Read the Report

Executive Summary

Despite growing efforts to drastically cut carbon dioxide (CO2) emissions and address climate change, energy outlooks project that the world will continue to rely on certain products that are currently carbon-intensive to produce but have limited alternatives, such as aviation fuels and concrete. Recycling CO2 into valuable chemicals, fuels, and materials has emerged as an opportunity to reduce the emissions of these products. In this way, CO2 recycling is a potential cornerstone of a circular carbon economy that can support a net-zero future. However, CO2 recycling processes have largely remained costly and difficult to deploy, underscoring the need for supportive policies informed by analysis of the current state and future challenges of CO2 recycling.

This report, part of the Carbon Management Research Initiative at Columbia University’s Center on Global Policy, examines 19 CO2 recycling pathways to understand the opportunities and the technical and economic limits of CO2 recycling products gaining market entry and reaching global scale. The pathways studied consume renewable (low-carbon) electricity and use chemical feedstocks derived from electrochemical pathways powered by renewable energy. Across these CO2 recycling pathways, the authors evaluated current globally representative production costs, sensitivities to cost drivers, carbon abatement potential, critical infrastructure and feedstock needs, and the effect of subsidies. Based on this analysis, the paper concludes with targeted policy recommendations to support CO2 recycling innovation and deployment.

Key findings of the analysis include the following:

  • CO2 recycling pathways could deliver deep emissions reductions. When supplied by low-carbon electricity and chemical feedstocks, CO2 recycling pathways have the combined potential to abate 6.8 gigatonnes of CO2 per year (GtCO2/yr) when displacing conventional production methods.
  • Some CO2 recycling pathways have reached market parity today, while the costs of remaining pathways are high. Electrochemical carbon monoxide (CO) production, ethanol from lignocellulosic biomass, concrete carbonation curing, and the CarbonCure concrete process all have an estimated cost of production (ECOP) lower than the product selling price. These pathways have a combined carbon abatement potential of 1.6 GtCO2/yr. Most remaining pathways have an ECOP of 2.5 to 7.5 times greater than the product selling price. In particular locations and contexts, ECOP may be substantially lower, but these costs are representative of CO2 recycling at global scale.
  • Catalyst performance and input prices are the main cost drivers. The largest component of ECOP is electricity and chemical feedstock costs, and the main cost drivers are those who influence these two cost components. For electrochemical pathways, ECOP is most sensitive to catalyst product selectivity (the ability of the catalyst to avoid unwanted side reactions), catalyst energy efficiency, and electricity price. For thermochemical pathways, the largest cost drivers are product selectivity, chemical feedstock price, and the price of the electricity used to make the feedstocks.
  • CO2 recycling at the scale of current global markets would require enormous new capacity of critical infrastructure. Each pathway at global scale would consume thousands of terawatt hours of electricity, 30–100 million metric tons (Mt) of hydrogen, and up to 2,000 Mt of CO2 annually. This would require trillions of dollars of infrastructure per pathway to generate and deliver these inputs, including a combined 8,400 gigawatts (GW) of renewable energy capacity and 8,000 GW of electrolyzer capacity across all pathways.

Based on these findings, the authors recommend the following set of policy actions:

  • Ensure CO2 recycling pathways are fed by low-carbon inputs. Without low-carbon electricity and feedstocks, CO2 recycling could potentially be more carbon-intensive than conventional production.
  • Prioritize certain pathways strategically. CO2 recycling methane and ethane production are extremely uneconomic and should be deprioritized. All other pathways are more economically promising and could be the focus of a targeted innovation agenda to reduce costs. In addition, the following pathways that have an ECOP less than 5 times the selling price could be prioritized for early market growth: electrochemical CO production, green hydrogen, ethanol from lignocellulosic biomass, concrete carbonation curing pathways, CO2 recycling urea production, and CO2 hydrogenation to light olefins, methanol, or jet fuel.
  • Target research, development, and demonstration (RD&D) to catalyst innovation to bring down ECOP and reduce input demand. Policy makers can promote RD&D to improve the selectivity and energy efficiency of CO2 recycling catalysts. By decreasing a pathway’s consumption of electricity and feedstocks, these innovations would both decrease ECOP and alleviate the sizable critical infrastructure needs.
  • Create demand pull for early market CO2 recycling products. Governments can use demand pull policies such as public procurement standards to bolster early markets for the most mature CO2 recycling pathways.
  • Promote build-out of critical infrastructure. To provide for the substantial infrastructure needs of CO2 recycling, policy makers can seek to remove barriers to and catalyze investment in building renewables installations, transmission lines, electrolyzers, and CO2 transport pipelines.

