Carbon Utilization Infrastructure, Markets, and Research and Development: A Final Report (2024)

3

Life Cycle, Techno-Economic, and Societal/Equity Assessments of CO₂ Utilization Processes, Technologies, and Systems

3.1 INTRODUCTION

CO₂ utilization technologies and infrastructure will incur environmental impacts, potentially different from or differently distributed than those from existing technologies and infrastructure for chemicals and materials production. Developers, funders, and policy makers will consider environmental justice needs associated with their deployment and use. Thus, as CO₂ utilization technologies are being developed, decision-support tools are important to ascertain not just economic competitiveness but also the environmental sustainability and equity outcomes associated with their deployment. Such tools can quantify and inform the progress of emerging CO₂ utilization technologies and strategies for reaching net zero and a circular carbon economy. This chapter reviews methodologies and requirements to assess the contribution of CO₂ utilization to a net-zero emissions future—namely, techno-economic assessment (TEA), life cycle assessment (LCA), and societal/equity assessments.

TEA, LCA, and equity assessments provide vital information for decision makers considering the potential impacts of CO₂ utilization projects and related infrastructure, including implications on sustainability; they are broadly recognized as critical tools for evaluating existing and emerging technologies (Moni et al. 2019). TEA informs the economic viability of technologies, providing valuable information regarding the competitiveness of a product. LCA quantifies the environmental burdens associated with products, processes, or services from materials extraction through end of life. LCA and TEA can identify areas needing further attention and are an integral component of applied research and development (R&D) (Lettner and Hesser 2019), often used in a combined toolkit. Equity assessments seek to minimize unintended adverse outcomes while maximizing opportunities and positive outcomes from policies, programs, or processes, particularly for those facing inequity or disparities (Bradley et al. 2022b; Cremonese et al. 2022). Life cycle thinking can inform equity assessments to ensure that what appears to be a more benign choice does not result in unintended burden-shifting across the value chain. Social LCA (s-LCA) is a particular type of LCA aimed at establishing the human and societal impacts of a product’s life cycle. These assessments are complementary: TEA and LCA were not designed to robustly examine questions of equity, and equity assessments do not capture the suite of impacts quantified in LCA and TEA. The Department of Energy (DOE) requires LCA and TEA to be completed as part of applied R&D programs (Skone et al. 2022) and is operationalizing environmental justice, energy justice, and equity into its CO₂ utilization R&D program in line with the Biden administration’s

Justice40 initiative (Clark 2023; E.O. 14008).¹ Federal tools (Argonne National Laboratory’s Greenhouse Gases, Regulated Emissions, and Energy Use in Technologies LCA models and DOE Office of Technology Transitions’ Commercial Adoption Readiness Assessment Tool) and databases (the National Renewable Energy Laboratory Life Cycle Inventory and the National Energy Technology Laboratory Unit Process Library) exist to support these requirements. However, only certain programs within DOE require TEA and LCA estimates to be reported to program managers as part of the funding proposal and iteratively improved through project completion. This inconsistency can result in contradictory information and data gaps.

Carbon utilization technologies range in their technological development from early technology readiness level (TRL)² through pilot, demonstration, and full commercialization (IEA 2021; Sick et al. 2022). Because many CO₂ utilization technologies have yet to operate at commercial scale, LCA, TEA, and equity assessments are inherently challenging, and decision makers question how to gain useful insights into these earlier TRL technologies compared to those that are reaching demonstration and beyond (Cremonese et al. 2022; Goglio et al. 2020; Langhorst et al. 2022). Key issues facing LCA and TEA of early-stage technologies include paucity of data, incomplete understanding of scale up, a lack (in some cases) of comparator or proxy data, and uncertainty in how the technology will be deployed and in future market conditions. Recent studies (Langhorst et al. 2022; Newman et al. 2023, Figure 3) have shown demonstrable progress in managing these challenges by providing detailed guidance on how to streamline TEA-LCA, conduct analyses on emerging technologies, and interpret the results (Sick et al. 2023). Present efforts to improve global guidance have been limited aside from AssessCCUS, which includes collaborators from the United States, Canada, the United Kingdom, Germany, and Japan.

The committee’s first report considered the full life cycle of CO₂ utilization processes when accounting for net CO₂ emissions from projects (NASEM 2023a). In doing so, the committee considered the durability of the carbon storage (product lifetime of greater or less than 100 years) alongside whether the carbon source was atmospheric, oceanic, biogenic, or fossil, noting that these factors determine whether CO₂ utilization is capable of being net-negative, net-zero, or net-positive in emissions. The committee also identified carbon accounting across the value chain as an important component of assessing sustainability for CO₂ utilization technologies, recommending that DOE “fund research to quantify the dynamic impact of CO₂-derived products, for example, their specific lifetime, on the CO₂ balance in the atmosphere” and that the United States “incorporate knowledge acquired from European projects and regulatory activities in addressing circular carbon economies and net-negative emissions” (Recommendation 3.1, NASEM 2023a). DOE also should support national laboratories, academia, and industry in performing TEA, LCA, and integrated systems analysis to identify CO₂ utilization approaches that are technologically feasible, sustainable, economically viable, and take into account relevant regulatory and policy frameworks, environmental justice impacts, and factors that may influence societal acceptance (Recommendation 6.1, NASEM 2023a).

This chapter reviews the current state of knowledge of LCA, TEA, and equity assessments for CO₂ utilization technology and infrastructure, providing results from recent studies as well as an overall synthesis of how these tools can inform decisions across TRLs, in concept, research, scale up, deployment, and diffusion. The chapter examines the challenges faced in recent studies, pointing to where methodological and data improvements are necessary to ensure that results can provide reliable decision-support. It assesses progress of CO₂ utilization technologies in meeting net-zero goals by reviewing current LCA and TEA results and identifying opportuni-

___________________

¹ The Justice40 initiative has the goal of ensuring that 40 percent of the overall benefits of certain federal investments go toward disadvantaged communities. Categories covered by Justice40 include climate change, clean energy and energy efficiency, clean transit, affordable and sustainable housing, training and workforce development, remediation and reduction of legacy pollution, and the development of critical clean water and wastewater infrastructure (E.O. 14008).

² TRL defines stages from basic research to maturity for full-scale market introduction. The stages begin with basic research, where principles are observed and reported, technology is conceived, and applications formulated. They proceed through research to prove feasibility, technology development, technology demonstration, and system commissioning, and end in system operations, where an actual system is operated in the final form over the full range of operating conditions. (See Table 2-3 in Chapter 2, which is modified from DOE [2011], for more detail about each TRL.)

ties from LCA, TEA, and equity assessments to improve decision-support for circularity, R&D investments, and infrastructure build-out.

3.1.1 Technology Assessment Approaches

Technology assessment aims to anticipate the current and future performance of a (novel) technology by integrating knowledge of the benefits and risks of emerging technologies, contributing to the formation of public and political opinion, and supplying effective, pragmatic, and sustainable options for decision making (UNCTAD 2022). In the case of CO₂ capture and utilization technologies, such assessments are far from straightforward, owing to both technology and market factors. Technology factors include the inherent uncertainties associated with low TRL; the time horizon of weeks to centuries in which environmental impacts are foreseen, depending on the lifetime of the final product and other factors; and the different impacts of interest, including global warming, and land and water use. Factors associated with the markets include the changing needs of society, such as shifts and upgrades in material end-use application and performance; the large and diverse number of actors in the value chains; the differences in scales at which technologies could be deployed (from niche to large markets); and the potential for CO₂ utilization technologies to disrupt the fossil-based incumbent. This heterogeneity strongly affects the future usage and performance of CO₂ utilization technologies. For example, producing sustainable aviation fuels from CO₂ requires large amounts of hydrogen and energy, both ideally sourced without CO₂ emissions. Furthermore, production facilities will be quite similar to petrochemical plants and refineries, and thus public opposition might become a factor in deployments.

Technology assessment transcends a purely technical and scientific evaluation, requiring a broad evaluation of the environmental, economic, and societal context and impacts of a technology. Holistic evaluation relies on three interconnected types of assessments:

• Techno-Economic Assessment (TEA) is an integrated assessment of the technical performance and economic feasibility of a technology. It combines process modeling and engineering design with an economic evaluation to assess the (future) viability of a process, system configuration, or product, also projecting if it will change the competitive landscape. As emerging technologies are often noncompetitive with incumbent technologies at their inception, understanding key drivers of performance, uncertainties associated with initial cost estimates, and potential pathways to improved performance are key results of early-stage TEAs (Buchner et al. 2018).
• Environmental Life Cycle Assessment (e-LCA) is an analytical tool that aims to provide insight into the environmental performance of technologies, processes, products, or services. It allows a comprehensive and systematic comparison of systems (e.g., a new technology versus a baseline condition) in order to identify potential shifting of environmental burdens across different environmental life cycle phases (e.g., raw material extraction, production, end use) or environmental compartments (e.g., soil, water, air). e-LCA is the most common form of LCA and often what is being referred to when “LCA” is used.
• Equity and Social Impact Assessment aims to support and advance societally equitable outcomes resulting from the deployment of a technology, product, or service. It provides information for decision makers about different forms of equity, such as intergenerational or temporal (i.e., equity across time), spatial or geographical (equity across space), racial or ethnic, socioeconomic, gender, cultural, and democratic (decision-making) equity (Parson and Mottee 2023). There are different approaches to conducting equity and societal impact assessment, including social life cycle assessment (s-LCA), which builds on e-LCA methodology and aims to quantify the impact of a product or process on society along the full life cycle. In contrast with e-LCA, the methodological development, application, and harmonization of s-LCA is still at a preliminary stage (McCord et al. 2023; Sala et al. 2015).

Box 3-1 defines terminology used in technology assessments.

Goal and scope definition is a common initial stage for both TEA and LCA. For integrated TEA-LCA, alignment in the goal, scope, data, and system elements provide consistency in results (Langhorst et al. 2022, Part E; Mahmud et al. 2021). Different goals lead to different comparisons, with varying data requirements and inventory creation efforts (inputs and outputs to TEA-LCA are compiled in inventories that document data for the analyses in line with the goal and scope definition). Conversely, the inventory also impacts the goal, especially if data are not available. Importantly, the assessment goal is specific to the individual study and the practitioner’s perspective. Even when focusing on the same product system, the assessment goal can vary between studies, depending on factors such as the scope and size of the project, technological maturity, geographical region, and time horizon.

