Recycling Carbon
by alan krupnick and emily joiner
alan krupnick is a senior fellow at Resources for the Future in Washington and director of RFF’s Industry and Fuels Program. emily joiner is a senior research analyst at RFF.
Illustrations by james steinberg
Published April 21, 2023
The straightforward way to contain climate change is to dump less carbon dioxide from tailpipes and smokestacks into the air. There’s a very different means to the same end, though — one that has long been given short shrift by the environmental community, which says it gives polluters permission to continue polluting. But the times they may be a-changin’.
The idea of grabbing the CO2 after the fuel has been used, then recycling it or locking it up permanently — a process known as carbon capture, utilization and storage (CCUS) — is now in the spotlight in part because the technology is moving rapidly. And it just got a big boost with the 2022 Inflation Reduction Act (IRA), which expanded the existing federal tax credit (known as 45Q by the nerds in the audience) that’s offered for every ton of captured carbon safely used or stored.
CCUS is most attractive today because it can be grafted onto especially carbon-intensive activities like fuel refining as well as energy-gobbling industrial processes like cementmaking and electricity generation, bringing them closer to net-zero emissions territory. Indeed, it is now considered by many to be an essential part of a portfolio of solutions if we are to have any hope of limiting warming to 2 (let alone 1.5) degrees centigrade.
But it’s hardly going to be all wine and roses from here on out. Despite billions of dollars of investment in R&D and the aforementioned rich incentives for adoption, CCUS has yet to prove it is ready for the large-scale deployment needed to make a substantial difference. What follows is a look at where we are, where we’re going and why. Pardon the dive into technology — hopefully, it will be worth getting your feet wet.
How It Works
“Point-source” carbon capture typically happens like this. A gas stream — the effluent that would ordinarily go up a smokestack, for instance — is fed into an absorption chamber where the CO2 content is absorbed, usually with help from a solvent. The CO2-rich absorption material is then fed into a “desorption” chamber where it undergoes a temperature or pressure change, releasing the captured CO2. This is the stage of carbon capture that places the greatest additional energy demands on the system.
Identifying a winning technology is difficult because each application is unique, and the technologies are still evolving. For example, post combustion capture may be as cheap as $40 dollars per ton of CO2 or as expensive as $100.
While the solvent is regenerated for repeat use, the captured CO2 is cooled and pressurized to make it possible to transport efficiently. The capture may happen before the gas has been combusted, or more commonly, after combustion has occurred, leading to the two methods termed “pre-combustion” and “post-combustion” capture. Post-combustion carbon capture, which filters the exhaust emission stream, can be used with electricity generation from natural gas and coal, as well as for industrial processes like aluminum smelting, ammonia production, steel production and cement production.
The process of CO2 capture in its most commercial forms can be broadly described as absorption of the CO2 followed by desorption and regeneration of the solvent. But the specific techniques — as well as the efficiency and financial costs — depend on what’s in the emission stream and the type of solvent or capture material used.
That first consideration is most crucial since the concentration of CO2 in the gas stream drives other costs in CCUS, notably the cost of energy inputs and the cost of the capture material over its useful life. Where the gas stream contains more CO2 the solvent can more easily capture it, making it practical to use smaller capture units that cost less and consume less energy. Then, too, gas streams containing less CO2 typically have greater levels of other pollutants mixed in — most notably sulfur dioxide and nitrous oxide. And these pollutants burden the system by requiring more frequent replenishment of the solvent/ capture material.
Identifying a winning technology is difficult because each application is unique, and the technologies are still evolving. Pre-combustion capture has seen the most commercial deployment, being utilized in both natural gas processing and ethanol (alcohol) production. Post-combustion capture has applications in a greater number of industries and in power generation, but its costs are inflated by typically lower CO2 concentrations. Costs even within one technological category can vary widely. For example, post-combustion capture may be as cheap as $40 dollars per ton of CO2 or as expensive as $100.
The CO2 can reach its destination via truck, train, ship or pipeline. For aboveground transport, the CO2 must be liquefied via refrigeration and stored in insulated containers. Shipping liquefied CO2 is most economical over very long distances, though ideally the captured CO2 should never have to travel very far to its destination.
