Like an old racehorse, the West Seminole oilfield, a 12-square-mile patch of dirt on Texas’s far western flank, has been trudging along for years, kept kicking by an elixir its jockey shoots into its rump.

On a recent afternoon, as storm clouds blanketed the field and a winter wind howled, dozens of rusty pump jacks rocked up and down, groaning with each revolution as they sucked out more black gold. The drug that the field’s operator, Occidental Petroleum, injects into West Seminole loosens the oil in the stone beneath the sagebrush—forcing from the rock ever more of the hydrocarbon treasure locked inside its geologic pores. The magic medicine is an old industrial gas with a new image problem: carbon dioxide.

For decades, Occidental has been pumping massive quantities of CO₂ into the ground, juicing the flow of oil in aging fields that have lost the oomph nature originally gave them. The CO₂ frees more oil to rise to the surface, where it can be sold and burned. The petroleum industry has used this turbocharging technique—called “enhanced oil recovery”—elsewhere. But Houston-based Occidental is a global expert. Across thousands of square miles of eastern New Mexico and western Texas, on the iconic swath of land called the Permian Basin, the company nicknamed Oxy has built a multibillion-dollar web of infrastructure to manage vast quantities of the CO₂. The Permian rocks’ structure makes them particularly giving of their oil when their spongelike holes are coaxed with the greenhouse gas in liquid-like form.

Oxy buys CO₂ drilled from natural formations in Colorado and New Mexico, then sends it through thousands of miles of pipelines, goaded by massive pump stations. At the end of the line, the company blasts it into thousands of wells that fan out like toy soldiers across the Permian, a booming chunk of the American West that now accounts, extraordinarily, for some 5% of global oil production. Oxy’s setup amounts to a ruthlessly efficient assembly line for hydrocarbons.

Now, amid rising consumer anger about global warming and ballooning government subsidies for companies working to solve it, Oxy is attempting a stunning CO₂ pivot. It hopes to stop pumping into its fields CO₂ extracted from the earth, and instead deploy CO₂ sucked from man-made sources: from power plants, factories, and even thin air.

The company’s ambition is to build into a core business a process that has long been little more than a science project: “carbon capture and storage,” or CCS. It involves chemically snagging CO₂, typically as it’s wafting out of smokestacks but also from ambient air itself, and then injecting it into subterranean rocks. The goal: Rather than continue to dump CO₂ into the atmosphere, where it’s thickening a chemical blanket that’s warming the earth, humanity can bury it underground, ostensibly forever.

Countless difficulties imperil the CCS dream. Influential environmentalists oppose it, arguing it diverts attention from renewable energy. Beyond principle, technical dilemmas loom. One, now fueling a technological race, is how to slash the cost of capturing CO₂, which remains too expensive to work without subsidy. Another, now evolving into a high-stakes lobbying fight, is how far regulators should go in forcing oil companies to prove that CO₂ they’re sending into rock stays safely where it’s put.

No member of Big Oil is gunning harder for CCS than Oxy. Even before concerns about the coronavirus—and a global petroleum-price war—sent oil-company shares tumbling this March, Oxy was trading around a 15-year low, burdened by debt from a recent acquisition. And its CO₂-fueled position in the Permian is core to its viability. So from its Houston headquarters, whose walls are hung with glamour shots of oil rigs, Oxy is making a series of bold CCS bets. One is on a contraption of massive fans that would suck CO₂ from the air. Another is a bid to establish what amounts to a green-minded polluters’ club: a network in which some of America’s biggest industrial emitters would capture and sell CO₂ to Oxy, delivering it through an as-yet-unbuilt pipeline bankrolled with the help of American taxpayers.

If the plans work, Oxy contends, the company will be able to accurately say that much of its oil is “carbon negative”—actually good for the planet, because extracting it would involve the safe disposal of more CO₂ than burning it would emit. More immediately accretive to Oxy’s bottom line, the move would boost its oil sales and also capture subsidies governments are rolling out to companies that can prove they’re capturing man-made, or “anthropogenic,” CO₂. For Oxy, such a masterstroke might repaint its brown corporate image green. It also could gird the company for a future in which, more oil executives are coming to believe, they will have to somehow neutralize emissions from their hydrocarbons in order to continue selling them.