The goal for a TEA is to address techno-economic questions, such as the cost or potential profitability of a new technology, process, product, plant, or project. TEA is often carried out for a specific audience (e.g., assessment of a CO₂ utilization reaction concept for a funding agency, assessment of a CO₂ utilization plant concept for industry managers, assessment of CO₂ utilization technology options for policy makers). The goal informs the details needed to define the scope, and the goal and scope together then frame all subsequent work phases of the study. The TEA goal

also interacts with the subsequent work phase of data inventory creation (i.e., a listing of all relevant components, cost). The TEA scope describes what aspects of a product (or service) will be assessed and how a product (e.g., a plastic) or service (e.g., energy storage) will be compared to competing solutions. The first step of the TEA scope phase is to identify and describe the analyzed product system; the central elements are precise, quantitative descriptions of the function(s) of the new technology and selection of the comparison metrics in the form of functional units and reference flows that are needed to meet the functional unit. The next step is to specify the analyzed system elements and define system boundaries, followed by selecting benchmark systems for comparison. Last, the technological maturity of the product system will be used to select suitable assessment indicators. Since operational data are not available for technologies in development, one CO₂ utilization-specific challenge with the scoping phase is that many CO₂-derived products provide similar but nonidentical performance to benchmark products.

As noted by the International Organization for Standardization (ISO) 14044 standard for LCA, the goal and scope stages determine the intended application of the study, the reasons for carrying out the study, the intended audience of the study and whether the results are to be used in comparative assertions disclosed to public (Euro

pean Committee for Standardization 2022a). These stages of LCA are carried out similarly to TEA, but with an environmental focus that includes identifying impacts of interest for the analysis.

The system boundary defines the limits of the product system and describes which system elements belong to it. For example, details of producing and delivering raw material streams—for example, CO₂, could be included in the assessment to analyze limitations or opportunities for improvements, or they could be treated as fixed input. More detail on transportation-specific considerations can be found in the committee’s first report (NASEM 2023a). Material and energy flows crossing the system boundary are referred to as “input flows” and “output flows.” A product system can have one or multiple input or output flows (e.g., coproducts or by-products, waste streams, various feedstocks, and various inputs for waste treatment); the latter often are referred to as multifunctional product systems or as having “multifunctionality.” System boundaries can be defined for product systems and comparative benchmark systems and are derived from the assessment goal and product functions. System boundaries allow for a transparent and process-based comparison of the product and benchmark systems. They set the basis for reviewing what is included in a LCA or TEA study and for comparing different studies with each other. System boundaries must be consistent throughout the study.

The system boundaries, geographic and temporal representativeness, and choice of functional unit are key decisions that can impact LCA results of CO₂ utilization technologies (Hauschild et al. 2018). As for TEA, the system boundaries determine what elements will be included in the analysis (e.g., capture and transport of CO₂; inputs such as the required energy, hydrogen, and other raw materials; through plant decommissioning) (NASEM 2019).

3.1.2 Importance of Evaluating Carbon Utilization Systems and Individual Products and Processes in a Net-Zero Future

As discussed in Chapter 2, the economic potential for CO₂-derived products is substantial. The latest report of the Intergovernmental Panel on Climate Change (IPCC) indicated that “In order to reach net zero CO₂ emissions for the carbon needed in society (e.g., plastics, wood, aviation, fuels, solvents), it is important to close the use loops for carbon and carbon dioxide” (Bashmakov et al. 2022, p. 1163). The need for chemicals, fuels, and materials cannot be sustainably achieved without modifying their production, use, and disposal. CO₂ utilization technologies are a class of options to reduce dependence on nonrenewable resources while contributing to a net-zero future. However, the potential to remove CO₂ from the atmosphere or avoid emissions in the first place depends on the source of CO₂, the process and product that would be displaced, and the duration of storage (Mason et al. 2023). Schematically, Figure 3-1 summarizes the best-possible high-level outcomes of utilizing captured CO₂. These outcomes, which reflect the impact of deploying a CO₂ utilization technology on ambient concentrations of CO₂ in the atmosphere, are critical to decision making. The National Academies’ report Gaseous Carbon Waste Streams Utilization summarized key considerations for performing LCA on different types of CO₂ utilization products (NASEM 2019, Table 8-2), a frame through which the technologies and products discussed in Chapters 5–8 can be viewed.

TEA, LCA, and equity assessments provide valuable feedback across TRLs as CO₂ utilization technologies move from proof-of-concept through prototype to eventual deployment. Owing to the wide range of outcomes, TEA and LCA can offer critical input to the design of early-stage technologies. The tools provide different types of decision-support at different stages of technological advancement (Figure 3-2). At early stages of technology readiness, results from LCA, TEA, and equity assessments are much more uncertain with much less information to understand actual operations and localized impacts. Recognizing these results as uncertain, they nonetheless can offer valuable insights into strategic investments in applied R&D to increase competitiveness or decrease negative impacts of the design. As technology advances toward deployment, additional spatial, temporal, and operational data will enable more accurate results for decision making. After substantial technical changes are made or more accurate data is obtained, repeated assessments will allow practitioners to obtain updated information and guidance for decisions on whether and how to proceed.

3.2 TECHNO-ECONOMIC ASSESSMENTS FOR CARBON UTILIZATION

TEAs are routinely conducted for CO₂ utilization technologies across TRLs, resulting in a wide range of published estimates. The committee reviewed recent TEAs and reviews of published literature to document the stage for the state of the art in the field. Early-stage technologies are still in development; therefore, such estimates

will change as technological advances are accomplished. Figure 3-3 shows the different components of a TEA for assessment of a future CO₂ utilization plant, illustrating how data and uncertainties in the performance of the plant’s core technology (a CO₂ electrolyzer, which is at low TRL) impact the design and cost evaluation of the full plant (Vos et al. 2024). Although the figure is specific to low-temperature electrolyzers, the building blocks of the TEA are relevant for any technology. The representativeness of data inputs and levels of uncertainty in results for low-TRL technologies have implications for other process units within the system boundaries (e.g., the need to purify the streams). In turn, assumptions of technology performance prior to deployment influence expected material and equipment requirements, and subsequently the energy requirements, which are at the core of the cost estimations (e.g., levelized cost of production, payback time, or net present value).

Despite their inherent uncertainty, the results of early TEAs provide insights that can be used to shape the future development and deployment of a technology. For example, a review of production cost estimates for electrocatalytic conversion of CO₂ to different products showed that the technology is not generally competitive with fossil fuel alternatives, except possibly in a select few cases. At the time of this review, market prices for carbon monoxide and formate were $0.18 and $0.66/kg (Grim et al. 2020),³ while production costs from CO₂ utilization technologies were estimated to be $0.39 (0.18–0.64) and $0.96 (0.10–2.63)/kg, respectively (Jordaan and Wang 2021, ranges in parentheses are minimum and maximum values).⁴ Production cost estimates will evolve over time

___________________

³ The carbon monoxide price is the average for 2014–2018, and the formate price is the average of 2014 and 2016.

⁴ Study estimates refer to 2014–2022.

with technological advances and depend on the scope of the analysis, markets, prices, and local and geographical factors (Jordaan and Wang 2021). Such estimates provide value in showing the difference between published production costs and market prices, as well as by pointing to how variable estimates can be, depending on technologies, design, and assumptions. If the gap between production costs and market prices becomes narrow or nonexistent, a more comprehensive TEA can help identify the entry point for a first-of-a-kind (FOAK) plant. TEA estimates for FOAK plants will be less competitive than nth-of-a-kind (NOAK) plants, as further discussed in Section 3.2.3.3.

TEA provides critical information about how a CO₂ conversion technology can be competitive under specific technological and market conditions, informing both R&D and policy. For example, results can indicate how much the production costs must decrease to break even with present market prices (Ruttinger et al. 2022). A recent study completed consistent TEAs across five major electricity-driven CO₂ conversion technologies (low- and high-temperature electrolysis, microbial electrosynthesis, biological and thermochemical conversion with electrolytic hydrogen) and 11 distinct carbonaceous products, leveraging results to compare the minimum selling prices (i.e., required price to break even with production costs) to market prices (Huang et al. 2021; see Figure 3-4). Only one product, polyhydroxybutyrate, had a minimum selling price lower than market prices, but several other products show promise if technological advancements are realized (Figure 3-4). Such estimates will evolve over time as technology advances, as market conditions change, and if certain technologies are deployed in specific geographies. Low prices of renewable electricity and a price on carbon can play a substantial role in encouraging deployment as technologies approach competitiveness.

Owing to the evolving nature of these technologies and their estimated costs, the committee developed additional guidance on conducting TEAs across TRLs to produce valuable results for informing applied R&D through deployment.

3.2.1 What TEA Can and Cannot Do in Assessing CO₂ Utilization Technologies

TEAs usually are performed with a focus on the production phase and from a producer’s point of view, but they can be expanded to include up- or downstream aspects—for example, the use and disposal phases of a product, as is done in LCC. Doing so might be useful if such aspects have an outsized impact on cost, which may be especially salient for CO₂ utilization technologies where accounting for the fate of the carbon in the product is important.

While related, assessment and decision making are separate efforts. A TEA is based generally on information from process design, and the results can be used as feedback or recommendations for improved design. However, TEA does not include technical development activities, such as chemical process design, but it builds on and feeds back into them. The studies are context-specific with respect to factors such as location, time horizon, or access to information and thus require specific assumptions in order to be meaningful. TEA can support project-specific decision making in both technological and economic contexts, such as specific R&D work or investment decisions. Reliably applying TEA results in a generalized context, such as for global policy making, can be quite challenging and therefore requires exercising considerable caution. At its core, TEA is about single-actor costs and how to minimize them, with some consideration of how this relates to the market of the product or function.