For pipeline transport, CO2 is most often cooled and compressed to its “supercritical” phase, which is about 500 times denser than it is as a gas. Pipelines are enormously expensive to build and therefore are the most economical when several carbon capture facilities in the same area can make use of them. As CO2 must be in its dense supercritical state to be injected into the ground, pipeline transit is the most apt when the destination is a permanent storage site.
Oxyfuel Combustion
The amount of energy that can be extracted in burning fossil fuels depends in part on the temperatures achieved in combustion. And one way to get higher temperatures is to burn the fuel in pure oxygen rather than air. So “oxyfuel combustion” is relatively climate-friendly because it generates less CO2 per BTU. The catch: pure oxygen is expensive and the process for extracting it from air is itself energy-intensive.
But what if you wanted to build a more efficient power plant that also captured the CO2 it produces? Burning fuel in pure oxygen greatly simplifies carbon capture since the CO2 in the effluent stream is much more concentrated. In effect, oxyfuel combustion offers a two-fer, increasing the energy-efficiency of the power plant and making it easier to recycle the carbon from the exhaust.
Oxyfuel technology is less far along than other technologies facilitating carbon capture. But the R&D is promising. And climate scientists see it as a way to increase the potential value of biomass-based fuels. Stay tuned.
Once the captured CO2 has reached the underground storage site, it is injected into special rock formations with adequate porosity that are found at roughly one-kilometer depth. These storage sites are subject to a host of regulations to prevent contamination of drinking water and to “permanently” (see below) prevent leakage. Essentially, the places natural gas and oil can be found underground make good geologic homes for CO2.
The alternative to indefinite underground storage is use CO2 as an input in industrial production or in finished products. Turning first to use in processes, the most ubiquitous is in “enhanced” oil recovery (EOR). When wells are first sunk, the oil is under pressure — think of the gushers in old-time movies when the protagonist strikes it rich. But as oil is extracted, the pressure drops. Injecting CO2 can be used to repressurize the well and extend its economic life. And once the oil is above ground, the CO2 can be recaptured and either reinjected or stored.
CO2 also has value in products, the classic example being carbonated beverages. The downside here — as in another common use of CO2, processing jet fuel — is that the gas is released with use and can’t be recaptured. At the other end of the spectrum are some technologies in development for using CO2 in making cement. The current process emits huge quantities of CO2, while the new technology actually absorbs CO2 in the cementmaking process.
Where Government Fits In
The most important policy initiative influencing CCUS in the U.S. is the aforementioned 45Q tax credit. Before the expansion of 45Q last year, eligibility requirements for firms looking to access the credit were more stringent. Depending on the source, facilities had to meet minimum capture thresholds between 100,000 to 500,000 tons of CO2 per year. Post-IRA, these capture requirements have been lowered sharply and the per-ton credit has been increased (and applicants can take the credit in cash rather than find a partner with tax liabilities to write off).
The infrastructure issues are formidable. There are only 5,500 miles of pipeline suitable for transporting CO2, and other pipelines can’t be repurposed. And, while there are plenty of geologic storage sites, there are bound to be land use and environmental hurdles.
Uncle Sam now pays $60 per ton (previously $35) for capturing CO2 used in enhanced oil recovery, while CO2 destined for geologic storage earns an $85 per ton credit (previously $50). At pre-IRA levels, the industry response was tepid. But given the involvement of industry players in drafting the expanded policy, and independent estimates that $85 is well within the range of the cost of capture in power generation and most industrial processes, the richer 45Q credits offer hope that CCUS will be economically viable on a large scale.
As the infomercials say: wait, there’s more. The 48C tax credit, also in the IRA, provides a 30 percent credit for the manufacture of equipment used for CCUS (as well as for other clean energy products). Then there’s the 45V tax credit in the IRA, which subsidizes production of hydrogen — provided most of the associated CO2 emissions are captured. Meanwhile, Washington also supports R&D and pilot projects on CCUS, with a multibillion- dollar grab bag of initiatives through the Department of Energy.
With all this federal cash on the table, will CCUS take off? It’s a massive undertaking, but not beyond our technological and economic capacities.
That said, the infrastructure issues are formidable. There are only 5,500 miles of pipeline suitable for transporting CO2, and other pipelines can’t be repurposed. NIMBY obstacles with CO2 may prove as difficult as they are with fossil fuel pipelines. And, while there are plenty of geologic storage sites, there are bound to be land use and environmental hurdles.
