Have a question?

Ask us!
Ask MIT Climate

Is capturing carbon dioxide from the air worth the energy it takes to do it?

The ideal energy source for “direct air capture” of CO2 would be very clean, very cheap, and available around the clock—an extremely rare combination.

 

October 27, 2026

For roughly the last decade, a handful of companies have taken an unusually literal approach to fighting climate change. The warming of our planet is caused by the buildup of carbon dioxide (CO2) in the atmosphere, belched up there whenever we burn fossil fuels or cut down forests. With “direct air capture” (DAC) plants, some startups are sucking a little of that climate pollution back out of the air.

“Direct air capture uses some sort of machine that you push the atmospheric air through, and that equipment removes the CO2 from the air and puts the depleted air back out,” says Howard Herzog, senior research engineer at the MIT Energy Initiative, who has been studying related carbon capture technologies for over 30 years.

Several different DAC technologies have been tested, but most share the same basic steps. First, atmospheric air is blown through enormous fans, the better to process a lot of it quickly. The air comes in contact with a chemical that bonds with CO2. Then conditions like temperature or pressure are changed to get the chemical to release the CO2 as a concentrated stream. Finally, the offending CO2 is injected underground, storing it permanently where it can’t warm the planet.

This process demonstrably works, says Herzog, but there’s a big catch. “You're always going to need a lot of energy,” to run fans, heat chemicals, and compress and transport CO2. Exactly how much energy DAC requires is controversial—only small, early-stage plants are running today, and there’s conflicting public data on how efficient they are—but even optimistic estimates suggest that each ton of CO2 they put into the ground takes over 1,000 kilowatt-hours of electricity:1 more than the average American household uses in a month.2

That makes the math behind DAC challenging. For one thing, it would be easy for a DAC plant to leave the climate worse than when it switched on. “If you hooked up one of these machines to a coal-fired power plant, you'd actually be putting more CO2 in the atmosphere than you remove,” says Herzog. “So you really need low-carbon energy to run them,” like wind or solar power.

DAC’s energy needs are picky in other ways. Because DAC plants are expensive to build, and because they need to process so much air to capture a meaningful amount of CO2, a DAC company that wants to make a return on its investment3 can’t afford to switch off its equipment when the sun and wind don’t cooperate. “Your energy has to be available 24/7,” says Herzog. “So while doing it with renewables would be nice, they're not available 24/7.” That leaves more expensive or niche options, like nuclear reactors or renewables plus batteries.

DAC operators, of course, know this, and build plants where this kind of round-the-clock clean energy is available. For example, the world’s largest DAC plant,4 nicknamed Mammoth, is located in Iceland, where it can run on abundant geothermal power, a type of clean energy used almost exclusively in volcanically active places where high underground heat is found close to the Earth’s surface.

But relying on rare cases like this, says Herzog, is not enough to make a meaningful difference to the climate. Mammoth, he points out, has a capacity of up to 36,000 tons of CO2 removal a year:5 roughly one one-millionth the amount that humans add to the atmosphere each year.6 “If you want this to take a significant amount of carbon dioxide out of the atmosphere,” he says, “you really want to be operating at the billion ton level, in which case you need to go beyond the niche opportunities.”

That points to a second big challenge for DAC: In most times and places, power is scarce and expensive. “We don't have a lot of cheap, reliable, non-carbon energy lying around,” says Herzog. “If we did, we wouldn’t have a climate problem.” In a 2024 study, he and his colleagues estimated that using DAC at a scale of 10 billion tons of CO2 a year (roughly one quarter of human emissions) would require 40% of the world’s current electricity supply7—or nearly all of the world’s clean electricity.8

If we’re planning to build clean power at that scale, Herzog says, we should ask whether sucking CO2 out of the air is the best use of it. By adding a nuclear reactor or large solar-and-battery farm to the electric grid instead of using it to power DAC, “we can back out a coal-fired power plant. And that, in the short term, would have a much bigger benefit.”

