The air we breathe has a carbon problem. But in Hellisheidi, Iceland, a geothermally active plateau just outside of Reykjavik, a new technology is taking a small but mighty step toward fixing it.

A plant called Orca, built by Climeworks, is the first-ever facility where CO2 is being filtered directly from the air and stored permanently underground.

Orca’s carbon-capturing devices resemble giant transistor radios. They fit right into an already larger-than-life Icelandic landscape, where the wind blows fierce even on a rare day when sunlight gleams off the icy mountaintops.

Though the plant has only been operational since September, its air-straining technology, known as direct air capture, has been a point of contention among environmentalists for much longer. Vacuuming up carbon dioxide was once considered a last resort, but it’s looking like we’re headed toward a future where last resorts are a must-have.

“The combination of direct air capture and storage is very likely what the world will need at a massive scale if we want to be compliant with Paris climate targets,” said Jan Wurzbacher, the CEO and co-founder of Climeworks.

Carbon removal, by math and magic

By “Paris” targets, Wurzbacher was referring to the global goal of limiting emissions to two degrees Celsius (or ideally 1.5 degrees), established under the 2015 Paris Agreement. To meet that goal, the United Nations has estimated that 10 billion tons of carbon dioxide will need to be removed from the atmosphere annually by 2050. That number is a best-case scenario, assuming that aggressive cuts in emissions are achieved through other means. Without enough cuts, the need for carbon removal could be even higher.

“It’s relatively simple climate math,” Wurzbacher explained on a video call from Zurich, Switzerland, where Climeworks is based. “By mid-century, we need to remove 10 billion tons of CO2, if everything else goes well. We might end up needing to remove 20 billion tons, because we can’t ramp down fast enough coal power plants and other stuff.”

Direct air capture technology is one among many options for removing excess CO2. There are natural methods, like planting trees, and there are technologies that capture CO2 directly from smokestacks and other emission sources. Compared to capturing CO2 at the source, it’s more challenging and costly to pull CO2 literally out of thin air, but a benefit of direct air capture is that it doesn’t require finding and stopping every single polluter. It’s a solution that works across the globe.

“When you do direct air capture, you don’t need to go where the CO2 is, because air is everywhere,” said Wurzbacher.

The Orca plant consists of eight shipping container-sized boxes, which Climeworks calls collectors. On the front of each box, there are slats, kind of like large venetian blinds. On the back, there are 12 fans that pull air through the box. Within the collectors, the CO2 molecules hit the surface of a specially developed filter material, where molecules, called amines, selectively grab onto them.

That point of contact is a magical moment. The rest of the air continues out the other side of the collectors, but the carbon sticks tight to the amines. In that moment, the CO2 goes from the chaotic fray of the atmosphere to the ordered grip of humanity, potentially remaining under control for thousands of years to come. With the application of heat, the CO2 is released from the amines, and then it gets pumped into nearby volcanic rocks, where it forms long-lasting carbonate minerals.

Currently, removing a ton of CO2 at Orca costs between $600 and $800, which is prohibitive for most potential payers. Early customers have been companies and individuals willing to pay a premium, such as Microsoft, Stripe, Swiss Re and even the band Coldplay (which hired Climeworks to cancel out some of the emissions from its upcoming world tour).

Climeworks is aiming to get that cost down to between $100 and 200. The US Department of Energy recently set a similar goal of bringing the cost of technological carbon removal to under $100 per ton. At those lower price points, direct air capture would be on par with other ambitious measures to reduce emissions.

For now, the magic moment of capture may not be cheap, but at least it works.

“Orca has gone from zero to one,” said Dr. Julio Friedmann, a senior research fellow at Columbia University “At this point we know, if we had to, we could make more Orcas. We imagine the cost will reduce, we imagine the performance will improve, et cetera, but we now have a single-unit device that pulls four thousand tons of CO2 from the air every year.”

In addition to the high cost, Orca has been criticized for the miniscule quantity of CO2 that it captures. Four thousand tons is paltry compared to the 10 billion tons that need to be removed within several decades. At our current level of emissions, humanity is canceling out Orca’s yearly efforts every three seconds.

However, it may be helpful to reframe this amount by considering it relative to other means of removing carbon from the atmosphere. Growing an acre of redwood forest also removes about 4,000 tons of CO2, but it takes much longer than a year, and it can only be done in a limited number of places.

Friedmann framed the elevator pitch for direct air capture in similar terms: “It does the work of two hundred thousand trees, in one thousand times less space.”

Activists like Greta Thunberg have dismissed engineered solutions like direct air capture as “technologies that barely exist” and have instead promoted nature-based solutions, but it’s possible to pursue both strategies. If the United Nations estimates are correct, there will be room for multiple solutions to suck up the extra carbon.

