DIRECT AIR CAPTURE IS NOT OUR FUTURE
THE FUTURE OF ENERGY AND DECARBONIZATION
DIRECT AIR CAPTURE SUCKS
DAC is not our future, short version:
If we allocated EIGHT TIMES all the renewable energy in the world to Direct Air Carbon capture (DAC), we still WOULDN’T REDUCE CO2 BY A SINGLE MOLECULE; we would ONLY just be keeping up with new emissions.
VACUUM CLEANING THE SKY
The idea of DAC is to run great big machines that pump vast quantities of air and use chemical reactions that remove the carbon and then sequester it underground. Think ‘vacuum cleaning the sky’ and you get the picture.
Companies such as Carbon Engineering and Climeworks are already running these machines. Customers like Microsoft are already paying them to capture carbon on their behalf. Fossil-fuel executives like Occidental Petroleum’s CEO Vicki Hollub are saying “direct capture technology is going to be the technology that helps to preserve our industry over time. This gives our industry a license to continue to operate for the 60, 70, 80 years” and are even pointing to sequestered CO2 as a means to frack more oil and gas out of the ground. International leaders are pointing to them as a way to keep burning and exporting fossil-fuels.
MOST PEOPLE LOOK AT THE DOLLARS
One way, the most popular way, to look at the viability of DAC as a future solution might be totalling up the dollars. If we can stomach the cost, the logic goes, then maybe we can scale it.
In 2019 I was lucky enough to meet and then interview the wonderful David Keith, a world-leading climate physicist and the CTO of Carbon Engineering. At the time his predicted costs for the commercial version were in the order of USD 200 for each ton of CO2 captured, and his aspiration was to eventually get costs down to USD 100 for each ton of CO2 captured. David was extremely frank about both the weaknesses and the strengths of his technology, and I believe he thought it achievable. $100/ton also happens to be the stretch goal of the US government’s Carbon Negative Shot, and based on the snippets of data I’ve picked up since, it still seems to reasonably combine plausibility with ambition for a best case.
Multiplying this best case of $100/ton by 37,149,786,000 tons (global CO2 emissions in 2022) gets you a starting point of USD 3.7 Trillion per year. That’s operating costs of course, just enough to pay for this year’s CO2 without removing a single molecule of last year’s CO2.
On top of that you’ll need to add the following:
Capital costs to setup tens of thousands of DAC plants around the planet (big!) and chemical plants and transportation and infrastructure to support the DAC plants
Annual incremental increases to both operating costs and capital costs keep pace with the NEW coal and gas-fired power plants coming online
Double and redouble all of the above, because presumably you’d like to do more than just ‘keep up;’ you’d like to at some point actually reduce the atmospheric concentration of C02?
Eventually you’ll arrive at a number in the tens of trillions. You might decide it’s too big for DAC to scale. Or, you might decide it’s within the bounds of acceptability, that businesses and taxpayers of richer countries might be willing to pay it when the super-storms are hitting their houses often enough.
So, that’s one way.
But it’s the wrong way.
THE KILLER IS ENERGY
Back in 2019, I also asked David Keith how much energy was required to power the DAC process, and his answer to this was far more interesting than the dollars: his then-experimental plant burned natural gas and used “about 5 GJ” of gas to power the process per ton of CO2 captured.
Uh-oh. This says immediately that DAC is a game of vicious diminishing returns.
DAC takes energy. A lot of energy. That energy has to come from somewhere. There is no escape. The physics won’t change.
Now, quite obviously commercial versions of DAC are powered by renewables or you’re spewing CO2 while you’re sucking CO2, but a Terrawatt allocated to DAC is of course a Terrawatt NOT allocated to reducing the corresponding emissions of replacing fossil fuels, which puts you right back where you started in terms of real consequences for the atmosphere. Argh!
HOW MUCH ENERGY?
Now let’s start calculating energy numbers. Using my most conservative estimates (see bottom of this post for more notes and data sources) …
Energy cost to capture 1 ton CO2 = 1,833 kWh
Global CO2 emissions in 2022 = 37,149,786,000 tons
Multiplying the above yields total energy requirement to keep up with new emissions: 68,095,557,738,000 kWh (68,096 TWh)
Total global electricity produced by renewables in 2022 (solar, wind, hydro, geothermal, bioenergy, other) was 8,533 TWh
Dividing total energy requirement by total renewables production tells you that you’ll need a minimum of 798% or eight times global renewables, just to keep up with new emissions and before REMOVING A SINGLE MOLECULE OF LAST YEAR’S EMISSIONS.
And like the dollar calculations earlier, we still have to add …
Energy to setup tens of thousands of DAC plants, and infrastructure/chemical plants to support those DAC plants
Energy to build 8 x global renewables capacity (not 7x because presumably we’ll want to keep using renewables for their original purposes?)
