CCUS: The Industrial Plumbing of Decarbonisation
Every forge makes more than iron. It makes smoke, and ash, and the sour air that gathers in the valley below — and the smiths who thought only of the blade left that reckoning to those downwind. The wiser houses dug channels for it first, and called the digging part of the craft.
Why This Exists At All
Most decarbonisation is a substitution story: replace the coal plant with solar, the boiler with a heat pump, the petrol engine with a battery. But a stubborn share of global emissions has no such swap available — not because the will is missing, but because the chemistry itself produces the carbon dioxide.
Making cement releases CO₂ when limestone is heated, regardless of how the kiln is powered. Steelmaking, ammonia, refining and a range of chemical processes have similar problems. Together, hard-to-abate industry and power account for something on the order of 30% of global CO₂ emissions, and for much of it there is no electrification pathway at all. That is the gap CCUS exists to fill: capture the carbon dioxide at the point it is produced, move it, and put it somewhere permanent.
Mechanically it is unglamorous and entirely familiar — a capture unit, a pipeline, a well. It is plumbing. And that framing is useful, because it points at the right question: not whether the technology works (it does, and has for decades in oil and gas), but whether anyone will pay for a pipe network whose only product is the absence of emissions.
The Scale Problem, Stated Honestly
Start with the number that governs everything else. Global operational capture and storage capacity reached roughly 50 million tonnes of CO₂ a year by early 2025. Global emissions run in the tens of billions of tonnes annually. The industry, after decades of development, handles a rounding error.
The pipeline of projects is genuinely expanding — more than 600 projects in development, growing around 15% a year, with investment tripling to over $6 billion. But even on the optimistic assumption that every announced project proceeds, capture capacity would reach roughly 430 Mt a year by 2030, against something like a gigatonne a year contemplated in net-zero pathways for the energy sector. The gap between what exists and what is called for is roughly twentyfold.
The same discipline applied to the interconnection queue applies here. An “announced” CCUS project is not a built one: the sector has a long history of cancellations, and a pipeline figure that assumes full conversion is a marketing number, not a forecast. The useful question is which projects have a concentrated CO₂ source, secured storage, a permitted transport route, and a credit-worthy counterparty — the rest is optionality. Most of the announced 600 will not be built.
The One Variable That Decides Everything: Concentration
Here is the analytical key to the entire sector, and the thing most coverage skips. The cost of capturing a tonne of CO₂ depends overwhelmingly on how concentrated it already is in the gas stream you are capturing it from. Separation is the expensive part, and separation gets dramatically harder as the target gets more dilute.
An ethanol fermenter or an ammonia plant produces a nearly pure CO₂ stream — capture is close to compression and dehydration, and costs very little. A cement kiln’s flue gas runs somewhere in the range of 14–33% CO₂ — harder, but workable. A power-plant exhaust is more dilute again. And ambient air is roughly 0.04% CO₂, which is why direct air capture, the technology that attracts the most attention, is by far the most expensive: verified operating costs were estimated at around $600–800 per tonne in mid-2026.
CCUS is not one industry. It is a cost curve, and the tax credit is a flat line drawn across it.
The US 45Q credit pays roughly $85 per tonne for CO₂ put into dedicated geological storage and about $60 where it is used, including enhanced oil recovery. Lay that flat payment across the cost curve and the sector sorts itself instantly: anything to the left of the line — high-purity industrial sources — is profitable today. Anything to the right needs either a much higher credit (direct air capture gets about $180) or a different revenue source entirely. Do not evaluate “CCUS.” Evaluate where on the cost curve a specific project sits relative to its available credit.
Policy Isn’t the Backdrop. It Is the Revenue.
In most infrastructure, policy shapes the return. In CCUS it very nearly is the return — captured CO₂ has almost no natural buyer, so with narrow exceptions the cash flow is a government payment. That makes this the purest illustration of the primer’s principle anywhere in the section.
Three design details matter more than the headline rate. The credit was raised sharply and extended, with construction start required before 2033. Capture-rate and utilisation thresholds mean a project must genuinely perform to qualify. And crucially, transferability — the ability to sell credits to third parties — solved the sector’s oldest financing problem, since developers without large tax bills could never previously monetise the incentive. That one provision did more to unlock project finance than any technical advance.
Elsewhere the mechanism differs but the principle holds. The UK has committed on the order of £21.7 billion to CCUS clusters, with Teesside and HyNet under construction. Germany passed legislation enabling commercial-scale storage and transport and launched a multi-billion-euro carbon contracts-for-difference auction covering steel, cement and chemicals. Norway’s Northern Lights has proven something genuinely new — an open-access storage service, taking CO₂ shipped from a German cement plant, the first cross-border arrangement of its kind. That model, storage sold as a utility service rather than built per-project, may prove more consequential than any single capture plant.
Why Projects Still Die
Even with generous credits, the failure rate is high — and the causes are almost never the capture technology. They are the two ends of the pipe.
- Transport. CO₂ pipelines need rights-of-way across many landowners and jurisdictions, and have met the same organised opposition as any other pipeline. Major US CO₂ pipeline projects have run into sustained legal and permitting resistance.
- Storage permitting. Injection wells require specific federal permits, and while some states have accelerated approvals and gained authority to issue them directly, others have imposed moratoriums. Where you can legally inject is now a first-order siting constraint.
- Public consent. Communities asked to host pipelines and injection wells frequently object, and the benefit to them is abstract in a way that a wind farm’s lease payment is not.
- Policy reversal. The US Department of Energy cancelled roughly $3 billion of industrial demonstration grants, including over a billion earmarked for CCUS, with further awards at risk — a reminder that grant-dependent projects carry the same political-durability risk that felled offshore wind.
An unexpected new customer
One genuinely fresh demand driver deserves note: hyperscale data centres. The operators building the AI infrastructure covered in the Build thread hold firm net-zero commitments while consuming enormous and growing quantities of power. That creates real appetite for low-carbon firm generation — and gas-with-capture is one of the few options that is dispatchable, buildable this decade, and defensible against a carbon target. Whether it proves cheaper than the alternatives is unsettled, but the buyer is new, credit-worthy and motivated.
Reading It Through the Frameworks
How does it get paid? Predominantly by a government credit, sometimes supplemented by a product sale or a voluntary carbon-market buyer. That is contracted-ish revenue with a sovereign-policy counterparty — strong while the policy holds, and exposed to precisely the reversal risk that has already hit US grant programmes.
Where is the moat? Not in capture equipment, which is engineering others can replicate. It is in storage — permitted, characterised, legally-secured pore space is genuinely scarce and cannot be manufactured — and in shared transport networks, which are natural monopolies with classic toll-on-a-flow economics. The Northern Lights model makes this explicit: the durable asset is the hole in the ground and the pipe to it, not the capture unit.
CCUS is real, necessary and much smaller than its coverage suggests. For cement, steel and a handful of chemical processes there is no other route to deep abatement, which makes the demand structural. But operational capacity remains a rounding error against the requirement, most announced projects will never be built, and the revenue is a tax credit rather than a customer.
Judge any project on two things and you will be right more often than the consensus. First, where its CO₂ source sits on the concentration cost curve relative to the credit available to it. Second, whether it owns any part of the storage or transport network, because that — not the capture unit — is where the toll booth is. The plumbing is genuine infrastructure. The question is only ever who pays to keep the channel dug.
The channel was never the pride of the works, and no visitor was taken to see it. But when the rains came and the valley did not choke, the house that had dug it prospered — and those that had not paid dearly, and late, for the digging of their own.
Leave a Reply