Negative Emissions Technologies (NETS): Feasibility Study - Technical Appendices
Technical annex for study that estimates the maximum Negative Emissions Technologies (NETs) potential achievable in Scotland, 2030 - 2050.
Appendix 6. DACCS
Liquid Solvent DACCS
Configuration
Carbon Engineering’s process is split into 4 key stages[23], [52]: 1.) the air contactor unit, where ambient CO2 absorbs into the hydroxide KOH solvent to form a K2CO3 salt; 2.) the pellet reactor, where the carbonate salt reacts with another solvent (Ca(OH)2) in a fluidised bed to form CaCO3, which in turn regenerates the KOH solvent; 3.) the calciner, where CaCO3 is heated to 900degC and thermally decomposes to release the captured CO2, which is subsequently compressed to 150 bar ready for storage; 4.) the slaker, where the Ca(OH)2 solvent is regenerated by hydrating the CaO exiting the calciner. This cyclic configuration enables continued operation. The process heat and power demands are met through onsite natural gas combustion within a NGCC unit, where the CO2 emissions are captured.
There are alternative process configurations where the NGCC unit has been replaced with grid electricity[23], with the most ambitious being a fully electrified system which meets all heat and power demands. This comes at the expense of a high electricity penalty (1,535 kWhel/tCO2)[52], but at the benefit of downsizing several process units.
Loading capacity
The CO2 loading capacity on the solvent must be limited to a maximum concentration of 30%, due to the corrosive nature of hydroxide bases. This in turn reduces the CO2 flux potential for absorption. This is reflected by the fact that the CO2 flux of a strong NaOH base is ~0.52 molCO2.min-1.m-3, which is an order of magnitude lower than compared to solid amine adsorbents. In order to maximise this loading capacity, Brentwood XF12560 packing is used to increase surface contact area[51].
Temperature
The sorbent regeneration temperature is considerably high (at 900degC), which can at present only be achieved through natural gas combustion[23]. However, there is scope to investigate alternative low carbon fuels via the Dreamcatcher project[4], [5].
Energy requirement
In the baseline case, where all energy demands are met through the NGCC unit, 2.45 MWh/tCO2 of natural gas is required. If instead electrical demands are met through the grid, then natural gas usage drops to 1.46 MWh/tCO2 coupled with heat recovery and 366 kWh/tCO2 of electricity[23]. The majority of this power demand (70%) is allocated to the CO2 compressors[48]. There is also the instance of having a fully electrified system, which requires 1,535 kWh/tCO2 of electricity[52].
Modularity
Compared to solid absorbent DACCS, the modularity of Carbon Engineering’s capture units is poor, with capture capacities having to be greater than 10 ktCO2/year[51]. Therefore, it is preferential to deploy liquid solvent DACCS at larger scales.
New research
The reaction kinetics of the solvent are being improved through biomimetic catalysts (e.g., using carbonic anhydrase, which hydrate and dehydrate CO2 orders of magnitude faster than amines or water)[51].
Solid Adsorbent DACCS
Configuration
When regenerating the sorbent bed to release CO2, either Temperature Swing Adsorption (TSA), Pressure Swing Adsorption (PSA), or Vacuum Swing Adsorption (VSA) is used. The benefit of TSA is that it exhibits lower steam requirements (0.2–0.4 kg steam/kgCO2) and is cheap, whilst PSA experiences quicker CO2 adsorption/desorption times, but at the expense of higher costs and safety[51]. VSA provides similar benefits to PSA and is safer; however, it is the most expensive option[51].
Climework’s pilot plant is described in Fasihi et al’s work[52]. Ambient air is drawn into modular contactor units using fans, where CO2 and moisture adsorb onto the solid surface of a special cellulose fibre that is supported by amines; thus enabling CO2 capture and providing sufficient water for onsite use (0.8–2 t/tCO2). The remaining air in the adsorbent bed is then purged by reducing pressures or inserting steam into the system. The CO2 is then released by heating to temperatures of 80-120degC, compressed, and is ready for storage or utilisation. Finally, the sorbent bed is cooled to ambient conditions before re-use. A whole adsorption/desorption cycle takes between 4-6hrs[52]. A key benefit of this process is its ability to use a wide range of sorbents and low temperature heat sources, such as waste heat, geothermal energy, and heat pumps. However, there are also drawbacks associated with building the large surface area structures that exhibit low-pressure drops, as they exhibit high capital costs[48].
Global Thermostat follows a similar process, with the key differences being a lower regeneration temperature requirement (85–95 degC), the adsorbent being an amino-polymer, and the full cycle time being shorter (~100s)[52]. The heat and electricity requirements are also lower (please see Table 14).
Loading capacity
Compared to liquid hydroxide solvents, the use of solid adsorbents enables much higher CO2 loading by weight (3.53 mol CO2 min-1.m-3). Therefore, the flux of CO2 adsorption is less limited[51]. A hierarchical adsorbent pore structure of micro and mesopores is used to maximise adsorption[51].
