Edited by: Mai Bui, Imperial College London, United Kingdom
Reviewed by: Joan Ramón Morante, Institut de Recerca de l'Energia de Catalunya, Spain; Stefano Stendardo, Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA), Italy
This article was submitted to Carbon Capture, Storage, and Utilization, a section of the journal Frontiers in Energy Research
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
In this study, the process of carbon dioxide (CO2) capture directly from ambient air in a conventional monoethanolamine (MEA) absorption process was simulated and optimized using a rate-based model in Aspen Plus. The process aimed to capture a specific amount (148.25 Nm3/h) of CO2 from the air, which was determined by a potential application aiming to produce synthetic methane from the output of a 2.7 MW electrolyser (593 Nm3/h H2). We investigated the technical performance of the process by conducting a sensitivity analysis around different parameters such as air humidity, capture rate defined as a ratio of moles of CO2 captured during the process to the total mole of CO2 in the feed stream, CO2 loading of lean and rich absorption liquids and reboiler temperature, and evaluated the energy consumption and overall cost in this system. In order to meet the design requirement for standard packed columns, the rich absorption liquid was circulated to the top of the absorber. A capture rate of 50% was selected in this process as a baseline. At higher capture rates, the required energy per ton of captured CO2 increases due to a higher steam stripping rate, required in the desorber, and at lower capture rates, the size of equipment, in particular, absorber and blowers increases due to the need for processing a significantly larger volume of air at the given CO2 production volume. At the base case scenario, a reboiler duty of 10.7 GJ/tCO2 and an electrical energy requirement of 1.4 MWh/tCO2 were obtained. The absorber diameter and height obtained were 10.4 and 4.4 m, respectively. The desorber is found to be relatively small at 0.54 m in diameter and 3.0 m in height. A wash water section installed at top of the absorber decreased the MEA loss to 0.28 kg/ton CO2. However, this increased capital cost by around 60% resulting in CO2 capture costs of $1,691 per ton CO2 for the MEA base scenario. Based on the techno-economic analysis, assuming a non-volatile absorbent rather than MEA thereby avoiding a wash water section, and using an absorption column built from cheaper materials, the estimated cost per ton of CO2 produced was reduced to $676/tCO2. The overall cost range was between $273 and $1,227 per ton of CO2 depending on different economic parameters such as electricity ($20–$200/MWh) and heat price ($2–$20/GJ), plant life (15–25 years) and capital expenditure (±30%). In order to reduce the cost further, the use of innovative cheap gas-liquid contactors that operate at lower liquid to gas ratios is crucial.
Ongoing use of fossil fuel over the past one and half century led to an increase in the concentration of CO2 in atmosphere from around 280 ppm to just above 400 ppm (Lindsey,
Among all proposed negative emission technologies (McLaren,
Several technologies are proposed and investigated in the literature for the direct capture of CO2 from ambient air. Carbon Engineering, Climeworks and Global Thermostat are the three major companies that developed technologies for the large scale capturing of CO2 from air. Carbon Engineering uses a potassium hydroxide solution to capture CO2 and the energy intensive calcination process to regenerate the solvent. On the other hand, the other two companies, Climeworks and Global Thermostat use solid sorbents to capture CO2. They use high temperature and low pressure to regenerate the sorbent. A summary of these studies is shown in
Major technologies on direct air capture.
