A San Francisco company is borrowing the plumbing of the petroleum age and running it in reverse. The plan: turn American farm waste into a thick black liquid, pump it a mile below the surface, and leave it there.
In an industrial park near the San Francisco shoreline, a small group of engineers spends most days on the phone with farmers in Kansas and well operators in Colorado. They are not selling fertilizer. They are not buying mineral rights. They are buying tons - specifically, tons of carbon dioxide that nobody else has figured out how to remove yet.
Charm Industrial is, by most useful measures, one of the largest carbon dioxide removal operators on Earth. That is a low bar. The industry is roughly a decade old and most of its competitors are still on slide decks. Charm is on a forklift, hauling a substance called bio-oil out of a pyrolyzer in the corn belt and into an injection well in the high plains.
The company sells permanent carbon removal to buyers like Microsoft, Boeing, JPMorgan Chase, Shopify and the Frontier coalition. It publishes a public ledger of every ton delivered, batch by batch, with measurement and verification baked in. There is no rainforest accounting magic. There is a barrel, a well, and a receipt.
In 2015, Peter Reinhardt was running Segment, the customer-data company he had built into a billion-dollar concern. He wanted to offset the firm's emissions. He looked into Indonesian rainforest credits. The closer he looked, the worse they got.
This is a familiar story by now. Voluntary carbon markets have been pulled apart by investigative journalists for years - credits that didn't represent real removals, projects that would have happened anyway, forests that burned the year after they were certified. The market mostly survived on the assumption that the alternative was worse.
The problem he kept landing on was permanence. Trees decay. Soils get tilled. Direct-air-capture machines, while elegant, demand staggering amounts of electricity. Whatever Charm built, the carbon had to stay down. That ruled out most of what the climate establishment was selling.
Reinhardt teamed up with Kevin Meissner in 2017, and the two were soon joined by Shaun Meehan and Kelly Hering - all four hardware-minded operators who had spent time inside Planet Labs or its orbit. They started where most climate startups now would not: at the cheapest, most boring end of the periodic table.
The early idea was hydrogen. The company pivoted. By 2020, Meehan had stitched together two unfashionable facts. The first: agricultural residues - the stalks, husks and stover left in fields after harvest - can be cooked in the absence of oxygen and turned into a liquid called bio-oil. The chemistry, called fast pyrolysis, dates to the 1980s. The second: that bio-oil is acidic, viscous, and behaves remarkably like the heavy crudes the U.S. oil and gas industry has been pumping in and out of geological formations for a hundred years.
It was an unfashionable bet at an unfashionable time. The hydrogen people thought it was off-topic. The direct-air-capture people thought it was a sideshow. The petroleum engineers thought it was charming, in a noncommittal way. Charm took the name. Then it took the orders.
Charm's core product looks deceptively simple. A mobile pyrolyzer - a steel oven the size of a shipping container - sits near a farm. Corn stover and other crop residues go in. In seconds, at around 500 degrees Celsius and in the absence of oxygen, the biomass cracks into three things: a gas that the pyrolyzer recycles for its own heat, a solid carbon called biochar, and the headline product, bio-oil.
Bio-oil is dark, dense and faintly sweet-smelling. Crucially, it is stable for geological time when stored in the right rock formations. Charm trucks it to permitted injection wells, where it is pumped underground and, in effect, returned to the family of carbon-rich liquids that the United States has been removing from the ground at industrial scale since 1859.
Fast pyrolysis converts ag residues into a stable bio-oil. Injected into permitted Class I, II or V wells.
The solid carbon co-product. Returned to soils for additional storage and tilth.
Public, ton-by-ton record of every removal delivered. Verified by third parties.
Charm's own hardware - portable, deployable near the biomass to slash trucking emissions.
The fastest way to lose an audience in climate tech is to talk about future capacity. Charm prefers past tense. Its public Carbon Ledger lists deliveries, batches, well IDs and verification partners. CarbonPlan, the nonprofit that grades carbon removal projects with a forensic accountant's energy, has rated Charm as best-in-class on permanence and additionality.
Charm's buyers read like a roll call of the carbon-curious wing of the Fortune 500: Stripe, Alphabet, Meta, Shopify, McKinsey, JPMorgan, Microsoft, TD Bank, Boeing, Workday, H&M. None of them are in the business of buying mediocre offsets quietly any longer. The companies that show up on Charm's ledger want their disclosures to survive an audit.
Charm's mission statement is four words long. "Put oil back underground." It is a joke and a thesis at the same time. The joke is on the petroleum industry, whose business model Charm has gleefully inverted - the wells, the trucks, the chemistry, the regulatory permits, even the workforce. The thesis is more serious. If humanity wants to undo a century of carbon-rich liquid coming up, it might want a company whose entire focus is sending carbon-rich liquid back down.
The company is also unusually candid about what it does not solve. Bio-oil is not a substitute for cutting emissions. It is not a license to keep flying. It is a tool for the unavoidable residue of an industrial economy - the cement, the steel, the long-haul flights nobody has electrified yet. In a sane carbon economy, Charm scales until it is no longer needed, and then it scales some more, because the historical overhang of CO2 is its own permanent customer.
The next decade of climate technology will be won, mostly, by companies whose product can be measured, audited and paid for in dollars per ton. Charm is one of a small handful that has crossed that line. It is not a research project. It is not a thesis startup. It is a company with a P&L, a fleet, and a queue of buyers who do not have the patience for vibes-based carbon accounting.
The interesting question is not whether Charm scales. It is what happens to the country's quietest infrastructure - the wells, the trucks, the regulators - when the cheapest thing to put underground stops being a fossil and starts being a stalk. That is the second half of the bet. The first half is already moving.
Back at the San Francisco office, the engineers are still on the phone with the farmers. The shipping container is still cooking corn stover somewhere in Kansas. The wellhead in Colorado is still humming. None of it is dramatic. That is the point. The story you would write about Charm a decade from now is not a story about heroism. It is a story about the new normal, sitting on top of the old one.