Temperature
The temperatures required to regenerate the sorbent (80-120degC) are significantly lower than compared to liquid DACCS[51]. This is due to adsorbent bonds formed between CO2 and the sorbent surface being weaker than the chemical bonds formed during absorption.
Energy requirement
For the Climeworks process, the electrical demands are lower than that of Carbon Engineering (200–300 kWh/tCO2), to supply the fan and control systems. However, heat demands are greater (1.5 -2.0 MWh/tCO2)[52]. Despite this, the temperature of the heat is much lower and hence is easier to source.
Modularity
Climeworks’ technology is provided in small modular units, with the maximum potential of one unit being 50 tCO2/year. This modularity helps the technology be manufactured and deployed at scale[51].
New research
There’s bountiful research investigating alternative adsorbents that show potential in reducing thermal energy demands. This includes metal–organic frameworks (MOFs), zeolites, activated carbons, etc. In terms of design, a moving bed adsorber is being considered over the more traditional fixed bed, which helps reduce pressure drops and cycle times[51].
Other DACCS
Cryogenic DACCS
Takes advantage of the sublimation point of CO2. The CO2 extracted from the air is converted into a solid or sublimated to produce a high purity gas stream[51].
Moisture/humidity swing adsorption (MSA)
MSA uses anionic exchange resins to capture and evolve CO2. These sorbents will bind to CO2 in arid conditions and evolve CO2 when contacted with water, which has the potential to decrease energy requirements but at the expense of increased water consumption[51]. After CO2 is removed, the system is heated to 45 degC to dry the resin sheets. The electrical energy demands range from 316-326 kWhel/tCO2, depending on whether a fan is used to draw in ambient air within the contactor[52].
Electro-swing adsorption
In this process CO2 binds to a polyanthraquinone-carbon nanotube composite upon charging and is released upon discharge, eliminating the need for thermal energy, and producing a high purity CO2 stream[51].
Molecular filters
Nano sized molecular filters are used to capture CO2 from the air powered by solar energy. The technology is expected to only require 333 kWhel/tCO2 of electricity, where pure CO2 is delivered at 100 bar, at a cost of 14 €/tCO2[52].
Alternative feedstocks
Manufactured alkaline feedstocks (e.g., MgO) and aqueous amino acids could absorb CO2, where CO2 is regenerated by crystallization of an insoluble carbonate salt with a guanidine compound[51].
DAC pilot projects
Liquid solvent DAC
Carbon Engineering is the only company active in liquid solvent-based DAC, who use an aqueous KOH solvent to capture carbon. At present, they have a demonstration and pilot plant capturing a combined 1,365 tCO2/year, which costs 132-191 £/t CO2[227]*. The company’s goal is to establish broad commercial deployment of synthetic fuels produced from captured carbon and green hydrogen.
Solid sorbent DAC
Climeworks is the most well-known solid sorbent-based DAC company, who use an adsorbent made of special cellulose fibre supported by amines to capture carbon[52]. Their first demonstration project was in 2014, in collaboration with Audi and Sunfire, where ambient CO2 was captured and converted to synthetic diesel[52]. Since then, an additional 14 projects have been undertaken, most of which utilise the captured carbon to either produce e-fuels, aid plant growth in horticultural sites, or carbonate beverages. Their first CO2 storage project was a pilot plant located in Iceland, capturing 50tCO2/year in 2017. This site was scaled up in capacity in 2021, to provide 4,000 tCO2/year of negative emissions at a cost of 411-494 £/tCO2[51]*. This latest project is named Orca and is the largest operating DACCS facility in the world.
Global Thermostat uses an amino-polymer adsorbent to capture CO2[52]. The company has pilot and commercial demonstration plants operating since 2010 in the United States, which provide a combined capture capacity of 1500 tCO2/year.
Smaller scale DAC companies include Antecy, a Netherlands based company who have developed a detailed DAC design in collaboration with Shell that utilises a K2CO3 adsorbent, and Hydrocell Ltd, a Finnish company that has built a 1.387 tCO2/year pilot unit that exhibits low regeneration temperatures (70–80 degC) and is a NETs producer of water (1.9 t/tCO2)[52].
Moisture Swing Adsorption (MSA)
MSA DAC is less developed, and only has two small-scale companies developing the technology: Skytree (founded in the Netherlands) and Infinitree (founded in the United States). Skytree are building upon electrostatic absorption and moisturising desorption, and Infinitree utilise an ion exchange sorbent. Early markets for both companies are urban farming projects, for which captured CO2 is used to assist in plant growth[52].
*Values converted from USD to GBP using conversion of 1 USD = 0.83 GBP.
Contact
Email: NETs@gov.scot
There is a problem
Thanks for your feedback