3.2.2 Use of TEA for Technologies at Different TRLs

Appropriate use of TEA results to direct R&D and deployment can be challenging for decision makers in industry and policy when indicators (e.g., net present value) are estimated using inconsistent methods or applied outside the intended context. Objective comparative analysis may not be possible without consistent and systematic methodology (Zimmermann and Schomäcker 2017). To resolve these challenges, the Global CO₂ Initiative led the development of community-directed harmonized guidelines for TEA and LCA specifically for CO₂ utilization (Langhorst et al. 2022). Opinions differ as to the usefulness of conducting TEA for TRLs below 3, and thus a TRL-dependent approach to TEA will be important (Buchner et al. 2018). Assessments for low TRL necessarily will be very limiting, driven by the lack of sufficiently accurate data, technology details, and information about potential applications, and can provide only high-level insights. Although cost estimation for TRL 1 is not advisable, a qualitative assessment could be conducted and used to recommend technology pathways. At TRL 2, limited quantitative assessments will be possible on mass, energy, and value efficiencies for general guidance. At higher TRLs, progressively more information will be available to allow the inclusion of more accurate data for additional elements that impact the cost of goods sold. Applying TEA results conducted at the earliest TRLs may limit innovation in basic science.

3.2.3 Critical Issues for CO₂ Utilization TEA

The level of detail available in a TEA depends on the TRL of the technology. Limitations in data availability and/or their respective accuracy are a key issue at low TRL; often relevant information has not been obtained in early-stage laboratory experiments (Buchner et al. 2018). Furthermore, projecting the equipment needs from bench-scale operation to scale up production processes to pilot-plant and full-scale deployment sizes is challenging. Capital expenditure estimation will be sensitive to the kind of equipment used and the estimation methods employed—for example, the AACE International Cost Estimate Classification System.⁵ The greater the specificity in output information required, the higher the assessment effort in terms of time, complexity, detail, and work needed to gather data. In many cases, a technology assessment will include a multicriteria decision analysis in the interpretation phase (Chauvy et al. 2020; McCord et al. 2021), necessary for examining trade-offs and helping to prioritize the selection of options. Multiple-attribute decision making is useful for choosing between a set of specific and preselected variables that determine alternative solutions for the purpose of ideally narrowing down to one choice. By contrast, multiple-objective decision making is useful for a set of variables that will produce an infinite number of solutions—the Pareto group (with solutions on the Pareto frontier being considered optimal; Marler and Arora 2004).

___________________

⁵ See AACE International, “Homepage,” https://web.aacei.org, accessed August 5, 2024.

3.2.3.1 Transparency and Consistency Challenges with System Boundaries

The system boundaries in TEA and LCA are often inconsistent; most TEAs are gate-to-gate studies, while LCAs have broader system boundaries. This discrepancy has implications for the analysis of results. Differences in system boundaries can be a significant determinant of estimated CO₂ avoidance costs. To illustrate this point, Tanzer et al. (2023) assessed the impact of system boundaries in a biological carbon capture and storage (CCS) case study. The study used biochar to replace coal as clinker kiln fuel in a cement plant with CCS and included the uptake of CO₂ by concrete over time. System boundary choice varied net CO₂ estimates from −660 to +16 kg CO₂(eq)/t cement, and aligning boundaries shrank the avoidance cost range from 48–321 to 157–193 €/t CO₂(eq) (Tanzer et al. 2023).

3.2.3.2 CO₂ Purity

The CO₂ stream will contain multiple source-dependent impurities that may negatively impact the performance and costs of CO₂ utilization technologies. As indicated in Table 3-1, electrocatalytic and thermocatalytic routes are very sensitive to impurities. Impurities can impact electroreduction in several ways—for instance, they can compete with CO₂ for electrons or adsorb on or react with the catalyst surface and deleteriously modify its properties (Harman and Wang 2022). For conventional thermocatalytic processes, a major challenge is the presence of impurities such as H₂S, NH₃, carbonyl sulfide and alkali halides, which can result in catalyst deactivation owing to carbon deposition, sintering, pore blockage, and sulfur poisoning (Pattnaik et al. 2022). More on CO₂ stream purification can be found in Chapter 11, Section 11.1.2, noting that transport has less stringent requirements for purity than conversion. Appendix H contains further tables from the committee’s first report with information on typical CO₂ stream impurities from different sources and CO₂ purity requirements for different transportation modes.

The cost to purify CO₂ streams depends on several factors, including process efficiency, operating conditions, energy requirements, safety considerations, and whether there is experience with similar applications in the industry. As such, impurities can significantly alter the business case. Vos et al. (2023) conducted a TEA of the pretreatment units needed to purify CO₂ from a bioethanol plant then used in a solid-oxide electrolysis unit to produce syngas; compositions and tolerance assumptions are shown in Table 3-2. The TEA indicates high capital expenditure (Capex) (almost 3 million euros in bare equipment costs) and energy costs, the former driven by units removing sulfur and alcohols and the latter mostly driven by the cryogenic distillation step used to remove noncondensable gases.

TABLE 3-1 Overview of Impurities of Concern by CO₂ Utilization Route

^a Although use of CO₂ in the food and beverage industry is out-of-scope for this report, these purity levels have been included as a point of comparison, because many CO₂ utilization studies assume that the input CO₂ is food-grade purity. Purity levels are taken from EIGA (2018).

NOTE: ppm = parts-per-million; ppmv = parts-per-million by volume; ppmw = parts-per-million by weight.

SOURCE: Modified from NASEM (2023a).

TABLE 3-2 Composition of the CO₂ Stream by Component from a Bioethanol Plant, Electrolytic Reactor Degradation Type, and Tolerance Limits for Solid-Oxide Electrolysis

^a All amounts in column 2 are in ppm, except for CO₂, which is given in percent.

SOURCE: Adapted from Vos et al. (2023).

To date, most experimental studies of CO₂ utilization technologies have used near-pure CO₂. There is a lack of understanding of the potential impact of impurities or partially converted feedstocks in technology performance and of mitigation strategies that go beyond developing more efficient pretreatment of streams. Such strategies can include developing more resilient or more easily regenerated catalysts and/or new approaches/reactions where impurities can act as promoters or useful reactants (Harman and Wang 2022). When industrial configurations are considered, potential impurities created in the conversion process are important to the downstream separations associated with recycle streams (Sarswat et al. 2022).

3.2.3.3 Scalability

Scaling a new technology to cost-competitive, industrial-scale production capacity requires upscaling from bench-top experiments to a full-size plant. The plant setup has to be scaled up physically to allow industrial-scale production volumes to meet market demands. Matching the capacity of the CO₂ source and the utilization plant would avoid unnecessary CO₂ transportation. This scale up is not simply making everything physically larger; it usually requires significant changes in plant design to achieve the necessary mass flows, temperature and pressure conditions, and more.⁶ Scale up also requires integration of utilities and supporting processes such as heating and cooling to maximize the energy efficiency of an operation, often not restricted to the operation of one manufacturing process, but more frequently for the operation of a collection of manufacturing processes on a site

___________________

⁶ Some processes do not follow rules of scale, such as those dependent on surface area for photo-driven reactions. Although the photo-driven reactions may not follow economies of scale, the downstream processes required to purify the products to market requirements tend to do so.

to reuse heat. The resulting FOAK plants are more costly than NOAK plants, as indicated in Figures 3-5 and 3-6, owing to overdesign of equipment, redundancy of equipment to ensure that the plant will operate as desired, and design size limits or nonstandard material use in early-stage equipment.

Learning curves provide information about the speed at which costs would decrease in relation to the cumulative installed capacity. The FOAK cost and installed capacity, as well as learning rate are needed to estimate the shape of the learning curve (Figure 3-5). At the FOAK stage, learning rates are not yet established, and therefore data from other technologies are often used. For example, IEAGHG (2023) uses chlor-alkali and polymer electrolyte membrane (PEM)

fuel cells to estimate learning for low- temperature CO₂ electroconversion and a solid oxide fuel cell for high-temperature CO₂ electroconversion. Incorporating learning rates allows Capex data from a FOAK plant to be used to derive cost projections for a NOAK plant (Greig et al. 2014; Rubin 2014, 2016; Van der Spek 2017).

Full-scale NOAK plants require process improvements and cost reductions both in terms of Capex to build the plants and operational expenses. Once multiple plants are in operation with the same or similar technologies, learning curve strategies for process improvements and cost reductions can be employed.

3.2.3.4 Geographic and Temporal Relevance

Local conditions can have a significant impact on the feasibility and viability of a new technology. Overall supply chain factors—for example, raw material availability, cost, transportation modes, and product distribution channels—are highly context-specific. An International Energy Agency (IEA) study found significant differences in the median projected costs of electricity generation technologies among regions and stressed that technology competitiveness depends on national and local conditions (IEA 2020b). To account for such differences, location factors can be used to convert construction costs of industrial plants among countries. Although these factors reflect average differences and values may differ within countries, they provide more realistic comparisons of TEAs across regions. As an example, the location factors published by Intratec for 2018 indicated that building a plant in China would be 16 percent less expensive than building a plant in the United States, while building the same plant in Germany would be 7 percent more expensive (Intratec Solutions 2023). These differences are not owing to exchange rates but rather reflect differences in labor costs, infrastructure availability, and so on.

Therefore, factoring regionality into assessments enables TEA to address the impact of local policy support or inhibitions and to quantify competing demands for the same resources (Jiang et al. 2020; Kähler et al. 2023; Ravikumar et al. 2020; Thielges et al. 2022). Alternative carbon sources also might experience strong seasonal variations in availability and cost, especially if derived from biomass, or might require critical raw materials. TEA uses commodity prices of chemicals and materials, which can be highly volatile; therefore, the year used for the analysis significantly affects the result. The price of steel, for example, quadrupled between 2020 and 2021 and remains at least 50 percent higher than in 2020 (Trading Economics 2024).

3.2.3.5 Addressing Uncertainty

Models used in TEA and LCA describe mathematical relationships between input variables and the desired output, which is any result or indicator of interest for a reference base case and additional scenarios that are crucial for the subsequent decision. Uncertainty and sensitivity analyses are key elements in assessments like TEA and LCA and need to be examined to put results in quantitative context. Uncertainty and sensitivity analyses are distinctly different in how they are conducted and what they mean, but are often confused with each other (see Box 3-1).

