Thanks for another great piece, Joe! Unsurprisingly, I respectfully disagree. 🇰🇪
So many thoughts, but let me try just shoot a few rapid-fire facts through.
1) Clean energy in the US is massively supply-constrained: e.g., the US built only 222 miles of high-voltage transmission in 2024 despite clear demand & policy signals, so pretending that clean energy abundance everywhere is around the corner is probably misconceived. That solar plant a DAC player might sign a PPA with would likely get built anyway, and would probably go further reducing, vs. removing emissions.
2) I disagree that 1 TWh is 'small'. On a grid that still has coal (e.g., Louisiana), that same clean energy could have been used to retire a coal plant. 1 TWh of coal = ~1 MT of CO2 = roughly the annual emissions of my home town, Nairobi.
3) Scaling DAC will need 1000s of TWh of 24/7 clean energy + not all places have the same endowment for this + DAC is a highly location-specific industry. DAC's performance varies a lot with humidity, temperature, precipitation, etc., so building it today where it doesn't make much sense to scale it significantly limits our learning rate.
4) Treating clean energy as an afterthought to DAC deployment creates severe quality issues for our industry's only product - i.e., supposedly very high-quality carbon credits. Unless DAC power needs are matched with 24/7 clean energy (which few places have in excess), we don't have the atmospheric impact we claim, which seriously risks the social license we need to scale.
Ultimately, all DAC deployment is a good thing, but at least as far as the very limited pool of price-insensitive CDR demand is concerned, I think we shouldn't paint a false dichotomy. It's not a question of "to do DAC today you have to be pragmatic about clean energy", but there are actually DAC companies that are both in the field AND have their clean energy sorted out. Buyers would be right to prioritise those. Those Octavia folks seem to be doing a good job, for example. 🙂
Martin, agree, and we are largely making the same point: that DAC companies cannot credibly go-to-market without a strong plan for procuring clean energy.
I'm obviously a huge fan of the strategic decisions Octavia has made, including how your location eliminates a bunch of tough questions that North American DAC developers are facing. In the case of your geography (and Climeworks plants in Iceland) your point of view makes a lot of sense.
That said, I do believe that a healthy DAC industry will be geographically (and technologically) diverse. For those companies that are choosing to build in the US, in spite of the tradeoffs you outlined, it is still worth developing in areas that are imperfect but have considerable advantages. Louisiana, for example, is ahead of most of the US on transport and storage. And more partners paying for and advocating for clean energy on these grids will accelerate the renewable energy transition. Yes, the US has a long way to go.
I agree in particular with your point that clean energy cannot be a DAC afterthought. But on the question of 24/7 energy matching vs. annual energy matching vs. emissions-based energy matching vs. various other approaches, verifiers and customers have different points of view of what good looks like.
In a world with real tradeoffs, perfect is the enemy of good. But that doesn't mean I'm advocating for truly bad behavior, like powering DAC plants with entirely fossil fuels with no plan to mitigate. Just the opposite.
I like the premise that we can use around 1 TWh annually to see how a few early commercial DAC plants scale up. What scares me a bit more is powering a mature DAC industry. Assuming we improve energy efficiency, we might still be looking at 500-1,000 TWh per billion tons of removal, which is far from trivial. At the same time, renewable energy technologies are only getting cheaper, so ideally DAC developers can finance renewable energy projects specifically for their needs without pricing out their buyers! Either way, I guess this just means the bottleneck won't be energy demand, but CDR demand, as usual...
So can you provide a couple of simple answers: How much carbon dioxide is generated by coal to generate 1 TWh? And how much CO2 will be removed with 1 TWh of 100% clean energy (wind/solar)?
Would it not make more sense to try to integrate DAC with one of the major users of electricity -hvac- a demand which is likely to increase as the world warms? There are already innovative ideas to use high=rise buildings overall air conditioning process to also remove CO2 at the same time. Surely there must be cost savings if DAC could 'piggy-back' on 'hvac;?
Interesting idea. Current DAC (and all carbon capture) technologies benefit tremendously from economies of scale. We're not yet at the point where small modular DAC units are affordable. But we could get there.
Would note that, contrary to intuition, the most energy intensive part of the DAC is not absorbing the CO2 from the atmosphere (e.g. the loud fans in HVAC machines) but separating CO2 from the sorbent, which why these plants have a lot in common with large chemical engineering facilities rather than super-sized HVAC units.
The current discussion seems to be overly focused on large-scale engineering facilities of the kind you mention using chemical sorbent -with significant heat demands. However I believe concerns such as Verdox are promoting the development of 'electro-swing' technology for DAC. Using chemical substances such as quinones whose polarity can be reversed to attract and release the CO2 molecule depending on the electrical charge(+/-). Presumably small modular units using this technology (which could integrate well with hvac) are on the way? As an afterthought the 'electro-swing' idea is i believe a core element to the notion of the 'DAC Railcar' being developed by a Texas outfit -CO2Rail. Who knows but the idea might be even taken aloft to enable commercial aircraft to do the same?
Edward that's an excellent point, I hadn't considered electro-swing when responding to your initial comment, but I'm also excited about electro-chemical DAC. It will be interesting to see how Verdox/RepAir/others develop. Potential here for the facilities to be (1) smaller and (2) more energy efficient.
And this point, however, it's all potential and promotion -- the tech is at an earlier stage than the more well-known players, and these electro-chemical companies need to demonstrate low power-output DAC on a small scale before I take their claims seriously.
Have more to learn here so don't hesitate to point me in the right direction if you've seen new data.