Despite these high hurdles, Herzog is a relative optimist about DAC. His case for the technology is that, as we get more and more of our energy from clean sources, our remaining climate pollution will come from the hardest and most expensive sectors to break away from fossil fuels. (Think of things like steel manufacturing, or jet fuel for airplanes.) In the future, if clean electricity is cheap enough, and clean jet fuel remains stubbornly expensive, it might be easier to deal with the last remnants of our climate pollution by extracting them from the air than by stopping them at the source. The goal of early investment in DAC, then, is to make the technology ready for when that day comes.

“There's been a lot of excitement and talk about direct air capture, but I think it's got to be tempered a little,” says Herzog. “It's not going to be a white knight coming in and solving our problems. We absolutely have to reduce our CO2 emissions. And once we get to a much lower level and we want to get to net zero, then direct air capture can help us get over the finish line.”

 

Thank you to Bill Sellicks of the United Kingdom for the question.

Submit your own question to Ask MIT Climate

Get the latest from Ask MIT Climate monthly in your inbox

Read more Ask MIT Climate

 

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license (CC BY-NC-SA 4.0).
Footnotes

1 For a sense of the range of recent estimates, see, e.g.:
Howard Herzog, “Chapter 6: Direct Air Capture,” in Greenhouse Gas Removal Technologies, ed. Mai Bui and Niall MacDowell (Royal Society of Chemistry, 2022), https://doi.org/10.1039/9781839165245-00115.
Terlouw, Tom, et. al. “Life cycle assessment of direct air carbon capture and storage with low-carbon energy sources.” Environmental Science & Technology 55 (2021). https://doi.org/10.1021/acs.est.1c03263.
Jia, Charles, and Donald Kirk. “The inconvenient truth about direct air capture: Realigning climate strategy with physical reality.” iScience 29 (2026). https://doi.org/10.1016/j.isci.2026.116567.
Fasihi, Mahdi, Olga Efimova, and Christian Breyer. “Techno-economic assessment of CO2 direct air capture plants.” Journal of Cleaner Production 224 (2019). https://doi.org/10.1016/j.jclepro.2019.03.086.
Kashtan, Yannai, et al. “Direct air capture has substantial health and climate opportunity costs.” Communications Sustainability 1 (2026). https://doi.org/10.1038/s44458-026-00068-0.
2 U.S. Energy Information Administration: “How much electricity does an American home use?” Updated January 8, 2024.

3 The business model for direct air capture is itself a sticky question, since storing CO2 underground is not a conventionally valuable product. But there are customers willing to pay by the ton to get CO2 out of the atmosphere—a model known as carbon offsets—so the trick for a DAC company is to sell those offsets as cheaply as it can.

4 Measured by capacity, or the amount of CO2 it could process if running at full tilt. In fact, the plant operator’s recent disclosures indicate it is operating very far from capacity. See its audits by the Puro Registry (accessed October 26, 2026), or recent news coverage: Akshat Rathi and Alastair Marsh, “Carbon Removal Pioneer Climeworks Increases CO2 Capture Sixfold,” Bloomberg, September 10, 2026.

5 “Climeworks switches on world’s largest direct air capture plant.” Climeworks, May 8, 2024.

6 Friedlingstein, Pierre, et al. “Global Carbon Budget 2025.” Earth System Science Data 18 (2025). https://doi.org/10.5194/essd-18-3211-2026.

7 Herzog, Howard, et al. “Getting real about capturing carbon from the air.” One Earth 7 (2024). https://doi.org/10.1016/j.oneear.2024.08.011.

8 Ember Energy: Electricity Data Explorer. Accessed October 26, 2026.

Aaron Krol
by Aaron Krol, MIT Climate Portal Writing Team
Howard Herzog
featuring guest expert Howard Herzog, senior research engineer, MIT Energy Initiative
Related MIT Groups
More Resources for Learning
MIT Climate Portal: "How much carbon dioxide would we have to remove from the air to counteract climate change?"
MIT Climate Portal: "What is the most efficient way to remove CO2 from the atmosphere?"
MIT Energy Initiative: "Technologies to remove carbon dioxide from the air: A reality check"
International Energy Agency: "Direct Air Capture 2022" (Report)
Direct Air Capture Coalition: Global DAC Deployments (Interactive Map)
MIT Press: Carbon Removal (Book)
Topics
Carbon Capture
Carbon Removal

Want to learn more?