Double and redouble all of the above, because presumably you’d like to do more than ‘keep up;’ you’d like to at some point actually reduce atmospheric concentration of C02?
Does that sound like the worst investment proposition you’ve ever heard? Maybe the worst in human history?
THIS is why scientists label DAC variously as unscalable, greenwashing, myth-making, a fairytale, magical thinking, a monstrous dead-end, a pernicious excuse to do nothing, dangerous, and a scam perpetuated by fossil-fuel companies that solidifies our global fate.
WE ALREADY HAVE THE TECHNOLOGIES
Of course, if you took the simple route and didn’t run any DAC or indeed any new technology and simply used new renewables production to replace fossil fuels, you’d have NO vicious circle and would eliminate emissions at source, not perfectly, but efficiently and permanently and with a solid and extremely well documented ROI.
We already have all the technologies and pathways we need to avoid the worst impacts of climate change, if only we choose to take them. It’s a tough road, and a complex message. We’ll need to change a lot of behaviours and mature beyond “perpetual growth” as a success metaphor (big changes!) but we can do it, and the commercial opportunities happen to be vast and positive. Take note: decarbonisation will create more new wealth than any other endeavour in human history.
DAC IS NOT A VALID PATHWAY
Because of the ENERGY requirements …
Direct air capture offers no pathway out of climate change.
Direct air capture is not even a ‘promising approach’ to climate change.
Direct air capture cannot cancel fossil fuel CO2 emissions by ‘vacuum-cleaning the sky.’
The ONLY sensible allocation of renewables is to replacing fossil fuels, NOT running DAC
Direct air capture is not our future
And just in case you had any notion that we can install such vast quantities of renewables that we can have our direct air capture cake and our renewables future too … we can’t. We don’t have that kind of install rate (while we are installing renewables faster and faster, we also continue – shockingly – to build new coal, oil and gas power plants, which is why emissions are still increasing) and we certainly don’t have the time. We need to start phasing down emissions NOW, not in some dystopian planet-encrusted-with-giant-vacuum-cleaners future 50 years from now.
To be triply clear: there is no pathway to a (non-catastrophic) future where we can still burn fossil fuels.
IS THERE ANY LEGITIMACY to carbon capture?
Some: I can support limited R & D in “carbon capture at source” (i.e. direct from the combustion process of power plants or factories) which is more concentrated and therefore more energy efficient and could perhaps help mitigate emissions in steelmaking, for example, while it makes the challenging transition to hydrogen. But note: carbon capture at source still requires additional energy! There is no free ride! And ANY extra energy used is still more efficiently applied to replacing fossil-fuels to avoid making emissions in the first place.
As for direct air capture, none.
improve my CALCULATIONs
My goal is always to communicate our future, and especially pathways to a better future, as objectively and scientifically as I can, but I do make mistakes. I encourage you to check my math, and to substitute your own assumptions and do your own calculations, and I positively welcome dissenting voices that come from a rational and scientific background. I have posted this publicly on LinkedIn. Always feel free to engage with me there, or to point me to more accurate calculations and better data sources.
NOTES on DATA SOURCES
Different estimates are sometimes available for the same data point. In all cases, I’ve tried to be conservative with my assumptions and as favorable as possible to DAC advocates. Click hyperlinks for sources.
For energy requirement per ton of CO2. David Keith’s 2019 estimate in his conversation with me was “about 5Gj/ton” for the energy required to remove one ton of CO2 using DAC, which converts to 1,389 kWh/ton, but this did not include the sequestration of the carbon. Since then, Carbon Engineering (David Keith’s company) has set up commercially and says it requires energy inputs of 8.8 Gj/ton which converts to 2,444 kWh. Climeworks has been quoted as placing their long-term projections based on current technology assumptions for the DAC process at 2,000 kWh/ton. SAPEA (Science Advice Police For European Academies) says the thermodynamic minimum energy required to extract CO2 from ambient air (ie the theoretical minimum based purely on breaking and making chemical bonds) is 250 kWh/ton and the energy consumption requirement for actual DAC plant operations is 2000 kWh / ton. The International Energy Agency estimates the best case DAC energy requirement at 6.6 Gj/ton CO2 which converts to 1,833 kWh. Being generous to DAC advocates, and discounting David’s early and very informal estimate to me which didn’t include sequestration, I’m using the best case of 1,833 kWH / ton.
Global CO2 emissions. Global CO2 emissions from all sources and including land use changes totalled 41,459,737,000 tons (41.5 Gt) in 2022. If we exclude land use changes, the total was 37,149,786,000 tons (37.1 Gt) in 2022. Again, I’m being generous to DAC advocates and using the lower figure.
Total global electricity produced in 2022 (all sources) was 28,528 TWh and total global electricity production from renewables in 2022 (solar, wind, hydro, geothermal, bioenergy, other) = 8,533 TWh