Eight massive metal containers that initially appear to have been borrowed from a shipping dock are located out on the Hellisheiði plateau in southwest Iceland, where the terrain consists entirely of black lava fields and the steam from geothermal vents drifts across the horizon in thick white columns. They are mechanically patient, silent devices with fans and sophisticated filters that draw air through a chemical process to remove carbon dioxide. After being captured, the CO2 is combined with water and injected deep into the volcanic rock beneath, where it eventually mineralizes and turns into stone. The Orca plant had a capacity of 4,000 tonnes of CO2 annually when Climeworks and Carbfix opened it in September 2021. It is simple to write off a figure like that as symbolic when compared to the 31.5 billion tonnes of annual emissions worldwide. It’s more difficult to ignore what came next.
The direct air capture industry has long needed evidence that the technology can scale, maintain operations, and produce actual, quantifiable removal rather than projected or modeled estimates, and the milestone of one million tonnes of carbon dioxide cumulatively removed from the atmosphere at the Iceland site represents just that. Before lunchtime on any given Tuesday, the world emits far more than one million tonnes. Any serious discussion about this technology should include that sincere admission. However, it’s not a goal, a plan, or an estimate from a consulting report; rather, it’s a tangible figure generated by a physical process in a real place. The distinction is more important than it may seem in an industry where there are many of those.
Climeworks / Carbfix — Iceland Direct Air Capture
| Company / Partnership | Climeworks AG (Switzerland) + Carbfix (Iceland) |
| Location | Hellisheiði geothermal area, Iceland |
| First Plant (Orca) — Opened | September 9, 2021 · Capacity: ~4,000 tonnes CO₂/year |
| Cumulative Capture Milestone | 1 million tonnes CO₂ removed from atmosphere |
| Technology | Direct Air Capture (DAC) — high-tech filters + fans; CO₂ mixed with water, pumped underground → mineralizes to rock |
| Energy Source | 100% geothermal renewable energy (nearby power plant) |
| Global DAC Context (IEA, 2021) | 15 plants operating globally · ~9,000 tonnes/yr total at launch |
| Global Annual CO₂ Emissions | ~31.5 billion tonnes/year (IEA) |
| Notable Corporate Client | Swiss Re (10-year carbon removal purchase agreement) |
| US Comparable Project | Occidental — 1M tonne/year DAC plant, Texas (under development) |
At Hellisheiði, the process is virtually irreversible. When CO2 is combined with water and pumped into Icelandic basalt, it transforms into carbonate minerals over a period of months to years, becoming a part of the rock itself, rather than remaining as a gas waiting to escape. The mineralization process is the safest method of carbon storage currently accessible at scale, according to Carbfix, the Icelandic company in charge of the underground injection and storage. Iceland is especially well-suited for this due to the geochemistry of basaltic rock, which helps explain why the island serves as a sort of natural laboratory for the technology. Another level of coherence is added by the geothermal energy that powers the entire operation, which uses the earth’s heat to draw carbon from the sky and bury it beneath the same earth. Standing on the lava field makes it easier to understand its logic than reading about it in a briefing document.
Since the technology first garnered significant attention, the cost issue has continued to cast a shadow over direct air capture. According to preliminary estimates, the cost of removing a tonne of CO2 through direct air capture ranges from $250 to $600, depending on the plant and the accounting methodology. This is significantly more expensive than some alternative methods, such as reforestation, which has serious drawbacks of its own. Claiming that the economics of direct air capture will follow a trajectory similar to solar panels or battery storage, where successive generations of deployment produce significant cost reductions, Climeworks has been working to lower costs through scale. Building the demand-side revenue that makes future investment credible has included a ten-year carbon removal purchase agreement with Swiss Re and a subscription model that allows individuals and businesses to pay for monthly carbon removal.
Occidental’s strategy in West Texas provides a useful analogy. Instead of using Climeworks‘ modular container approach, the Houston-based oil company is building a direct air capture facility close to its oilfields with an annual removal capacity of one million tonnes. The unit economics of removing carbon from ambient air are fundamentally more challenging than capturing it at the point of emission from a smokestack, where concentrations are much higher. Both initiatives represent significant commitments from very different types of organizations, and they are facing the same fundamental challenge. The effectiveness of direct air capture is not the question. This is confirmed by the milestone in Iceland. The question is whether it can become affordable and quick enough to make a significant contribution to the volumes of carbon removal that climate scientists claim are required.
It’s difficult to ignore Iceland’s unique significance in this narrative. The nation has a geological profile that makes carbon mineralization exceptionally favorable, runs almost entirely on renewable energy, and has a research ecosystem that has been addressing these issues for many years. The eight shipping-container units at the Orca plant are not a glamorous sight. They don’t attract the same level of attention as wind turbine installations or solar farms. They simply run steadily through a lava field, drawing invisible gas from the atmosphere and returning it to the ground where it originated. Even after removing a million tonnes over a number of years, there is still a long way to go. However, compared to when those eight containers first turned on direct air capture, the road now has a confirmed beginning point.