Nor is it entirely clear how the regulation of storage sites will play out. The EPA must either delegate regulation to a state or approve injection wells under its authority to administer the Clean Water Act. And the Bureau of Land Management has a protocol for processing applications for storage under BLM-managed lands. But to date there are only two states with delegated authority (Wyoming and North Dakota) and one EPA-approved well being filled with CO2. Others are in the permit stage, though, with most located in California or Louisiana.
Another big issue is liability for leaks from the geologic formations. For states with “split estate” ownership, where above- and belowground property rights may differ, the issue of who is responsible for verifying the integrity of the site is particularly knotty. Montana, Indiana, Louisiana, Wyoming, Nebraska and North Dakota have all addressed the issue, often as part of comprehensive legislation to enable CCUS. In contrast, California has not yet decided on subsurface property rights, which could prove to be a major obstacle.
A number of states have set up mechanisms to transfer monitoring responsibility for geologic storage sites to them after injection has been completed and a long “latency period” has passed in which the sites are monitored by the private owner for integrity. States will fund their ongoing oversight from fees collected from the companies doing the injecting. Project managers will thus opt when possible for states with lax requirements — at least more lax than California, which has a 100-year monitoring requirement for geologic storage. Consider, too, that leaks spell double-trouble for the liable party since the 45Q rules require tax credits to be repaid on a ton-for-ton basis.
The View from Ground Level
Some CCUS projects were underway before the richer subsidies were approved. There are currently 136 CCUS projects in progress in the U.S., with 23 operational CCUS facilities, 13 of which are commercial and 10 of which are demonstration facilities. Most of them produce ethanol or fertilizer or process natural gas. Most also use their captured CO2 for enhanced recovery at nearby oilfields — though some, notably the ethanol facilities, are pursuing geologic storage. Thirty more ethanol plants will start operation in 2024, joining a five-state CCS consortium that will store the CO2 in a massive sequestration site in North Dakota.
Carbon capture at power generation facilities and from other industrial processes is less common, though several projects are in development and will receive a boost from 45Q. Regional efforts are in the early stages, tackling tasks such siting, identifying markets and bringing together stakeholders.
The low-hanging fruit going forward will be in expanding carbon capture for activities with high-concentration emission streams, such as fertilizer production and other processes using natural gas. Expansion of carbon capture in these industries will also lead to opportunities for a simultaneous ramp-up of hydrogen production. Carbon capture, in fact, may be viewed as a stage in the supply chains for hydrogen and other natural gas derivatives like methanol. Meanwhile, R&D efforts spearheaded by DOE will likely bring down capital and operating costs, creating the potential for carbon-capture opportunities in power generation and hard-to-decarbonize industries like cement, steel and paper. Also don’t expect a lot of retrofitting in existing industrial plants. It is cheaper to build in carbon capture from the start.
The CCUS landscape is less developed outside the U.S. The non-governmental Clean Air Task Force lists 73 CCUS projects in Europe, only three of which are operational, and 15 CCUS projects in the Middle East and North Africa region, five of which are operational.
Direct Air Capture
This article (and most of the R&D on carbon capture) focuses on cleaning up the effluent from pre- and post-combustion industrial activities. But what if you could grab CO2 directly from the air and use it or store it underground?
You can — and it is most notably being done today in Iceland by a Swiss company called Climeworks. The technology is straightforward but the cost is ginormous, in large part because there isn’t much CO2 in air (a little bit, alas, goes a long way in warming the atmosphere). Climeworks is economically viable only because (a) it has access to dirt-cheap, carbon-free energy from Iceland’s geothermal plants and (b) it is charging a handful of big companies like Microsoft €1000 per ton of carbon sequestered — far, far more than the tax credits being offered by Uncle Sam.
The IRA does actually provide incentives for Direct Air Capture at $130 per ton if the CO2 is used for EOR and $180 if it’s put into the ground. Given that the geologic storage subsidy is low compared to the cost, it’s unlikely that approaches similar to Climeworks will spring up in the U.S. anytime soon (though DAC plants for EOR are in the works in the U.S.).
