Reducing the concentration of carbon dioxide (CO2) in the atmosphere to combat climate change is a global challenge. Direct air capture (DAC) is a new set of technologies that directly removes CO2 from the air; therefore, DAC can address emissions from any source. This paper begins by reviewing the literature on negative emission technologies to summarize the most recent technological developments. We then conduct a life cycle assessment on one of the most recently developed technologies, the direct air capture and utilization (DAC-U) system. DAC-U systems, like photovoltaic systems, can be installed in various locations, including homes, offices, and mega-plants, resulting in a highly scalable, on-site system that can be deployed in various ways. Based on the life cycle assessment results, this article presents the CO2 capture and reduction potential of the DAC-U system, with a focus on installations in households, and examines the willingness to adopt the system in Japan. Our results show that the DAC-U system functions as a negative emission technology and demonstrate the large CO2 capture and reduction potential of the DAC-U system through household-level installations.
Research Article
Direct Air Capture and Utilization: Life Cycle Assessment and CO2 Reduction Potential Estimation
https://doi.org/10.21203/rs.3.rs-3892596/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Negative emission technology
direct air capture and utilization
life cycle assessment
CO2 reduction potential
willingness to adopt
As many nations strive for the ambitious goal of carbon neutrality, academics and energy policymakers are increasingly focusing on the combination of technologies required to achieve this goal (LV et al. 2022; Zhang and Chen 2022). Concerns include cost issues, carbon reduction capacity, and individual technology’s deployment potential. Moreover, there is widespread agreement that renewable energy can play a significant role in achieving carbon-neutral goals.
Although renewable energy can offset carbon emissions by assuming that the role of electricity generation from fossil fuels and carbon capture and storage (CCS) can also play a role in sequestering these emissions, carbon neutrality is difficult to achieve with this suite of technologies alone due to carbon emissions from the production of renewable energy technologies and the energy expended in capturing carbon dioxide (CO2) in the case of CCS. Negative emission technologies (NETs) are critical to achieving a net-zero emissions energy system because they account for emissions from difficult-to-control industries like iron and cement production (Ozkan, Akhavi et al. 2022). Converting captured CO2 to useful fuels, such as methanol, is one way to reduce carbon in the atmosphere, potentially reaching ambitious Paris Agreement goals and achieving national-level carbon neutrality goals (Li et al. 2022).
In addition to using carbon captured from fossil fuel energy plants for fuel production, the concept of direct air capture (DAC) has recently gained attention as a tool to reduce atmospheric CO2. In particular, the DAC and utilization (DAC-U) concept can be realized using this captured CO2 to generate fuel. DAC approaches currently in use include solid and liquid sorbents, ion exchange materials, and direct capture from air, which are frequently combined with renewable energy or waste heat and used for renewable fuel generation, enhanced oil recovery, synthetic gasoline, and agricultural and building processes (Ozkan, Nayak et al. 2022). This study focuses on a novel approach to DAC-U: directly capturing and concentrating CO2 from the air via membranes before methane generation via thermochemical conversion processes.
Although there are several approaches to achieving carbon neutrality, including combining renewables with innovations such as DAC-U, these approaches and scenarios must be critically evaluated in terms of their life cycle emissions when production and operation are considered. In addition to a comprehensive assessment of the carbon reduction capabilities of these technologies, societal perspectives on their acceptability and likelihood of adoption must be considered at multiple scales.
This paper aims to evaluate the environmental and social implications of the proposed membrane and thermochemical conversion DAC-U system. Regarding the environment, the global warming potential (GWP) of DAC-U combined with the deployment of renewable energy is assessed using life cycle analysis (LCA). Furthermore, from a social point of view, the trends in energy system participation are applied to the DAC-U, considering government statistics and stakeholder input from Japanese households.
The rest of this paper is organized as follows. Section 2 provides a literature review on NETs to clarify the standing point of the DAC-U system. Section 3 discusses LCA methodologies, as well as the willingness to adopt LCA and its effectiveness. Section 4 discusses the results Section 5 summarizes the discussion and conclusions.
Negative emissions refer to the reduction of CO2, not only in comparison to alternative approaches or technologies but also in terms of a device’s lifetime emissions falling to a level below net zero. Negative emissions can be achieved by using natural resources to exploit them, developing through the development of advanced technologies, and using hybrid approaches, as described below.
Natural NET refers to maintaining, restoring, and managing ecosystems to reduce CO2 in the atmosphere and increase CO2 absorption. Figure 1 shows a classification of NETs.
The first layer is classified as natural, technological, and hybrid NETs. Natural NET is distinguished into three categories: reforestation and improved forest management, wetland and coastal restoration, and soil carbon restoration. A recent study estimated that natural NET will reduce CO2 emissions by 14 GtCO2/y by 2030 (Griscom et al. 2017); however, natural NET faces resource constraints, such as land, water, and nutrients; it also has the disadvantage of reducing biodiversity and albedo (Smith et al. 2016). Ocean fertilization, for example, is a natural NET that involves introducing nutrients into the upper ocean to promote algal growth and CO2 uptake; however, the efficiency of ocean fertilization is low and may harm ecosystems.
