Carbon Capture
The Machine That Eats CO₂ — and Why It's Both Brilliant and Controversial
We now have technology that can pull carbon dioxide directly out of thin air — and the most honest engineers building it will tell you it might already be too late for that to matter.
The Idea
Direct Air Capture — DAC — is exactly what it sounds like: industrial machines that suck ambient air across chemical filters, strip out the CO₂, and either store it underground or use it to make synthetic fuels and materials. It sounds like a moonshot, but the physics are real and the first commercial plants are already running. The catch is one of concentration. CO₂ makes up roughly 0.04% of the atmosphere. Capturing something that dilute requires moving enormous volumes of air, which demands enormous amounts of energy. Current DAC plants consume somewhere between 1,500 and 2,500 kilowatt-hours of energy to remove one tonne of CO₂ — which means that if that electricity comes from fossil fuels, you're barely breaking even, or worse. This is why DAC sits at a strange intersection of genuine promise and legitimate scepticism. Proponents argue the energy problem is solvable as renewables get cheaper and more abundant. Critics argue that every unit of clean energy spent on carbon capture is a unit not spent on directly displacing fossil fuels — and that the math only closes in a world where we've already decarbonised the grid. There's also the question of permanence. Storing CO₂ as a gas underground carries geological risk. Mineralising it into rock — as one Icelandic plant does — is more permanent, but slower and more expensive. The technology is not one thing; it's a family of approaches at very different stages of maturity, cost, and credibility.
In the World
In 2021, a company called Climeworks switched on Orca, a DAC plant in Iceland, capable of capturing 4,000 tonnes of CO₂ per year. It was, at the time, the largest direct air capture facility in the world — and it was immediately humbling. Global annual emissions sit around 37 billion tonnes. Orca handled roughly one ten-millionth of that. But Iceland was chosen for a reason. The plant runs on geothermal energy — essentially free, clean heat from the earth — which dissolves the energy problem that haunts DAC everywhere else. And the captured CO₂ is pumped underground into basaltic rock, where it reacts chemically and mineralises into stone within two years. No tanks, no leaks, no monitoring in perpetuity. Just rock. In 2023, Climeworks opened a successor plant nearby called Mammoth, with ten times the capacity. The cost per tonne at Orca was around 1,000 euros — Climeworks aims to get it below 300 by 2030 and below 100 by the mid-2030s. That cost curve is the whole argument. Climeworks is essentially betting that DAC follows the same trajectory as solar panels — laughably expensive in the early years, then suddenly competitive as manufacturing scales and engineering matures. Solar took roughly forty years. We probably don't have forty years. But the Iceland story suggests that in the right conditions, with the right geology and the right energy source, DAC is not a fantasy — it's an engineering problem with a plausible, if narrow, path to viability.
Why It Matters
Carbon capture has become a politically loaded term — weaponised by fossil fuel companies as a reason to delay emissions cuts, and dismissed by some climate advocates as a distraction. Neither response is quite right, and the nuance here is worth holding. DAC almost certainly cannot save us if used as a substitute for cutting emissions. The numbers are too brutal, the energy requirements too steep, the costs too high to scale fast enough to compensate for continued burning. But in a world that does cut emissions sharply, DAC may be essential for the residual CO₂ — from aviation, agriculture, heavy industry — that we genuinely cannot eliminate any other way. What this means practically is that DAC is best understood not as a silver bullet but as a tool in a constrained toolkit. Knowing that changes how you read the news. When a government funds a DAC project, the right question isn't 'is this good or bad?' — it's 'is this being used alongside aggressive emissions cuts, or instead of them?' That distinction is where the real policy fight lives, and it's one most coverage never quite reaches.
A Question to Ponder
If a technology can genuinely solve a problem but is also routinely used as an excuse not to solve it another way, how do you decide whether to support it?
Get a new one of these every morning.
Start learning with Thinkable