Introduction

To mitigate the severe consequences of the climate crisis, the world must reach net-zero greenhouse gas emissions by midcentury and net-negative emissions shortly thereafter.[1] Current decarbonization efforts have focused on efficiency improvements (e.g., US CAFE standards and appliance standards), increased build-out of renewable electricity (e.g., Germany’s Energiewende), and electrification of heating and automobiles. Many studies indicate that despite recent progress, existing policies will leave substantial residual emissions in some sectors, such as heavy-duty transportation or heavy industry, that are either expensive or infeasible to completely decarbonize.[2] Moreover, many products in our day-to-day lives, be they chemicals, fuels, or materials, contain substantial carbon volumes themselves that will eventually oxidize and form CO2 in the atmosphere.

One way to abate emissions from these sectors involves CO2 recycling, wherein CO2 is converted into valuable chemicals, fuels, and materials and used in commercial products.[3] The CO2 used can be sourced from point-source emitters such as fossil power plants or be drawn directly out of the atmosphere or oceans. This can be considered a component of a circular carbon economy that includes CO2 reduction, reuse, recycling, and removal.[4] By consuming CO2 as a feedstock, these pathways could replace difficult-to-decarbonize products and processes with low-carbon, carbon neutral, or net carbon-negative alternatives. Since CO2 recycling products can help reduce sources of residual emissions, as in the potential case of producing net neutral fuels for the transportation sector, these pathways will be crucial to achieving net-zero emissions.

CO2 recycling pathways include many processes and products with different costs, viability, and carbon footprints. To contribute meaningfully toward deep decarbonization, CO2 recycling pathways must be deployed on the gigatonne scale and must operate with extremely low-carbon energy and material inputs.[5] Investment and commercialization of CO2 recycling requires a measured approach informed by chemistry, engineering, and economics. A misunderstanding of the merits and limitations of each pathway risks poor use of capital, lost time, and ineffective carbon abatement measures, limiting the abatement achieved through CO2 recycling pathways. Coordinating an effort of this scale and complexity will require a comprehensive quantitative analysis of the techno-economics of key CO2 recycling pathways and the critical infrastructure needed to support them. Several previous studies have investigated the techno-economics of various CO2 utilization pathways,[6] but they often do not address a wide range of pathways and do not integrate techno-economic findings with infrastructure needs or broader policy implications.

In this study, we investigate the techno-economics, critical infrastructure needs, and effects of subsidies for 19 of the chemical and industrial CO2 recycling processes with the largest volume of product demand. These include eight electrochemical pathways and eleven thermochemical pathways. We evaluate the potential for these CO2 recycling pathways to replace conventional production methods and supply current global demand for their products. We model these pathways to consume renewable electricity and use chemical feedstocks, such as hydrogen, that are produced electrochemically with renewable electricity. We evaluate several characteristics of each pathway: the current estimated cost of production (ECOP, in $/metric ton product), global carbon abatement potential, marginal cost of carbon abatement, effective carbon price and scale of subsidies needed to make the pathways profitable, and the critical infrastructure necessary to deploy the pathway at global scale. We also use a cost sensitivity analysis across a range of techno-economic inputs to identify specific innovations that can drive cost reductions for the pathways. We make policy and investment recommendations based on these findings to accelerate deployment of CO2 recycling across sectors at reasonable costs and ultimately realize gigatonne-scale CO2 recycling for deep emissions reductions.