Technological NETs are measures to reduce CO2 emissions from the atmosphere through various geoengineering schemes. These include indirect ocean capture, enhanced weathering, and ocean alkalinity modification (enhancement), which involves increasing the ocean’s ability to absorb atmospheric CO2 by adding alkalinity (House et al. 2007; Kheshgi 1995). The ocean plays an important role in CO2 reduction because it absorbed approximately 30% of total CO2 emissions after the Industrial Revolution (Webb et al. 2021). Previous research has investigated ocean alkalinity modification methods and their effectiveness (Köhler 2020; Renforth & Henderson 2017). According to the scenario analysis by Ilyina et al. (2013), although significantly adding alkalinity to large areas is required to avoid further ocean acidification, rapid increases in alkalinity over a broad ocean area may have an unexpected negative impact on the natural environment. Indirect ocean capture, which chemically treats seawater and returns decarbonized and alkalized seawater to the sea (de Lannoy et al. 2018; Eisaman et al. 2018), is another method to improve the ocean’s ability to absorb CO2. However, this method is still in the experimental stage.
Another NET that has recently piqued the interest of researchers is enhanced weathering, which has the potential to reduce CO2 in the atmosphere by improving the chemical weathering of rocks. Enhanced rock weathering has great potential in terms of performance and cost (Beerling et al. 2018, 2020), but its environmental impact is still unclear.
To reduce CO2, hybrid NETs combine natural and technological approaches, and bioenergy with carbon capture and storage (BECCS) is recognized as the most mature technology (de Lannoy et al. 2018). BECCS combines bioenergy applications and CCS; using energy derived from biomass combustion and capturing and storing CO2 emitted during combustion results in net CO2 removal (Bui et al. 2018). Although BECCS has the potential to play an important role in mitigating climate change (Duan et al. 2021), promotional challenges, such as efficiency and CO2 transportation, remain (Smith et al. 2016).
DAC systems (also known as postcombustion capture systems) remove CO2 directly from the atmosphere and are classified into three types: liquid solvent, solid sorbent, and membrane-based systems. Ishimoto et al. (2017) provided an overview of DAC estimation, proposing a wide range of cost estimates ranging from USD 100/tCO2 to USD 1000/tCO2.
The liquid-solvent DAC system includes an air contactor and a regeneration facility. In the air contactor, CO2 from the air reacts with potassium hydroxide solution (KOH) to produce water and potassium carbonate (K2CO3), as follows:
2KOH + CO →HO + KCO
The potassium carbonate solution is then dissolved in caustic soda, which reacts with calcium hydroxide (Ca[OH]2) to form a calcium carbonate (CaCO3) precipitate. Water is removed from the CaCO3 slurry by a clarificatory and filter press, and the precipitated CaCO3 is sent to a calcination furnace, where it is heated to about 900°C in an oxygen calcination furnace using natural gas to produce solid calcium oxide (CaO). Finally, CO2 gas can be compressed and stored indefinitely. Socolow et al. (2011) showed the cost of a liquid solvent DAC system at USD 641–819/tCO2. Meanwhile, Mazzotti et al. (2013) and Zeman (2014) estimate costs of USD 528–579/tCO2 and USD 309–580/tCO2 under optimized operation systems.
DAC technology employs a solid sorbent system and a liquid solvent. Adsorption and desorption are the two processes for capturing CO2 (National Academies of Sciences, Engineering, and Medicine [NASEM] 2019). In the adsorption process, solid sorbents from gas mixtures capture CO2 via physisorption or chemisorption (McQueen et al. 2021). Physical adsorption by intramolecular forces between CO2 molecules and the surface of solid sorbents is referred to as physisorption, whereas chemisorption is a chemical reaction in which CO2 and specific sites on the sorbent form covalent bonds. Following adsorption, heating, and vacuuming can be used to desorb CO2 from the solid sorbent. According to NASEM (2019), the net removed cost of a solid sorbent DAC system is USD 124–407/tCO2.
The third DAC system under consideration employs membrane-based gas separation. There are many different types of membrane separation, each of which can separate different gases (i.e., N2/H2, natural gas sweetening, and CO2 capture), including free-standing, hollow fiber, and spiral wound membranes that separate CO2 from natural gas (Llosa Tanco et al. 2018). Membranes must have two qualities: permeance (i.e., the passage of gases through the membrane) and selectivity (i.e., the ability to filter the desired gas from a mixture of gases).
Membrane-based DAC systems have shown potential and have some advantages over solid sorbent and liquid solvent systems, such as a smaller footprint and simpler operation (Fujikawa et al. 2021). Zanco et al. (2021) evaluated the following postcombustion DAC systems: aqueous piperazine solution absorption, Zeolite 13X adsorption in conventional fixed beds, and polymeric multistage membrane separation. Their results indicate that the absorption system is the most cost-effective for the majority of plant sizes and recovery rates; however, the membrane-based system may also become cost-effective at smaller scales in the future.