Carbon Engineering Company | Absorption | KOH solution | Calcination process (High temperature) | Synthetic fuels | 1,000,000 |
Keith et al., |
Climeworks Company |
Adsorption | amine based porous sorbents | High temperature | Greenhouse, Synthetic fuels | 900 | Climeworks press release, |
Global Thermostat Company | Adsorption | amine based monoliths | High temperature and low pressure (vacuum) | Beverage and food | 4,000 |
Li et al., |
Arizona State University and Columbia University |
Adsorption | Anionic exchange resin | High moisture | – | Not declared | Lackner, |
Others |
Adsorption/Absorption | Amine-functionized polymer/Aqueous amino acid solutions/Ion exchange | High temperature/low pressure/precipitation/ | Fuels/longer space mission/Greenhouse | <1 | Sandalow et al., |
The use of aqueous alkaline sorbents such as NaOH, Ca(OH)2, and KOH, based on the concepts of the conventional MEA absorption process, for the DAC application, was claimed to provide a simpler contact between air and sorbents. It was also anticipated that the system can operate continuously with a very long contactor's lifetime as the absorbing liquid in such systems is not subject to degradation reactions. The high cost for the regeneration of the aqueous solutions and water loss are the main disadvantages of such liquid-based absorption systems (Lackner,
While testing CO2 capture systems at large scale is expensive, process simulation programs such as Aspen Plus have been widely used to evaluate the process configurations and identify the optimum operating conditions. There are a number of studies that used Aspen Plus to model CO2 capture from flue gases in MEA-based absorption processes (Desideri and Paolucci,
In this work, using a rate-based model in Aspen Plus, we conducted a comprehensive analysis on the performance of a conventional MEA-based absorption process for capturing CO2 from air, with particular focus on reducing the thermal and electrical energy consumption and, ultimately, the overall cost. The process performance is evaluated against the various parameters such as air humidity, the CO2 loadings of lean and rich solutions, capture rate and reboiler temperature. As a result, a benchmark condition for DAC technologies that use a chemical absorption process is determined and a parametric techno-economic assessment is conducted for this baseline case study. Further analysis of the study results has resulted in the identification of areas for efficiency improvement and cost reduction.
An MEA-based air capture process was designed to capture CO2 from air.
Representation of the MEA-based air capture process using
A conventional packed column was utilized for CO2 absorption using a 30 wt% MEA solution, which is typically used for CO2 capture from flue gas. It is important to note that the CO2 concentration in air is ~300 times lower than that in the flue gas from coal-fired power station. Hence the absorption liquid flow could be much lower than that in a flue gas capture system. However, this would give rise to an extremely low liquid to gas ratio (L/G) that could not be achieved in a traditional packed column. In terms of operability, the process design used a liquid recirculation process in which most of the rich CO2 absorption liquid exiting absorber is pumped to the top of absorber in order to achieve a standard L/G of about 2.5 in the absorber, at a 70% flooding ratio. A small portion of the rich absorption liquid is sent to the desorber for CO2 removal. The regenerated lean absorption liquid from the desorber was cooled and mixed with the absorption liquid exiting the absorber, and fed back to the top of absorber for continuous CO2 absorption. The desorption process can be modified using a cold rich-split configuration, i.e., sending a small portion of cold rich absorption liquid to the top of desorber, as shown in
A rigorous, rate-based MEA model developed in Aspen Plus was used to simulate the air capture process. The MEA model has been well-validated in previous work at CSIRO (Li et al.,
A series of simulation activities were undertaken to examine the effect of various technical variables on the reboiler duty and equipment size to determine the optimized conditions for a standard MEA-based air capture process. In this optimization, the aim is to minimize the reboiler duty at any given conditions, for example different capture rates etc. The generic simulation conditions used for the CO2 absorption and desorption are provided in
Conditions for the MEA-based CO2 capture process simulation.