Selected CO2 Recycling Pathways

The CO2 recycling pathways analyzed in this study are listed in Table 1. We investigated 8 electrochemical pathways and 11 thermochemical pathways. Each pathway is defined by the pairing of a chemical process (e.g., CO2 hydrogenation) and its desired product (e.g., methanol). Several of the chemical processes can be adapted to target various products, and in these cases, we evaluate each process-product pairing as a distinct pathway. The electrochemical pathways include water electrolysis to produce hydrogen and electrochemical CO2 reduction, in which CO2 and water react to produce a range of hydrocarbon products. The thermochemical pathways include CO2 hydrogenation, which uses CO2 and H2 feedstocks to produce various hydrocarbons, and Fischer-Tropsch (F-T) synthesis, which uses syngas (CO and H2) to create various hydrocarbons. The ethanol from lignocellulosic biomass pathway breaks down and ferments the sugars in lignocellulosic biomass feedstocks, such as agricultural residues, to produce ethanol. The urea production pathway is modeled as a Bosch-Meiser process that uses green ammonia and externally sourced feedstock CO2. The concrete production pathways are each modifications to one step of conventional concrete production in order to incorporate CO2 into the concrete. We model these as identical to conventional concrete production, save for the modified CO2 recycling step. A detailed description of each pathway is included in the appendix.

Methodology

We designed our analysis in the context of using CO2 recycling pathways to supply current global demand for their products. Our analysis is outlined here and detailed in the appendix. The general categories of analysis for each CO2 recycling pathway were current production costs, sensitivities to cost drivers, carbon abatement potential, critical infrastructure and feedstock needs, and the effect of subsidies.

Estimated Cost of Production (ECOP) and Sensitivity Analysis

The current estimated cost of production (ECOP, in $/metric ton product) for each pathway was determined by estimating the electricity cost, feedstock cost (hydrogen, CO2, CO, biomass, and/or ammonia), fixed operation and maintenance (O&M) cost, and capital cost per ton of product produced. The key techno-economic inputs used for calculating ECOP were electrical energy efficiency, product selectivity and yield, capacity factor, capital cost, electricity price, and feedstock prices. Product selectivity broadly refers to the percentage of input feedstock and/or energy that goes toward making the desired product as opposed to undesired byproducts. More detailed definitions of the different selectivity metrics we used and how they were calculated can be found in the detailed methodology in the appendix.

To determine the ratio of a certain chemical feedstock consumed by a pathway (in tons of feedstock consumed per ton of product), we divided the ideal feedstock consumption ratio by the appropriate selectivity metric. The ideal feedstock consumption ratio is defined by the reaction’s stoichiometry (the ratio of molecules of feedstock to molecules of product in the chemical reaction equation). Dividing by selectivity accounts for the additional feedstock that is consumed to produce reaction byproducts. To calculate CO2 consumption through this method, we used the carbon selectivity of the pathway—the selectivity to the desired product among all carbon-containing products. This allowed us to exclude the hydrogen evolution side reaction in electrochemical CO2 reduction, which does not consume CO2. Similarly, to find the H2 consumption, we used the hydrogen selectivity, which gives the percentage of input hydrogen that goes to the desired product. This excludes CO2 conversion to CO in many thermochemical pathways, which does not consume H2.

For a cost sensitivity analysis, the value of each techno-economic input was increased and decreased by 20 percent, with all else remaining fixed, and the resulting percent change in ECOP was noted. In designing the tool for these analyses, we varied input parameters broadly to understand sensitivities; a future analysis will detail the sensitivities to inputs.