Another study(Yun et al. 2021) compared postcombustion DAC processes based on absorption and membranes for the iron and steel industry. The simulation results showed that as the concentration of CO2 in the flue gas increases, the membrane-based DAC system becomes more cost-effective than the absorption-based system. To be cost-effective for post-combustion CO2 capture, polymeric membranes must have a CO2/N2 selectivity greater than 200 while maintaining a relatively high permeability (Al-Mamoori et al. 2017).
The nanomembrane technology is a free-standing membrane system. For example, Ariyoshi et al. (2021) developed a robust hyper-permeable nanomembrane out of polydimethylsiloxane and cellulose nanofibers (PDMS–CNF). With a 17-nm thick defect-free PDMS layer of 17-nm thickness, their robust PDMS–CNF nanomembrane achieves superior permeation of 50,000 gas permeation units.
Nano membrane-based DAC systems are more scalable and can be installed in various locations. Because of its scalability, membrane-based DAC technology is also useful for DAC-U systems. Based on the performance of state-of-the-art polymer separation membranes, Fujikawa et al. (2021) demonstrated that CO2 in the air (0.04%) could be concentrated to more than 40% through multistage membrane separation. This membrane-based CO2 capture approach, representing the concept of “ubiquitous CO2 capture,” can be implemented at various sizes and scales in any location. This technology approach could become a viable method of recycling CO2 and eventually lead to a circular CO2 society.
Life Cycle Assessment of the DAC-U System
Description of the DAC-U system
The DAC-U is a functional system that captures CO2 directly from the air, enriches it 1,000 times to approximately 40% by volume, and converts it into carbon fuel methane (CH4) for use at home. The DAC-U is made up of two parts: a capture unit and a conversion unit. The capture unit uses membrane separation technology. A thermal conversion method is used for the Sabatier reaction, a methanation process, as shown in Eq. 1.
CO2 methanation is carried out with the help of a Ru-based catalyst made of aluminum oxide, CaO, and ruthenium. Due to the exothermic nature of the methanation reaction, some heat is produced during the conversion process. This heat is conducive to sustaining the thermal conversion process.
The methods are presented in two parts: (1) for the LCA calculations for the proposed DAC-U system; and (2) the methodology used to determine willingness to adopt the proposed technology in Japan, based on LCA inventory data and a national survey.
Life cycle assessment
LCA assesses the environmental impact of a product throughout its life cycle, from resource extraction to the production of materials, product parts, and the product itself, as well as its use and disposal. The application of LCA in the early stages of research and development (R&D) has recently generated considerable interest (Cucurachi et al. 2018). The European Union has even acknowledged it as a necessary component of R&D projects for funding proposals (Blanco et al. 2020). LCA methodology is divided into four stages: (i) definition of the goals and scope clarification; (ii) life cycle inventory analysis; (iii) life cycle impact assessment; and (iv) interpretation.
The first phase determines the objective of the study, potential applications, and other critical components, such as system boundaries, functional unit (FU), assumptions, and limitations. The second phase concerns data collection, necessitating an inventory of the system’s input and output data. The third phase consists of three mandatory steps: (i) selection of impact categories, indicators, and characterization models; (ii) classification, which entails associating the inventory data with the specified impact categories; and (iii) characterization, which involves determining the results of each category indicator by converting life cycle inventory parts into common units (using a characterization factor) and aggregating the converted results within the same impact category. Finally, during the interpretation phase, the LCA results are reported and discussed to identify relevant issues and provide conclusions, recommendations, and information useful for decision-making.
Goal and system boundaries
This study evaluates the environmental impact of a DAC and conversion unit based on the NEDO Moonshot project prototype. The system under consideration is divided into two subsystems: the capture unit (SS1) and the conversion unit (SS2). SS1 employs a membrane separation module to separate CO2 from air selectively. The operation of SS1 requires four stages, each with its own vacuum pump, and the first stage requires a fan. SS2 includes a module that uses a furnace with two flow reactors and four valves to convert enriched CO2 into methane. The LCA study included all of the remaining components of the DAC-U system, including vacuum pumps, fans, and connectors. Figure 2 shows the system’s primary energy and material flows; this study used a cradle-to-gate approach (with a cradle-to-grave approach in the future), including raw materials extraction, equipment production, operation, and maintenance. This study’s analysis did not consider the disposal stage and capital goods. In LCA studies, the FU quantifies the function of the product system and provides a reference unit. The FU for this study is defined based on the quantity of captured CO2 during the lifetime of DAC-U as 5,475 kg of captured CO2.