The Climeworks approach, then, is at best a teaser for better things to come, when the technology for scrubbing CO2 from air is much cheaper. Still, it’s a pretty cool teaser.
Environmental Pushback
The environmental movement has a deep antipathy toward fossil fuels because of their many destructive impacts on the environment, not least of which is related to climate change from the release of the carbon in plants and animals sequestered underground eons ago. Of course, technological change combined with market forces — the fracking revolution — has been largely credited for rapidly diminishing coal use for electricity generation. But, while releasing more energy per ton of CO2 emitted, the natural gas that now out-competes coal is still a large source of greenhouse gas emissions. Moreover, while petroleum also is a major source of CO2 emissions, the promise of electric vehicles is focusing more of the environmentalists’ ire on natural gas.
The natural gas sector sells itself as a bridge fuel to a low-carbon economy. The comeback is that carbon capture technology could make this bridge way too long, and that CCUS will soak up funds needed to develop a hydrogen/ electric economy based on renewable fuels.
As economists, we are inclined to leave environmental ideology to others. In our view, if natural gas remains competitive with renewables and in other uses after the application of CCUS technology, so be it. But that’s a big “if.”
The Enhanced Oil Recovery Detour
One of the loudest critiques of CCUS comes from those opposed to the utilization of CO2 in enhanced oil recovery. The idea is that EOR makes more (and cheaper) oil available, and therefore adds to CO2 emissions. But this concern doesn’t stand up well to scrutiny.
First, even if CO2 were not available as the injection medium, enhanced oil recovery would still be economically viable using other substances. Second, EOR-based oil produced in the United States is a trivial share of the global market, so its elimination would have trivial effects on the global price, and therefore global output. And in any event, EOR is still a sideshow even if the market for oil worked perfectly: the global oil price is influenced in an outsized way by administrative decisions made by OPEC countries, and ultimately by the cheapest and largest oil producing state, Saudi Arabia. If the Saudis opposed the modest price increase linked to an end to enhanced recovery, they could make additional oil available from its reserves and the price would drop. In short, the EOR pathway is real, but a distraction from the main event.
Two Steps Forward…
CO2 leakage in the process of capture, transport and utilization/storage is also of concern beyond the oil recovery context. Can a ton of sequestered CO2 really be considered a ton if it doesn’t account for the energy expended to capture and prepare it for transport? Careful calculation of the ancillary emissions produced over the entire capture to utilization/ storage life cycle is needed to quell concerns that CCUS will not deliver on its promised levels of reduction.
In addition to the increase in energy demand from facilities with carbon capture technology, there is also a significant increase in demand for water, which is used for cooling. This has been flagged as problematic for deploying CCUS on a large scale, particularly in areas — including much of the American West — that struggle with water scarcity.
Critics emphasize that expensive pilot studies have failed to demonstrate that carbon capture at scale in electricity production is economically viable.
Environmental Justice
Environmental justice advocates also have a bone to pick with CCUS. Their concerns start with the reality that exposures to air pollution have been disproportionately borne by disaddvantaged communities. And that given those communities’ lack of access to health care, and other consequences of the historical marginalization of communities of color and those living in poverty, the consequences of exposure to toxins are magnified. CCUS may thus add to already disproportionate exposure because it takes electricity to run the carbon capture equipment, and so, unless the grid itself is “clean,” the air pollution exposures in marginalized communities from additional electricity generation can only grow.
In addition, carbon-capture technologies themselves create a concerning pollution load. The most common capture technology is based on the use of amine as a solvent, which emits ammonia, itself a pollutant but also a precursor to the formation of the more overtly harmful small particulate matter especially damaging to airways.
Then there’s the issue of the pipelines. Deploying large-scale carbon capture will depend on construction of many more thousands of miles of pipelines that would operate at very high pressures and very low temperatures. And that poses obvious safety and health risks.
Why is this of special interest to disadvantaged communities? Pipeline siting will be subject to the same political pressures as other NIMBY issues. And if the past is prologue, there is little reason to believe that marginalized communities will fare better in this context than they have in the siting of everything from oil refineries to waste dumps. Resistance to CO2 pipelines may mirror the fight against the Dakota Access Pipeline, the 1,200 mile pipeline connecting the oil fields of North Dakota to a transit hub in Illinois. Opposition to the pipelines needed to create the Summit Carbon Solutions’ bioethanol hub is already developing in Iowa, with environmental and Native American activists as well as farmers uniting to prevent it breaking down.
