Mantel.

Somewhere in Cambridge, a steel box the size of a shipping container has been quietly swallowing carbon dioxide for nearly two years. Inside, a salt that would be solid in your kitchen runs as a liquid, drinking CO2 out of hot industrial exhaust and handing back high-grade heat as a thank-you. This is Mantel. It is not trying to save the world with a slogan. It is trying to make the boring math of carbon capture finally work.

Mantel sells a chemistry, not a fantasy. The company's molten borate materials sit where emissions are actually born - inside boilers, kilns and furnaces - and capture more than 95% of the CO2 that passes through. The hardware looks industrial because the customers are industrial: cement, steel, pulp and paper, power plants, refineries, increasingly data centers. The places nobody knows how to clean up are exactly the places Mantel is aiming at.

Carbon capture had a heat problem

Here is the inconvenient part nobody puts on a pitch deck. Conventional carbon capture hates heat. The liquid amines and solid sorbents that dominate the industry work best when flue gas is cooled down first, then need large amounts of energy to be regenerated afterward. So you spend energy cooling the exhaust, more energy stripping the CO2 back off, and you end up with capture that is expensive enough to make most plant managers politely decline.

Heavy industry, meanwhile, is hot by definition. A cement kiln does not run at room temperature. A steel furnace is not asking to be cooled to suit a chemist. The mismatch between how capture wants to work and how industry actually works is the reason so many decarbonization plans quietly stall at the spreadsheet stage.

A cold email, a hot idea

In 2016, Cameron Halliday cold-emailed MIT professor Alan Hatton asking to spend a summer on carbon capture. That email is, more or less, the origin of the company. By 2019, Halliday was testing lithium-sodium ortho-borate salts and watching them do something the field had struggled with for years: absorb over 95% of CO2, then do it again, and again, with almost no degradation after 50, 100, and even 1,000 cycles. The salts stayed liquid at high temperature, sidestepping the brittle cracking that wrecks solid sorbents.

Halliday finished a famously demanding joint PhD-MBA at MIT, met his co-founders in the school's Climate and Energy Ventures class, and in 2022 turned the lab result into a company. The founding team is small and technical: Cameron Halliday as CEO, Sean Robertson as CTO, and Danielle Rapson as COO. The bet they placed is almost rude in its simplicity - stop fighting the heat, and start using it.

Three people, one molten idea, and a chemistry department's worth of stubbornness. // The founding team, per MIT and company records.

Salt that pays its own energy bill

The mechanism is elegant. Mantel's molten borate absorbs CO2 directly from hot exhaust, and the absorption reaction itself releases high-grade heat. Instead of treating that heat as waste, Mantel captures it and turns it into steam - a thing factories actually want to buy. The recovered energy offsets what it costs to regenerate the salt and release a pure stream of CO2 for transport or storage. Capture stops being a pure expense and starts behaving like a process with a byproduct worth money.

The numbers Mantel cites are the whole argument. The company says its system can use roughly 3% of the net energy of state-of-the-art capture, and cut the cost of capturing a ton of CO2 by more than half. In an industry where capture cost is the difference between a press release and an actual installation, halving it is not a footnote.

It also helps that the material refuses to wear out. The reason most capture chemistries struggle is that the thing doing the capturing degrades - solid sorbents crack under thermal stress, liquids break down, and performance drifts downward with every cycle until the economics quietly collapse. Mantel's salts stayed liquid and stayed effective through 50 cycles, then 100, then 1,000, with the kind of flat performance curve that engineers find suspicious until they run it themselves. A material that does not degrade is a material you can plan a twenty-year plant around, and that durability is as much the product as the chemistry itself.

Four numbers doing the heavy lifting that adjectives usually try to. // Figures per Mantel and MIT News.

When oil majors fund the cleanup crew

In September 2024, Mantel closed a $30 million Series A. The notable part is who co-led it: Shell Ventures and Eni Next, the venture arms of two oil and gas majors. There is a certain irony in the companies most associated with emissions writing checks to a startup whose entire purpose is catching carbon - but it is also a useful signal. Energy incumbents have seen a great deal of capture technology, and they put money behind this one. The round also drew bp Ventures, Engine Ventures, New Climate Ventures, Hartree, Arosa Ventures, Vale Ventures, Newlab and MCJ Collective.

Money is one kind of proof; a factory is another. Mantel has partnered with Kruger Inc. to build its next-generation system at a plant in Quebec, moving the technology out of an incubator and onto a real industrial site for a multi-year test. The company has also said it is in conversations with roughly 100 industrial partners worldwide - the kind of pipeline that suggests the cost math is landing with the people who actually sign capture contracts.

Affordable enough to actually happen

Mantel's mission is unglamorous on purpose: make carbon capture cheap and reliable enough that hard-to-abate industries can hit net zero without going bankrupt. The hard-to-abate sectors - cement, steel, chemicals - produce a large share of global emissions precisely because nobody has handed them an affordable way to clean up. A capture technology that recovers its own energy as steam is, in that light, less a moral statement and more a business proposition that happens to be good for the atmosphere.

The boring revolution

The future of decarbonization will probably not be won by the most dramatic technology. It will be won by whatever is cheapest to install on a Tuesday at a cement plant in a country with thin margins and a tight deadline. Mantel is betting that a liquid salt, a recovered stream of steam, and a halved cost per ton is that unglamorous winner. If it is right, the heavy industries we have spent decades calling impossible to clean become merely expensive - and then merely ordinary.

There is also a timing argument worth taking seriously. Regulation, carbon pricing and customer pressure are all pushing heavy industry toward emissions targets that, for now, most plants have no realistic way to meet. The companies that figure out affordable capture first will not just clean up their own output; they will sell that capability to everyone else scrambling to comply. Mantel's roughly 100 partner conversations are a hint that emitters already see capture as something they will have to buy, not just admire from a distance.

Back to that steel box in Cambridge. For nearly two years it has been doing the least romantic thing imaginable: catching carbon, releasing heat, and refusing to break down. It is not a metaphor. It is the prototype of a future where capturing CO2 is just another line item that pays for part of itself. Mantel started with a cold email and a salt that stays liquid when everything else cracks. The bet now is whether the rest of heavy industry decides to pour it.