Data
The materials and energies required for the DAC-U’s manufacture, maintenance, and operation are inventoried based on the Moonshot project listed in Table 1.
| Item | Subsystem | Unit | Amount |
|---|---|---|---|
| Polystyrene Sulfonate sodium salt | SS1 | Kg | 4.455 |
| Polydimethylsiloxane | SS1 | Kg | 1.485 |
| Hexane (for PDMS) | SS1 | Kg | 28.215 |
| Polyactive | SS1 | Kg | 0.0891 |
| Chloroform | SS1 | Kg | 17.731 |
| Polystyrene plate | SS1 | Kg | 86.130 |
| Glass plate 5x5 cm | SS1 | Kg | 41.58 |
| Polyactylonitrile | SS1 | Kg | 3.394 |
| Nonwoven PET nonwoven mechanical support | SS1 | Kg | 6.788 |
| Polyimide | SS1 | Kg | 89.100 |
| Urethane dimethacrylate | SS1 | Kg | 81.675 |
| Methacrylate monomer (s) | SS1 | Kg | 27.225 |
| Steel sheet | SS1 | Kg | 5 |
| Aluminum oxide Al2O3 | SS2 | gram | 75 |
| Calcium Oxide | SS2 | gram | 20 |
| Ruthenium | SS2 | gram | 5 |
| Pyrex tube | SS2 | gram | 600 |
| Nichrome | SS2 | gram | 31.106 |
| Stainless steel | SS2 | gram | 634.05 |
| Aluminosilicate | SS2 | gram | 45.00 |
| Solar-PV electricity (for membrane fabrication) | SS1 | MWh | 2.1038 |
| Solar-PV electricity (for operation in the base scenario) | SS1 | MWh | 64.3899 |
| Solar-PV electricity (for operation in the office scenario) | SS1 | MWh | 33.3033 |
| Solar-PV Electricity (for operation in the classroom scenario) | SS1 | MWh | 26.8481 |
| Solar-PV electricity (for operation in the conversion process) | SS2 | MWh | 19.272 |
Three scenarios are defined: base, office, and classroom, based on different situations for the device’s location with varying CO2 concentrations. We assume that the base, office, and classroom (class) scenario have a CO2 concentration of 400, 800, and 1,000 ppm, respectively, which affects the amount of electricity required to capture and enrich CO2. The operation stage data for both subsystems were inventoried based on Aspen simulations. The inventory data for each process were adapted from the IDEA v2 database.
Several simplifying assumptions are made to render the analysis more traceable. First, the capacity of the unit is set to capture and convert 1 kg of CO2 per day, and the device’s lifetime is assumed to be 15 years, with the separation membrane replaced once a year. Second, the device’s materials and mass are used to calculate manufacturing emissions. Furthermore, solar panels generate all of the electricity used to run the system and hydrogen is supplied externally. This study assumes that Japan will have a mature hydrogen infrastructure when the DAC-U system is launched to the market.
Willingness to adopt and carbon reduction potential
Willingness to adopt
We used questionnaire data to investigate the household-level adoption potential of the DAC-U system and relationships with personal attributes using an ordered logit model. In 2017, we surveyed 4,148 respondents in Japan, asking if they were willing to adopt a DAC-U system. The survey was conducted for all generations, and the sampling procedure was designed to randomly select respondents while keeping the gender and age distributions of the respondents consistent to reflect the Japanese population. For the questions related to the purchasing intention (willingness to adopt) DAC-U system, descriptions of a DAC-U system were provided first, and respondents were asked they would like to introduce a DAC-U system that could reduce their energy costs in the long run. Then, the respondents were asked to respond using the following: (1) Purchase for sure; (2) Positively consider purchasing under certain conditions; (3) I don’t know; (4) Purchase under certain conditions; and (5) Not likely to purchase. We code the survey responses as 1, 2, 3, 4, and 5, in which the values increase with the order. As a result, we decide to analyze ordinal responses and perform an ordered logit analysis. Although the individuals’ “willingness to adopt” typically indicates whether they would adopt or not, our study uses the term to indicate whether people would accept DAC-U systems and how they could become more favorable toward purchasing DAC-U systems. This approach to such treatment of dependent variables enables us to use complete information in the survey. Table 2 displays the distribution of the respondents.
| Choice | Frequency | Percentage |
|---|---|---|
| (1) Purchase for sure | 276 | 6.65% |
| (2) Positively consider purchasing under certain conditions | 489 | 11.79% |
| (3) I don’t know | 1,763 | 42.50% |
| (4) Purchase under certain conditions | 1,269 | 30.59% |
| (5) Not likely to purchase | 351 | 8.46% |
| Total | 4,418 | 100% |
We also included sociodemographic variables such as income, gender, age, dummy variables on marriage, and whether a respondent has a child. Moreover, we established the occupations of the respondents. Table 3 shows the descriptive statistics.