“It’s at the order of billions of tons that we will need this. I’m quite sure of that,” said Wurzbacher.

Sometimes, smaller is better

Criticisms of Orca’s small size also miss another key point: starting small is a unique learning opportunity.

Like with any technology, direct air capture can improve with iteration, becoming more efficient and less expensive over time. The Climeworks CO2 collectors are modular, meaning that the way to collect more CO2 is to add more collectors, rather than making the collectors themselves any bigger. With a modular product, it’s cheaper and easier to iterate than with one that’s larger and more bespoke.

“By the time we build big plants or multiples of what we have now, we are quite certain that it works and how we keep operating,” said Nathalie Casas, the head of technology at Climeworks. “That’s the beauty of the modular approach.”

A bigger plant is already in the works, according to Wurzbacher, though the location has not yet been determined. It will be ten times larger than the Orca plant, so there will be 80 shipping container-sized collectors, instead of just eight, capturing 40,000 tons of CO2 per year.

An advantage of building a plant the size of Orca is that any tweaks to the collector design only need to be repeated eight times.

“If you build eighty containers, it’s a whole different story,” Wurzbacher said.

Climeworks’ small- to medium-scale approach stands in contrast to that of another major company commercializing direct air capture technology, Carbon Engineering, which is currently building a plant in Texas that will capture half a million tons of CO2. That plant is scheduled to go online towards the middle of this decade.

“Climeworks is going into the swimming pool step by step as opposed to cannonballing in all at once,” said Colin McCormick, chief innovation officer at Carbon Direct, an investment and advisory firm focused on carbon removal.

It remains to be seen if one, both, or neither of these approaches will be able to achieve carbon removal at scale and at cost. Direct air capture is still in its infancy, but it has parallels to some of the heavy hitters of sustainable technology. Both solar photovoltaic (PV) panels and wind turbines started off as long shots several decades ago, but now they’re huge and growing industries at the vanguard of the energy transition.

The solar comparison is particularly apt, because the panels were working with novel materials to achieve something theoretically possible but commercially unproven.

According to McCormick, there’s a minimum amount of energy needed to capture carbon dioxide from ambient air, and it’s substantial, but Climeworks and Carbon Engineering are both using about ten times more energy than that minimum, so there’s a lot of room for improvement.

“We’re wildly off from 100% efficiency, and that’s okay,” he said. “Early solar panels were a few percent efficient.”

There are many ways that Climeworks is seeking to boost efficiency and reduce costs. A big one is to improve the filter material, so that it captures more CO2 and lasts longer. The company is also sorting out how to streamline production, so that the modular units are cheaper to build. Then there are the relatively fixed costs, like pipelines for CO2, that will naturally decrease as the plants get bigger.

The technical challenges ahead may seem daunting, but the team at Climeworks is unfazed. In fact, Wurzbacher considers the current state of direct air capture to be much more favorable than that of wind and solar in the 1970s and 1980s.

“If you compare it to where solar PV or wind started, they had to do larger factors of cost reduction,” said Wurzbacher. “It’s actually good news that it’s only on the order of a factor of ten that we have to do.”

Achieving these cost reductions requires learning that can only take place in real-world conditions, not a laboratory or an office.

“It’s not a software startup,” said Wurzbacher. “We’re putting steel and concrete in harsh environments. Weird things happen that you can only predict maybe to ninety percent, but the last ten percent you can never predict, and then you learn it, so it’s so important to get stuff out in the field.”

The geothermal power plant, which supplies heat and energy to Orca, as well as much of Reyjavik.

The geothermal power plant, which supplies heat and energy to Orca, as well as much of Reyjavik.

Why Iceland?

You’d be hard-pressed to find a field site more majestic than Orca. The plant sits on the edge of a grassy plain, just beneath craggy black peaks accentuated with crisp white snow. But the Climeworks team didn’t choose to build Orca in that spot for the scenery. The site in Hellisheidi offers two elements that are essential for direct air capture: cheap renewable energy and a place to put CO2. Both of those elements are products of Iceland’s unique volcanic geology.

Orca is located directly next to the Hellisheidi Geothermal Power Plant, which is one of Iceland’s biggest sources of geothermal energy. The plant draws hot water from over a mile below the ground, where it’s naturally warmed by a volcanic hotspot. The geothermal process produces heat and electricity, both of which are key inputs to direct air capture.

The electricity is used to move air through the collector, and the heat is used to release captured CO2 from the filter material, which happens at around 100 degrees Celsius, the temperature of boiling water.

“Geothermal is particularly good to start with, because it’s 24/7; it’s heat and electricity, so it’s really well suited to what we are doing,” said Wurzbacher.