Carbon Abatement Metrics and Effect of Subsidies

The intensive carbon abatement (tCO2 abated/metric ton product) of each pathway was defined as the net decrease in CO2 emissions that would result from displacing a conventional production process with a CO2 recycling pathway that produces the same product. Since a CO2 recycling product and its corresponding conventional product are functionally identical, we assume that the emissions associated with the combustion or use of the products (gate to grave emissions) are equivalent. Under this assumption, the change in emissions from displacing conventional processes with CO2 recycling arises entirely within the cradle to gate life cycle scope. Thus, carbon abatement was calculated through a cradle to gate life cycle assessment incorporating the CO2 consumed in the chemical reaction, the emissions associated with the generation of renewable electricity consumed in the process, the natural gas-derived process heat emissions, the emissions associated with carbon capture to produce CO2 feedstock, byproduct CO2 generation, and the counterfactual cradle to gate emissions of producing the product using a conventional pathway. We multiplied the intensive carbon abatement of a pathway by the global demand for its product to obtain the global carbon abatement potential of the pathway in million metric tons of CO2 per year (MtCO2/yr).

The marginal abatement cost (MAC) was found by dividing ECOP by intensive carbon abatement for each pathway. The marginal abatement revenue ($ of revenue/metric ton product) was calculated by instead dividing the product selling price by the intensive carbon abatement of the pathway. By subtracting the marginal abatement revenue from MAC, we determined the net marginal abatement cost after revenues, which we took as equivalent to the effective carbon price needed for the pathway to reach market parity. Using these effective carbon prices, we plotted the sum of the carbon abatement potentials of all pathways that have reached market parity as a function of carbon price. In the case of repeated products, we excluded the pathway with the higher MAC after revenues. We included both concrete production pathways because they apply to different types of concrete. Separately, we calculated the gross subsidies needed to close the cost-price gap for each pathway by multiplying the difference between selling price and ECOP by the global demand for the product.

Global Critical Infrastructure Needs and Resource Consumption

The current global consumption of each pathway’s product was used as a basis for all extensive calculations. Using this basis, the critical infrastructure requirements of renewable electricity generation, electricity transmission, electrolyzer capacity, and CO2 pipeline transport networks were analyzed for each global-scale process with the goal of providing some coarse estimate of potential costs and infrastructure needs. The global TWh of renewable electricity required for electrochemical pathways was calculated using the electricity consumption per metric ton of product for each pathway, the global consumption volume, the electrical energy efficiency, and the faradaic efficiency—the selectivity of input electrical current to the desired product. For thermochemical pathways, the global renewable electricity consumption is the electricity consumed to produce the pathway’s feedstocks plus the comparatively smaller amount of direct electricity consumption of the thermochemical plant. Since the vast majority of conventional chemical production and industrial energy use does not come from electricity,[7] and current electricity mixes are mostly fossil-based, we assume that the conventional processes being displaced by CO2 recycling use no renewable electricity. Therefore, all renewable electricity consumption from deploying CO2 recycling has full additionality. The intensive hydrogen consumption of the thermochemical pathways (tH2/metric ton product) was found by dividing the stoichiometric hydrogen consumption of the pathway by the hydrogen selectivity. This value was then multiplied by global demand for the product to determine the global hydrogen consumption of the pathway.

For each pathway, we determined the necessary renewable electricity generation capacity in gigawatts (GW) using the global electricity consumption and the capacity factor. We similarly calculated the needed electrolyzer capacity in GW using the electrolyzer electricity consumption associated with the pathway and the capacity factor. We translated these GW capacity values into total capital costs by multiplying the capacity by the unit capital cost ($/kW). The capital cost of the necessary CO2 transport pipeline infrastructure for each pathway is based on the global CO2 consumption of the process. The global CO2 consumption was found by multiplying the global product demand by the units of feedstock CO2 consumed to produce one unit of product. The CO2 transport capital cost was then calculated by multiplying global CO2 consumption by a CO2 pipeline network capital cost of $42/tCO2/yr capacity.[8] This capital cost is based on a CO2 pipeline network capable of transporting 20 MtCO2/yr, consisting of a 500 km central spine with two 10 km distribution pipelines on each end.[9] This provides a minimal cost estimate, as lower capacity, smaller diameter pipelines would cost more.