Uncertainty analyses determine the uncertainty associated with the model output, which in TEA can be a calculated profitability indicator (e.g., net present value or individual rate of return). The overall uncertainty is determined from the propagation of errors in input data as well as uncertainties in the model that describes the technology itself. It can even depend on the context in which the assessment is conducted—for example, the assumed market and or the environmental conditions of operations may result in uncertainty in estimates for net present value. Typically, each input variable has uncertainty associated with its value, which can be expressed with probability distributions that show the likelihood of the variable having a certain value. These distributions are then used as model input for a simulation that determines the uncertainty in the output. For multiple input variables with known or assumed probability distributions, the combined influence of all input variables on the output of the assessment can be determined rigorously with methods such as Monte Carlo simulations, which provide statistical (not physical) uncertainties associated with the model (Figure 3-7).

Sensitivity analyses (SAs) determine how sensitive the model output is to variations in one or more input variables. Uncertainty analyses and SAs are complementary, as SAs reveal how any uncertainty within the output is constructed and identify the key input variables that contribute most to the uncertainty. Figure 3-8 shows how a systematic variation of input variable values is used to determine their significance. SAs are especially powerful

in providing insights into which variables have the highest impact on the output—for example, cost—and where resources should be allocated to develop improvements or alternatives. Likewise, SAs can identify variables that have minute impact on the output—that is, low sensitivity—and can be deprioritized in future work, improving the technology. Methods to conduct SAs include one-at-a-time sensitivity analysis, one- and multiple-way sensitivity analysis, scatterplot analysis, variance-based methods, and density-based methods.

Given the variety of techniques available, guidelines for selecting an uncertainty analysis method for TEAs are gaining traction in literature. An example in Appendix J (Roussanaly et al. 2021) shows a decision tree recommending the type of uncertainty analysis based on purpose—“what if” or “what will.” “What if” uncertainty analyses address diagnostic questions and provide insights into changes in output resulting from changes in inputs. “What will” analyses are prognostic in nature and focus on understanding the conditions under which a result may be obtained with a certain probability (Roussanaly et al. 2021; Rubin 2019; Saltelli et al. 2008). Once a “what if” or “what will” analysis is chosen, an SA can then be performed. Expert elicitation approaches can be used to complement SAs. For instance, pedigree matrices⁷ can be used to evaluate the knowledge base of a model or data (Edelen and Ingwersen 2016; Fernández-Dacosta et al. 2017; Pinto et al. 2024); an example of how to use these matrices in ex-ante TEA and LCA of a CO₂ to polyols plants is presented in Fernández-Dacosta et al. (2017).

3.2.3.6 Competing Technologies

TEA provides insights into the economic performance of a technology, and, in most cases, is discussed against a reference, also called a counterfactual case. For example, economic results from comparing a CO₂-derived product with a fossil-based product will be quite different from those comparing a CO₂-derived product to a bio-based product or a waste-based product (Singh et al. 2023). Complicating the interpretation of such comparisons, environmental impacts will be substantially different—land use may be a challenge for certain bio-based products—pointing to the importance of complementing TEA with LCA results. For CO₂ utilization technologies at early TRL, selecting the most likely competitor technology is not straightforward. Figure 3-9 shows a selection of 69 process routes to produce ethylene from alternative carbon sources (e.g., biomass, CO₂, waste). The variety of sources and process options to form just one product indicates that for early-stage TEA, reference routes beyond the fossil-based counterpart also need to be investigated to obtain robust insights into when a CO₂ utilization route performs better or worse under different scenarios. Furthermore, such analyses must be done on an equal basis, such that the system boundaries, temporal and geographic assumptions, and level of detail used to carry out the TEAs are as similar as possible so that the results can be compared fairly. If the TEA of the CO₂ technology is carried out for a given date in the future (e.g., 2040), potential changes in the reference technology (e.g., decline in cost owing to learning) must be accounted for.

3.3 LIFE CYCLE ASSESSMENTS FOR CARBON UTILIZATION

3.3.1 Summary of Recent LCA Reviews and Publications

The committee examined systematic reviews and recently published LCAs to determine the state of the art being applied to CO₂ utilization. Such systematic reviews are limited and cannot address the entirety of products and technologies; however, they do point to important conclusions about the state of LCA of CO₂ utilization. First, studies conducting LCAs of CO₂ utilization for the same product can examine many different types of technologies, and these technologies may incur different environmental impacts. Table 3-3 (in this chapter) and Tables J-1 and J-2 (in Appendix J) show compiled LCA results for CO₂ emissions released to produce methanol, dimethyl ether (DME), and dimethyl carbonate (DMC) from Garcia-Garcia et al. (2021), demonstrating the wide variety of technologies and processes that have been examined. Second, the system boundaries of published studies are highly variable and may not be consistent—for example, studies may include only a part of the supply chain, or they may examine impacts from materials extraction to the product as it leaves the production facilities

___________________

⁷ A pedigree matrix analyzes the strengths and weaknesses of available information or knowledge base on an ordinal scale—for example, 1–5, low–high (Fernández-Dacosta et al. 2017).

(cradle-to-gate). Differences in results can be attributed to these variations in system boundaries or assumptions (e.g., source of hydrogen and energy) as well as whether feedstock-related emissions are considered. Results for the production of methane from CO₂ via the Sabatier reaction and a nickel-based catalyst with cradle-to-gate system boundaries showed large variations in methods (e.g., allocation and application of emissions credit) and energy source for electrolysis (e.g., specific renewable technologies and grid mixes), resulting in a wide range of results from −1.99 to +16.7 kg_CO2 kg⁻_CH¹₄ and from −0.04 to +0.3 kg_CO2 MJ⁻_CH¹₄, where the latter metric is relative to the lower heating value of methane in megajoules. Methane produced directly from fossil sources had emissions of 0.54 kg_CO2 kg⁻_CH¹₄ and 0.15 kg_CO2 MJ⁻_CH¹₄ respectively. Third, LCA results often focus on CO₂ emissions alone rather than a more comprehensive set of environmental impacts. The latter is recommended by ISO 14040 (European Committee for Standardization 2022a). As a result, important environmental impacts may be overlooked, and such studies can make conclusions only about climate-related impacts, not the overall sustainability of the products.

A meta-analysis of LCA studies examining a more comprehensive set of impacts highlighted the importance of including impacts beyond CO₂ alone: no CO₂ utilization-based chemical production alternative performed better in all impact categories than the conventional production technology (Thonemann 2020). For example, while formic acid production from CO₂ did not yield the lowest global warming potential (GWP), it showed significantly lower environmental impacts in most other impact categories. Depending on the source of heat and electricity, the global warming impact for formic acid could be reduced by as much as 95 percent compared to conventional production. Use of wind power resulted in lower environmental impacts across most impact categories compared to grid mixes that include fossil fuels. Given the challenges in completing consistent and comparable LCAs of CO₂ utilization technologies, the committee compiled additional guidance on LCA for CO₂ utilization, clarifying how to conduct these assessments and how decision makers can make use of results across TRLs.

TABLE 3-3 Compiled LCA Results for CO₂ Emissions Using Different System Boundaries, Assumptions, and Processes for Methanol Production from CO₂

^a Standard production (non-CO₂ utilization) processes for comparison.

^b Range contingent on hydrogen and electricity sources and other assumptions.

NOTE: Bi-reforming = mix of conventional steam (water) and dry (CO₂) reforming of methane to form synthesis gas that can be directly transformed into methanol; Tri-reforming = bi-reforming + partial oxidation of methane simultaneously.

SOURCE: Adapted from Garcia-Garcia (2021).

3.3.2 What LCA Can and Cannot Do in Assessing CO₂ Utilization Technologies

3.3.2.1 Dependencies

The accuracy and certainty of LCA results depend on the representativeness of the input data, which will be more uncertain for earlier-stage technologies. Recognizing uncertainty in the results informs R&D and design for early TRL technologies. For later-stage technologies, specific details about operation and siting are more available. Results for a CO₂ utilization technology deployed in one location and time horizon likely cannot be directly applied to a different geographical and temporal context, as data inputs may not be representative (e.g., electricity and heat generation, hydrogen production, and infrastructure requirements may be subject to large variability). For technologies that will be deployed in the future, present LCA results may not capture environmental changes

TABLE 3-4 Environmental Impact Categories, Scale of Impact, and Environmental Releases Used to Estimate the Impacts

^a The environmental releases are estimated in the inventory analysis, and they can contribute to specific impact categories.

SOURCE: Matthews et al. (2014), www.lcatextbook.com. CC BY-SA 4.0.

necessary to best represent future impacts. This challenge is particularly salient for impacts that require local or regional data, particularly when they change over time, as noted in Table 3-4.

Characterization factors transform the data inventories into impacts in the impact assessment stage of LCA. For example, GWP is a characterization factor used to translate the mass of each greenhouse gas (GHG) emission (e.g., kilogram of CO₂, CH₄, N₂O) into a common unit of mass of carbon dioxide equivalent (e.g., kilogram of CO₂ equivalent). Even global impact categories, such as climate change, are subject to important temporal factors specific to LCA of CO₂ utilization technologies. ISO 14040 notes the following temporal factors, each of which are relevant for CO₂ utilization: time horizon, discounting, temporal resolution of the inventory, time-dependent characterization, dynamic weighting, and time-dependent normalization (European Committee for Standardization 2022a; Lueddeckens et al. 2020). In a static LCA, for example, all emissions or avoided emissions are assumed to occur in 1 year, and a 100-year GWP is applied to estimate results. However, emissions inventories from CO₂ utilization technologies will occur at different times; thus, the emissions release or avoidance would imply a 100-year time horizon applied from the time of the event if using GWP 100. A dynamic LCA can provide more accurate results. Additionally, LCAs of presently deployed and future CO₂ utilization technologies will face different challenges in the uncertainty of impacts. Impact assessment models and the associated characterization factors can rapidly become outdated, so the accuracy of LCA results is tied to the data vintage associated with the most recent update. For example, a recent consensus was reached on how to characterize impacts of water consumption in terms of available water remaining per unit of surface in a given watershed relative to the world average, after demands from humans and ecosystems have been met (Boulay et al. 2018). While a forward-looking model is being developed to quantify future impacts of water consumption (Baustert et al. 2022), CO₂ utilization technologies to be deployed in the future may not be well represented for impact categories that are subject to large variability in environmental conditions over time (e.g., impacts related to nitrogen and other environmental loads). Understanding potential life cycle impacts is a critical aspect of examining the sustainability outcomes of CO₂ utilization technologies;

however, quantifying uncertainties in the results can inform decision makers about potential trade-offs and critical areas that warrant detailed investigation as more representative data become available (e.g., as facilities are sited).

