Antora Energy

Overview

Antora Energy is a San Jose, California company that builds thermal batteries for industrial heat. Electricity heats blocks of solid carbon; an insulated enclosure holds the heat; the heat comes back out either as process heat and steam for a factory or, through thermophotovoltaic cells, as direct-current electricity. That second discharge path separates Antora from Rondo Energy and Electrified Thermal Solutions, both of which discharge heat only, and it is the basis of the data-center pitch the company began making in 2026. Antora was founded in 2018 by chief executive Andrew Ponec and Justin Briggs, both out of Stanford, with David Bierman, whose MIT doctorate is in thermophotovoltaics. Ponec's previous company, the solar power electronics startup Dragonfly Systems, was bought by SunPower in 2014.

The company is better capitalized than it is proven. It closed a $550 million Series C on July 30, 2026, which Canary Media reported brings total corporate and project financing to roughly $1 billion. Against that sits one 5 MWh pilot, running at a Wellhead Electric site near Fresno since September 2023, and one flagship: a 50 MW / 5 GWh system at POET Bioprocessing's ethanol plant in Big Stone City, South Dakota, which entered commissioning in May 2026 and is expected to reach full operation later in the year. Everything beyond those two assets is a signed agreement Antora will not describe, a federal award still under negotiation, or a stated ambition.

The technology

The charging step is resistive heating. Grid or on-site power runs through solid carbon blocks and warms them, and the timing is chosen to catch cheap or surplus hours. The blocks then sit inside insulation, though Antora has published no standby loss data. Discharge works by radiation. Insulating shutters open, the glowing block radiates light onto a heat exchanger, and the exchanger raises steam or hot gas for the host plant. Antora's September 2023 pilot announcement claimed storage above 1,800°C, which it called a record for a full-scale thermal battery; its 2026 materials describe temperatures up to 2,400°C. The company has not publicly reconciled the two numbers, and the second may be a design ceiling rather than a demonstrated result.

Thermophotovoltaic (TPV) cells are III-V semiconductor devices that convert radiated infrared and visible light into DC electricity with no moving parts. Antora calls its discharge unit the Solid-State Heat Engine. The company opened a dedicated TPV cell line at its then-Sunnyvale headquarters in January 2023 with initial capacity of 2 MW of cells a year, and said the cells convert heat to electricity at greater than 40% efficiency. That figure is a company claim, and no independent verification of it has been published. The 41.1% (± 1) TPV efficiency that carries peer review belongs to a different group: LaPotin et al., published in Naturein April 2022, working at MIT's Department of Mechanical Engineering and NREL, using a 1.4/1.2 eV tandem cell at 2.39 W/cm² against a 2,400°C emitter. It is not Antora's measurement and should not be read as validating Antora's.

Antora also says the carbon storage block uses no critical minerals. The claim covers the block, not the III-V cells, which depend on gallium- and indium-class inputs and specialized epitaxy.

No credible public source gives a round-trip efficiency figure, an installed cost, or a delivered heat cost for an Antora system. That absence is easy to fill badly, because vendors in this sector quote round-trip efficiency two incompatible ways. Electricity-in to heat-out can exceed 95%, since resistive heating is close to lossless, which is the basis of Rondo Energy's claimed figure above 97%. Electricity-in to electricity-out is capped by the TPV conversion step, so roughly 40% on the best published cells and lower once storage losses, optical coupling, and parasitic loads are counted. Those two numbers measure different things and cannot be compared.

The consequence for the data-center pitch is an inference from the published efficiency figures, not a claim any source makes: the 2024 NREL Annual Technology Baseline assumes 85% round-trip efficiency for utility-scale lithium-ion, so a TPV-discharged thermal battery would return under half the electricity a lithium-ion system returns from the same input. Any advantage over lithium-ion for a data-center load has to come from the price of the charging power, the achievable duration, and the capital cost per kWh. Efficiency will not supply it.

Deployments

Antora's deployment record has three tiers. One pilot has operated for nearly 3 years, one commercial-scale project is in commissioning, and everything else is announced, contracted on undisclosed terms, or still being negotiated with a federal agency.

Funding & investors

Antora has raised three disclosed equity rounds. The Series A, more than $50 million announced on February 16, 2022, was co-led by Breakthrough Energy Ventures and Lowercarbon Capital, with Shell Ventures, BHP Ventures, Grok Ventures, Trust Ventures, Overture VC, Impact Science Ventures, and existing investor Fifty Years. The $150 million Series B, announced February 22, 2024, was led by Decarbonization Partners, the BlackRock and Temasek partnership, joined by Emerson Collective, GS Futures, The Nature Conservancy, and a subsidiary of NextEra Energy Resources. The $550 million Series C closed on July 30, 2026, described by the company as oversubscribed and co-led by G2 Venture Partners and Eclipse, with Ribbit Capital, Salesforce Ventures, Activate Capital, John Doerr, Westly Group, StepStone Group, and Liberty Mutual Strategic Ventures new to the register.

The headline rounds sum to roughly $750 million of equity, which is arithmetic, not a published figure. At the Series B, Antora said the total investment it had attracted to that point exceeded $230 million, implying about $30 million of earlier or non-equity capital beyond the two round totals. Canary Media put total corporate and project financing at around $1 billion after the Series C, which counts project debt and equity together and should not be read as money raised into the company. Grok Ventures, the family office of Mike Cannon-Brookes, provided project financing for Big Stone in an undisclosed amount. No valuation has been disclosed for any round; a roughly $500 million figure that appears on third-party research sites dates to the 2024 Series B and is an estimate, not a company disclosure.

Antora says the Series C will fund large-scale project deployment, added production capacity, a second US manufacturing hub in an undisclosed location, and domestic supply chain work. Its San Jose campus reached more than 100,000 square feet across three buildings in spring 2026, roughly double the prior footprint. Antora has never published a nameplate manufacturing capacity for thermal batteries in GWh or units per year. The only public capacity number is the 2 MW per year TPV cell line from 2023.

Federal support is modest next to the private rounds. In November 2023 Antora announced more than $4 million from the California Energy Commission and ARPA-E to expand TPV production. In June 2024 ARPA-E selected the company for up to $14.5 million under its SCALEUP program for the combined heat and power product and pilot-scale TPV manufacturing. A SCALEUP selection begins award negotiations and is not money received, and no public confirmation that the award was finalized, or of amounts obligated, could be located.

Tax credit eligibility is unsettled. Section 48(c)(6)(C) of the Internal Revenue Code defines thermal energy storage property around building HVAC systems and heating or cooling a building interior, and it is not clear that a heat-only industrial unit fits. A unit configured to discharge electricity through TPV looks closer to the Section 48(c)(6)(A)(i) branch covering property that stores and delivers energy for conversion to electricity. That reading is inference from the statutory text. No IRS guidance, private letter ruling, or Antora statement addressing the point could be located, and no evidence was found that Antora has claimed or received Section 45X or Section 48C benefits.

Market & competition

The gap between electricity and gas prices is the business case. US industrial customers paid an average retail electricity price of 8.71 cents per kWh in May 2026 and 8.83 cents year to date, per EIA's Electric Power Monthly Table 5.3 released July 23, 2026, with 2025 and 2026 values flagged as preliminary. Natural gas delivered to US industrial consumers averaged $5.23 per thousand cubic feet in 2025, per EIA data released July 31, 2026. Converting both to a common energy basis at 3,412 Btu per kWh and roughly 1,027 Btu per cubic foot of pipeline gas, that is about $25.50 per MMBtu for electricity against about $5.09 for gas, a gap of roughly 5 to 1 before accounting for the combustion losses in a boiler, which narrow it somewhat. That conversion is arithmetic on the two EIA series, and the series cover different periods: one monthly in 2026, one annual for 2025. At average tariffs, an electric thermal battery does not compete with a gas boiler on fuel cost. It works only on power priced well below the industrial average, which means surplus-hour pricing, a bespoke flexible-load rate, or behind-the-meter generation.

That is why the rate design at Big Stone does as much work as the hardware. Antora and Otter Tail Power developed a thermal market energy pricing rider that lets the battery charge selectively when local generation is in surplus, priced off day-ahead trading, and penalizes draws outside the baseline agreement. The South Dakota Public Utilities Commission approved it 3-0 on July 15, 2025, in docket EL25-017, effective August 1, 2025. Canary Media also reported approvals in Minnesota and North Dakota, and the rider appears in Otter Tail rate books in both states, though those dockets could not be located.

The competitive set is small and moving fast. Rondo Energy, founded in 2020 in Alameda, California, uses firebrick and delivers heat only, above 1,000°C, with a claimed electricity-to-heat round-trip efficiency above 97%. On October 16, 2025, Rondo brought a 100 MWh unit into commercial operation at a Holmes Western Oil Corporation site in California, charged entirely by 20 MW of on-site off-grid solar, a different answer to the same charging-cost problem Antora addressed with a tariff. Rondo also signed a memorandum of understanding with Aramco in May 2024 covering potential gigawatt-scale deployment.

Electrified Thermal Solutions, an MIT spinout from 2021, was selected for more than $40 million of DOE funding in 2024 and commissioned its first commercial-scale 20 MWh Joule Hive unit at Southwest Research Institute in San Antonio in January 2026. The unit has a peak brick temperature of 1,800°C and adjustable hot gas delivery to 1,500°C. Fourth Power, also founded in 2021, is the closest technical analogue, pairing graphite and liquid tin with TPV, though it targets grid electricity storage rather than industrial heat. It raised a $19 million Series A in 2023 and a further $20 million announced in September 2025.

On the public record as of August 2026, Antora is the only one of these with a GWh-scale project in commissioning, an in-house TPV cell line, and a system that can discharge as either heat or power. Rondo has the operating hours. Big Stone at 5,000 MWh is 50 times the energy capacity of Rondo's Holmes Western unit, and Rondo's unit is fully commercial while Antora's is still being commissioned.

Key considerations

One ethanol counterparty accounts for Antora's entire commercial-scale portfolio, and that project is in commissioning rather than in full operation. The 2023 Fresno pilot is 5 MWh. Set against that, the company raised $550 million while declining to describe the pipeline behind the round: no counterparty names, no confirmation that any agreement is binding, no contracted capacity, no timing. That pattern sometimes converts into a book of business and sometimes does not, and the public record does not say which case this is.

The tariff that makes Big Stone work is also the hardest thing about repeating it. A thermal market energy pricing rider is a negotiated instrument approved by a specific state commission for a specific utility. Every new market means a new filing, a new utility counterparty, and a different commission, each with its own view of how a large flexible load should be priced against the interests of other ratepayers. The scale ambition compounds the problem: chief operating officer Justin Briggs told Canary Media that future industrial projects would be similar in size to Big Stone and that power-sector systems could be 5 to 10 times the scale of its 5 GWh, which would put them at 25 to 50 GWh. Antora has not published a charging or interconnection rating for Big Stone, so the load such projects would add cannot be derived from public figures. Interconnection queue position and local capacity would bind that plan as tightly as anything in the technology.

The performance record is thin for a company of this size. No round-trip efficiency figure, no installed cost, no delivered heat cost, no third-party test validation, and no published discharge temperature-versus-time curve. That last gap matters because Pasquale Romano, chief executive of the competing thermochemical storage firm Redoxblox, argued in Energy-Storage.news on December 31, 2025, that thermal storage vendors advertise peak internal temperatures while large-block and packed-bed systems show declining temperature through discharge, which can force costly retrofits at the host plant. Romano has a commercial interest in that argument, and Antora has not published data that would settle it.

The federal money is contingent. The ARPA-E SCALEUP award was a selection for negotiation, and the Summit cement project remains under award negotiation on DOE's own page after Summit changed hands. The second US factory is an intention so far, with no site, size, capex, or date disclosed.

Supply chain adds one more variable, since the III-V cells depend on gallium- and indium-class inputs and specialized epitaxy while the prohibited foreign entity rules introduced by the One Big Beautiful Bill Act begin to apply in 2026. For Section 48E storage, IRS Notice 2026-15 requires at least 55% of direct cost to come from non-prohibited sources for projects beginning construction in 2026, rising to 75% from 2030. That test binds the taxpayer claiming the credit, normally the project owner, rather than Antora as equipment vendor.

Sources