Thanks for another great piece, Joe! Unsurprisingly, I respectfully disagree. 🇰🇪
So many thoughts, but let me try just shoot a few rapid-fire facts through.
1) Clean energy in the US is massively supply-constrained: e.g., the US built only 222 miles of high-voltage transmission in 2024 despite clear demand & policy signals, so pretending that clean energy abundance everywhere is around the corner is probably misconceived. That solar plant a DAC player might sign a PPA with would likely get built anyway, and would probably go further reducing, vs. removing emissions.
2) I disagree that 1 TWh is 'small'. On a grid that still has coal (e.g., Louisiana), that same clean energy could have been used to retire a coal plant. 1 TWh of coal = ~1 MT of CO2 = roughly the annual emissions of my home town, Nairobi.
3) Scaling DAC will need 1000s of TWh of 24/7 clean energy + not all places have the same endowment for this + DAC is a highly location-specific industry. DAC's performance varies a lot with humidity, temperature, precipitation, etc., so building it today where it doesn't make much sense to scale it significantly limits our learning rate.
4) Treating clean energy as an afterthought to DAC deployment creates severe quality issues for our industry's only product - i.e., supposedly very high-quality carbon credits. Unless DAC power needs are matched with 24/7 clean energy (which few places have in excess), we don't have the atmospheric impact we claim, which seriously risks the social license we need to scale.
Ultimately, all DAC deployment is a good thing, but at least as far as the very limited pool of price-insensitive CDR demand is concerned, I think we shouldn't paint a false dichotomy. It's not a question of "to do DAC today you have to be pragmatic about clean energy", but there are actually DAC companies that are both in the field AND have their clean energy sorted out. Buyers would be right to prioritise those. Those Octavia folks seem to be doing a good job, for example. 🙂
Martin, agree, and we are largely making the same point: that DAC companies cannot credibly go-to-market without a strong plan for procuring clean energy.
I'm obviously a huge fan of the strategic decisions Octavia has made, including how your location eliminates a bunch of tough questions that North American DAC developers are facing. In the case of your geography (and Climeworks plants in Iceland) your point of view makes a lot of sense.
That said, I do believe that a healthy DAC industry will be geographically (and technologically) diverse. For those companies that are choosing to build in the US, in spite of the tradeoffs you outlined, it is still worth developing in areas that are imperfect but have considerable advantages. Louisiana, for example, is ahead of most of the US on transport and storage. And more partners paying for and advocating for clean energy on these grids will accelerate the renewable energy transition. Yes, the US has a long way to go.
I agree in particular with your point that clean energy cannot be a DAC afterthought. But on the question of 24/7 energy matching vs. annual energy matching vs. emissions-based energy matching vs. various other approaches, verifiers and customers have different points of view of what good looks like.
In a world with real tradeoffs, perfect is the enemy of good. But that doesn't mean I'm advocating for truly bad behavior, like powering DAC plants with entirely fossil fuels with no plan to mitigate. Just the opposite.
Fair points! Thanks as always for a great discussion, Joe!
I like the premise that we can use around 1 TWh annually to see how a few early commercial DAC plants scale up. What scares me a bit more is powering a mature DAC industry. Assuming we improve energy efficiency, we might still be looking at 500-1,000 TWh per billion tons of removal, which is far from trivial. At the same time, renewable energy technologies are only getting cheaper, so ideally DAC developers can finance renewable energy projects specifically for their needs without pricing out their buyers! Either way, I guess this just means the bottleneck won't be energy demand, but CDR demand, as usual...
Totally - full-scale DAC needs entirely renewable energy!
So can you provide a couple of simple answers: How much carbon dioxide is generated by coal to generate 1 TWh? And how much CO2 will be removed with 1 TWh of 100% clean energy (wind/solar)?
Would it not make more sense to try to integrate DAC with one of the major users of electricity -hvac- a demand which is likely to increase as the world warms? There are already innovative ideas to use high=rise buildings overall air conditioning process to also remove CO2 at the same time. Surely there must be cost savings if DAC could 'piggy-back' on 'hvac;?
Interesting idea. Current DAC (and all carbon capture) technologies benefit tremendously from economies of scale. We're not yet at the point where small modular DAC units are affordable. But we could get there.
Would note that, contrary to intuition, the most energy intensive part of the DAC is not absorbing the CO2 from the atmosphere (e.g. the loud fans in HVAC machines) but separating CO2 from the sorbent, which why these plants have a lot in common with large chemical engineering facilities rather than super-sized HVAC units.
The current discussion seems to be overly focused on large-scale engineering facilities of the kind you mention using chemical sorbent -with significant heat demands. However I believe concerns such as Verdox are promoting the development of 'electro-swing' technology for DAC. Using chemical substances such as quinones whose polarity can be reversed to attract and release the CO2 molecule depending on the electrical charge(+/-). Presumably small modular units using this technology (which could integrate well with hvac) are on the way? As an afterthought the 'electro-swing' idea is i believe a core element to the notion of the 'DAC Railcar' being developed by a Texas outfit -CO2Rail. Who knows but the idea might be even taken aloft to enable commercial aircraft to do the same?
Edward that's an excellent point, I hadn't considered electro-swing when responding to your initial comment, but I'm also excited about electro-chemical DAC. It will be interesting to see how Verdox/RepAir/others develop. Potential here for the facilities to be (1) smaller and (2) more energy efficient.
And this point, however, it's all potential and promotion -- the tech is at an earlier stage than the more well-known players, and these electro-chemical companies need to demonstrate low power-output DAC on a small scale before I take their claims seriously.
Have more to learn here so don't hesitate to point me in the right direction if you've seen new data.