Listen to this episode of the Ask MIT Climate podcast on direct air capture.

LHF: [00:00:00] Hello, and welcome to Today I Learned: Climate, the show where you learn about climate change from real scientists and experts. I’m your host Laur Hesse Fisher of the MIT Environmental Solutions Initiative. We’ve had people ask us, OK, if climate change is caused by adding too much CO2 in the atmosphere, can we just suck it back out? Won’t that solve our climate change problem?

To answer this question, we spoke with someone who would know.

NM: [00:00:27] My name is Niall Mac Dowell. I'm a professor at Imperial College London, and I've been working for carbon management for about 15 years, particularly, recently, with a focus on greenhouse gas removal, or taking CO2 out of the atmosphere.

LHF: [00:00:40] Prof. Mac Dowell shared with us that this is actually pretty tricky to do at a large scale.

To understand why… well... think about the sky. Maybe you can even see it right now. That’s our atmosphere, and even though you can’t see them, there’s a whole mix of gases swirling around up there: nitrogen, oxygen, argon, water vapor … and CO2 is just one of them.

 So Prof. Mac Dowell asked us to imagine that each molecule of CO2 in the atmosphere is a red marble, and every other molecule is a blue marble.

NM: [00:01:19] So if you imagine a bucket of marbles and if we have a hundred marbles and all of the marbles are red and you're given the task of getting five red marbles. It's already easy. You just grabbed five red marbles. There's no work to do.

 If we imagine now the atmosphere, the air, it's not a hundred marbles, it's a million marbles and of this million, only about 400 of them are red, and everything else is blue. So you can spend a lot of energy, you will have to do a lot of work to search through all the blue marbles to find the red marbles. [skip] For every million tons of CO2 that you want to recover from the atmosphere, you will have to handle — you have to physically move — between 5 and 7 billion — with a B — tons of air, and a ton of air is as heavy as a ton of rock. So it's a big effort.

LHF: [00:02:10] Wow. So, is this really possible?

NM: [00:02:14] Yeah, sure. It's technically eminently feasible. As we've discussed, it requires a lot of energy, it takes a lot of work, to sort through all of the atmospheric marbles to find the carbon, but you can totally do it.

LHF: [00:02:29] And, in fact, people are doing it. Or, at least, trying it out. The technology is called “direct air capture,” because you’re capturing CO2 directly out of the air. There’s a company with three facilities in Europe, there’s a different company with one up in Canada, there’s an oil company testing it out in Texas.

NM: [00:02:50] Every time I turned around, there's news of some, somebody else, you know, proposing a DAC pilot, pilots can come in many shapes and sizes.

LHF: [00:02:59] So if I were to tour the facility, what would it look like?

NM: [00:03:03] When this has been trialed in different places — you're  talking about a bank of units, a bit like shipping containers. So say, you know, two meters by two meters, this kind of thing. And that has a big fan on it, and that fan is sucking air through and it's just sucking the air through and blowing it over some kind of contactor, which will react with the CO2 and that's directly pulling the carbon out of the atmosphere.

LHF: [00:03:28] What is the contactor? Is that a chemical solution of some kind?

NM: [00:03:33] Yeah. So very simply CO2 is an acid. So, whatever you want it to react with, will want to be some kind of base or a caustic or an alkaline material.

LHF: [00:03:44] OK so you have this, like, chemical solution that acts like a sponge, pulling out the CO2 from the air. But then, you gotta wring the sponge and get the CO2 out.

NM: [00:03:56] So you do that simply by adding energy. So if it's in the liquid form, you have to effectively boil it out. And this allows you to recover a pure string of CO2, which you can compress, transport and store.

LHF: [00:04:09] Once you have all this extracted CO2, you have to put it somewhere where it won’t go back up in the atmosphere.

NM: [00:04:17] So this is typically underground. In Texas, for example, people are talking about having direct air capture technologies to directly transport the CO2 into oil fields that are right nearby.