What Could Go Wrong with the Technology?
We’ve already had a preview of the commercial viability of CCUS. Petra Nova, a coal-powered generation plant in Texas was retrofitted with a post-combustion carbon capture system at an estimated cost of $1 billion. As the first postcombustion power plant in operation, it started capture in 2017 — but was out of service by May 2020. That’s because the economics of Petra Nova carbon capture depended on demand for CO2 in oil recovery, and the demand for injection material cratered, along with the demand for oil, during the pandemic. Adding to its woes, Petra Nova’s performance fell about 15 percent short of CO2 capture goals during operation.
It might not take much to stall the train even as it approaches its destination. Environmental groups have already helped derail a bill in New Mexico to promote hydrogen production by capturing CO2. Pipeline siting is a particularly acute point of vulnerability.
More recent attempts to retrofit power plants such as the Kemper project in Mississippi and the San Juan Generating Facility in New Mexico have also foundered due to prohibitive capital costs and ownership disputes. Critics emphasize that expensive pilot studies have failed to demonstrate that carbon capture at scale in electricity production is economically viable, and that electricity from renewables remains the cheaper and greener option.
Leaks (Again)
Though hardly the most volatile substance that travels by pipeline, CO2 is an asphyxiant that necessitates careful transport when concentrated. In 2020, a CO2 pipeline leak in Louisiana sent 49 people to the hospital and necessitated the evacuation of the immediate area. Though no serious injuries resulted, the incident catalyzed opposition to impending pipeline expansion and led the Transportation Department to announce new safety rules for CO2 pipelines last year.
Perverse Incentives
As noted by many, 45Q subsidies reward CCUS rather than CO2 emissions reductions. And that is a distinction with a clear difference. Like any subsidy, 45Q subsidies can reward bad behavior — here, for example, increasing the competitive position of coal in electricity generation because coal contains more CO2 to be captured. Consider, too, that 45Q ignores the life cycle emissions implications of CCUS systems. And even if one wanted to take note of such second-order effects, this is often very difficult. For instance, carbon capture operations where CO2 emissions from natural gas processing are being captured may be a net negative for greenhouse gas emissions if the natural gas being processed comes from wells leaking climatewarming methane into the air.
What’s Next?
Given all the promise and perils of CCUS, what should we be looking for in the future?
IRS rules. The Internal Revenue Service is responsible for writing the rules to implement the tax credits for CCUS projects (45Q) and the tax credit for clean hydrogen projects using CCUS (45V). Look for how the rules balance the goals of making it easy to qualify for the credits against assuring that the projects create lasting, sizable reductions in CO2.
The industry response to 45Q and 45V. Even though the increase in the tax credits owes much to industry lobbying, those same players might largely take a pass if the rules prove onerous. Earlier versions of 45Q didn’t work well, as the IRS was obliged to claw back about half of the tax credits over the question of leaks from enhanced recovery applications. What applies to 45Q applies to 45V in spades, as the regulatory landscape is less plowed, and the terrain is less well understood.
Cheaper, more reliable technologies? There are a host of issues that remain open — everything from liability for leaks to pipeline safety — and even issues of bureaucratic capacity to get the money out the door. Then there are technology uncertainties. Will we see the same kind of steady downward movement on costs as has been enjoyed in wind and solar power? More efficient technology will be needed to make many carbon capture applications commercially viable.
Antipathy to CCUS. It might not take much to stall the train even as it approaches its destination. As already noted, some environmental groups are deeply skeptical about the inherent desirability of containing carbon emissions after the fact. And these groups have already helped derail a bill in New Mexico to promote hydrogen production by capturing CO2. Pipeline siting is a particularly acute point of vulnerability.
* * *
We believe the perfect should not be the enemy of the good. We are past the point at which we can reject plausible paths to zero carbon without a serious try. And CCUS can’t be dismissed, because it does hold the promise of making significant reductions in carbon footprints around the world. While it is hardly time for starry-eyed celebration of a group of technologies that are both expensive and unproven, there is still very good reason to make the investment to find out. And the IRA legislation is doing just that.