Papex

Open problem · Chemistry · Open

Economical direct air capture

Can new solvents or other techniques make capturing carbon dioxide directly from the air economical?

Summary

Direct air capture removes carbon dioxide from ambient air, but because the gas is so dilute, capture needs a lot of energy and published cost estimates differ widely. Whether new sorbents or processes can bring the cost and energy demand down far enough is unresolved.

Why it is open

Carbon dioxide makes up about 420 parts per million of air, roughly 0.04%, compared with 3–20% in the flue gases of power plants and industrial processes [11]. Direct air capture (DAC) uses chemical sorbents that bind CO2 from air and are then regenerated in repeated cycles, releasing a concentrated stream [3]. Removing one tonne of CO2 means processing about 1.8 million cubic metres of air [11].

Thermodynamics sets a floor on the energy: a 2019 committee of the US National Academies of Sciences, Engineering, and Medicine put the minimum work of capturing 75% of the CO2 from air at 400 parts per million, as a stream of 98% purity, at 0.45 GJ per tonne (20 kJ per mole) [5]. For liquid-solvent designs it estimated the actual work at 8.2–11 GJ per tonne, an exergy efficiency (the ratio of minimum to actual work) of 4.1–6.2% [5]. How far real processes can close that gap is disputed: from the performance of operating commercial separation systems, Kurt House and colleagues argued in 2011 that, unless DAC significantly outperforms them, it is likely to require more than 400 kJ of work per mole [2].

"Economical" has no agreed bar. John Young and colleagues measure costs against a target of $100 per tonne [8]. Howard Herzog and colleagues, in a 2024 paper summarised by MIT News, report that recent modelling studies assume DAC costs of $100–200 per tonne, which they consider unrealistic [11]. Costs are also quoted on different bases (per tonne captured, per tonne removed net of the plant's own emissions, or per tonne avoided), and the National Academies committee warned that comparing them without first normalising conditions and boundaries is misleading [5]. The problem is therefore partly chemical, finding sorbents and cycles that need less energy [3], and partly a question of engineering economics: whether costs fall with deployment, and how far [8, 9].

What is known

Early estimates (2011). The American Physical Society's 2011 assessment estimated $641–819 per tonne avoided for a benchmark liquid system; the National Academies committee later wrote that this design, adapted from the capture of CO2 in flue gas, "introduced key limitations" [5]. House and colleagues estimated total system costs on the order of $1,000 per tonne, from experience with large-scale trace gas removal systems [2]; the committee noted that this estimate was not based on a specific technology [5].

Process designs (2018–2019). David Keith and colleagues, all four of them affiliated with the company Carbon Engineering, described a plant capturing about 1 million tonnes a year with a potassium hydroxide solution and a calcium loop that regenerates it, with pilot-plant data and a capital cost breakdown developed with an independent engineering firm [4]. Delivering CO2 at 15 MPa, the design needs either 8.81 GJ of natural gas, or 5.25 GJ of gas and 366 kWh of electricity, per tonne; its levelised cost ranges from $94 to $232 per tonne depending on financial assumptions, energy costs and the choice of inputs and outputs [4]. In its generic analyses the National Academies committee found, for a liquid-solvent system powered by natural gas, capture costs of $147–264 per tonne and net-removed costs of $199–357 [5]. For solid sorbents it found that processes not optimised for DAC fall in House's range of $1,000 or more, that processes with physically realistic parameters designed for DAC could cost $100–600 per tonne, and that large plants using known materials could approach $100 in the most promising scenarios, "though no such large-scale, continuously operating installation exists at this time" [5].

Projections (2019–2024). Mahdi Fasihi and colleagues projected that low-temperature solid-sorbent DAC powered by solar, wind and batteries under Moroccan conditions would cost, in their conservative scenario, €222 per tonne in 2020 and €54 by 2050, or €133 and €32 with free waste heat [6]. Using a costing method consistent with each technology's readiness level, Young and colleagues found that DAC with storage is unlikely to reach $100 per tonne, its costs falling to $100–600 at the scale of a billion tonnes a year, and that coupling DAC to intermittent renewable electricity with a low capacity factor is uneconomical [8]. Katrin Sievert and colleagues assigned experience rates to plant components according to their similarity to mature technologies. At a cumulative capacity of 1 billion tonnes a year, they projected costs per tonne net removed, including CO2 transport and storage, of $341 ($226–544 at 90% confidence) for liquid solvents, $374 ($281–579) for solid sorbents and $371 ($230–835) for calcium oxide ambient weathering [9].