Vicki Hollub, Oxy’s CEO, tells me that she has concluded that her company’s “social license to operate”—the consumer and political buy-in any firm needs to sell its product—depends on doing “all that we can do to address climate change.” The trick will be to do so while satisfying shareholders. Hollub argues Oxy’s CO₂ expertise is a competitive edge in an industry ever more concerned about a warming planet. “The world will transition out of fossil fuels, but it’s going to take quite a long time,” she says. Along the way, “the last barrel of oil produced in the world” should come not from a new well but from an aging field, helped by CO₂. “The carbon footprint is lower,” she says, “and it’s just more efficient for the planet.” Like lemonade from lemons, her environmental argument leverages her corporate portfolio.

Anthony Cottone, senior director for strategic development at Oxy Low Carbon Ventures, a unit the company created to engineer this shift, is even more direct. Buffeted by today’s push for lower-carbon energy, “oil companies need a reason to exist,” says Cottone, who contends that reason should be to serve up CCS. “It’s a second chance.”

****

As consumers eye a torrent of natural disasters indicative of a warming world, scientific experts say curbing carbon emissions enough to avert the worst consequences of climate change will require deploying CCS at grand scale. The existential task, according to the Intergovernmental Panel on Climate Change, or IPCC, a scientific body whose pronouncements frame conventional wisdom on global warming, is to slash emissions by 2050 to “net zero,” meaning any remaining emissions would have to be offset by pulling CO₂ out of the air.

So far, emissions continue to rise. Although renewable energy use is growing fast, the International Energy Agency projects fossil fuels, which provided 81% of global energy in 2018, still will provide 74% in 2040. If there’s any chance to meaningfully decarbonize the economy, it will require decoupling the emission of greenhouse gases from the burning of hydrocarbons. And that will mean grabbing as much as possible of the approximately 33 billion metric tons of CO₂ that fossil fuel combustion produces annually and siphoning it away.

That is why governments are racing to support CCS—from Sacramento to Washington, from Texas to North Dakota, from Canada to the U.K. to Norway and beyond. Some of them view the technology as a way to sustain their own fossil-fueled economies. Others see it as a bridge to a future when renewable energy is cheap and robust enough to power the world largely on its own.

Carbon capture, in short, presents heavy industry and all who profit from it with the possibility of switching in the public eye from climate villains to climate saviors—without abandoning their reliance on fossil fuel, and largely at public expense. It’s a potential new lease on life at a time when markets—far more than policymakers—are pushing polluters to clean up their acts. In January, BlackRock, the world’s largest asset manager, announced it will reorient its investments to trim their climate risk, including reducing fossil fuel exposure. In February, BP, the British oil behemoth, said it would aim for net-zero carbon emissions by 2050, a plan that depends largely on CCS.

With real money suddenly on the line, the race to scale up CCS has entered a cutthroat, meaning very real, phase. The competition is among the surest signs that global capitalism is undergoing a fundamental transformation to adapt to a warming world. Whether the metamorphosis will actually check climate change is, alas, another question.

****

The promise and the difficulty of large-scale CCS loom in cement and steel over the outskirts of southwest Houston. The W.A. Parish Generating Station was the seventh-largest CO₂ emitter in the U.S. in 2018. It also houses one of the largest carbon-capture projects in the world.

As I approach Parish on a drizzly winter morning, I try to wrap my head around the notion that CCS might anchor a green future. My first glimpse, from several miles away on the highway, is of the plant’s four dominant smokestacks. They tower some 500 feet, dwarfing anything else on the horizon. When I reach the entrance, where the red, white, and blue sign boasts “Pride & Power,” I see a black pile of coal so massive that two yellow earth-moving machines crawling atop it, grunting as they spread the solid fuel, look from my vantage point like toy trucks. The coal arrives from Wyoming, typically twice daily, in trains about 120 cars long.

The carbon-capture project—dubbed Petra Nova, Latin for “new rock”—starts with a 15-foot-diameter pipe that sucks, from the smokestack of one of Parish’s four coal-fired power-generating units, some of the waste gas it coughs out. CO₂ constitutes about 13% of that waste gas, which whooshes through the pipe and into a 300-foot-tall tower. That tower is a vertical maze of tubes, inside which the gas mixes with a chemical, amine, that grabs the CO₂. As the rest of the gas heads into the sky, the amine, in another chemical reaction, releases the CO₂, which flows into huge compressors. Pressurized, the CO₂ enters a pipeline, travels 81 miles to an oilfield near Vanderbilt, Texas, and is injected into the earth—with the aim of helping produce a lot of sellable oil.