| Variable | Mean | Std. dev. | Min | Max |
|---|---|---|---|---|
| Annual income (10,000 JPY) | 3.774 | 2.41424 | 1 | 10 |
| Age | 51.946 | 15.588 | 20 | 93 |
| Female dummy (= 1 if female) | 0.42 | .49 | 0 | 1 |
| Married dummy (= 1 if married) | 0.70 | .46 | 0 | 1 |
| Child dummy (= 1 if responded has a child) | 0.64 | .48 | 0 | 1 |
The function of purchasing a DAC-U system takes the following form:
$${Y}_{J}={\varvec{X}}^{\varvec{{\prime }}}\beta +{ϵ}_{J}, J=\text{1,2},\text{3,4},5$$
2
\({Y}_{J}\) represents the ordered response of the survey answer J in five categories (1, 2, 3, 4, and 5), \({\varvec{X}}^{\varvec{{\prime }}}\) is a vector of controls, and \({ϵ}_{J}\) is the error term. \(\beta\) is a vector coefficient of interest. The signs and magnitudes of \(\beta\) demonstrate the impact of individual variables on the choices of the DAC-U system. The ordered response is \({Y}_{J}=J\) if
$${a}_{J-1}\le {Y}_{J}\le {a}_{j}$$
3
where \(j=\text{1,2},\text{3,4},5,{a}_{1}=-\infty\), \({a}_{J-1}\le {a}_{j}, {a}_{5}=\infty\), which follows the cumulative distribution function of the logistic distribution:
$$F=\frac{1}{1+\text{exp}\left(-a\left({\alpha }_{j}-{\varvec{X}}^{\varvec{{\prime }}}\beta \right)\right)} \left(4\right)$$
The marginal effect \({\delta }_{J}\) of \({x}_{J}\) for the J-th response is given by the following:
$${\delta }_{j}=\frac{\partial \text{Pr}\left(Y=J|X\right)}{\partial {x}_{J}} \left(5\right)$$
The result of the ordered logit is presented in Section 4.2.1. We also estimate marginal effects to investigate how the probability of selecting the adoption level of the DAC-U system varies, relative to each unit, and changes according to income level. The results are shown in Fig. 4. Section 4.2 depicts the results. Estimates, rather than causation effects, explain the correlations and interactions between independent variables and purchase decisions, rather than causation effects.
CO2 capture and reduction potential
We constructed two scenarios cases to estimate the DAC-U’s CO2 absorption and reduction potential: (A) The DAC-U system is installed in all detached houses in Japan, along with a roof DAC-U system; (B) the DAC-U system is installed in some detached houses, along with a roof DAC-U system (installation ratio ranging from 1–11% among the 1,747 municipalities in Japan). The installation ratio is determined at the municipal level, based on the installation data of residential fuel cells (which could be perceived as technology similar to the DAC-U system) provided by the Ministry of Economy, Trade, and Industry (METI 2021) in each of the 1,747 municipalities.
We first estimate the amount of annual CO2 capture potential and CO2 reduction potential (reflecting the results of the LCA) at the municipality level and then present the CO2 capture and reduction potential at the prefectural level (47 prefectures) by combining the municipal-level estimation results for cases (A) and (B).
The electricity required for the DAC-U to absorb 1 kg of CO2 in detached houses is calculated using the inventory data for the base scenario (Table 1), which is 15.28 kWh/kgCO2. The LCA results show that in the base scenario, 0.96797 kg of CO2 is emitted to capture 1 kg of CO2. Therefore, considering the system’s life cycle, the CO2 reduction potential factor is set toto 0.03202. The annual solar power generation potential was calculated in detached houses in 1,747 municipalities using the number of detached houses (excluding unoccupied houses) from the Housing and Land Survey (Statistics Bureau, Ministry of Internal Affairs and Communications 2018) and the coefficient of photovoltaic (PV) power generation by region coefficient provided by the Ministry of Environment (MoE 2021). Following the assumption made by MoE (2021), we set the installed capacity per house at 4kW. Figure 3 depicts the amount of annual generation potential provided by the PV power provided for the DAC-U system in cases (A) and (B).
Using the above parameters and the estimated generation potential, we estimated the amount of CO2 absorption and the negative emission amount (from the LCA standpoint) at the prefectural level for cases (A) and (B).
The results are also presented in two sections, beginning with LCA and progressing to the willingness to adopt and effectiveness of the DAC-U, as was the case in the methodology.
LCA
Because this study is primarily concerned with CO2 emissions, the impact category of GWP is evaluated using IPCC GWP 20a (Intergovernmental Panel on Climate Change, Global warming potential). The SimaPro 9.1 software assesses the environmental impact of the system under study. Table 4 shows the results of the DAC-U system’s environmental characterization, as well as the contribution of each subsystem to the GWP. In the base, office, and classroom scenarios, the capture unit (SS1) contributes 97.9%, 97.8%, and 97.7%, respectively, accounting for most of CO2 emissions for the system. Manufacturing the capture unit emits the most CO2 (approximately half of SS1 emissions), and maintenance comes second (approximately 35%).
| Scenario | SS1 (Capturing Unit) | SS2 (Conversion Unit) | Total GWP of the DAC-U system | |||
|---|---|---|---|---|---|---|
| Manufacturing | Operation | Maintenance (replacing membrane) | Manufacturing | Operation | ||
| Baseline | 2421.00 | 426.30 | 1480.88 | 7.34 | 85.07 | 4420.59 |
| Office | 2421.00 | 220.40 | 1480.88 | 7.34 | 85.07 | 4214.69 |
| Class | 2421.00 | 177.87 | 1480.88 | 7.34 | 85.07 | 4172.15 |
Regarding SS1 emissions, two processes stand out: 1) the use of four vacuum pumps in the capturing stages (1,630 kg CO2); and 2) the use of polyimide or Kapton tape to assemble membrane modules, which are replaced 14 times over the lifetime of the DAC-U (945 kg CO2). Both materials are produced using heavy fuel oil, which has a significant environmental impact. Improving these LCA hotspots is a critical component of the Moonshot project. Fit-for-purpose vacuum pumps may be used in the future, and physical LCA can be performed to improve accuracy. Furthermore, more sustainable materials can be sought as polyimide or Kapton tape substitutes.
