Verdox captures CO₂ from a real smelter, in a two-month trial

Almost everything that leaves the chimney of Hydro's aluminium smelter at Sunndal, in Norway, is not carbon dioxide. CO₂ makes up only about 1% of the exhaust. For two months, Verdox's electrochemical process worked on that gas, with the contaminants that real smelter gas carries.

The difficulty is mostly a matter of dilution. The material used by Verdox contains quinone, a chemical compound that captures CO₂ in one electrical state and releases it in another. A capture surface therefore cannot do both at once. A steady smelter needs several banks of modules working at different stages of the cycle, so that fresh exhaust keeps arriving while others release concentrated CO₂. Hydro's investment and continued work with Verdox are what make this trial more than a laboratory result, or a test run on a prepared gas stream.

The trial answers one question and leaves the others open. Verdox publishes an energy figure of about 417 kWh per tonne of CO₂, roughly 1.5 GJ. But the public material does not say which loads it includes, such as the fans and ducts needed to move a very large volume of dilute gas, the gas cleaning, or the compression. Nor does it say how long the capture material survives years of exposure to smelter gas. Laboratory cycling tests show the chemistry, not the maintenance record of a module in service. Replacement frequency drives capital cost, downtime and waste, so it weighs as much as the electricity used.

The captured gas is also only part of the chain. It must be separated, dried and compressed to a specification that a pipeline or storage operator will accept, then transported and stored in suitable geological formations, under contracts and monitoring. The trial does not show any of those steps. The aluminium industry is also attacking the source of the emissions: anode consumption accounts for roughly three quarters of a smelter's direct carbon emissions, according to Hydro and Rio Tinto. Rio Tinto and Alcoa's ELYSIS venture started a commercial-size inert-anode cell at the Alma smelter in 2025, and Hydro's HalZero programme, which uses aluminium chloride, is at an earlier stage.

For whom, and when? For an existing smelter with decades of life left, where switching the process is difficult, capture can make sense if clean electricity and affordable storage are available. A plant close to a major replacement of its production lines faces a harder choice, because money spent on capture competes with money that could change the process itself. The sources do not give a date for commercial installations. Other sources of CO₂ are easier: fermentation and biogas upgrading produce a more concentrated stream, which needs less separation before storage.