We also determined the total infrastructure costs, GW capacities, and carbon abatement potential summed across all CO2 recycling pathways. In the case of multiple pathways that produce the same product, we included only the pathway with the largest value being summed and excluded the others from the total. We also excluded ethane pathways from the total, since the current leading use of ethane as a feedstock to produce ethylene would be eliminated if ethylene was produced entirely through CO2 recycling. Similarly, we excluded electrochemical CO production from the total since the majority of global demand for CO is as a component of syngas, and other pathways separately account for syngas production. We included both concrete production pathways since they apply to different types of concrete. Finally, we calculated the totals with and without methane production, since the extremely large global demand for methane eclipses the contribution of all other pathways and because CO2 recycling methane pathways were found to have limited prospects of supplying large-scale methane demand.

Inputs and Assumptions

We designed both the electrochemical and thermochemical CO2 recycling pathways to consume low-carbon electricity, here modeled as renewable energy, and assumed the H2, CO, and/or ammonia consumed as a reactant in thermochemical pathways are supplied by onsite electrochemical processes powered by renewable energy. All numerical assumptions are detailed in Table A1, A2, and A3 in the appendix. We chose globally representative input values to correspond to the global scale of this analysis.

We assume the feedstock CO2 is sourced from carbon capture at point-source emitters and delivered to the CO2 recycling plant via a CO2 pipeline network at a total cost of $50 per metric ton of CO2 (tCO2).[10] While lower carbon capture costs may be available in certain contexts, these lower costs will not be accessible on average for the global scope and scale of this study. For the thermochemical pathways that consume electrochemically generated feedstocks, we use our own calculated ECOP of the electrochemical feedstock production pathway (e.g., water electrolysis for feedstock green H2) as the feedstock price.

Though similar studies often assume renewable electricity prices based on the levelized cost of energy (LCOE) for renewable generators or average power purchase agreement (PPA) prices for renewables, these methods do not reflect the duty cycles of industrial use, contributions of network costs (costs of transmission and distribution), and electricity taxes to the end-use industrial electricity price paid by producers. While producers may have access to PPA-range renewable electricity prices at certain times, in limited contexts, and with low capacity factors, these low prices will not be available on average for the global production scale evaluated in this study. Therefore, for our estimate of the renewable electricity price paid for global-scale CO2 recycling processes, we must incorporate the duty cycle requirements and contributions of network costs and taxes beyond renewable PPA prices.

To estimate the renewable electricity price used in this study, we take renewable PPA prices in a region to be equivalent to the wholesale price of renewable electricity. We calculate the corresponding industrial price of renewable electricity by increasing the PPA price to reflect the additional contributions of network costs and taxes to industrial prices. Wholesale electricity prices account for 10–70 percent of industrial electricity prices in different countries.[11] We assume that renewable PPA prices on average account for the same percentage of the final renewable electricity industrial price in a particular country. To estimate the renewable electricity price used in this study, we assume that CO2 recycling producers have access to industrial electricity markets and that wholesale electricity prices represent 50 percent of the price of industrial electricity on a global basis. We use an average global renewables PPA value of 4.75 ¢/kWh[12] and divide it by 0.5 to obtain an industrial electricity price for renewables of 9.5 ¢/kWh ($0.095/kWh). In keeping with this framing, we assume a capacity factor of 50 percent for electrochemical pathways supplied by renewable electricity, which was chosen to represent a combination of wind and solar with complementary output profiles.[13] Since thermochemical pathways have very small direct electricity consumption and therefore are not as limited by the availability of renewable resource, we assume a higher capacity factor of 90 percent.

An obvious condition from this cost framing is that the true costs of recycling CO2 will vary substantially across regions and jurisdictions and that some locations may provide high-capacity electricity at much lower costs. These sensitivities and ranges will be published separately.