The carbon-capture machinery is working as planned, says NRG, the power producer that helped launch the project and owns Parish. In 2017 and 2018, Petra Nova’s first two full years of operation, it captured about 8% of the approximately 32 million tons of CO₂ that the Parish plant produced. The rest has risen into the atmosphere. But that small victory was the objective, Judith Lagano, NRG’s senior vice president of asset management, tells me during a tour of the plant: “It’s doing what it was supposed to do.”

Financially, however, Petra Nova, whose $1 billion price tag was defrayed by $195 million in federal grants, has underperformed. Global oil prices and the quantity of oil the CO₂ has wrung from the field near Vanderbilt have fallen below expectations and triggered $209 million in “impairment losses” for NRG in 2016 and 2017.

Today, Lagano says, NRG has new hopes for Petra Nova’s profitability. One reason is Texas know-how: NRG and its partners expect, as they’re getting more familiar with the oilfield, to boost production. A more immediate reason is Washington politicking: A newly generous carbon-capture tax break is about to start flowing.

****

The tax credit, passed in 2018, succeeds a much-smaller prior version. It results from six years of oil-industry lobbying, in large part by Oxy. When I ask Oxy’s Hollub whether that push was defensive, to ensure her industry didn’t lose ground to renewables, or offensive, to win Oxy market share over oil rivals, she doesn’t flinch. “It was, without a doubt, an offensive move,” she says.

The subsidy gives operators of CCS projects, for 12 years, a tax credit ramping up to $35 for every metric ton of man-made CO₂ they catch and deploy to boost oil production. It gives them a credit reaching $50 for every metric ton they capture and shoot into a different sort of geology, filled not with oil but with salty water. Studies suggest underground “saline aquifers” could be stuffed with quantities of CO₂ that dwarf those in oilfields—indeed, that could entomb global emissions through the end of the century.

Once Oxy won the subsidy, it raced to take advantage of it. In 2018, it created Oxy Low Carbon Ventures, essentially a low-carbon skunkworks. Its marching orders were “to replace natural CO₂ with anthropogenic CO₂” in the Permian, says Robert Zeller III, a buttoned-down chemical engineer and Oxy lifer who was tapped as the unit’s vice president for technology. Zeller and I are speaking at Oxy’s headquarters, in a windowless and mirthless conference room. Oxy won’t let me meet with Zeller in his office, in part for fear, the spokeswoman overseeing the discussion explains, that I’ll see something on the walls that’s too revealing.

In plain English, says Zeller, his unit’s mandate was: “You have a blank palate. Go figure it out.” The team figured it out—through some calculations that bowled them over. Burning a barrel of oil typically coughs out about 8,000 cubic feet of CO₂, Zeller says. The team realized that Oxy typically shoots about the same quantity of CO₂ into the Permian to produce each barrel of enhanced-recovery oil. That meant that if Oxy could capture enough CO₂ to replace the natural CO₂ it was using, it could call that oil “carbon-neutral.” The equation would be even more compelling in certain regions where the geology requires more CO₂ to push out every barrel. CCS-produced oil from those formations could be not just carbon-neutral, but actually carbon-negative.

“That,” Zeller says, “was the eureka.”

To capture CO₂, Oxy is pursuing two parallel paths. The sexy one aspires to grab the gas from thin air. Studies have concluded that if enough CO₂ could be economically vacuumed out of the sky, the climate could benefit materially. A handful of “direct air capture” startups have since become investor darlings. Bill Gates was an early backer of a Canadian firm called Carbon Engineering, which has built a pilot plant in British Columbia. Oxy invested in Carbon Engineering in January 2019, after concluding its technology is particularly scalable in oilfields. (U.S.-based oil major Chevron also invested in the firm. Neither oil giant will say how much it put in.)

The challenge is to slash the technology’s cost. Zeller says Carbon Engineering’s first commercial-scale plant will be up and running in the Permian by 2023. It will be sized to take from the sky 1 million metric tons of CO₂ yearly, roughly as much as is coughed out by 216,000 cars. The plant’s likely cost: $1 billion. “The first of a kind is going to cost too much,” he says, but Oxy intends to position potentially dozens more such facilities in the region, at which point Zeller expects the cost for each to fall at least 30%.