Table 5 demonstrates a scenario in which Japan already has a mature hydrogen supply system, which is when the DAC-U system is released to the public, and which does not account for hydrogen consumption during the conversion process (SS2).
| SS1 | SS2 | Total | Methane offset | Total with methane offset | Hydrogen supply | Total | |
|---|---|---|---|---|---|---|---|
| Base | 4328.41 | 92.41 | 4420.589 | −355.88 | 4064.71 | 1234.929 | 5299.64 |
| Office | 4122.28 | 92.41 | 4214.688 | −355.88 | 3858.81 | 1234.929 | 5093.74 |
| Classroom | 4079.75 | 92.41 | 4172.155 | −355.88 | 3816.27 | 1234.929 | 5051.20 |
We also detail the total carbon emission assessment results, including the offset for methane that would otherwise have to be imported into Japan and the cost of hydrogen supply. The DAC-U is expected to emit 421.5 kg of methane over its lifetime. We offset the CO2 emitted by importing LNG to Japan, which is equivalent to the DAC-U, because methane is combusted whether imported or not. The DAC-U system absorbs 5,475 kg of CO2 and emits 4,064, 3,858, and 3,816 kg of CO2 in the three scenarios. In terms of the hydrogen used in the thermal conversion unit to convert CO2 to methane, we calculated the total GWP emissions, 1,234 kg, over a 15-year period using hydrogen gas from the IDEA v2 database. Figure 4 compares the DAC-U’s GWP emissions with and without a hydrogen supply.
In all three scenarios, it was estimated that the DAC-U system could provide a negative GWP emission result; however, it should also be noted that our results did not consider disposal or recycling processes after the defined lifetime of 15 years. Future research that can account for these processes will be considered within the carbon budget in future research.
Willingness to adopt and CO2 reduction potential
Willingness to adopt
Table 5 shows the results of Eq. (2), thus indicating the results of the ordered logit results. Model (1) excludes survey respondents who answer “I don’t know,” whereas in the survey, Model (2) accepts treating the response as valid. The specifications include the prefecture-, area-, and year-fixed effects. We also account for the differences in the survey respondents’ occupations. The coefficients in Table 6 indicate how each variable relates to the willingness to adopt a DAC-U system. Therefore, a positive coefficient indicates that a one-unit increase of the variable by would increase the willingness to adopt by the degree of the coefficient.
We find that income positively correlates with the adoption rates; a 1% increase in the household income positively correlates with the adoption rate by 0.154% in Model (1) and 0.134% in Model (2). Meanwhile, age accounts for 0.447%. Conversely, age is negatively correlated with adoption rates by − 0.447% in Model (1) and negatively correlated with adoption rates in Model (2) by − 0.387%. In both models, women are less likely to adopt, and whether a respondent is married or has a child appears to be statistically insignificant in terms of DAC-U adoptions.
| Model (1) | Model (2) | |
|---|---|---|
| In (Household income) | 0.154** | 0.134** |
| (0.0753) | (0.0562) | |
| ln (Age) | −0.447** | −0.387*** |
| (0.180) | (0.136) | |
| Female | −0.284** | −0.172* |
| (0.129) | (0.0933) | |
| Married | 0.151 | 0.123 |
| (0.129) | (0.0918) | |
| Child | (0.0752) | 0.0856 |
| (0.121) | (0.0867) | |
| Prefecture-fixed effects | Yes, | Yes, |
| Area Fixed Effects | Yes, | Yes, |
| Fixed jobs effects | Yes, | Yes, |
| N | 2028 | 3436 |
| Pseudo R-sq | 0.018 | 0.010 |
| Note: Standard errors are in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01. | ||
Based on the estimates in Table 6, we calculate the proportion of people using DAC-U systems, as in Eq. (5) (marginal probability), the proportion of the population belonging to the Japanese government’s income level categories. We plot the introduction probability of each income group. The results provided in Fig. 5 show that the adoption probability increases with income levels, particularly income levels 4–10.
Our estimated coefficient of household income in Table 5 shows a positive coefficient; therefore, a lower, smaller probability of people belonging to income level groups 12–20 does not indicate a refusal or unwillingness to adopt DAC-U systems. Instead, it indicates that the number of people included in this category is lower and smaller than in other income level groups, such as income group level 7.
CO2 capture and reduction potential
Figure 6 depicts the annual CO2 capture potential in Japan prefectures for cases (A) and (B), respectively; the total annual CO2 capture potential by households (living in detached houses) was estimated to be 8,727.2 and 698.2 (ktCO2/year), respectively. In 2021, the CO2 emissions from the household sector totaled 167,351 ktCO2, accounting for 16% of total CO2 emissions in Japan. Analysis of the CO2 capture potential of the DAC-U system at the household level shows that the DAC-U system can capture between 5.21% and 0.42% of CO2 emissions from the household sector annually.