Energy and market prices (2024). According to MIT News, Herzog and colleagues found that most proposed processes need the equivalent of at least 1.2 MWh of electricity per tonne, including compression, so that electricity at $0.10 per kWh would alone cost $120 per tonne, and they consider such a low electricity price questionable [11]. The same article, from November 2024, reports a market price of $1,500 per tonne for removal credits sold by the company running the largest plant then in operation, an industry price rather than a cost estimate [11].

Main approaches

Liquid solvents. An alkaline solution absorbs CO2 and the resulting carbonate is heated to release it and regenerate the solvent [5]. Strongest evidence: a published design with pilot-plant data and a costed plant of 1 million tonnes a year, from the company developing it [4]. Main difficulty: regeneration needs temperatures of up to 900 °C, and with natural gas as the heat source the committee's net-removed cost was $199–357 per tonne [5].

Solid sorbents. Supported amine materials and metal–organic frameworks (MOFs) bind CO2 [3] and release it on heating, alone or combined with vacuum, at temperatures below 150 °C [5]. Strongest evidence: in 2024 Zihui Zhou and colleagues reported a covalent organic framework, COF-999, that fully retained its performance over more than 100 capture cycles in the open air of Berkeley, California, and regenerates at 60 °C [10], a laboratory result from one group. Main difficulty: the committee's cost scenarios varied sorbent life from 0.25 to 5 years and sorbent price from $15 to $100 per kilogram [5], and Eloy Sanz-Pérez and colleagues concluded that new materials must be developed "with a practical sorbent–air contacting process in mind" [3].

Moisture swing. An anion exchange resin absorbs CO2 when dry and releases it when wet, a cycle driven by moisture rather than by changes of temperature or pressure [1]. Strongest evidence: wetting the resin raises the equilibrium partial pressure of CO2 over it by two orders of magnitude [1]. Main difficulty: the one humidity-swing company in the committee's 2019 list was at laboratory scale, producing 3–5% CO2 for algae rather than the roughly 98% purity the committee assumed for storage [5].

Electrochemical capture. In the cell of Sahag Voskian and T. Alan Hatton, reduced quinones in an electrode bind CO2 when the cell is charged and release it on discharge [7]. Strongest evidence: a faradaic efficiency above 90%, work of 40–90 kJ per mole of CO2 and less than 30% loss of capacity after 7,000 cycles [7]. Main difficulty: the tests used inlet streams of 0.6% (6,000 parts per million) to 10% CO2 [7], well above the concentration in air [11].

Learning-curve projections. Costs are projected to fall with experience, as those of solar photovoltaics and batteries have fallen [9]. Strongest evidence: Sievert and colleagues validated their component-based method ex post before applying it [9]; the most optimistic projections cited here reach €32–54 per tonne by 2050 [6]. Main difficulty: the projections assume deployment at the scale of billions of tonnes, which Herzog and colleagues call "highly uncertain" [11], and costings tied to readiness level or to components found $100 per tonne unlikely [8] or outside their 90% ranges [9].

Open sub-questions

  1. COF-999 over a sorbent's life. Does COF-999 keep its open-air capacity over thousands of cycles, and at what lifetime and price would it fall in the lower part of the committee's $100–600 range for realistic sorbent processes [5, 10]?
  2. Electro-swing at air concentration. Does a quinone electro-swing cell keep a faradaic efficiency above 90% and work within 40–90 kJ per mole when fed air at about 420 parts per million rather than 0.6–10% CO2 [7, 11]?
  3. Purity from moisture swing. Can a moisture swing resin deliver CO2 at the roughly 98% purity assumed for storage without a thermal or vacuum step that removes its energy advantage [1, 5]?
  4. Component experience rates. As plants are built, do measured component costs follow the experience rates that Sievert and colleagues assigned, and does their 90% range narrow [9]?
  5. Intermittent power. Can heat or electricity storage make DAC coupled to low capacity factor renewable electricity economical, contrary to the finding of Young and colleagues [8]?
  6. Net against captured cost. For a plant using about 1.2 MWh per tonne, how does the gap between capture cost and net-removed cost vary with the carbon intensity of its electricity [5, 11]?

Work on this in Papex

If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.

Key references

  1. Wang, T., Lackner, K. S. & Wright, A. (2011). Moisture swing sorbent for carbon dioxide capture from ambient air. Environmental Science & Technology 45(15), 6670–6675. 10.1021/es201180v · Checked 15 Oct 2026
  2. House, K. Z., Baclig, A. C., Ranjan, M., van Nierop, E. A., Wilcox, J. & Herzog, H. J. (2011). Economic and energetic analysis of capturing CO2 from ambient air. Proceedings of the National Academy of Sciences 108(51), 20428–20433. 10.1073/pnas.1012253108 · Checked 15 Oct 2026
  3. Sanz-Pérez, E. S., Murdock, C. R., Didas, S. A. & Jones, C. W. (2016). Direct capture of CO2 from ambient air. Chemical Reviews 116(19), 11840–11876. 10.1021/acs.chemrev.6b00173 · Checked 15 Oct 2026
  4. Keith, D. W., Holmes, G., St. Angelo, D. & Heidel, K. (2018). A process for capturing CO2 from the atmosphere. Joule 2(8), 1573–1594. 10.1016/j.joule.2018.05.006 · Checked 15 Oct 2026
  5. National Academies of Sciences, Engineering, and Medicine (2019). Negative Emissions Technologies and Reliable Sequestration: A Research Agenda, chapter 5, Direct Air Capture. The National Academies Press, Washington, DC. 10.17226/25259 · Checked 15 Oct 2026
  6. Fasihi, M., Efimova, O. & Breyer, C. (2019). Techno-economic assessment of CO2 direct air capture plants. Journal of Cleaner Production 224, 957–980. 10.1016/j.jclepro.2019.03.086 · Checked 15 Oct 2026
  7. Voskian, S. & Hatton, T. A. (2019). Faradaic electro-swing reactive adsorption for CO2 capture. Energy & Environmental Science 12(12), 3530–3547. 10.1039/C9EE02412C · Checked 15 Oct 2026
  8. Young, J., McQueen, N., Charalambous, C., Foteinis, S., Hawrot, O., Ojeda, M. et al. (2023). The cost of direct air capture and storage can be reduced via strategic deployment but is unlikely to fall below stated cost targets. One Earth 6(7), 899–917. 10.1016/j.oneear.2023.06.004 · Checked 15 Oct 2026
  9. Sievert, K., Schmidt, T. S. & Steffen, B. (2024). Considering technology characteristics to project future costs of direct air capture. Joule 8(4), 979–999. 10.1016/j.joule.2024.02.005 · Checked 15 Oct 2026
  10. Zhou, Z., Ma, T., Zhang, H., Chheda, S., Li, H., Wang, K. et al. (2024). Carbon dioxide capture from open air using covalent organic frameworks. Nature 635(8037), 96–101; author correction 636(8043), E10. 10.1038/s41586-024-08080-x · Checked 15 Oct 2026
  11. Stauffer, N. W. (2024). Reality check on technologies to remove carbon dioxide from the air. MIT News, 20 November 2024; an account of Herzog, H., Morris, J., Gurgel, A. & Paltsev, S. (2024), Getting real about capturing carbon from the air, One Earth 7(9), 1477–1480 (doi:10.1016/j.oneear.2024.08.011, whose text was not reachable). Source · Checked 15 Oct 2026

Page reviewed 15 Oct 2026

Source

Open, as listed by Wikipedia · Unsolved problems in chemistry. Checked 08 Oct 2026.

The status is the source’s. Papex does not decide whether a problem is solved.

Records on this problem

No public record addresses this problem yet.

More in Chemistry