Oxy’s less-sexy strategy is to lock down a supply of anthropogenic CO₂ from smokestacks, Petra Nova–style. It has mapped the locations of America’s biggest emitters: power plants, steel plants, petrochemical plants, and more. Oxy executives won’t show me their map—it’s presumably one of the things on their walls I’m not supposed to see. But emissions data is publicly available, and as CCS enthusiasm spreads, such maps line the walls of more and more offices in the oil patch. They show hotspots in Midwest coal-burning country and throughout the Texas and Louisiana oil colossus. Oxy is now scrambling to line up deals for the CO₂ these polluters are releasing into the atmosphere.

It’s also now lobbying Washington for another subsidy, this one for a pipeline to ferry human-spewed CO₂ to the Permian. In case the government doesn’t concur, Hollub tells me, Oxy is wooing infrastructure-investment funds. Her ultimate goal: Cheaper CO₂, because it and electricity are the biggest costs in Oxy’s Permian operation. By “the mid to latter part of the 2020s,” Hollub says, a shift to anthropogenic gas should cut Oxy’s CO₂ costs by 20% to 30%.

Many other big oil companies, consultants and bankers tell me, are also angling for CO₂ pipeline incentives. “They can’t let anyone know they’re jockeying,” explains one well-placed consultant who doesn’t want to be identified, “because then people would know they’re not doing it for society—they’re doing it for themselves.”

****

To monetize CCS, companies need permission from regulators to shoot CO₂ underground. The rules governing such storage are young, they differ among jurisdictions, and they have yet to be tested. This spring, they’re prompting lobbyist combat in Washington, where the Internal Revenue Service is formulating rules to govern who qualifies for the federal tax credit—and in California, where a potentially bigger financial prize awaits.

In natural CO₂ deposits, rock securely traps gas much as, elsewhere, it traps oil. The IPCC, the global climate-science group, says more than 99% of injected CO₂ is “likely” to stay put for 1,000 years in a well-designed and managed project. What’s more, the IPCC says, CO₂ tends to grow more “immobilized” over time, and thus a project could retain essentially all of its CO₂ “for up to millions of years.”

The question is how to ensure CCS is done well. Regulators want to prevent CO₂ from seeping into other underground areas—either into someone else’s property, or into geologic layers, such as drinking-water sources, where it could cause harm. They also want to make sure it doesn’t waft back up, years or even centuries later. If sequestered CO₂ leaked into the sky, its climate-saving value would be negated, and the polluter that pocketed subsidies for capturing it would have contributed no lasting benefit.

The industry’s D.C. drama boils down to whether an independent third party should verify that oil companies have safely stored as much CO₂ as they claim. An industry group that Oxy helps fund, the Carbon Capture Coalition, whose roster also includes Shell, recently wrote to the IRS that such verification is “vitally important” to the viability of the credit. That puts them at odds with another group, the Energy Advance Center, which lists ExxonMobil and Denbury Resources, both of which have extensive CCS operations, among its members; it opposes mandatory third-party verification. BP recently quit that group, and the company endorsed mandatory verification.

Beyond Washington, however, Oxy is trying to loosen some rules. Even richer than the federal tax break is a subsidy that California is rolling out. When companies can substantiate that they’ve used low-carbon processes to produce transportation fuel they sell in California, the state grants them “carbon credits,” which they can sell to other, less-green fuel producers. A credit representing one metric ton of CO₂ saved currently sells for about $200. And that’s cash, not merely a credit against taxes. Oxy has tussled with California over what it must do to monitor man-made CO₂ that Oxy wants to inject at West Seminole. The source of that gas: two ethanol plants in the Texas Panhandle.

Plainview, Texas, lies between Lubbock and Amarillo, on a highway whose speed limit is 75 miles per hour but doesn’t seem to much matter. The town’s name fits: On the early morning I visit, the pancake-flat landscape, offset by a vast orange sky, affords what seems an endless vista. Dominating the horizon on the eastern edge of town are the seven 807,000-gallon fermentation tanks at an ethanol plant owned by Frisco, Texas-based White Energy.

