Want CNET to notify you of price drops and the latest stories?

Machines that suck CO2 out of the air promise to reverse emissions. Will they work?

Carbon dioxide removal is proving costly and difficult to scale up.

Jackson Ryan Science Editor
Jackson Ryan is CNET's award-winning science editor. He used to be a scientist but he realized he was not very happy sitting at a lab bench all day. Now he has the best job in the world, telling stories about space, the planet, climate change and the people working at the frontiers of human knowledge. He also owns a lot of ugly Christmas sweaters. Tweet him.
Jackson Ryan
6 min read
Orca facility Iceland carbon dioxide
Halldor Kolbeins/Getty

On March 19, the flat-topped Icelandic volcano known as Fagradalsfjall provided one of the most Instagrammable eruptions of all time. People flocked to the site to watch lava slither down the volcano's side for the first time in 800 years. The last time the volcano erupted, in the 13th century, carbon dioxide levels in the atmosphere were well below 300 parts per million (ppm).

But around 100 years ago, humanity saw carbon dioxide levels breach that mark. They've continued to rise exponentially throughout the 20th century, and six years ago they smashed through the 400ppm mark. 

The cause is, unequivocally, human activity. Burning fossil fuels sends carbon dioxide into the atmosphere where it lingers for centuries (maybe even millennia). The molecule is good at trapping heat and forms something of a blanket over the Earth. 

The most recent report, published by the Intergovernmental Panel on Climate Change in August, revealed that our excess carbon dioxide emissions have resulted in a 1.1 degree Celsius increase in temperatures since preindustrial times. Scientists say temperatures will continue to rise as carbon dioxide levels increase, resulting in more extreme weather events, more heat, more drought and a catastrophic decline in biodiversity.

Fagradalsfjall volcano

The Fagradalsfjall volcano in Iceland began erupting in March. 

Saul Loeb/Getty

If humans have pumped all that excess carbon dioxide into the air, why don't we just try to vacuum it back up? Literally. Why not suck carbon dioxide out of the air? This concept, known as "direct air capture," or DAC, has been discussed for decades, but testing and deploying machines to perform the task has proven challenging, mostly because they are costly and inefficient.

As world leaders, activists and academics meet in Glasgow for COP26, the UN's premier climate summit, CNET Science has been examining some of the technological advances being developed to help tackle the climate crisis. While technology might help us adapt or mitigate the effects of climate change, alone it's not a solution to the problem. 

With direct air capture facilities coming online and looking to expand, can we expect them to be a viable tool to reverse carbon emissions? Or are we getting sucked into the spin?

Ghost in the machine

Imagine 1 million particles of "air." The vast majority of these particles are nitrogen and, to a lesser extent, oxygen. Only about 412 particles are carbon dioxide, the heat-trapping greenhouse gas. 

This is a simplistic view of air, but it helps describe the complex task a DAC machine has to carry out -- taking in millions and millions of particles of air and sifting through them to grab carbon dioxide.

To do so, DAC facilities use a series of huge fans to suck in ambient air and push it through a filter laced with chemicals that carbon dioxide reacts with and sticks to. Think of it as a specialized kind of flypaper. The CO2 gets trapped, while the other components of air pass right through. 

The Climeworks system sucks in ambient air and traps CO2. The captured CO2 is mixed with water and pumped deep into the Earth where it can stay for millennia.


Heat, pressure or other chemicals can unstick the concentrated carbon dioxide. You could squirrel away this stock of CO2 underground, mix it with water and inject it into the Earth where it mineralizes and turns to stone. Voila! You've just removed CO2 from the atmosphere.

But since carbon dioxide makes up such a tiny fraction of the air we breathe, DAC facilities need to take a whole lot in. This requires energy. The heating of the filter to free the concentrated CO2 also requires energy. If that energy is provided by fossil fuels, well… you can see the conundrum. 

There are 19 direct air capture facilities in operation around the world, according to the International Energy Agency. Fifteen of these are operated by Swiss company Climeworks, and its most recent DAC facility highlights both the promise of vacuuming up CO2 and the remaining hurdles to large-scale builds.