3.3.2.2 Key Uncertainties in Quantifying and Interpreting Impacts

LCA seeks to quantify specific categories of environmental impacts based on a cause-effect chain linked to environmental releases from different processes. Impacts are quantified either as midpoint indicators or endpoint indicators. Midpoint impacts are considered intermediate indicators of environmental impacts that quantify changes in the environment caused by emissions or resource use but do not reflect the full consequence (e.g., GHG emissions expressed as CO₂ equivalents) (Kowalczyk et al. 2023). Endpoint indicators show much more aggregated environmental impacts, generally represented in three categories of damages: (1) effect on human health, (2) biodiversity, and (3) resource scarcity (Hauschild and Huibregts 2015). While endpoint impacts can more easily be interpreted by broad audiences, they are subject to greater uncertainty (Hauschild and Huijbregts 2015). Specific to CO₂ utilization, uncertainty will be further challenged by scarcity in life cycle inventory data (i.e., the data inputs and outputs for different life cycle stages), which will translate into greater uncertainties in impact assessment (Figure 3-10).

3.3.2.3 Consequential LCA

There is an increasing need to use LCAs to quantify impacts of system-level change—for example, if a CO₂derived product has the potential to capture a substantial portion of a market, understanding the environmental impacts associated with the change in supply is valuable. Attributional LCA determines the share of global environmental burdens associated with a product by quantifying the environmentally relevant physical flows to and from the life cycle of the product (Ekvall 2019). Consequential LCA, on the other hand, estimates how a product affects global environmental burdens by determining how environmentally relevant physical flows

change in response to decisions, making it especially relevant for policy makers. Consequential LCA is particularly relevant for CO₂ utilization if product systems can be impacted by large-scale deployment—for example, gigatonne-scale carbon mineralization—as there is increasing potential for displacement of more carbon intensive products (NASEM 2019).

3.3.3 Use of LCA for Approaches and Technologies at Different TRLs

The need for LCA of early-stage, emerging technologies is widely recognized and has become a standard expectation in DOE’s applied R&D projects on technologies across TRLs (Moni et al. 2020). Results generally show reductions in climate impacts compared to baseline technologies (Garcia-Garcia et al. 2021). Early-stage LCA can inform R&D, particularly through an iterative analysis that can be considered part of the R&D process (Lettner and Hesser 2019), incorporating knowledge from LCA into the design of the technology.

LCA results using databases representing present conditions, unit processes, and sectors are unlikely to be fully representative of the conditions where and when low TRL technologies may be deployed in the future. LCAs may overestimate impacts, as early-stage, emerging technologies are likely to have higher energy and material requirements; however, auxiliary requirements may be overlooked (Müller et al. 2020). LCAs of early-stage technologies typically use process simulation, manual calculations (e.g., stoichiometric equations), and proxies (Langhorst et al. 2022; Tsoy et al. 2020), and results depend on the benchmark data derived from these methods. LCA results for early-stage technologies are subject to higher levels of uncertainty compared to those for later-stage technologies. Despite this, LCAs of early-stage technologies are very useful in identifying critical pain points that have large influence on impact categories. This information could be used to prioritize and focus research.

Pilot and demonstration projects can provide useful baseline data representing present conditions; however, full scale up and deployment may be subject to different environmental and market conditions. Fully commercialized technologies can be characterized in the most representative sense, but CO₂ utilization technologies tend to be earlier stage. The impacts embodied in the required materials and the operational efficiencies will be project dependent on characteristics and infrastructure such as plant size, CO₂ transport, compression, and storage. The more the pilot or demonstration project resembles the proposed project or operations, the more representative the LCA’s environmental impacts will be.

3.3.4 Critical Issues for CO₂ Utilization LCA

Present ISO LCA standards require updates to be more suitable to carbon capture and circular systems (Pinto et al. 2024), which the committee extends here to CO₂ utilization. Recently, the National Energy Technology Laboratory (NETL) developed specific guidance for CO₂ utilization (Skone et al. 2022)—in part based on NASEM (2019)—with additional directions on software, data, and tools to complete LCA in accordance with ISO for applicants to funding programs (Sick et al. 2020). Additional key challenges are described in the following sections.

3.3.4.1 System Boundaries and Choice of Functional Units

Decisions made in the goal and scope definition, where the system boundaries and functional units are chosen, affect LCA results for CO₂ utilization, and there is a lack of consensus on what to include within the system boundaries. Some experts recommend that CO₂ be considered as a regular feedstock with its own production emissions (von der Assen et al. 2013); however, the system boundaries may include the source of the CO₂ emissions (i.e., the emissions reductions). As with TEA, transparency in the choice of system boundaries provides clarity on whether different LCA results can be compared.

The functional unit is the denominator in a life cycle result, the basis for reference for the product system that represents the quantifiable function of the product under investigation (European Committee for Standardization 2022b). It plays a major role in determining the results for LCA, which is particularly important for CO₂ utilization with its potential impact across many different products. For example, one LCA compared CO₂ utilization technologies based on the treatment of 1 kg of CO₂ (Thonemann and Pizzol 2019). Because the marginal suppliers were those

supplying CO₂ and H₂ rather than the products that may be displaced, the results are less relevant to the end products of each CO₂ utilization technology. Other studies recommended comparing LCA results for CO₂ utilization based on the products they are displacing, in terms of the analysis and functional units selected (Garcia-Garcia et al. 2021). A deeper integration of LCA with supply chain analysis for CO₂ utilization can enable an even more comprehensive understanding of these elements. NETL distinguishes between cradle-to-gate and cradle-to-grave, where the latter accounts for the end use of the product, and the former stops after the product is made (Skone et al. 2022).

3.3.4.2 Source of CO₂

The source of CO₂ (e.g., point sources versus direct air capture [DAC]) is an important consideration for the LCA of CO₂ utilization technologies (Müller et al. 2020). The life cycle carbon emissions of CO₂ capture have been reported to vary from negative to positive, meaning the CO₂ source is an important assumption. The challenge is thus determining the most environmentally beneficial sources, especially because the purity of CO₂ streams varies widely, generally from 5 to 35 percent by concentration but close to pure CO₂ in some cases (von der Assen et al. 2016). Some research suggests prioritizing point sources, as they result in the highest emissions benefits, noting that even low-purity CO₂ streams from cement production provide higher benefits than DAC (von der Assen et al. 2016). Published studies consider CO₂ differently: treating it as a negative emissions stream, an available pure CO₂ stream, or a co-product where emissions are split between the CO₂ and other products (Müller et al. 2020). Treating the CO₂ as a negative emission in LCA risks double counting emissions benefits (Lenzen 2008). DOE has recognized the challenge in overlooking the CO₂ source and assuming a negative emission: emissions benefits are attractive to both source and sink. Systems expansion can eliminate the risk of double counting by including the CO₂ source in the LCA—such an expansion is recommended by several studies (Cooney et al. 2022; Müller et al. 2020; Singh et al. 2023).

3.3.4.3 Duration of Carbon Storage

Durability of carbon abatement is a crucial factor for determining the life cycle impacts of CO₂-derived products. For example, synthetic fuels made from captured CO₂ are recombusted typically within 1 year, which rereleases CO₂ to the atmosphere. Therefore, synthetic fuels made from fossil CO₂ do not provide a net reduction in fossil-CO₂ emissions, except for global economic scenarios where production of synthetic fuel decreases oil and gas production owing to a drop in demand. The latter pathway readily occurs in the current global economy, which is 80 percent fossil-based, but will diminish in future net-zero scenarios in which energy increasingly is replaced by renewable or low-carbon nuclear sources.

Carbon-based products such as polyethylene, the dominant plastic, have a theoretical sequestration life of more than 1000 years in a landfill. However, 80 percent of waste is incinerated in the European Union, and 20 percent is incinerated in the United States, numbers that are expected to grow over concerns about land use and methane emissions. Thus, the European Union considers average sequestration life of polyethylene to be less than 100 years. de Kleijne et al. (2022) conclude that CO₂-based production of polyethylene is not compatible with the Paris Accord. Application of LCA to CO₂ utilization technologies is often challenged by making assumptions about the duration of storage, as actual duration is unknown. Carbon storage duration is also an important consideration for uncertainty and/or sensitivity analyses. Research is needed both to better understand storage duration through studies of the actual end of life of products and to better represent waste management in LCA and TEA of CO₂ utilization technologies.

3.3.4.4 Geographical and Temporal Representativeness

Geographical and temporal representativeness are widely recognized as critical data quality indicators for the accuracy of LCA results (Müller et al. 2016), creating challenges for early-stage technologies that have yet to be deployed. The results from one LCA in one region should not be applied to another region, as the environmental impacts often depend on local conditions. Characterization factors in typical LCAs have been identified as representative of large regions, but unsuitable at more granular scales such as counties (Pinto et al. 2024).

While LCA of earlier-stage technologies can provide useful information about potential impacts of deployment, these results should not be assumed as facilities are sited; rather, new LCAs need to be completed as relevant information becomes available. Once sites are selected, more detailed LCAs can support environmentally superior procurement decisions.

Temporal representativeness hinges on questions of key importance to the operation of the technology under examination. For example, use of static GWPs in LCA, as is typical, means that changes in emissions are considered as one pulse at the beginning of the assessment, rather than gradual emissions changes over the lifetime of operations (Langhorst et al. 2022; Levasseur et al. 2010, 2012). The selection of time horizon may overlook important emissions if not aligned with the duration of time the carbon is stored in the product (which is already uncertain). CO₂ storage duration is an important factor in the selection of time horizon; if the time horizon is too short, important emissions may be omitted from the analysis (von der Assen et al. 2013).

Whether emissions are reduced or avoided is also a critical question in LCA (Finkbeiner and Bach 2021). Typically, a service or pathway is replaced by an alternative, and it is important that system boundaries show the global cradle-to-grave impact of competing options (Finkbeiner and Bach 2021). LCAs often stop at intermediate boundaries (e.g., vehicle tank to tailpipe and not considering upstream supply chain) and underreport emissions relative to competing options. The value of the option varies with scenario (e.g., electric vehicles powered by coal plants today versus by renewable energy tomorrow), such that scenarios must be clearly stated. Avoiding petrochemical products through displacement with CO₂-derived products may realize emissions reductions in certain scenarios (Ruttinger et al. 2022).