CO2 concentration in inlet air, ppm | 400 |
Inlet air temperature, °C | 25 |
CO2 capture rate, ton/h | 0.291 |
MEA concentration, wt% | 30 |
L/G for absorber with circulation, ton/ton | 2.54 |
Packing materials in absorber and desorber | Mellapak 250X |
Flooding capacity of absorber/desorber (%) | 70/65 |
Desorber pressure (bar) | 2 |
Temperature approach of cross heat exchanger, K | 10 |
Capture rate, % | 20, 35, 50, 70, 90 |
Rich loading, mol/mol | 0.27, 0.30, 0.33, 0.36 |
Lean loading, mol/mol | 0.15, 0.20, 0.25 |
Air relative humidity, % | 30, 50, 70, 90 |
Once the optimized conditions were determined using the standard absorbent regeneration process, the technical performance was further optimized using the rich-split configuration and the addition of a wash water column (
With the determination of technical parameters and equipment size of the air capture process, the Aspen Capital Cost Estimator (ACCE) was utilized to estimate the direct costs for the equipment, materials and labor for construction of the air capture plant. The other costings associated with the supportive materials and labor, facilities, engineering, contractors and contingencies for plant construction were estimated based on the direct costs using multiplicative factors. These multiplicative factors were derived from a 2007 report from the Department of Energy, USA (DOE/NETL,
In this section, the effects of various parameters on the air capture process are reported. The heat of regeneration and the size of absorber are mainly discussed as they are often the two major costs in CO2 capture process. It should be mentioned again that this process aims to capture a specific amount of CO2 from air (around 0.291 t/h).
Here it is assumed that a constant amount of CO2 was produced from ambient air using the MEA-based absorption process as the capture rate of the process varied from 20 to 90%. The simulation results revealed that at a capture rate of 90% the reboiler duty was 21.9 GJ/tCO2, almost double the one required in a process with a capture rate of 20% (
The variation of
An air capture system operating at a low capture rate can also result in an extra number of energy and cost components, such as much higher MEA loss, a more significant circulation rate of rich absorption liquid to the absorber, and a significantly high energy requirement for air movement through the absorption column.
The effect of the CO2 loading of lean and rich absorption liquids on the reboiler duty and the size of the absorber were examined at a 50% capture rate.
The effects of CO2 loading of rich absorption liquid on the performance of the process were also investigated. In this set of conditions, a lean loading of 0.2 and a humidity of 100% were used. As evident in
The effect of humidity of inlet air on the air capture process was also evaluated. This is important as the very large volumes of air that pass through the absorber may lead to high rates of water vaporization, depending on humidity of the inlet air. The results indicate that the reboiler duty decreased from 14.4 GJ/tCO2 to around 9.9 GJ/tCO2 when the relative humidity of air increased from 30 to 90% (
The humidity of the inlet air had no effect on the height and diameter of the absorber.
Next, the effect of reboiler temperature on the performance of the air capture process was examined. A lean loading of 0.2, a rich loading of 0.35 and a capture rate of 50% were considered for this case.
Reboiler duty at different reboiler temperature.
According to the results obtained from the sensitivity analysis shown above, a benchmark condition was selected for the capture of CO2 from air using the MEA absorption process. This point was based on a capture efficiency of 50%, a lean loading and a rich loading of 0.2 and 0.35, respectively, and a reboiler temperature of 123°C. The rich split configuration was used in this case study due to its superior energy performance. A detailed analysis was conducted for this base case scenario, which included determination of thermal and electrical energy requirements in this system. The total electricity consumption for air blowers and liquid pumps was calculated to be 1.452 MWh/tCO2, and the reboiler duty was calculated to be 10.7 GJ/tCO2. The absorber height and diameter were calculated to be 6.3 and 10.4 m, respectively. A wash water section, 10.4 m in diameter and 5 m high, was also added to decrease the evaporative loss of MEA to around 0.28 kg/tCO2 (0.04 ppm in exhaust air) during the absorption process. The desorber dimensions are significantly smaller than the absorber dimensions which reflects the large difference in gas and liquid flow rates. The final simulation results for this base case scenario is indicated in
The operating conditions and process simulation results for a base case scenario in the air capture process.
