot Battery Technology Commercial Development (Novotny et al., Energies 15(2), 647) MDPI, 2022
  3. Advancing Pumped Thermal Energy Storage Performance and Cost Using Silica Storage Media (Applied Energy 387, 125567) National Laboratory of the Rockies (formerly NREL), 2025
  4. Design-Point Techno-Economics of Brayton Cycle PTES for Combined Heat and Power (Neises and McTigue, GTP-24-1410) NREL (now National Laboratory of the Rockies) / OSTI, 2024
  5. World-first 560 C Carnot battery unveiled as SPIC launches ultra-high-temperature heat storage (news report of company claims) ESS News, 2026