3.3.4.5 Comprehensiveness of Impacts

The ISO standard recommends that LCAs include a comprehensive set of environmental impacts—for example, energy use, GHG emissions, and water consumption (Royal Society of Chemistry n.d.). LCA results for earlier-stage technologies may have limited accuracy for specific impact categories that depend on geographical representativeness. Estimates for the impacts of technologies and services can vary widely relative to the database used in assessment, as they may be too generic if the technology is earlier stage or if the characterization factors do not represent the environmental conditions at sites where facilities are being deployed. As technologies near deployment, geographical representativeness becomes increasingly important for interpreting the results for impacted communities. Complex trade-offs on weighting factors among categories may depend on local stakeholder values and drive substantial complexity that goes beyond results provided by LCA.

3.3.4.6 Addressing Uncertainty

Owing to the emergent nature of most CO₂ utilization technologies, uncertainty analyses are of particular importance. Uncertainty for early-stage technology necessitates an analysis of how well data inventories and impacts represent the time at which the technologies will be deployed and the regions where facilities may be sited (if it is possible to discern). Reasonableness checks against published literature results can support the identification of uncertainties. The methods outlined in Section 3.2.3.5 on addressing uncertainty in TEA also can be applied to address uncertainty in LCA results. For example, specific parameters may be identified as uncertain and impactful to cost or environment during the design stage (e.g., process simulation), which can be tested in sensitivity analyses informed by known operational performance and/or experimental results. Sensitivity analysis also can be informed by parameters found to be uncertain in experimental results for early-stage technologies. LCA-related assumptions (e.g., allocation) must be examined if they are used in the analysis (von der Assen et al. 2014). As discussed above, the duration of carbon storage is often unknown, yielding uncertainty in the long-term climate benefits of the technology under investigation. Scenarios for different fates can be tested to quantify how results may be impacted owing to specific outcomes. Break-even analysis⁸ can play an important role in determining the potential extent of environmental risks in the face of uncertainty.

___________________

⁸ An environmental break-even point refers to the point at which the environmental benefits of a product are reversed compared to another when varying a parameter in a parametric analysis (Messagie et al. 2012).

3.4 ASSESSING SOCIETAL IMPACT, PERCEPTION, AND EQUITY

As discussed further in Chapter 4, maintaining or developing social license is an essential component of technology development, particularly for technologies associated with products and industries that have been connected to past and ongoing societal harms. Developing an industry around CO₂ utilization will require proactive engagement with equity and justice concerns. These social considerations cannot be limited only to the site and direct operations of the carbon management project. Meaningfully addressing equity and justice will require a comprehensive look at the processes and products engaged by a project. LCA, and particularly s-LCA, can provide a structured framework for incorporating assessments of equity and societal impacts into the evaluation of a process or product.⁹ s-LCA has been applied to fossil and nonfossil alternatives; yet less research has been completed that is focused specifically on CO₂ utilization (Ekener et al. 2018; Fortier et al. 2019; Iribarren et al. 2022). While s-LCA has been more directly applied, there are documented limitations of LCA and decision making in addressing equity to date (Bozeman et al. 2022).

3.4.1 What Equity-Specific Tools for Assessment Can and Cannot Do in Assessing the CO₂ Utilization Value Chain

3.4.1.1 Equity Evaluation

Assessments of a product or technology seek to minimize uncertainty about the quantification of positive and negative impacts and form an interpretation basis for decision makers to consider when determining whether to proceed with a project. s-LCA is a means of integrating assessments of equity, justice, and other issues of social license into the assessment of a technology. Specifically, s-LCAs connect the methodological approaches of LCA to quantify a technology’s or a system’s environmental, societal, and equity impacts (Bouillass et al. 2021; Bozeman et al. 2022; Sala et al. 2015; UNEP 2020). Figure 3-11 compares (traditional) e-LCA and s-LCA throughout the standard phases of an LCA.

While the standardized phases of LCA provide a template for creating a comprehensive impact assessment across the life cycle of a technology or process, LCA and equity/justice efforts are not meaningfully linked in relevant policy but instead are treated as separate considerations. For example, with the 45Q carbon utilization tax credit, DOE engages LCA for environmental assessment, but not for societal implications. Instead, DOE addresses societal implications for project funding by requiring Community Benefits Plans as a means of attaining Justice40 Initiative outcomes (see Chapter 4 for more detail), which does not connect a proposed project’s equity and justice implications to the relevant product and process life cycle.

An s-LCA looks at the direct effect of evaluated products on stakeholders and the indirect impact of stakeholders on a product’s socioeconomic processes (Yang et al. 2020). These stakeholders can be grouped into five categories—workers/employees, local community, society, consumers, and value chain actors—to discuss related subcategories (i.e., fair salary, health and safety, and social benefits) (Toniolo et al. 2020), with particular indicators and requirements detailed in Table 3-5. These indicators and requirements inform an s-LCA approach to understanding and accounting for equity/justice impacts, but there are other frameworks grounded in other disciplines and fields of practice that may be appropriate to employ. For example, frameworks from economics suggest including nonmarginal changes, modeling approaches that provide information on the environmental burdens that occur, because of changes in demand (Almeida et al. 2020), using frameworks based on theories of resource allocation and competitive equilibrium for public goods (Foley 1966), and a life cycle cost (LCC) analysis framework that analyzes the cost-effectiveness of an investment over its economic lifetime (Norris 2001). Social psychologists advocate for frameworks based on equity theory, wherein individuals are the most satisfied when they experience a relatively equal exchange of resources, rather than being greatly overbenefited or underbenefited (Van Dijk and Wilke 1993),

___________________

⁹ The committee’s first report connected the broad impacts of carbon capture and utilization to cost-benefit analyses (CBAs), noting that the distributional effects of a project can be integrated with the framework of a CBA (although there may be additional normative criteria employed to reflect societal views) (NASEM 2023a). LCA provides a full life cycle framework for accounting and has established practices for reporting and comparing different impact categories stemming from the same product or process—providing an additional analytical approach for integrating social and equity considerations into an analysis.

while public health, political science, communications, and public policy scholars recommend including frameworks based on an individual’s core beliefs and values, or that of psychological constraint, along with the institutional framework of U.S. politics (Converse 2006; Druckman 2014; Feldman 1988, 2003).

3.4.1.2 Equity Assessment Tools and Uncertainties

No standardized s-LCA tools or methods are of similar quality to life cycle costing or e-LCAs (Reijnders 2022). Proposed methodologies give different weight to the role of local stakeholders and the need of a common social theory (Toniolo et al. 2020). Further challenging this issue, LCA and other decision-making tools have limited sociodemographic data, reducing the ability of practitioners to address important equity issues (Bozeman et al. 2022). Few studies have actively sought to improve s-LCA to include questions of distributional justice (Fortier et al. 2019), and the indicators presently used in s-LCA have been noted as insufficient to support the advancement of equity and justice (Bozeman et al. 2022; Greer 2023; Zeug et al. 2023). This limitation in s-LCA indicators points to the importance

TABLE 3-5 Stakeholder Categories, Indicators, and Corresponding Requirements

SOURCES: Adapted from Aparcana and Salhofer (2013); Arcese et al. (2013); Dreyer et al. (2010); Edvinsson et al. (2013); Feldman (2003); Foley (1966); Macombe and Loeillet (2014); Mattioda et al. (2017); Rajan et al. (2013); Toniolo et al. (2020); UNEP (2009); Van Dijk and Wilke (1993).

of advancing s-LCA methodology (Fortier et al. 2019) as well as combining s-LCA with more advanced tools that were developed specifically for equity. Community engagement tactics, for example, can help determine the weight of specific indicators in the s-LCA framework to further understand local outcomes.

Parameterizing social aspects generally requires judgment and cannot be determined by mathematics alone, so weighing what is critically important and considering community input in doing so is necessary to determine what metrics should be included and measured. The 2020 update to the United Nations Environment Programme and the Society for Environmental Toxicology and Chemistry Life Cycle Initiative s-LCA guidelines includes two approaches to data analysis: using generic databases or collecting and applying site specific information (UNEP 2020). For the former, determining localized issues of equity will be challenging. Equity assessments necessitate the collection of expert information including from impacted communities (Bradley et al. 2022a), meaning that generic s-LCA databases are unlikely to be sufficient in overcoming impacts. The guidelines emphasize the importance of site visits and working with relevant organizations for implementing s-LCA where site-specific information is used (UNEP 2020). Approaches that engage local communities hold promise for better integration with equity assessments but have yet to be implemented and will be challenging prior to proposing sites for facilities. Community engagement can provide insights to support equity beyond impact assessments and siting by including construction, operation, and decommissioning (Elmallah and Rand 2022). Participatory approaches that include residents of historically underserved communities also can improve the inclusion of minoritized¹⁰ people in infrastructure site selection (Hasala et al. 2020).

Appropriately implementing s-LCA requires understanding its capabilities as well as the elements outside its scope. Specifically, s-LCA can inform improvements to a product or system but cannot provide solutions for sustainable consumption or living or ensure equitable outcomes (UNEP 2009). While some s-LCA procedures use public engagement to address communities’ concerns and embed procedural justice principles into development processes (Bozeman et al. 2022; Fortier et al. 2019; Grubert 2023), incorporating public engagement and other participatory processes sometimes creates challenges related to societal acceptance and trust. The use of geospatial tools with social data layers, where possible, can support more equitable siting and inform aspects of the LCA—for example, resource consumption—to avoid increasing burdens or impacts on marginalized communities. Examples of such geospatial tools include EPA’s EJScreen and the Council on Environmental Quality’s (CEQ’s) list of tools and resources (CEQ n.d.; EPA 2024).¹¹ Further discussion of community engagement and environmental justice can be found in Chapter 4.