Inlet air temperature (C) | 25 |
Air relative humidity (%) | 100 |
CO2 production (t/h) | 0.291 |
CO2 concentration in inlet air (ppm) | 400 |
CO2 capture efficiency (%) | 50 |
Flow rate of air (t/h) | 943 |
MEA concentration (%) | 30 |
Flooding capacity of absorber/desorber (%) | 70/65 |
L/G in absorber | 2.54 |
Type of packing in the absorber and desorber | M250X |
CO2 rich loading | 0.35 |
CO2 lean loading | 0.2 |
Desorber pressure (bar) | 2 |
Temperature approach on cold side (K) | 10 |
Absorption liquid flow to desorber (t/h) | 9.113 |
Absorber recirculation liquid flow (t/h) | 2396.5 |
Specific liquid flow to desorber (t/tCO2) | 31.3 |
Specific absorber recirculation flow (t/tCO2) | 8247.3 |
Condenser temperature (°C) | 31.2 |
Reboiler temperature (°C) | 123.1 |
Reboiler duty (GJ/tCO2) | 10.70 |
Total electricity (MWh/tCO2) | 1.452 |
MEA evaporation in absorber (kg/tCO2) | 53.4 |
Wash water flow (kg/s) | 612 |
MEA loss after water wash (kg/tCO2) | 0.28 |
Absorber packing diameter (m) | 10.36 |
Absorber packing height (m) | 4.43 |
Absorber volume (m3) | 373.2 |
Absorber pressure drop (kPa) | 0.73 |
Washing column pressure drop (kPa) | 0.26 |
Washing column packing diameter (m) | 10.36 |
Washing column packing height (m) | 5 |
Cross heat exchanger surface area (m2) | 102.59 |
Desorber diameter (m) | 0.544 |
Desorber packing height (m) | 3 |
Desorber (m3) | 0.70 |
Recirculation pump (MWh/tCO2) | 0.504 |
Washing pump work (MWh/tCO2) | 0.238 |
Blower work (MWh/tCO2) | 0.703 |
Pumps to/from desorber (MWh/tCO2) | 0.007 |
Total (MWh/tCO2) | 1.452 |
Using the Aspen Capital Cost Estimator with the base case design specified in
Economic performance of MEA-based air capture. All the costs are on a basis of 1st Qtr 2016 US$.
Washing column | 4.38 |
Absorber | 4.22 |
Desorber | 0.13 |
Blowers and fans | 1.66 |
Heat exchangers | 0.39 |
Pumps | 0.30 |
Tanks | 0.40 |
Other equipment | 0.22 |
Total direct costs | 11.70 |
Total indirect costs | 2.27 |
Engineering | 1.40 |
Contractor fees | 0.42 |
Contingencies | 3.49 |
Total plant costs | 19.27 |
Spare parts | 0.096 |
Total investment costs | 19.37 |
Annual O&M costs | 0.757 |
Annual heat costs | 0.213 |
Annual electricity costs | 0.286 |
Capital | 1,033 |
O&M | 396 |
Heat | 111 |
Electricity | 150 |
Total | 1,691 |
The total capital costs of the MEA-based air capture process were estimated to be $19.37 million for our benchmark MEA-based air capture process. The washing column ($4.38 million) and absorber ($4.22 million) constituted the largest and the second largest capital item among the total capital investment and accounted for 74% of total capital costs, as shown in
By comparison, the annual costs of O&M and energy, i.e., electricity for blowers and pumps and heat for absorbent regeneration, are lower than the capital costs. The electricity costs ($0.286 million/year) were higher than the heating costs ($0.213 million/year) owing to the large electricity consumption for air transfer through the columns and solution pumping and the high circulation for CO2-absorption and washing. It should be noted here that, in the base case, if less humid air than 100% was used, the added cost due to water evaporation in this process would be up to $4/tCO2, which is still negligible in comparison with other cost components shown in
For the base case, a plant life of 20 years, a discount rate of 8%, a capacity factor of 90%, heat price of $10/GJ and electricity price of $100/MWh were considered. A sensitivity analysis shows that with variation of equipment costs in the range of 30%, the CO2 capture cost varied between $1,262 and $2,120/ton CO2, a 34% decrease/increase with the 30% decrease/increase in capital costs. This again suggests that decreasing the process equipment costs would significantly reduce the cost of CO2 capture from air. This can be achieved by three aspects: (1) enhancing mass transfer between absorbent and CO2 through more efficient gas-liquid contacting in order to reduce the equipment size of CO2 absorption, (2) absorbents that have lower evaporation of absorbent (water and amine) to the air in order to minimize the size of equipment for emission control, and (3) seeking cheaper materials for equipment manufacture in order to reduce the equipment costs. In the following section, these elements for reducing the cost of the air capture process will be explored qualitatively.
