By Kurt Zenz House, Josh Goldman, Charles F. Harvey, December 15, 2023
Carbon Engineering pilot plant in Squamish, B.C. that captures carbon dioxide directly from the atmosphere. April 20, 2016 (Stephen Hui/Pembina Institute/Flickr)
This month elites from 198 nations gathered in the fossil-fuel-rich United Arab Emirates for the 28th annual Conference of the Parties to the United Nations Framework Agreement on Climate Change. Near the top of the agenda is the deployment of technologies to remove carbon dioxide, the principal greenhouse gas causing global warming, from the atmosphere. The week before the conference started, The Economist published an approximately 10,000-word special report on the topic, and the Financial Times reported that direct air capture of carbon dioxide is “grabbing investors attention.”
All year, the zeitgeist has been building toward technologies that separate carbon dioxide from air, referred to as direct air capture (DAC). In September, the United States Department of Energy awarded Occidental Petroleum a $600 million grant to build a DAC machine. As scientists and entrepreneurs who’ve dedicated our careers to help solve global warming, you might expect us to be happy.
We are not.
The reason is simple: Separating carbon dioxide from air, while technically straightforward, is outrageously expensive. In fighting climate change, the obvious question should always be: How can we avoid the most carbon dioxide per dollar invested?
Answering that question can be difficult: Should I buy an electric car or put solar panels on my roof or eat a plant-based diet? Should the government subsidize heat pumps or grid batteries or tree-planting? But, in this case, it’s easy: Air capture is among the most expensive of all climate mitigation options.
Oxy estimates that the project will separate 500,000 tons of carbon dioxide per year and cost about $1 billion to build. Adding in operations and maintenance, we, and others, estimate the total costs will be more than $500 per ton of avoided carbon dioxide.
It is easy to see that $500 per ton is extremely expensive; it implies $6 per gallon to clean up the carbon dioxide put in the atmosphere from burning gasoline. Dealing with all US emissions would cost about $3 trillion per year, every year—about 4 times what we spend on the entire US military—and would require building both air-handling capacity larger than the combined capacity of every HVAC system in the entire country as well as new power plants (all from carbon-free sources!) equal to twice the total power generation capacity of the US today. If this is what’s required to avoid climate change, we’re in big trouble.
Fortunately, there are abundant carbon mitigation opportunities that cost below $25, and even below $0, per ton of carbon dioxide avoided. So the same money we just gave to Oxy could easily have avoided well more than 20 times as much carbon.
Who could possibly have thought it was a good idea to spend so much taxpayer money on such an expensive approach?
Three groups: techno-optimists, who believe big investments will make air capture costs come down dramatically (they won’t); would-be climate central planners, who argue that we need air capture (if we ever do, it won’t be for 50 years, or more); and oil companies, who think they’re positioned to get fat subsidies (they’re right).
Techno-optimists are desperate to believe that air capture won’t always be this expensive: If we make big investments today, it will result in lower costs for air capture through innovation. But there are several reasons why this won’t happen.
First, industry has been separating dilute carbon dioxide from gas mixtures since World War II to purify natural gas and refresh air on submarines and spaceships. Over decades, private competition has driven innovation in these systems to maturity, so the low-hanging fruit was picked long ago.
Further, the major costs of air capture are not the kind that decrease with experience. Unlike high-throughput manufacturing of computer chips or solar panels, infrastructure costs have relentlessly increased with cumulative investment; similarly, industrial construction has consistently gotten more expensive. Air capture plants are large fluid-processing systems consisting of air-intake manifolds, absorption and desorption towers, liquid-handling tanks, and bespoke site-specific engineering. All of those components are built today, at scale, around the world, and they are not getting cheaper.