3.4.2 Use of S-LCA for Approaches and Technologies at Different TRLs

s-LCA is still under development and cannot provide comprehensive assessments for early-stage technologies. Including the potential effect of technology deployment on a local community will be difficult because even technical and economic indicators—for example, specific types of equipment, amounts of water needed, or costs of production—are highly uncertain. At best, the output of assessments can be used to begin informing the public on expected performance, likely land uses, noise, and so on. There is a risk that communities will react negatively because accurate, definitive information is lacking, potentially being misunderstood as attempts to evade the community’s critical concerns and needs (e.g., see Chailleux 2020 and Offermann-van Heek et al. 2018, 2020; see also Section 4.4.1.2 for more about how to strengthen public understanding of CO₂ utilization through

___________________

¹⁰ The word minoritized describes the dynamic with which the status of “minority” is imposed upon certain groups through the use of power and systems, not just the statistical status of being a minority. The term demonstrates that there are intentional power structures that have resulted in certain populations experiencing discrimination/disenfranchisement (e.g., see Benitez 2011, Stewart 2013, and Wright-Mair 2023).

¹¹ Many federal tools explicitly exclude race as a factor in establishing impacts on marginalized communities, a practice criticized by many equity and justice advocates (e.g., see Sadasivam 2023 and WHEJAC 2022). From a policy perspective, the race-neutral criteria will often survive legal pushback. For example, race was considered as a factor for the CEQ mapping tool. However, race was ultimately excluded owing to efforts to make the tool more robust and durable against efforts to either (1) dispose of the tool or (2) use its “focus” on race to delegitimize it to the public (Frank 2023; NASEM 2023b). Frank (2023) notes that criteria such as poverty levels and environmental risks “capture the breadth of ways that racial discrimination is felt in society.” However, equity and justice advocates still stress that policy “without explicit focus on race will not ultimately prioritize disadvantaged communities” (NASEM 2023b, p. 96). Policies and programs that acknowledge these shortcomings and incorporate supplemental equity and justice elements—such as meaningful community engagement or an explicit focus on environmental justice—could be beneficial for equity assessments. See Chapter 4 for more information about how to incorporate concepts of environmental justice into policy for the emerging CO₂ industry.

public engagement). s-LCA frameworks have not been designed to evaluate the societal impacts of early TRL technologies (i.e., technologies that may be deployed in a future sociotechnical system). Predicting how societal systems will develop is difficult, and because they are very context-specific (including geographic and temporal relevance), averages can be deceptive, as conditions may change drastically. Overall, there are ongoing discussions and coordination efforts on s-LCA with a focus on CO₂ conversion (McCord et al. 2023).

While not currently directed to s-LCA, Miller and Keoleian (2015) laid out a useful framework for transformative technologies that identified factors considered intrinsic, indirect, or external to the technology. Societal indicators could be incorporated into the indirect and external categories, delineating between what will be true for a given technology (intrinsic), and what may be true circumstantially (indirect and external).

3.4.3 S-LCA Goal, Scope, and System Boundaries

Following ISO standards and other guidelines, LCAs are not merely carbon accounting assessments but also provide information along multiple impact categories, including GHG emissions, water use, particulate matter emissions, eutrophication potential, and others. Given that results from LCAs are already structured to accommodate and present impacts in different categories, a congruent format exists for societal, equity, and justice indicators in the typical results reporting presentation. Toniolo et al. (2020) define two approaches to s-LCA: performance reference point methods (considering living and working conditions of workers at different life cycle phases) and impact pathways (considering the societal impacts using characterization models with indicators like those seen in e-LCAs).

The scope and level of data collection for different impact analyses and tools are detailed in Figure 3-12. The figure shows that s-LCA has both a significant system scope (the entirety or most of the product life cycle)

as well as data-gathering requirements from the process, facility/plant/site, and enterprise/management, in contrast to other analyses and tools that have more limited scopes and levels of data collection.

One database to support s-LCA is the Product Social Impact Life Cycle Assessment (PSILCA) database (PSILCA n.d.), which allows a practitioner to review industrial sectors and societal indicators and can be used to assess the societal impacts of products along life cycles (Ciroth and Eisfeldt 2016).

The end use of a CO₂-derived product is an important consideration in building public trust and acceptance of the CO₂ utilization sector. Products that contribute to a circular economy by reducing waste and enhancing sustainability are more likely to find public acceptance owing to their role in climate change mitigation, as discussed in Chapter 4. Climate change mitigation and the role of the end product in a sustainable value chain both embody principles and considerations around equity and justice—as communities currently overburdened by waste and pollution are unlikely to support systems that would set them up to bear the burdens of a new CO₂-derived product. As social and technical LCAs are completed, environmental and social justice principles support decision making that minimizes harm and does not exacerbate historical inequities.

Not all social and community considerations may fit into the traditional LCA framework, but life cycle thinking can help identify potential impacts at every aspect of the value chain (Clark 2023). When retrofitting a facility with carbon capture equipment or building a DAC facility, research into the surrounding community and its historical relationship to existing infrastructure, infrastructure development, and energy and water resources provides decision-support information critical to improving outcomes (Zuniga-Teran et al. 2021). The same can be said for the community near where the energy for a project is being provided or developed, as well as where the parts or materials for the final site will be manufactured. While environmental justice is a potential outcome of projects, the historical injustices in a particular community require careful consideration. Such injustices can be best understood through just practices that include community engagement and restorative justice (see Chapter 4). The global nature of product development, including the supply chains that may provide construction materials and the markets through which a CO₂-derived product might be distributed, also require a global climate justice application to ensure that procedures to reduce harm in one locality do not result in externalities that might create or exacerbate harms elsewhere.

To further develop the technical and localized societal aspects of an LCA, developers and researchers can consider the boundaries of what might comprise a community. For example, an LCA and a Refined LCA, detailing the initially proposed DAC plan and the final phase hub plan, respectively, are separately required from a Community Benefits Plan for the DAC Hubs Funding Opportunity Announcement (DOE 2023). The goal of the LCAs is to demonstrate robust accounting of full life cycle emissions. Opportunities to discuss the project’s relevance to the surrounding communities across the value chain are included in the portion of the LCA on impacts and discussion of potential co-benefits. If the captured CO₂ will be used for a product, the applicant must submit an LCA following the guidance document outlined by NETL (DOE 2023). This is a potential opportunity for a more robust discussion about incorporating social and equity considerations into LCA frameworks, especially when they are mandated by the federal government to receive funding. In the NETL guidance document, consideration of impacts from the product is limited to GWP and certain environmental impacts such as acidification, eutrophication, particulate matter emissions, and water consumption (Skone 2019). A narrow focus on traditional LCA impact categories will overlook potential societal concerns; considering s-LCA and equity considerations as part of funding decisions could play a substantial role in enacting environmental justice while moving toward tenets of a circular economy.

Use of resources like environmental product declarations (European Committee for Standardization 2020) conceivably could be adjusted and put in place to create LCA-type “scorecards” for certain products. Specific elements by which to judge the products across their lifetime (including disposal) would have to be agreed upon to further standardize and characterize their contribution to equity- and justice-related concerns and/or a circular economy. This is especially relevant if an incumbent product exists and the CO₂-derived version might have additional societal contributions that could either avoid or remedy injustices created by the existence or waste-stream of a particular product.

3.5 HANDLING CIRCULAR ECONOMY SYSTEMS

A key aspect when discussing CO₂ utilization technologies is their ability to achieve circular carbon chains. Circularity, in this context, refers to the transformation of atmospheric or biogenic CO₂ into a product, which would then rerelease the CO₂ at the end of its lifetime, making the flow of carbon circular. Such circularity

could enable compatibility of CO₂ utilization technologies with climate neutral targets, only if the CO₂ utilization technologies themselves resulted in no net life cycle emissions. Without lifetime extensions of products (i.e., sufficient duration of carbon storage), CO₂ utilization technologies using fossil emissions have already been questioned regarding their compatibility with climate goals as they may risk carbon lock-in (de Kleijne et al. 2022). If fossil CO₂ were embedded in products and looped through multiple uses over time, the storage of carbon in products could theoretically be extended to significant periods of time (e.g., >100 years). With such extended product lifetimes, technologies that use fossil CO₂ could conceivably be compatible with climate targets under certain conditions (e.g., the fossil carbon is not leaked into the atmosphere) (Malins et al. 2023; Ramírez 2022; Stegmann et al. 2022). Such use of carbon would be an inherent component, for example, with the bold systems changes required to improve the environmental sustainability of the plastics industry (Vidal et al. 2024). This strategy is rather controversial, as it will require stakeholders (producers, users, policy makers) to keep strict custody of the carbon through multiple value chains over long periods independently of the country where the products are produced and used. The controversy takes root not only in questions of how to develop practical and implementable tracking systems but also in noting that net-zero energy systems prioritize moving away from dependence on fossil fuels (e.g., ISO 2022).

Few studies systematically assess the potential and impacts of such a strategy, partly owing to limitations of existing methodologies to address this issue. On the one hand, there are limitations to the use of current LCA methodologies to assess the environmental performance of circular value chains that go beyond one or two loops. Such limitations are unsurprising, as LCA was designed to assess linear value chains. Methodological developments in LCA are being conducted to address this limitation, including the life cycle gaps framework that aims to measure the system losses—the so-called life cycle gaps—between an ideal closed system and the status quo (Dieterle et al. 2018). On the other hand, circular economy metrics are nascent and have been mostly developed to measure or simulate circularity at the macro and meso level (e.g., at the regional, national, or global level) rather than the product level. Further challenging the development of metrics has been the historical lack of standardization; the circular economy and related performance measurement have recently undergone standardization by ISO (2024). Multiple efforts are under way to develop product-level circularity indicators. Three examples are the Circularity Economy Index (Di Maio and Rem 2015), which focuses on recycling process efficiency; the Material Circularity Indicator (Ellen MacArthur Foundation and Granta 2015), which focuses on measuring the use of virgin material and resultant waste; and product-level circularity (Linder et al. 2017), which focuses on the ratio between recirculated and total economic product value. None of these indicators have been tested in potential products of CO₂ utilization.

In addition to resolving these methodological challenges, there are concerns with the feasibility of the option owing to challenges of traceability and access to reliable data, which are further compounded by the large number of datasets that would be required. Technically it is, for instance, possible to trace fossil (or biogenic) carbon flows using ¹⁴C methods (Palstra and Meijer 2010). This however would require that carbon flows are characterized, verified, and monitored throughout the different value chains over time, making this option difficult and likely to expensive to implement. One possibility that has been explored is the integration of blockchain technology and LCA (BC-LCA); however, the energy requirements warrant consideration. Frameworks proposed by Zhang et al. (2020) and Shou and Domenech (2022) focus on identifying key elements for BC-LCA, including mapping of value chains, identification of tracking methods, and data collection and validation protocols. The cases investigated in the literature examine products like leather for fashion and have yet to be applied to CO₂ utilization technologies.