As described, the wash water section and absorber constitute 45% of total cost in the process, hence using an alternative absorption liquid with low vapor pressure and replacing the packings and absorber materials with cheaper materials can drastically reduce the total cost of the process.
– Use of alternative absorption liquid
Owing to the contact between the large volume of air and the small amount of liquid in the air capture process, MEA is evaporated during the absorption process. As discussed in section Benchmark Conditions for the Capture Process, this necessitates the use of a large wash water section after the CO2 absorber, and hence creates an additional cost to the process. As shown in
Amino acid salts solutions have been proven to have a potential for absorbing CO2 with an effectiveness that matches that of MEA (Aronu et al.,
– Use of cheap plastic packings
Based on the techno-economic analysis in APS report (Socolow et al.,
– Use of cheap materials in absorber structure
The cost of the absorber structure was reported to be around $2,300 and $3,700 per inlet area of absorbers (m2) for cooling tower and contactors used in Carbon Engineering's process (Holmes and Keith,
The details of techno economic result based on the new design are given in
Economic performance of an improved amine-based air capture process.
Absorber | 0.70 |
Desorber | 0.13 |
Blowers and fans | 1.66 |
Heat exchangers | 0.39 |
Pumps | 0.23 |
Tanks | 0.40 |
Other equipment | 0.07 |
Total direct costs | 3.59 |
Total indirect costs | 0.70 |
Engineering | 0.43 |
Contractor fees | 0.13 |
Contingencies | 1.07 |
Total plant costs | 5.91 |
Spare parts | 0.03 |
Total investment costs | 5.94 |
Operating expenses | |
Annual O&M costs | 0.233 |
Annual heat costs | 0.213 |
Annual electricity costs | 0.241 |
Capital | 317 |
O&M | 122 |
Heat | 111 |
Electricity | 126 |
CO2 capture costs | 676 |
The parameters used for the base case economic analysis are: 20 year plant life, 8% discount rate, 90% availability factor, $10/GJ heat cost and $100/MWh electricity cost. The effects of various economic parameters, such as plant life, heat and electricity unit cost, discount rate and equipment costs for the air capture process, on the economic performance were analyzed, with the results summarized in
Effect of economic variables on the CO2 capture cost. The parameters used for the base case are: 20-year plant life, 8% discount rate, 90% capacity factor, $10/GJ heat price and $100/MWh electricity price.
A wide range of electricity prices from $20 to $200/MWh and heat prices from $2 to $20/GJ were considered in this study, providing various possibilities of electricity and heat sources applicable to the air capture process. The CO2 capture cost varied by 39% from $576 to $802/ton CO2 and 34% from $587 to $788/ton CO2, as the prices of electricity ($20-$200/MWh) and heat ($2–$20/GJ) varied by a factor of 10, respectively. The use of zero-carbon energy, e.g., renewables or nuclear, in the air capture process is most likely required to avoid any additional CO2 emission to air due to the energy consumption for the capture process.
Based on a sensitivity analysis on all variables, considering the most desirable and the most undesirable values for various economic parameters, the overall cost of capturing a ton of CO2 ranges from $273 to $1227.
There are only a few studies that reported the required energy and cost for the capture of CO2 from air. Using different technologies such as absorption and adsorption, the required thermal energy for the direct capture of CO2 from air was reported to be almost 4–8 times larger than the required electrical energy. The cost of capturing one ton of CO2 was reported to be between $100–1,000 (Keith et al.,
For the baseline case chosen in this study, the thermal energy required is 10.7 GJ/t CO2 which is almost 3 times greater than the required electricity (~1.03 MWh/tCO2). These are significantly larger than those reported in other studies. This is mainly due to different characteristics of this technology compared to others, for example the large recirculation of rich solvent is required here to prevent loading/flooding in the absorption column (it constitutes around 50% of total electricity requirement).