Under case (A), some prefectures, primarily large metropolitan areas or areas with many households, such as Tokyo, Hyogo, Fukuoka, and Hokkaido, have a high potential. Meanwhile, figure in case (B) shows that the capture potential of prefectures, such as Hokkaido and Nagano, decreases when compared to case (A), which could be explained by the differences in purchase ability derived from attributes, such as income level and age distribution, which are examined more closely in the analysis of willingness to adopt a DAC-U system.
Figure 7 also shows the annual CO2 reduction potential at the prefectural level in cases (A) and (B), estimated from a life cycle perspective and reflecting the LCA results presented in Section 3.
The estimation results show that the DAC-U system can reduce 279.5 and 22.4 ktCO2 annually in cases (A) and (B), respectively. The differences in reduction potential among prefectures largely mirror those of capture potential.
As NETs are increasingly recognized to be critical to achieving a net-zero-emission energy system, the concept of DAC has received significant attention as a potential tool to reduce atmospheric CO2. The DAC-U system represents an exciting emerging NET by not only capturing but also utilizing CO2 for the production of useful fuels. This study focused on a novel approach to the concept of DAC-U, in which a modular device captures and concentrates CO2 directly from air using membranes before generating methane via thermochemical conversion processes. It clarified the DAC-U’s unique features compared to peer NETs in a thorough literature review. Furthermore, the GWP of DAC-U was evaluated using LCA in conjunction with deploying renewable energy under multiple scenarios with CO2 concentrations of 400, 800, and 1,000 ppm, respectively. The LCA results revealed that the DAC-U system can be used as an effective NET in various CO2 concentration scenarios. One limitation of this study is that our results do not yet consider end-of-life processes (i.e., disposal or recycling) after the useful lifetime of the DAC-U system (estimated at 15 years). Future research could consider these processes to better understand the potential of the DAC-U system as a NET.
Finally, to investigate the capture and reduction potential of DAC-U’s CO2 further, we focused on Japanese households and estimated capture and reduction potentials at the municipal, prefectural, and national levels. Our estimations showed that the DAC-U system could capture up to 5% of current CO2 emissions from the household sector each year (as low as 0.42% under the conservative case). The CO2 reduction potential over the life cycle was estimated to be up to 279.5 ktCO2/year (22.4 ktCO2/year in the conservative case). We observed significant differences in the capture and reduction potential across prefectures in the two cases examined, which could be explained by differences in purchasing power between households across prefectures. To further investigate this, we used an ordered logit model to conduct a survey-based analysis of willingness to adopt emerging technologies and their relationship with personal attributes. The analysis on willingness to adopt revealed that income, age, and sex strongly affect willingness to adopt the DAC-U system. Using the analysis parameters, we presented the proportion of people in Japan who are likely to install DAC-U systems in their homes, taking into account the population’s income levels.
NETs are likely critical in limiting global warming by reducing atmospheric greenhouse gas concentrations. Carbon neutrality can be achieved using several approaches, including DAC-U. However, we must critically evaluate these approaches and scenarios in terms of life cycle emissions and production and operation inputs so that a suitable set of NETs can be integrated into the energy system through appropriate energy and climate change policies at the national level.
APS American Physical Society
CaO Calcium oxide
CCS Carbon capture and storage
DAC Direct air capture
FU Functional unit
GPU Gas permeation units
GWP Global warming potential
IOC Indirect ocean capture
LCA Life cycle analysis
METI Ministry of Economy, Trade, and Industry
NASEM National Academies of Sciences, Engineering, and Medicine
NET Negative emission technologies
R&D Research and development
Acknowledgements
The authors acknowledge the funding of the “Moonshot Research and Development Program” (JPNP18016), commissioned by the New Energy and Industrial Technology Development Organization (NEDO). Any opinions, findings, and conclusions expressed in this document are those of the authors and do not necessarily reflect the views of the funding agency.
- Al-Mamoori A, Krishnamurthy A, Rownaghi AA, Rezaei F. Carbon capture and utilization update. Energy Technol. 2017;5:834-49.
- Ariyoshi M, Fujikawa S, Kunitake T. Robust, hyper-permeable nanomembrane composites of poly(dimethylsiloxane) and cellulose nanofibers. ACS Appl Mater Interfaces. 2021;13:61189-95.
- Beerling DJ, Kantzas EP, Lomas MR, Wade P, Eufrasio RM, Renforth P, et al. Potential for large-scale CO2 removal via enhanced rock weathering with croplands. Nature. 2020;583:242-8.
- Beerling DJ, Leake JR, Long SP, Scholes JD, Ton J, Nelson PN, et al. Farming with crops and rocks to address global climate, food and soil security. Nat Plants. 2018;4:138-47.
- Blanco CF, Cucurachi S, Guinée JB, Vijver MG, Peijnenburg WJGM, Trattnig R, et al. Assessing the sustainability of emerging technologies: A probabilistic LCA method applied to advanced photovoltaics. J Cleaner Prod. 2020;259.
- Bui M, Adjiman CS, Bardow A, Anthony EJ, Boston A, Brown S, et al. Carbon capture and storage (CCS): The way forward. Energy Environ Sci. 2018;11:1062-176.