Finally, the laws of physics constrain how cheap air capture can get. The energy and cost required to separate one gas from a mixture increases as its concentration decreases. The unprecedented carbon dioxide level in our atmosphere, about 400 parts per million, is high enough to drive global warming, but still physically very small: just 0.04 percent. Carbon dioxide separation is much more effective at higher concentrations, like the gases emitted from power plants, which contain up to 400 times more concentrated levels of carbon dioxide than air. This higher concentration translates into 20 times lower capital costs and well more than six times lower operating costs for the same quantity of carbon dioxide separated. So, any air capture system would operate much more effectively by just plumbing the exhaust from a power plant into it.
And yet, carbon capture from smokestacks is already among the most expensive methods of avoiding carbon dioxide emissions. Few projects have succeeded at actually injecting carbon dioxide at all, costs have not declined for decades, and no project has cost less than $100 per ton avoided, unless it also produced natural gas or oil. Smokestack carbon dioxide capture has a grotesque history of failure with costs spiraling out of control—US taxpayers and Mississippi utility customers paid over a billion dollars for such a project that was recently demolished before capturing any carbon. Occidental Petroleum quietly sold off a large carbon capture plant that only ever operated at a small fraction of its nameplate capacity. Now the government is funding an even worse white elephant that’s guaranteed to be many times more expensive per unit of carbon dioxide.
The climate central planners argue that we have to build air capture now to achieve net zero carbon dioxide emissions because we cannot eliminate all uses of fossil fuels.
Air capture sounds appealing. Hard-to-decarbonize sectors like steel, cement, planes, and ships? Air capture can remove the emitted carbon dioxide from those sectors. Countries that keep burning coal and keep building coal plants? Air capture plants built and operated in America can offset emissions from those countries. Carbon dioxide levels already too high? Air capture can lower the concentration in the atmosphere below where it is today.
These are the reasons that central-planners love air capture, and why the US government is funding projects like Oxy’s new plant in Texas and providing unprecedented subsidies of up to $180/ton.
But it’s not possible to realize these benefits if air capture is too expensive—which, as we have discussed, it always will be. So investing in air capture today incurs a grave opportunity cost: If we spend $500 to separate one ton of carbon dioxide, that means that we didn’t spend that $500 elsewhere to avoid 20 tons, or more. Those extra 19 tons will remain in the atmosphere, warming the planet every year for thousands of years.
This year, and every year, government and private investors should deploy our limited money to maximize cumulative carbon avoidance. Every dollar invested in air capture—that would otherwise have been invested in solar, wind, EVs, grid batteries, nuclear, or even carbon capture on power plants—makes the planet hotter. There are abundant opportunities for lower-cost climate mitigation investments; it does not protect the environment to divert money to high-cost carbon dioxide solutions before we have depleted the huge inventory of much less expensive options. The air-capture planners worry about avoiding the last tons of carbon dioxide emissions (those hard-to-decarbonize sectors)—which we shouldn’t focus on until we have reached 100 percent zero-carbon electricity, a fully electrified vehicle fleet, and efficient buildings. It’s just dumb to build today something that we won’t need for 50 years, if ever.
Think of carbon dioxide in the atmosphere like water in a tub, and our mission is to lower the tub’s water level, but it keeps rising because the faucet is on full blast. The free-flowing tap is the global fleet of fossil fuel plants and cars pumping carbon dioxide into the atmosphere at ever greater rates. Investing in expensive, inefficient air capture rather than vastly cheaper measures like renewable energy and electric vehicles is like buying gold-plated thimbles to bail out the tub instead of turning off the faucet.
Oil companies like Oxy claim they’ll have one business extracting carbon as oil and natural gas and another business getting it back out of the air. The problem for Oxy is that it costs way too much to get that carbon dioxide out of the air.
Their solution? Get the taxpayer to build them an air capture machine.
But that’s not enough. Even once Oxy’s new air capture plant is built, it will cost hundreds of dollars per ton of carbon dioxide to operate. Separating and storing carbon dioxide doesn’t intrinsically generate any revenue; instead Oxy gets government subsidies for every ton of carbon dioxide sequestered and sells carbon credits on the voluntary market to other businesses, such as Amazon, for this same carbon dioxide. Oxy is selling carbon credits for sequestration that is already subsidized by the government. But even this combination of huge subsidies and sales of carbon credits likely won’t cover the operating costs. So, how could Oxy make money by running this plant?
Once separated from air, the carbon dioxide must be injected underground, and, if it’s injected into an old oilfield, more oil can be flushed out—a process known as enhanced oil recovery. This process has produced tens of billions of barrels of oil since it was pioneered 51 years ago in the very same Permian Basin where Oxy will build its air capture plant.
Oxy has long led the enhanced oil recovery business in West Texas, where it has a network of pipelines to transport carbon dioxide to oilfields, and they’ve been upfront about their goal to combine air capture with enhanced oil recovery to produce what they call low-carbon fuel or “net-zero oil.”
Likely because of fossil fuel industry lobbying, the Inflation Reduction Act’s subsidies for air capture extend to enhanced oil recovery projects at the gargantuan rate of $130/ton. US taxpayers could end up sending Oxy $65 million each year on top of the $600 million we just gave them. Oxy will then use the captured carbon dioxide to produce as much as two million barrels of extra oil per year; burning this oil will generate about twice as much carbon dioxide as they pulled out in the first place—hardly the carbon-negative project proponents would have you believe.
Unlike other climate technologies, the only way to make air capture a business is with oil production and perpetual giant subsidies. Misallocating resources to air capture makes the planet hotter. The only winners are the recipients of the subsidies and the builders of the boondoggles.
The Bulletin elevates expert voices above the noise. But as an independent, nonprofit media organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.
View Comments
Please enlighten us then, without capture or removal, how would the planet go on a pathway to 1.5C with minimal overshoot. IPCC AR6 WG3 scenarios for such pathways ALL include CCS, with a minimum of 300 Mt CO2 per yr in 2030 and median of scenarios around 1 gigaton per yr in 2030, globally. There are no scenarios without signifant CCS penetration unless you're comfortable with exceeding 1.5 and/or high overshoot. Or if there are realistic scenarios you have that keep us to Paris targets and dont include capture or removal, please share/link to your models. That would be more productive than a negative, ideological piece like this. Reality is we need everything: renewables, efficiency, ending deforestation, cut nonCO2 ghgs, and yes capture and removals.
All those scenarios are the result of lobbying by fossil fuel interests and other opportunists. But you have a valid question: How to get new zero? The only way I see is with the hydrogen energy economy. It's the only systems solution using mature technologies.
But even more than that, we need central management to go about the green energy transition, like we did in building the war machine for WWII. So the real problem is that the current paradigm requires each component of the infrastructure to be proven in the marketplace. That's foolish because there's not enough time and will satisfy only the same bunch of fossil fuel interests and opportunists. That's why the marketplace wasn't the arbiter in the war time economy. When in a crisis, humans reach for the most available tool, or weapon.
When attacked by a bear, we reach for the nearest weapon. You can't outrun a bear.
As the article describes, Direct Air Capture and Flue Capture followed by geological Storage or by Utilization for hydrocarbons’ extraction - range from being untenably expensive to being both massively expensive and highly counter-productive in terms of CO2 emission control. As opportunity costs they are damaging to society’s chance of controlling Climate Destabilization.
However, the article did not affirm or deny the urgent need of large scale Carbon Recovery from the atmosphere, running alongside accelerated efforts to displace fossil fuels, reverse deforestation and apply other modes of Emissions Control. Nor did it discuss the options for a scaleable sustainable mode of Carbon Recovery. In this light there is clearly a case for a follow-up article.
By a strange serendipity, there is one sustainable carbon sequestration technique that has been amply proven, at grand scale, and over more than a two millennia duration. Prehistoric farming communities across the Amazon Basin for centuries treated their land with charcoal steeped in very well matured compost. This resulted in areas of up to 400 ha.s of exceptionally fertile soil, with high microbial diversity and fungal dominance (by weight). Those areas are sought after and are prized by today’s farmers as they have retained their exceptional permanent fertility, as well as their charcoal content, and are known as "Terra Preta" [Dark Earth]
The optimum sylviculture for feedstock for charcoal production is that of traditional Native Coppice Forestry, where native deciduous trees are harvested on a moderate cycle of 7 to 28 years and allowed to regrow from the stumps. The regrowth is sheltered by the felling each year of one acre plots laid out at random across the land area (i.e. a 1,000 acre coppice on a 10yr cycle fells 100 plots each year). Owing to this and to the stumps’ large established rootball the regrowth is vigorous and the yield is around 20% higher than replanting for normal cohort forestry would give. An output of around 30% charcoal (25% pure carbon) by weight of feedstock is quite common.
A further serendipity is that before the fossil gas era tending Native Coppice Forestry was the UK’s Standard Industry Practice for the production of the liquid fuel methanol from the charcoal retorts’ hydrocarbon offgasses. The US NREL puts the present yield at 57% by weight of feedstock, i.e. from land harvesting 10 tonnes dry wood /ha /yr a yield of 5.7t /ha /yr of methanol would be expected.
Large scale afforestation for Native Coppice Forestry for coproduct charcoal and methanol would offer various merits as well as three distinct revenue streams :
No other mode of Carbon Recovery offers the long term security of sequestration, as well as massive rural employment and enhanced biodiversity, as well as improved Food Security, while also attracting sufficient revenues to become self-funding as the industry matures.
As yet there is little public knowledge of Native Coppice Forestry for charcoal and methanol, but that is something which The Bulletin could and surely should address.
From the figures you give (25% carbon per weight of feedstock, 10 tonnes dry wood /ha /yr) it looks like one hectare of trees could yield 2.5 tonnes carbon per yr. However, in the NREL publication "Where Wood Works" they say "on average, one-half of a bone-dry ton per acre can be sustainably removed from the forest." (pg. 13) Since one hectare is 2.47 acres and one ton is .9 tonnes, that works out to only 1.1 tonnes dry wood per hectare per year, which would be only 0.28 tonnes C /ha /yr. But for now, let's go ahead and assume the much more optimistic 10 tonnes per year.
Since wood does not start out dry, what would be the drying requirements (land area, energy inputs for each 10 tonnes dry wood output per year). And then this wood is turned into carbon... somehow. What equipment and land resources would that require, and what sources of energy might be needed? Are there any air quality considerations from this process? You also mention a processing step involving steeping in compost. How much compost would be needed to process 2.5 tonnes of carbon in a year? Does the composting process need any land area? Does growing the material to be composted need any land area? Does transporting the materials to be composted need any energy? Could we see some amount of wood or compost production shortfalls due to fires, heat waves, droughts, too much rainfall, or blights? Could the composting go anaerobic and release methane? And once the carbon has been processed into a soil-amendment for interment in farmland, how much farmland area would be used to sequester 2.5 tonnes of carbon? Does the transportation to farmland have any energy requirements? Does the interment process require any energy? You mention this approach would involve massive employment. Every other carbon removal option I know of tries to minimize the number of people needed per unit-weight of carbon. If this approach requires high labor input, isn't that a limiting drawback? And what would be the monetary source for the income for all these people?
We have an abundance of solutions we know will work at the small scale. The challenge is scalability. We will probably need net carbon removal of at least 10,000,000,000 tonnes C per year (much more would be better). Given the land, energy, and labor requirements of this soil amendment approach, how many tonnes C per yr do you think it could realistically handle?
Your quotation of the NREL remark of sustainable forestry yields includes the words "on average," and gives no indication of the forest location, of its type, of its management or of its origin. By contrast, the Native Coppice Forestry of which I wrote is an ancient sylviculture that is widely established on hill lands across Europe, and whose premium yields are extremely well proven, as is its sustainability under good management. If applied on lands within the humid subtropical regions, outputs of 10ts /ha /yr would be easily achievable.
A 30% yield of charcoal per tonne of feedstock in a well built retort - using wood that is commonly air dry at ~18%MC - is by no means exceptional. FYI wood pyrolysis for charcoal is an highly exothermic process - i.e. it uses no external energy resources - and the use of a retort also maximises the hydrocarbon offgasses available for reaction to methanol.
Technical developments include the research of microwave-powered pyrolysis which can reportedly yield over 40% charcoal by weight of feedstock, but would do so at the expense of the co-production of methanol.
In terms of sustainability of Native Coppice Forestry, the eminent ecologist, Prof. Alastair Fitter, found that "In-cycle native coppice accommodates the highest biodiversity of any European ecosystem" (Pers comm, York University,1994).
The theoretical upper limit of potential scale reflects the land area available for afforestation. The 2012 joint study of this by WRI & WFN identified 1,600Mha.s globally, without taking over old native forest, or farmland or special ecologies. If the average yield were as high as 10t /ha /yr, then the potential Carbon Recovery by charcoal sequestered in farmland would be around 4.0Gt C /yr. Given very rapid afforestation allowing full flow by 2050, the amount recovered by 2100 could be up to 200Gt C (or ~94 ppmv). As the upper limit of potential this is of course quite unlikely to occur.
While DAC is very expensive today and only a tiny amount has been deployed, the same thing would have been said about solar PV 25 years ago. Funding DAC is not about reducing the most emissions per dollar today. It is about developing and scaling a technology that is *required* for us to maintain a safe climate. As James Hansen has recently pointed out, we are effectively at 1.5ºC now and may pass 2ºC in the *2030s*! And 2ºC itself is catastrophic.
One of the most important things to know about climate change is that CO2 lasts in the atmosphere for hundreds to thousands of years, so things won't get better when and if we hit net zero. Whatever temperature we are at when we finally stop emitting GHGs, that is the temperature we have for the next 1000 years... if we are lucky and don't pass tipping points first!
So while renewables provide the most bang for the buck, DAC/CDR (and Sunlight Reflection Methods - SRM) are required to pass on a safe climate to our children. Renewables alone won't cut it. But we don't need to take renewable money to pay for DAC, we instead can take some of the $1 trillion in direct fossil fuel subsidies or the $6 trillion in indirect subsidies to pay for it. And if that's not enough, why not trade off cruise ships vs. DAC instead of solar vs. DAC?
While DAC is expensive, the cost of *not* deploying it is far higher than the cost of deploying it.
You give no reason for your assumption that DAC is essential because it is the only means of Carbon Recovery at scale.
Given that there are other, scaleable, highly preferable, well proven, potentially self funding means of Carbon Recovery, I think you need to reconsider.
Because the Earth has a lot of thermal inertia, it takes generations to heat up to the equilibrium point. So it was always known that the warming would continue for a very long time after we halted all our greenhouse gas emissions. The part we previously underestimated is how much shading/cooling effect we were getting from all those combustion particulates we were putting into the air--which turns out to be roughly half of the warming potential from the greenhouse gases that have already been released. Unlike CO2, those particulates clear out very quickly. So even if we could halt all emissions in the next ten minutes, the result would be that our pace of heating would accelerate, and in a matter of months, it would be roughly double the rate we are at now. And it would still take generations to reach equilibrium. Once we clean up our particulate pollution, we will only have three options to work with--solar radiation management, CO2 drawdown, or just trying to cope the best we can with much greater heat, and all the weather chaos and environmental devastation that would entail--especially if the accelerated heating is compounded by blowing through tipping points.
Out of those three options, I would really like to see us go big on CO2 drawdown. It seems like the least bad of the three. The big irony right now is that almost all of the opposition to developing our CO2 drawdown technologies is coming from the block of people who are the most worried about global warming.