As previously indicated in
The design of the Carbon Engineering gas-liquid contactor reduces the energy requirement and capital costs of blowers and fans. They use a horizontally oriented crossflow cooling tower, resulting in a low pressure drop, while in our study, we considered a vertically counter-current flow packed tower, significantly increasing the energy and capital costs of air movement in our system.
Based on the cost analysis described above, the other areas that can be improved in order to further reduce the capital and operating cost in the air capture system are:
If new air-liquid contactor is designed to optimize the contact between the large volume of air and small amount of liquid, this will eliminate the necessity for the large circulation rate/pumping of absorption liquids which consumes 0.54 MWh/tCO2 electricity (around 50% of total electricity requirement) and hence reduce the capital and operating cost. In this case, the electricity cost reported in
Development of new liquid absorbents and applying process modifications. If the reboiler temperature can be increased to 150°C, the absorbent heat requirement will be reduced by 60% to around 4 GJ/tCO2. The cost for heat reported in
The technical and economic aspects of a conventional MEA-based absorption process for capturing CO2 directly from air was simulated using a rate-base model in Aspen Plus. A benchmark condition was defined and further explored through a sensitivity analysis involving different parameters. It is concluded that:
– In an air capture system with a low capture rate, the heat required for the regeneration of absorption liquid per ton of captured CO2 is significantly lower. However, a large volume of air needs to be processed and this results in a larger unit and hence a higher capital cost.
– Owing to the very low concentration of CO2 in air, the amount of liquid required for the contact with air would be extremely low. To meet the operational requirement of a conventional packed column, a large liquid circulation around the absorber is required, resulting in a large electricity requirement.
– At the base case scenario, a capture rate of 50% was selected. The reboiler duty was 10.7 GJ/tCO2 and the electrical energy requirement was 1.4 MWh/tCO2. The absorber diameter and height were 10.4 and 4.4 m, respectively and the wash water section was 5 m in height.
– Owing to the large volume of air used in the air capture process, water and MEA losses due to evaporation were quite high. The addition of a wash water section to the absorber can reduce the MEA evaporation significantly down to 0.28 kg/ton of CO2 (0.04 ppm), at the expense of increased capital costs.
– Using the standard chemical engineering design which may not necessarily required for the air capture process, the total estimated cost for this process was around $1,690/ton CO2. The capital cost and operating cost were $1031/ton CO2 and $659/ton CO2, respectively.
– The wash water section and absorber accounted for around 74% of total capital cost of this process. Using an alternative absorption liquid with negligible vapor pressure which allows for the removal of the wash water section, the capital cost will be reduced by $393 per ton of CO2.
– The replacement of stainless steel packing and absorption column materials by cheaper materials like those used in cooling towers reduced the capital cost by another $323 per ton of CO2.
– The techno-economic analysis of the air capture process designed based on the cooling tower technology and non-volatile liquid absorbent showed that the cost for CO2 capture from air was, in the range of US$273 to US$1,227. Around 45% and 55% of this were attributed to the capital cost and operating cost, respectively.
In summary, this study provides an economic baseline for the air capture technologies that aim to use liquid-based capture process. Even though the energy requirement and cost of the baseline for such processes are still high with respect to the current CO2-price, even for the CO2-price in the framework of the Californian Low Carbon Fuel Standard (Low Carbon Fuel Standard,
The datasets for this article are not publicly available because the data ownership and control must stay with CSIRO for propriety reasons. Requests to access the datasets should be directed to Paul Feron (
AK drafted the manuscript and analyzed and integrated the research results. KJ performed the process modeling and did the techno-economic analysis. PF conceived and designed the research work and analyzed research results. All authors contributed to the review and editing of the manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This activity received funding from ARENA (
The Supplementary Material for this article can be found online at: