In November 2025, on a cement-plant lawn in Saraburi Province, Thailand, a small group of executives in hard hats cut a ribbon in front of what looks like a windowless industrial container. Inside it sits a stack of stoneware bricks the size of a barn, heated to over 1,000°C by electricity from a nearby floating solar farm. The bricks are now feeding 2.3 megawatts of continuous high-pressure steam into one of SCG’s cement kilns, twenty-four hours a day. SCG calls it Southeast Asia’s first commercial heat battery. Most of the people clipping the ribbon are from a seven-year-old startup in Alameda, California, that almost no one outside the climate-tech world has heard of.

That startup, Rondo Energy, holds 44 United States patent grants on industrial heat batteries. The next-most-prolific companies in the same broad category, electrically charged thermal storage built around a solid medium and discharged as hot gas or steam to industrial processes, are Antora Energy (which uses graphite blocks paired with thermophotovoltaic cells), 247Solar, and EnergyNest. Each holds five US grants since 2020. Stiesdal Storage, the Italian sand-storage firm Magaldi Power, Polar Night Energy of Finland, and the MIT spinout Electrified Thermal Solutions sit between one and six. Add every one of those competitors together and the combined US patent estate of the global heat-battery industry comes to twenty-seven. Rondo alone has 1.6 times that.

For a category that almost no major newspaper covers, this is an unusually lopsided land grab.

What the patents actually claim

Almost every Rondo patent in the database opens with the same two sentences: “An energy storage system converts variable renewable electricity to continuous heat at over 1,000°C. Intermittent electrical energy heats a solid medium.” Then the patent narrows. US 12,234,751 narrows to steam crackers, the towering furnaces at the heart of every petrochemical plant, where naphtha is heated to about 850°C to break carbon bonds and produce ethylene, the feedstock for polyethylene. US 12,146,425 narrows to alumina calcination, the high-temperature burnoff that turns mined bauxite into the smeltable powder aluminum producers feed into electrolytic cells. US 12,146,424 narrows to solid oxide electrolyzers, the high-temperature route to hydrogen that runs more efficiently the hotter you can keep the cells.

These are not arbitrary picks. They are three of the worst processes in the world to electrify. They demand temperatures that sit above what lithium-ion batteries, heat pumps, or resistive heaters can practically provide; they run continuously, not when the sun shines; and they sit at the heart of products no economy is willing to give up. A 2021 paper in Joule by Madeddu and colleagues, titled flatly “To decarbonize industry, we must decarbonize heat,” made the case in numbers: industrial heat is roughly a quarter of global final energy use, and most of it comes from burning fossil fuels. The Rondo patent stack is, in effect, a long methodical bet that the answer to that quarter of the energy economy is a pile of bricks.

The technical trick the patents keep returning to is geometry. US 12,320,277, granted June 2025, claims an array of bricks shot through with internal “radiation cavities,” voids designed so that electrical heating elements can dump energy into the brick stack as infrared radiation rather than as direct conduction. A flowing gas then carries the heat back out through structured channels in the same array, while a moving temperature boundary inside the stack, called a thermocline, keeps outlet temperature locked even as the bricks downstream of it cool off. Rondo claims 97 percent round-trip efficiency and 100-bar pressure delivery, both numbers worth reading carefully: a pressure of 100 bar is what a steam cracker actually wants. The patents are not just thermal storage; they are storage shaped to the inlet flange of an existing industrial process.

How this got built

The intellectual origin sits a continent away. In 2017, Charles Forsberg, a research scientist at MIT’s Department of Nuclear Science and Engineering, published the design for what he called FIRES, for FIrebrick Resistance-heated Energy Storage. Forsberg’s pitch was that nuclear plants and renewables both wanted somewhere cheap to dump electricity at off-peak hours, and that the world’s oldest high-temperature material, refractory brick, could store that energy at up to 1,800°C for a small fraction of the cost of any battery chemistry. His design was commercialized by a Cambridge spinout, Electrified Thermal Solutions, which holds one US grant in our search.

Rondo took a different fork. The company, founded in 2020 by Pete von Behrens and the former Ausra solar engineer John O’Donnell, kept the brick but moved the heating elements out of the brick body and into the cavities, then engineered the discharge path around continuous steam delivery rather than peak power generation. Their first commercial unit went into a Calgren ethanol distillery in Pixley, California in March 2023 at just 2 megawatt-hours. Two years later, a 100 MWh unit, fifty times larger, went online at a Holmes Western Oil enhanced-recovery field in California, operating ten weeks before being announced last October. Eight months after the first ground was broken, the SCG cement unit in Thailand started up.

The customer list is what gives the 44-patent estate its weight. Aramco, the world’s largest oil company, signed a memorandum of understanding with Rondo in May 2024 for a gigawatt-hour-scale deployment across Saudi refineries, hydrogen plants, and oil and gas fields, with Aramco’s senior VP of technology, Ali Al-Meshari, framing it as a path to the kingdom’s 2050 net-zero pledge. SABIC, the largest petrochemical company on earth, sits on Rondo’s strategic investor advisory board. So do Rio Tinto, Microsoft’s Climate Innovation Fund, and Thailand’s SCG, the cement giant that has begun building its own Rondo manufacturing line aimed at scaling production from 2 GWh per year to 90 GWh per year. John Doerr is in personally. The total raised is about $107 million across four rounds, modest by hardware-startup standards, but the strategic investors look less like venture financiers and more like the buyers of the eventual product.

What it means for everybody else

The story in patent filings is rarely about the inventor; it’s about who has to license what later. Industrial decarbonization will eat capital plans for the next two decades, and almost every newly built or retrofitted petrochemical plant, alumina refinery, or grid-scale hydrogen facility will need to answer the question of how to deliver high-temperature heat without burning gas or coal. If the answer turns out to be electrified bricks rather than green hydrogen, biomethane, or carbon capture, the engineering details (radiation cavities, thermocline discharge, integrated steam loops at 100 bar) are now mostly inside one company’s portfolio.

That does not guarantee anything. The 100 MWh unit at Holmes Western Oil has been running for less than a year, and a single seismic event near the California fault line it sits on could rewrite the cost curve. Solid oxide electrolyzers may not scale fast enough to need the heat-storage tie-in Rondo patented for them. Cheap geothermal or small modular nuclear reactors could undercut the case for storage entirely. But the fact that none of those things have happened yet, and that Aramco is studying a gigawatt-hour rollout in the meantime, is exactly the kind of asymmetry that matters to anyone planning the next refinery retrofit.

A pile of bricks at 1,000°C is the kind of object that journalists do not file stories about. It is also, on the current evidence, becoming the most heavily patented answer to a quarter of the world’s energy bill.

Method. Patent counts come from US utility grants in the USPTO bulk grant feed, filtered by assignee organization and grant date. Rondo’s portfolio was identified by assignee match on “Rondo Energy” and includes all 44 grants issued through April 28, 2026. Competitor counts use the same assignee-match approach across companies whose primary product is electrically charged thermal energy storage; pumped-heat-based grid-storage filings, including Alphabet’s Malta Inc. spinout, were not included in the comparison because they target a different application. Counts include continuations and divisionals, which inflate single-invention families. The Joule paper by Madeddu et al. (2021), “To decarbonize industry, we must decarbonize heat,” was identified through OpenAlex full-text search; deployment dates and capacities were verified via company press materials and Latitude Media reporting, October–November 2025.