Then there’s the CO2. The rock under Hellisheidi is porous basalt, less than 1 million years old, which means it’s pretty much brand new, geologically speaking. A company called Carbfix has figured out how to inject CO2 into this young rock so that it reacts to form carbonate minerals.

Carbfix is a subsidiary of Reykjavik Energy, the municipal-owned utility that operates the geothermal plant. For over five years, it had been using its technique to store the small amounts of CO2 released from the geothermal process, so the infrastructure to store the CO2 from Orca was already in place.

“This is a big part of why Orca is in Iceland,” said McCormick. “They have waste heat and zero-carbon power from the geothermal field. They have already-drilled injection holes and excellent geology for injecting CO2, so that location has everything you want.”

Carbon storage is an essential piece of the equation for carbon removal. There are many ways to do it, but the Carbfix method is particularly promising because the carbon dioxide turns quickly into rock. It mineralizes within two years—or more likely a few months—and it will remain in that solid state for thousands of years.

“The CO2 is not going anywhere, so basically once it’s underground, we know it’s going to stay underground,” said Kari Helgason, the head of research and innovation at Carbfix.

This time frame contrasts with other methods, like storage in abandoned oil wells, which requires indefinite monitoring to make sure that the CO2 does not escape.

Another benefit of the Carbfix method is that the cost is almost negligible, especially relative to the high cost of capturing carbon.

“If we receive pure CO2, it’s pretty cost efficient,” said Helgason. “What we’re doing with Climeworks, it’s ridiculously cheap.”

Fortunately, Iceland is comprised mostly of basalt, so the storage opportunities are almost limitless. Helgason estimates that each cubic kilometer of basalt can store one hundred million tons of CO2.

“The storage capacity is enormous,” he said.

And it’s not just Iceland where this massive storage capacity exists. Carbfix put together an online atlas that maps out regions round the world with potential for geologic carbon storage.

Wurzbacher noted that Hellisheidi and Carbfix were a perfect match for the Orca plant, but Climeworks is open to other locations for subsequent projects.

“What is not so perfect is the weather and wind in Iceland,” he said. “If you ask our commissioning team if they want to build the next plant in the same weather conditions, they might rather ask for going to Hawaii or someplace else with a lot of volcanic rocks.”

It takes a village to capture carbon

The partnership between Climeworks and Carbfix is an example of the collaborative innovation that’s needed for carbon removal to succeed. In an attempt to incentivize similar relationships, the US has allocated $3.5 billion for building four direct air capture “hubs,” where multiple companies will work together to capture and store CO2. The provision is part of the infrastructure bill recently passed in Congress.

“It’s really important to think of direct air capture as a partnership among a number of different groups,” said Rory Jacobson, deputy director of policy at Carbon180, a think tank focused on carbon removal.

Often, oil and gas companies have been part of direct air capture projects, usually in the role of financial backers. Carbon Engineering is partnering with Occidental Petroleum on its project in Texas, and another direct air capture company, Global Thermostat, is is partnering with Exxon Mobil on several smaller projects in the US.

For carbon capture startups, it makes sense to work with these larger companies, because they can write big checks and have a deep understanding of geology, but such partnerships have also fueled claims that carbon capture is a smokescreen for continued pollution.

“The thing that’s really promising about Orca is that there is no clear fossil fuel involvement at all in the project,” said Jacobson.

With or without help from the fossil fuel industry, direct air capture is likely to face limits in its ability to scale without support from government climate policy. The problem is that there’s no inherent market for removing carbon from the atmosphere, like there is for the electric power generated from wind and solar. Direct air capture needs to be funded by voluntary purchases (such as those that have so far supported Climeworks) or government incentives and directives.

“The voluntary market for carbon removal will bring us to millions of tons, maybe ten million tons, maybe more,” said Wurzbacher. “Public instruments will have to bring us from tens of millions of tons to billions of tons.”

Some incentives are already in place. The US currently has a policy called 45Q that pays $50 per ton for carbon capture and storage, and that payment would be expanded to $180 per ton under the Build Back Better bill, which was recently passed by the House of Representatives but remains in limbo in the Senate.

However, direct air capture will have a much bigger market if governments charge high carbon taxes or enact other measures that force industries to steeply reduce or remove their emissions right away. At the recent United Nations climate conference in Glasgow, world leaders weren’t announcing those kinds of bold moves. That may change as both public pressure and the effects of climate change continue to intensify. Neither shows signs of letting up any time soon.

If stronger policies materialize in the coming years, Climeworks will be ready, with direct air capture technology that’s likely to be much cheaper than it is today. The price will still be high, but the price of doing nothing could be far higher.

Orca plant is a 20-minute drive from Iceland's capital, Reykjavik,

The Orca plant is a 20-minute drive from Reykjavik.