Ethanol, made in the U.S. primarily from corn, has long been marketed as a clean fuel. But its production, a process of fermentation that resembles—and smells like—beermaking, produces gobs of CO₂. Soon after sunrise, I circle the plant with Brian Steenhard, White Energy’s chief executive. As we linger at the ethanol tanks, he points to a pipe atop one. It belches as much as 350,000 tons of CO₂ yearly. White Energy’s Hereford plant, an hour away, emits about the same amount.

White Energy and Oxy have applied to the California Air Resources Board, known as CARB, to receive credits for capping those pipes and sending the CO₂ to the West Seminole field for burial. Steenhard estimates that if regulators signed off, and if White Energy figured out how to ship to California all the ethanol it produced in Plainview and Here¬ford, the credits would generate between $360 million and $720 million over 12 years. The project could receive federal incentives over that same span, potentially offsetting some $275 million in taxes. CCS projects being hatched elsewhere would be far, far larger.

The obstacle: California has tougher requirements than the feds for ensuring injected CO₂ stays put. But those rules’ effects on the ground still are taking shape. Originally, for instance, CARB wanted Oxy to insert fiberoptic “geophones” into each West Seminole CO₂-injection well to listen for signals that the injection might be prompting an earthquake, Bill Raatz, Oxy’s chief geologist, tells me. Oxy argued that was “a complete nonstarter” for cost reasons, Raatz says, so Oxy asked CARB to let it put a geophone down no more than a handful of the wells and use cheaper aboveground seismic monitors elsewhere.

CARB officials say they won’t comment on their discussions with Oxy and may rule on the West Seminole application later this spring. In March, CARB was reviewing a 520-page document from Oxy that argues that the field would safely sequester CO₂. When I asked to see it, both Oxy and CARB refused to give it to me. Only after Fortune submitted an open-records request to CARB did the agency relent. And only after that did Oxy hand over a heavily redacted version that it had submitted to CARB. That version blacked out big chunks of information the company calls “business confidential.” What wasn’t redacted boils down to a four-word argument that Zeller, the Oxy technology executive, had lobbed at me in Houston. I asked him about the likelihood that man-made CO₂ at Oxy’s Permian fields might escape. Said the engineer: “It ain’t coming out.”

****

A roadside entrance to the West Seminole field is marked by a worn, rusty sign, pocked with what appear to be bullet holes. On sections of the field, among the pump jacks still rocking, chunks of old machinery lie atop the dry brush. The scene evokes an oil-patch graveyard.

Whether this acreage should become a graveyard of a different sort—a resting place for CO₂—is, fundamentally, a question about how humanity will wrestle with the wicked complexities of global warming. Proceeding will signal that a technology widely seen as crucial to addressing climate change may finally scale. Delay will leave the future of that technology in the same place where a lone bird is gliding above West Seminole on the gray afternoon I glimpse the field. It’s the place where, every day, more CO₂ is heading: up in the air.

****

Burying climate change

Carbon capture and sequestration (CCS) shows promise as a weapon against climate change, but the technology has barely begun to be deployed. Here, some facts to keep the science in perspective.

1. A growing slice of the pie

Curbing climate change in line with United Nations goals would require slashing global energy-related CO₂ emissions, now about 33 billion tons a year, to below 10 billion tons in 2050. CCS would have to shoulder about 9% of the carbon-cutting load between now and 2050, and 28% of cuts in 2050 itself, according to the International Energy Agency.

2. So far, slow going

In the United States, total energy-related CO₂ emissions were 5.3 billion tons in 2018. But the U.S. has developed only about 25 million metric tons of CO₂-storage capacity, according to a December 2019 study by the National Petroleum Council, a government advisory group, enough to bury less than one-half of 1% of those emissions.

3. Plenty of room

The U.S. has enough room in geologic formations to store hundreds of years’ worth of emissions from “stationary” sources like power plants and factories, according to the National Petroleum Council’s report.

Aging oilfields account for most current capacity; the greater potential, scientists say, is in underground formations onshore and offshore that naturally hold salty water, not oil.

CCS began in 2019 at Gorgon, an Australian project led by Chevron that’s capturing CO₂ from an underwater natural-gas field and injecting it into a formation under an island. When ramped up, it will be one of the world’s biggest CCS projects, injecting up to 4 million metric tons of CO₂ per year.

A version of this article appears in the April 2020 issue of Fortune.