- Cucurachi S, van der Giesen C, Guinée J. Ex-ante LCA of emerging technologies. Procedia CIRP. 2018;69:463-8.
- de Lannoy CF, Eisaman MD, Jose A, Karnitz SD, DeVaul RW, Hannun K, et al. Indirect ocean capture of atmospheric CO2: Part I. Prototype of a negative emissions technology. Int J Greenhouse Gas Control. 2018;70:243-53.
- Deutz S, Bardow A. Life-cycle assessment of an industrial direct air capture process based on temperature–vacuum swing adsorption. Nat Energy. 2021;6:203-13.
- Duan H, Zhou S, Jiang K, Bertram C, Harmsen M, Kriegler E, et al. Assessing China’s efforts to pursue the 1.5°C warming limit. Science. 2021;372:378-85.
- Eisaman MD, Rivest JLB, Karnitz SD, de Lannoy C, Jose A, DeVaul RW, et al. Indirect ocean capture of atmospheric CO2: Part II. Understanding the cost of negative emissions. Int J Greenhouse Gas Control. 2018;70:254-61.
- Fujikawa S, Selyanchyn R, Kunitake T. A new strategy for membrane-based direct air capture. Polym J. 2021;53:111-9.
- Griscom BW, Adams J, Ellis PW, Houghton RA, Lomax G, Miteva DA, et al. Natural climate solutions. Proc Natl Acad Sci U S A. 2017;114:11645-50.
- House KZ, House CH, Schrag DP, Aziz MJ. Electrochemical acceleration of chemical weathering as an energetically feasible approach to mitigating anthropogenic climate change. Environ Sci Technol. 2007;41:8464-70.
- Ilyina T, Wolf-Gladrow D, Munhoven G, Heinze C. Assessing the potential of calcium-based artificial ocean alkalinization to mitigate rising atmospheric CO2 and ocean acidification. Geophys Res Lett. 2013;40:5909-14.
- Ishimoto Y, Sugiyama M, Kato E, Moriyama R, Tsuzuki K, Kurosawa A. Putting costs of direct air capture in context (FCEA Working Paper Series: 002). SSRN Journal. 2017.
- Kheshgi HS. Sequestering atmospheric carbon dioxide by increasing ocean alkalinity. Energy. 1995;20:915-22.
- Köhler P. Anthropogenic CO2 of high emission scenario compensated after 3500 years of ocean alkalinization with an annually constant dissolution of 5 Pg of olivine. Front Clim. 2020;2.
- Llosa Tanco MA, Medrano JA, Gallucci F, Tanaka DA. Membrane optimization and process condition investigation for enhancing the CO2 separation from natural gas. In: Current trends and future developments on (Bio-) Membranes. Amsterdam, The Netherlands: Elsevier; 2018. p. 469-509.
- Mazzotti M, Baciocchi R, Desmond MJ, Socolow RH. Direct air capture of CO2 with chemicals: Optimization of a two-loop hydroxide carbonate system using a countercurrent air-liquid contactor. Clim Change. 2013;118:119-35.
- McQueen N, Gomes KV, McCormick C, Blumanthal K, Pisciotta M, Wilcox J. A review of direct air capture (DAC): Scaling up commercial technologies and innovating for the future. Prog Energy. 2021;3:032001.
- National Academies of Sciences, Engineering, and Medicine (NASEM). Negative emissions technologies and reliable sequestration: A research agenda. Washington DC: The National Academies Press; 2019.
- Renforth P, Henderson G. Assessing ocean alkalinity for carbon sequestration. Rev Geophys. 2017;55:636-74.
- Sandalow D, Friedmann J, McCormick C, McCoy S. Direct air capture of carbon dioxide: ICEF Roadmap 2018. 2018. https://www.icef.go.jp/pdf/summary/roadmap/icef2018_roadmap.pdf.
- Smith P, Davis SJ, Creutzig F, Fuss S, Minx J, Gabrielle B, et al. Biophysical and economic limits to negative CO2 emissions. Nat Clim Change. 2016;6:42-50.
- Socolow R, Desmond M, Aines R, Blackstock J, Bolland O, Kaarsberg T, et al. Direct air capture of CO2 with chemicals: A technology assessment for the APS panel on public affairs. American Physical Society (APS), 2011.
- Webb RM, Silverman-Roati K, Gerrard MB. Removing carbon dioxide through ocean alkalinity enhancement and seaweed cultivation: legal challenges and opportunities (Columbia public law research paper forthcoming). 2021. https://ssrn.com/abstract=3800494.
- Yun S, Jang MG, Kim JK. Techno-economic assessment and comparison of absorption and membrane CO2 capture processes for iron and steel industry. Energy. 2021;229.
- Zanco SE, Pérez-Calvo J-F, Gasós A, Cordiano B, Becattini V, Mazzotti M. Postcombustion CO2 capture: A comparative techno-economic assessment of three technologies using a solvent, an adsorbent, and a membrane. ACS Eng Au. 2021;1:50-72.
- Zeman F. Reducing the cost of ca-based direct air capture of CO2. Environ Sci Technol. 2014;48:11730.