Fig. 1: Simplified operating procedure for DAC. (Image Source: C.Cranmer, following Leonzio et al. [5]) |
Direct Air Capture (DAC) is a technology designed to extract carbon dioxide (CO2) directly from the ambient air. The process generally involves three main stages:
Air Intake: Large fans draw atmospheric air into the DAC system. [1]
CO2 Capture: The air passes through a medium that selectively captures CO2. There are two primary approaches:
Solid DAC (S-DAC): Uses solid sorbent filters that bind with CO2. When saturated, the filters are heated to release concentrated CO2.
Liquid DAC (L-DAC): Employs a chemical solution, often potassium hydroxide, to absorb CO2 from the air.
CO2 Separation and Purification: The captured CO2 is separated from the capture medium, purified, and compressed for further use or storage.
S-DAC systems, like those used by Climeworks, function by drawing air into a collector containing a specialized filter using a fan. This filter captures CO2 molecules from the air, and once the filter becomes saturated, the collector is sealed and heated to approximately 100C. This heating process releases the concentrated CO2, which is then collected for further use or storage. The system exemplifies a straightforward approach to direct air capture using solid materials to trap and release carbon dioxide.
L-DAC systems, employed by companies such as Carbon Engineering, take a different approach by using a liquid-based process. In these systems, air is brought into contact with a potassium hydroxide solution within an air contactor. The solution reacts with CO2 to form a carbonate salt, effectively binding the carbon dioxide. The CO2 is then extracted, purified, and compressed through a series of chemical steps. This method highlights a more chemically intensive process compared to S-DAC, involving the transformation and separation of carbon dioxide from a liquid medium.
The captured CO2 can in theory then be permanently stored underground or used in various applications, such as the production of synthetic fuels or in industrial processes. It's worth noting that DAC faces challenges due to the low concentration of CO2 in ambient air (about 0.04% or 400 ppm), which necessitates processing large volumes of air and contributes to high energy demands.
First, we can assess the theoretical minimum energy usage for a simplified DAC system using the principle of entropy.
To simplify, when something is "scrambled" - as in, CO2 molecules mixing with air molecules - energy is required to unscramble them. This "scrambling" is measured by a physical quantity called the entropy S, and entropy can never be destroyed. It can only be dumped as waste heat Q = T × S, where T is the Kelvin temperature of the heat dump. Since energy is conserved, this requires input of "work" equal to W = Q, measured in Joules.
The core physics challenge for DAC is that the entropy of mixing is very large. If I have N1 molecules of gas 1 and N2 molecules of gas 2 in a container thoroughly mixed, the entropy of mixing is:
S | = | - N1 kB ln( | N1 N1 + N2 |
) - N2 kB ln( | N2 N1 + N2 |
) |
where kB is Boltzmann's constant. A mole is NA = 6.022 × 1023 molecules. Therefore, if I have ν1 moles of gas 1 and ν2 moles of gas 2 the above entropy of mixing formula becomes
S | = | - ν1 R ln( | ν1 ν1 + ν2 |
) - ν2 R ln( | ν2 ν1 + ν2 |
) |
where R = 8.314 J °K-1 is the universal gas constant.
We can now solve this formula. The molar mass of CO2 is
The fraction of the molecules in the atmosphere that are CO2 is 427 ppm or
f | = | ν1 ν1 + ν2 |
= | 427 × 10-6 = 4.27 × 10-4 |
The mixing entropy per unit mass of CO2 in the air is then
S Mν ν1 |
= | - ( | R Mν |
) × [ ln(f) + ( | 1-f f |
) ln(1-f) ] |
≃ | - ( | R Mν |
) × [ ln(f) - (1 - f) ] |
The theoretical minimum amount of energy required to extract CO2 at temperature T = 300°K and deliver it as gas a 1 atmosphere of pressure is
E MCO2 |
= |
|
|||||||
= | - | 8.314 J mole-1 °K-1
× 300°K 0.044 kg mole-1 |
× [ ln(4.17 × 10-4) - (1 - 4.27 × 10-4 ) ] | ||||||
= | 4.97 × 105 J kg-1 |
or 0.5 GJ per tCO2. If you're delivering CO2 under pressure, say at 100 atmospheres, then there's additional compression energy of
Δ E MCO2 |
= | - ( | R T Mν |
) × ln(100) |
= | 2.61 × 105 J kg-1 |
or 0.26 GJ per ton of CO2. In summary, the theoretical minimum energy consumption for DAC is 0.5-0.76GJ per ton of CO2 (or approximately 140-210kWh per ton of CO2).