LHF: [00:04:29] If you want to know more about storing CO2 underground or using it to produce other things like building materials, check out our episode 7 from season 2 on carbon capture, which is about capturing the CO2 out of the smokestacks of manufacturing and power plants.

But shipping containers with big fans isn’t the only idea we have to take CO2 out of the atmosphere.

NM: [00:04:54] So. The important thing about greenhouse gas removal or carbon dioxide removal is that it's a portfolio of different approaches, some of which rely on engineered technologies — so this is direct air capture or some forms of what's known as enhanced weathering, so that's reacting CO2 with crushed rock, or bioenergy with carbon capture and storage pathways — so that could be turning biomass into heat, power, mobility, and then you capture whatever CO2 you can.

There are ways in which we can improve and change the management of our natural environment. So one good example of that, one really important example of that, is afforestation. So turning a landscape into a forest and carbon sink, and similarly changing the way in which we manage peatlands wetlands, wetland restoration.

LHF: [00:05:49] Put a pin in those last two —in our next episode, we’ll talk all about using forests to take CO2 out of the atmosphere.

Okay, let’s back up. The bottom line is that it really is possible to take CO2 out of the air. And that’s… sort of an intoxicating thought. Because if there’s one thing we’ve seen on this show, it’s that stopping our CO2 emissions [which Prof. Mac Dowell is going to call “mitigation,” “mitigating climate change”] requires us to change the way we get energy, the way we build, the way we travel, the way we grow food. So is carbon removal the easier way out of climate change?

NM: [00:06:32] In my personal opinion? No.

Greenhouse gas removal through any pathway, whether it's direct air capture, bioenergy with CCS, afforestation, or any other options, is not an alternative to mitigation.

LHF: [00:06:47] Why is that?

NM: [00:06:49] Very simply, most mitigation will be cheaper, just simply more cost-effective than greenhouse gas removal.

We are at, I think, the very early days of developing direct air capture technology. You know, the basic science is sort of there, right? We know how to capture the CO2.

The problem is that what we need to do is be actively removing lots of CO2 at the million tons per year scale minimum.

I mean, the analogy I think is that it's a bit maybe like saying we need to be able to break the sound barrier and Orville and Wilbur right now just managed to get their, you know, the first plane flying for 10 meters or whatever it was, you know, that's sort of where we are.

So it's far too early to rule anything out, but putting all your faith in the manifestation of some kind of technical unicorn, which will very, very cheaply reverse the impact of climate change, I think is brave. And I wouldn't do it.

LHF: [00:07:53] And this is the central challenge with carbon removal. If you look at it as an economist would -- how much it costs to remove or avoid a ton of CO2 --- direct air capture isn’t yet cheaper than pretty much any other option: building wind and solar, even capturing and storing CO2 from smokestacks is cheaper.

 So if it’s so expensive to take CO2 out of the atmosphere, and so much cheaper to avoid putting it there in the first place, why invest in CO2 removal at all? Well, it’s because a whole lot of CO2 has already accumulated in the atmosphere, and that CO2 is going to be warming the Earth for a long time—so we will need to remove it if we’re going to keep global warming in check.

In fact, almost every scenario scientists have come up with for how to keep global warming at relatively safe levels has large-scale carbon removal by the end of the century as one of the key tools in our toolkit.

NM: [00:08:57] We have to mitigate as fast as we can with every tool in our arsenal. And we will also very likely have to think about greenhouse gas removal options as well.

We will need everything. We need renewable energy. We need fuel switching. We need nuclear power. We need demand reduction. We need carbon capture and storage, and we will need all forms of greenhouse gas removal as and when they get going.

LHF: [00:09:24] We’re going to stay on this subject in our next episode, and talk about another way to suck out that CO2, using nature. But for now, I want to thank Prof. Mac Dowell for joining us, and remind you that you can always check out our website or Twitter @TILclimate for a bunch of resources we’ve pulled together: fun facts, other websites, our sources, and educator guides so teachers can use this podcast to introduce climate change in the classroom. As always, email us your questions at tilclimate@mit.edu, and thanks for listening.

Keep exploring

Check out these related Explainers, written by scientists and experts from MIT and beyond.

View All

Related Pieces