3.6 CONCLUSIONS

3.6.1 Findings and Recommendations

Finding 3-1: Appropriate use of techno-economic and life cycle assessments—Techno-economic and life cycle assessments (TEAs and LCAs) provide insights into the economic and environmental performance of a technology, and in most cases, are discussed against a reference technology. TEA and LCA are most impactful at Technology Readiness Level (TRL) 3 and above and can inform deployment, but they also can be useful to understand limi-

tations at earlier TRL levels and inform applied research and development (R&D). The accuracy and certainty of the results depend on the representativeness of the input data, which will be more uncertain for earlier-stage technologies; therefore, it is important to perform TEA and LCA iteratively as R&D progresses and incorporate the insights generated from TEA and LCA into the design of the technology.

Finding 3-2: Requirements for use of techno-economic and life cycle assessments—Requirements for use of techno-economic and life cycle assessments are not consistent across DOE applied research, development, and deployment programs. Contradictory information and data gaps will persist globally without efforts to develop broadly accepted, international guidance.

Finding 3-3: Integration of techno-economic and life cycle assessments—Integration of techno-economic and life cycle assessments (TEAs and LCAs) can provide insightful decision-support, allowing practitioners to simultaneously assess viability and evaluate potential negative impacts to avoid potential conflicting outcomes of a TEA and an LCA where the best-case scenarios for both are achieved for separate conditions.

Recommendation 3-1: Requirements of techno-economic and life cycle assessments for CO₂ utilization—The Department of Energy applied offices and other relevant funding agencies such as the Department of Defense and the National Science Foundation Directorate for Technology, Innovation and Partnerships should develop and maintain consistent requirements for techno-economic and life cycle assessments for technologies at Technology Readiness Level (TRL) 3+ as part of funded applied research and development projects to support increasing economic competitiveness and decreasing negative impacts on the environment. Requirements should inform the basis for collaboration through international agencies and organizations such as the International Energy Agency in improving global guidance. Life cycle and techno-economic assessments (LCAs and TEAs) are not recommended requirements for funding of early-stage TRL projects, as they could constrain innovation. CO₂ utilization assessments developed and maintained by relevant agencies for TRL 3+ should

1. Report uncertainties and assumptions in the data input and model choice, as well as in the results.
2. Periodically reevaluate assumptions and revisit assessments as the technology advances. If the data are no longer relevant, a new TEA/LCA should be completed.
3. Justify the reference against which the technology is compared and assess the technology against more than one option that includes not only a fossil-based reference but also alternative carbon sources.
4. Integrate TEA and LCA to inform deployment by providing consistent economic and environmental guidance.

Finding 3-4: CO₂ purity for techno-economic and life cycle assessments—CO₂ purity can play a significant role in the techno-economic and life cycle assessments (TEAs and LCAs) of CO₂ utilization technologies. The quality and cost of purification has often been overlooked in the CO₂ utilization literature. Further understanding of the implications of CO₂ purity for TEAs and LCAs, as well as for the scalability of the technology, is needed to steer technology deployment.

Recommendation 3-2: Research needs for CO₂ purity in techno-economic and life cycle assessments—The Department of Energy (DOE) and other relevant funding agencies should fund projects that examine the robustness of CO₂ utilization technologies to different CO₂ purities, as well as fund further research and development of CO₂ purification technologies. Insights from these projects should be disseminated to the larger community by DOE. DOE should require awardees of applied research and development funding for CO₂ utilization technologies to perform techno-economic and life cycle assessments that explicitly address the purity requirements of the CO₂ streams.

Finding 3-5: Performing techno-economic and life cycle assessments—Conducting techno-economic and life cycle assessments (TEAs and LCAs) is difficult and prone to mistakes if not performed by expert assessors. In general, public availability and selection of input data can be a serious concern. Additionally, International Organization for Standardization (ISO) LCA standards provide clear guidelines for neither CO₂ utilization technologies nor for the associated infrastructure across life cycle stages, which is further complicated by the implications of CO₂ being alternately considered as waste, product, or by-product. There is a lack of skilled experts and accessible tools to guide researchers and developers, to provide input data for assessments and guide the correct interpretation of results and their uncertainties.

Recommendation 3-3: Facilitating techno-economic and life cycle assessments—The Department of Energy (DOE) should facilitate the execution of techno-economic and life cycle assessments (TEAs and LCAs) for >5 Technology Readiness Level (TRL) technologies by

1. Providing data preparation guidance for novice users of DOE LCA tools so that the data can be more easily analyzed.
2. Developing open-source software that guides users through LCA and TEA decision trees, similar to graphical user interface–supported interview-based software tools for tax reporting purposes.
3. Further supporting the development of user guidelines for an integrated approach to TEA-LCA, including how results can support decisions based on TRL (e.g., guiding research and development for earlier stage and supporting more sustainable procurement decisions as construction commences).
4. Continuing support for the development of databases by DOE and its national laboratories to ensure that information is readily available to complete robust TEAs and LCAs for CO₂ utilization technology and the associated infrastructure across life cycle stages.

Finding 3-6: Non-CO₂-emissions impacts—A meta-analysis of life cycle assessments performed on CO₂-derived chemicals points to the importance of including impacts beyond CO₂ alone: no CO₂-utilization-based chemical production alternative available at this time performed better in all impact categories than the conventional production technology, and furthermore, production from CO₂ may not yield the lowest global warming potential while still achieving significantly reduced environmental impacts.

Finding 3-7: Social life cycle assessment tools—Social life cycle assessment (s-LCA) is a means for integrating aspects of equity and other issues of social license into the assessment of a technology; however, there are no standardized s-LCA tools or methods that are of similar quality to life cycle costing or environmental life cycle assessments. s-LCA is particularly challenging for technologies that are at low technology readiness level. s-LCA is valuable but is not a way of quantitatively measuring social license or justice.

Recommendation 3-4: Non-CO₂-emissions impacts within life cycle assessments—The Department of Energy and other relevant funding agencies such as the U.S. Environmental Protection Agency and the National Institute of Standards and Technology should support research into improving evaluation of non-CO₂-emissions impacts within life cycle assessments (LCAs) of CO₂ utilization technologies, including

1. Evaluating the appropriate but differentiated applications for global and local impact categories, as the latter generally involves data and information with high spatial and temporal granularity (e.g., processes versus facilities, technology readiness level of various components of the technology).
2. Evaluating appropriate applications of social LCA (s-LCA) and further developing s-LCA tools and their potential integration with environmental LCA and techno-economic assessments.

Finding 3-8: Use of social life cycle assessments—Potential community impacts are often site-specific, and social life cycle assessment (s-LCA) can be better applied nearer deployment. Using s-LCA to evaluate local impacts does not replace the need for life cycle assessments for determination of broader environmental impacts.

Recommendation 3-5: Life cycle thinking for equity assessments—It is challenging to evaluate equity within life cycle assessment because methods are underdeveloped; therefore, the Department of Energy (DOE) should prioritize assessment of supply chains through principles of life cycle thinking to enable equity assessments that extend beyond the physical borders of the project site. DOE should require life cycle thinking for equity assessments when project sites are being considered, in order to identify hotspots and integrate risk and societal assessments. Relevant agencies, such as DOE and the U.S. Environmental Protection Agency, should evaluate life cycle assessment tools for their applicability to equity assessments and environmental justice, based on technology readiness level, time to deployment, and challenges and opportunities for selecting the sites of facilities.

Finding 3-9: Circularity of carbon products—The use of circularity strategies that keep carbon in products through multiple cycles of use and recycling, are starting to be considered as a way to significantly extend the storage duration of carbon, including fossil carbon. There is, however, a lack of understanding of the technical, economic, environmental, social, and policy implications of such strategies. This includes assessing technological performance of products over multiple cycles; evaluation of potential leakage over time; need for fresh raw materials; life cycle impacts beyond global warming potential; design of new value chains; business cases and policy and regulatory mechanisms that can drive circularity of carbon products; monitoring, verification, and reporting; and social acceptance.

Recommendation 3-6: Implications of circularity on carbon storage—The Department of Energy and the National Institute of Standards and Technology should support research that examines the feasibility and impacts of extending the duration of carbon storage through circularity strategies of short-lived products. This includes

1. Building on state-of-the-art life cycle assessment approaches that are able to address circularity of CO₂-derived products over time.
2. Development of approaches and tools that allow the traceability and custody of carbon across value chains over time, including mapping of value chains, identification of cost-efficient tracking methods, and data collection and validation protocols.

3.6.2 Research Agenda for LCA-TEA Use with CO₂ Utilization Technologies

Table 3-6 presents the committee’s research agenda for LCA and TEA of CO₂ utilization technologies, including research needs (numbered by chapter), and related research agenda recommendations (a subset of research-related recommendations from the chapter). The table includes the relevant funding agencies or other actors; whether the need is for basic research, applied research, technology demonstration, or enabling technologies and processes for CO₂ utilization; the research themes) that the research need falls into; the relevant research area and product class covered by the research need; whether the relevant product(s) are long- or short-lived; and the source of the research need (chapter section, finding, or recommendation). The committee’s full research agenda can be found in Chapter 11.

TABLE 3-6 Research Agenda for Life Cycle and Techno-Economic Assessment of CO₂ Utilization Technologies

NOTE: ARPA-E = Advanced Research Projects Agency–Energy; BES = Basic Energy Sciences; DoD = Department of Defense; DOE = Department of Energy; DOT = Department of Transportation; EERE = Office of Energy Efficiency and Renewable Energy; EPA = U.S. Environmental Protection Agency; FECM = Office of Fossil Energy and Carbon Management; FHWA = Federal Highway Administration; NSF = National Science Foundation; OSMRE = Office of Surface Mining Reclamation and Enforcement; OST-R = Office of the Assistant Secretary for Research and Technology; USGS = U.S. Geological Survey.

3.7 REFERENCES