Last week I posted a high level summary of the cost assumptions associated with the direct air capture (DAC) systems built by Carbon Engineering. This technology shares a lot in common with a chemical plant: It is a big capex expenditure, and has to operate at high physical scale to achieve good economics. A lot of money has to be poured into the plant before it produces its first results.
Climeworks is taking a very different approach: They are building small, modular carbon captured systems, which can be ganged together for scale. Climeworks' approach is called "numbering up", to differentiate it from traditional "scaling up". Overall, numbering up can be a very good approach to driving down costs: The internal combustion engine produces power at a similar cost (measured in $/W) to a large natural gas turbine power plant. The advantage that a turbine has in massive scale is roughly matched by the advantages of mass production.
The core of Climeworks' DAC technology consists of a modular system that uses specially-designed fans to draw in air, which then flows through a filter coated with a proprietary amine-based sorbent. This sorbent material selectively binds to CO2 molecules, capturing them from the air while allowing other gases to pass through. Once the filter becomes saturated, the system switches to a regeneration mode, during which the captured CO2 is released from the sorbent by heating it to around 100°C. A vacuum is pulled on the system, and the CO2 is captured, then the cycle is repeated.
Climeworks' system operates at much lower energies than Carbon Engineering's system thanks to its lower temperature for releasing the CO2. Costs for energy should be <$20/ton captured, depending on the local cost of electricity. The desorption step (though not the vacuum) can also be powered using low grade waste heat or weak geothermal power rather electricity, lowering energy costs further. For this reason alone, the Climeworks approach appears competitive.
The biggest cost in Climeworks' system is the cost of the amine sorbents, which should be expected to run $5-$10/kg at scale, depending on the chemical sorbent used. The system requires quite a lot of sorbent - their can only absorb 10-17% of their mass as CO2. The cycle time to absorb and release incoming CO2 is measured in hours, so the sorbent can cycle a couple thousand times a year. But that still means that for a carbon capture capacity of a million tons of CO2/year might require $50 million dollars in upfront expenditure.
This is pretty small, and would lead to a long term cost of <$100/ton as long as the sorbent does not degrade. As it turns out, the sorbent does degrade. But how fast? I don't think anyone knows the answer to this yet. If the sorbent reaction with CO2 was 99.9% reversible, the total cost of carbon capture would be about $200/ton, making this a very profitable investment for EOR. At 99.8% efficiency, the cost spikes to over $300/ton. Details matter here. Climeworks and its copycats could have a very inexpensive solution for carbon capture, but the ultimate durability of their materials is as yet unknown, and will make all the difference.
Comments