870 cars

The crown jewel in Climeworks' operations is Orca, which lies just an hour away from Fagradalsfjall, at the Hellisheidi geothermal plant. It's the world's largest DAC facility. Orca opened two months ago, on Sept. 9, and has been sucking up CO2 ever since. 

The world's largest such plant to date, Orca captures around 4,000 tons of carbon dioxide each year -- a very small amount, equivalent to taking just 870 cars off the road. 

Climeworks has the backing of influential brands, like Shopify and Microsoft, which have purchased future removal of CO2 as the facility expands. The Climeworks team plans to scale up the Orca plant by 2024 and is shooting for a global rollout in 2027 which, it says, would see a hundredfold increase in removal. 

Iceland is an ideal location for Orca because it sits on top of a fault line between two tectonic plates, bringing heat and magma closer to the surface. Fagradalsfjall's eruption is a beautiful consequence of this location, and it also means there's an abundance of geothermal energy -- a renewable energy source that uses heat generated within the earth. DAC uses energy supplied by the Hellisheidi Geothermal Power Plant.

The volcanic landscape also means there's basalt rock deep below the surface. Once Orca removes CO2 from the air, a nearby facility, run by Icelandic company Carbfix, injects it into this basalt layer. Within two years, the CO2 turns to stone and can be locked away for millennia. 

It's important to note here that DAC is different from other carbon capture technologies, often referred to under the umbrella of "carbon capture, utilization and storage," or CCUS. These technologies have been developed and touted by fossil fuel industries as a way to try to capture carbon dioxide during burning of oil and gas -- that's a whole other issue covered very well by the Australian Broadcasting Corporation. Sometimes, DAC has been tarred with the same brush.

Nevertheless, large-scale facilities are being planned in the US and Scotland. A rival DAC company, Carbon Engineering, is hoping to eventually suck 1 million metric tons of CO2 out of the air at its facility in the US Permian Basin, the equivalent of taking around 200,000 cars off the road. 

That's a big number, but as of 2020 global emissions were still reaching around 31.6 billion metric tons per year, and it's not clear they've peaked. 

The spin and the cost

When Orca came online in September, climate scientist Peter Kalmus tweeted that he was rooting for it, but "only a fool would bet the planet on it."

Direct air capture is a promising technology, but there are a few major hurdles. The first is that the technology is prohibitively expensive. Taking one ton of carbon dioxide out of the atmosphere can cost between $100 and $600 -- or as much as $1,000. Carbon pricing, which forces fossil fuel polluters to pay for their emissions, is much cheaper. 

That means that, at present, it's more cost-effective to emit carbon dioxide than to pull it out of the air. "It's really not good bang for your buck," says Alia Armstead, a climate researcher at the Australia Institute, a Canberra-based think tank.

To make DAC truly carbon negative, you also need to find good renewable sources of energy to power your removal plants. That's easy in highly volcanic Iceland, but access to renewables isn't as easy across the world. Some critics suggest the money invested in DAC would be better going directly to renewable projects -- preventing carbon dioxide from entering the atmosphere in the first place.

Mark Jacobson, a civil and environmental engineer at Stanford University, told CNET in February that lobbyists for carbon capture technologies, including DAC removal, are "basically lying to the public about the benefits." He's also said government support for the tech is a subsidy that would keep the fossil fuel industry in business. 

But support from governments and private organizations is growing. Two weeks ago, the US Department of Energy announced $14.5 million to scale up DAC facilities, stating in a release that DAC is "critical to combatting the current climate crisis and achieving net-zero emissions by 2050." On Nov. 5, it announced a "Carbon Negative Earthshot" to accelerate research and innovation and to drive down removal costs. And earlier this year, Elon Musk's Foundation and XPrize fronted $100 million to help develop carbon capture technologies, with a winner expected to be announced in April 2025. 

While there is growing support for carbon removal technologies, scientists widely recognize that DAC cannot replace the need to decarbonize. In addition, technological progress has been slow. Even with increased support, it remains too slow to keep us from overshooting the climate targets set out in the Paris Agreement. 

"We need to invest in technologies that do work, and that can reduce emissions today, not in five or 10 years," says Armstead.