We can then look at this from a commercial perspective, to understand how this energy consumption occurs in practice. Energy usage for DAC systems is primarily divided into two main components:
Air Intake: Continuous fan operation is required to draw atmospheric air into the system. This process accounts for 20-40% of total energy use.
Sorbent Regeneration: This is the most energy-intensive step in the DAC process. It requires temperatures of 900°C for liquid solvents and 80-120°C for solid sorbents. [2]
Current DAC systems are characterized by significant energy demands, with total energy consumption typically ranging from 5.4-10.8 GJ (1500-3000 kWh) per ton of CO2 captured. Energy requirements vary significantly based on the technology used and operating conditions at the DAC site. There is also a tradeoff between upfront and backend energy requirements: certain materials allow for lower energy use during capture but result in a lower-purity CO2 output stream that then requires higher energy use for compression and injection. [3]
As a reference point, the average emissions per kWh of electricity produced in the US are 0.4kg CO2e. [4] If DAC was run off the average grid, if would generate 0.6-1.2tCO2e per ton of CO2 captured. This would obviously be highly inefficient, and so it is critical for DAC to source power solely from clean sources.
Additionally, there are questions over whether DAC is the most efficient use of energy. For example, Bioenergy with Carbon Capture and Storage (BECCS) can actually produce energy while capturing CO2, while Enhanced Rock Weathering requires very little energy (note that scalability is still in question for all of these technologies). If we believe that energy is a scarce resource - and this is likely, given the queue to connect new projects to the grid - then it may be the case that this energy would be better directed to other decarbonization projects.
The high energy demand of DAC systems poses a significant challenge to their widespread implementation and economic viability. Researchers are actively working on several fronts to reduce these energy requirements:
Improved fan efficiency and alternative airflow designs could potentially reduce fan energy consumption - one study suggests this approach could theoretically reduce energy consumption to 230kWh per ton CO2 (note: this is not proven and is therefore not reflected in most expert forecasts). [5]
Development of more efficient sorbent materials that require less energy for regeneration.
Many companies are aiming to integrate DAC with renewable energy sources to minimize its carbon footprint. However, renewable sources present challenges with intermittency, meaning that the DAC plant could not run at a high capacity factor. From a cost perspective, DAC needs to run near-constantly, so intermittent renewables would need to be combined with (often expensive) storage solutions or source power from sources like geothermal or nuclear.
$100/tCO2 is the often-stated cost target for DAC. A 2021 study considering the learning curves and modularity of DAC technology found the following: [6]
At the Gt scale (i.e., at full scale), costs estimates range from $100/tCO2 to $1,350t/CO2
At the Mt scale (along the way), costs estimates range from $250/tCO2 to $1,500t/CO2
Notably, there is significant variation based on the DAC technology considered. Electricity costs also play a major role, and this study only considered cases where DAC was paired with geothermal or nuclear energy.
© Caitlin Cranmer. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
[1] M. Zanatta, "Materials For Direct Air Capture and Integrated CO2 Conversion: Advancement, Challenges, and Prospects," ACS Mater. Au 3, 576 (2023).
[2] M. Ozkan et al., "Current Status and Pillars of Direct Air Capture Technologies," iScience 25, 103990 (2022).
[3] J. K. Soeherman, A. J. Jones, and P. J. Dauenhauer, "Overcoming the Entropy Penalty of Direct Air Capture for Efficient Gigatonne Removal of Carbon Dioxide,". ACS Eng. Au. 32, 114 (2023).
[4] "Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2019," U.S. Environmental Protection Agency, EPA-430-R-21-005, April 2021.
[5] G Leonzio, P. S. Fennell and N Shah, "A Comparative Study of Different Sorbents in the Context of Direct Air Capture (DAC): Evaluation of Key Performance Indicators and Comparisons," Appl. Sci. 12, 2618 (2022).
[6] J. Young et al., "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, 899 (2023).