The piece is almost entirely about carbon capture at the top of a smokestack, which is at best an emissions reduction play, and at worst a technique for extracting the last drops of oil from old wells. The authors don't mention the critical need to remove a big chunk of the roughly 1 Trillion tonnes of legacy CO2 that we've already spewed into the sky. A very telling point in the piece is that their reference (link) for the phrase "The climate central planners argue that we have to build air capture now" is not even close to referencing anything like "climate central planners" and instead links to a four-year-old article about one carbon capture (note, not "carbon removal") company partnering up with ExxonMobil. A better reference would have been Lawrence Livermore National Labs' detailed document called "Roads to Removal" that was released last Monday, where they could have guided their readers to learn the following from that 221-page document that experts spent two years preparing:
Removing CO2 from the atmosphere is critical to ensuring climate security and resilience. The destructive and profound climate disturbances caused by the past century’s excess greenhouse gas (GHG) emissions are unsustainable and already costing the United States trillions of dollars. To address this planetary emergency, human societies must immediately work to “decarbonize” and dramatically reduce GHG emissions, aiming to reach as close as possible to zero. But being pragmatic, we must recognize it will be impossible to decarbonize quickly enough or completely enough to avoid warming beyond 1.5°C. Thus, to sustain the goal of reducing global carbon emissions to net-zero by 2050, we will also need to directly remove CO2 from the air and subsequently store the carbon for as long as possible. The Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment (AR6) report states that globally we must remove 100–1000 billion tonnes of CO2 by 2100 to limit warming to 1.5°C.
The sheer scale of the task of Carbon Recovery indicates that neither charitable nor tax-payer funded technologies seem likely to meet the required scale of gigatonnes per year up to 2100. The prime hurdle is the priority given to spending on energy efficiency, emerging renewables, halting deforestation etc., as well as on loss and damage and adaption.
The nature-based sequestration options which rely on standing forests as carbon banks pose further deficiencies in terms of protection failures, of expanding drought and wildfire events, and of the global acceleration of trees’ metabolism (due to raised CO2) which is progressively cutting their carbon intake. See : “Brienen et al” (2020), "Forest carbon sink neutralized by pervasive growth-lifespan trade-offs" https://www.nature.com/articles/s41467-020-17966-z
It seems clear that while afforestation with native species can capture carbon rapidly and at low cost, it cannot be relied on to retain it, and it offers no significant revenue streams other than the sale of carbon credits, but these of course counter-act net Carbon Recovery tonne for tonne.
What is required is a carbon sequestration technology - which cuts the quantity to be processed by ~68% compared with the over-hyped CO2 sequestration. To be of value even at great scale, that technology needs to harness the highly benign role of soil carbon in the form of inert charcoal (~85% carbon) enhancing the fungal and microbial diversity in farm soil-ecologies, as well as acting as a potent soil-moisture regulator.
This use of forest-based carbon has the merits of rapid and low-cost carbon capture, with an exothermic conversion of wood in charcoal retorts, with a blending with compost or inoculant to form a valuable soil amendment, whose interment in farm soils then provides both a secure very long-term carbon sequestration and raised crop yields. When the technology is deployed at global scale, the latter benefit can strengthen the currently threatened global Food Security.
In an earlier post on this page this option is described in a little more detail, along with a description of the potential for co-production of the liquid fuel Green Methanol. It would be very interesting to know your thoughts on the option overall, as you clearly understand the need of an appropriate scalable technology, and that the need is plainly increasingly urgent.
The argument against carbon capture from air is well articulated in this article. There is one possible exception that was not addressed, however. It is capture by calcium deposits in the earth -- which permanently capture carbon in the form of calcium carbonates. This was discussed in the Scientific American ("The Carbon Rocks of Oman," July 2021, p. 44). While this article did not give a definitive estimate of the cost, but the cost per ton appears to be well below the costs discussed in this article.
Outstanding article. Makes nothing but sense.
Suppose white hydrogen becomes important. No need to transport it. Use it in place to capture CO2 from the atmosphere and dissociate it. The products are pure carbon, oxygen and water. Carbon mining in reverse. The carbon might serve as a soil amendment, like biochar.
In any case, pure carbon is easily and safely sequestered for as long as we will be around. No need for CO2 pipelines and underground storage, which are not safe, if we could even find enough space to store the CO2.
Cut the Gordian Knot. Capture and dissociate the CO2 with the clean and essentially free energy of white hydrogen. If we can find enough of it.