In our techno-economic analysis, we choose values for faradaic efficiency, carbon selectivity, and hydrogen selectivity that are 20 percent lower than those of the highest performance catalysts reported in the literature. The 20 percent decrease from the highest performance values was applied to reflect that many of the cited catalysis studies at the bench scale do not demonstrate industrially relevant production rates (measured as current density or space-time yield), and industrial demonstrations of these pathways today would not achieve bench-scale performance. We obtain energy efficiencies for electrochemical pathways from the same studies. Our efficiency and selectivity assumptions, along with the associated references, are summarized in Table A1. We assumed an electrolyzer capital cost of $1,000/kW for water electrolysis.[14] For electrochemical CO2 reduction pathways, since these technologies have not reached commercial scale and CO2 electrolyzer capital cost data is therefore not available, we modeled the electrolyzer based on water electrolyzers with a capital cost of $1,000/kW. For thermochemical pathways, we obtained capital cost estimates from plant simulations in the literature (Table A2). The annual fixed O&M costs were estimated as a small percentage of the total capital cost (Table A3).

We included the emissions associated with renewable electricity generation in our carbon abatement life cycle assessment. Renewable power has life cycle emissions of roughly 10–40 gCO2/kWh,[15] and we assume an average renewable electricity emissions intensity of 25 gCO2/kWh.

Results

A large and diverse set of pathways take CO2 as an input to produce value-added products. We reviewed 61 of such CO2 recycling pathways (Table A5) and further analyzed 19 of the pathways that produce products with the associated highest global demand in Mt/yr (Table 1). We evaluated green H2 production by water electrolysis—though this does not strictly qualify as CO2 recycling—because low-carbon hydrogen is a key required input for many CO2 recycling pathways. Our findings are detailed below, and key findings are summarized in Table A4 in the appendix.

Estimated Cost of Production

The current estimated cost of production (ECOP) gives the cost of producing a ton of product through a particular CO2 recycling pathway and provides a basis to compare the economics of each pathway. We calculated a globally representative ECOP for each CO2 recycling pathway. We define a pathway as “profitable” and at market parity if its ECOP is lower than the product selling price. The ECOP comprises capital cost (CAPEX), fixed operation and maintenance (O&M) costs, electricity costs, and costs of various feedstocks. Figure 1 shows the ECOP of each pathway and the breakdown of ECOP between each cost category. The findings show the ECOP for most pathways is well above the product selling price.

For electrochemical processes, ECOP values are high—most are above $4,000/metric ton product—and greatly exceed the corresponding product selling prices, which lie below $1,000/metric ton product. Electrochemical carbon monoxide (CO) production is a notable exception with a low ECOP of $546/metric ton product, below the CO selling price.

Electricity costs comprise the large majority of ECOP for electrochemical processes. Electricity is the main (or only) energy input used in electrochemical pathways to convert low-energy CO2 molecules into high-energy products. Due to relatively high estimated industrial renewable electricity prices today and large amounts of electricity input needed, electricity costs predominate in ECOP. While renewable electricity prices may fall in the future, we do not include projected prices here. Fixed O&M, CAPEX, and CO2 feedstock costs account for similar portions of the remainder of ECOP.

The thermochemical CO2 recycling pathways show slightly lower ECOPs than the electrochemical pathways, with most thermochemical ECOP values below $5,000/metric ton product and smaller margins between ECOP and selling price. Ethanol production by lignocellulosic biomass fermentation has a lower ECOP than the product selling price, making the process profitable. Similarly, the two concrete production CO2 recycling pathways, which modify a single step of conventional concrete production, add minimal costs to the conventional process, maintaining an ECOP below the selling price.

For thermochemical pathways, the cost of feedstock green hydrogen typically accounts for the majority of the ECOP. Hydrogen consumed as a reactant in thermochemical CO2 recycling reactions is often the largest source of energy input to the process. Since in this analysis all feedstock hydrogen is produced through renewable-powered electrolysis, the price of renewable electricity plays a prominent indirect role in thermochemical ECOPs. The cost of feedstock CO2 is typically a minor component of ECOP for thermochemical pathways. For Fischer-Tropsch synthesis processes, which use syngas—a mixture of carbon monoxide and hydrogen—as a feedstock, carbon monoxide and hydrogen costs are the largest component of ECOP. Once again, this feedstock syngas is modeled to be produced through electrochemical CO2 reduction, making the renewable electricity price an important indirect contributor to ECOP for Fischer-Tropsch.

While electricity costs comprise the largest part of electrochemical ECOPs and green hydrogen/CO are the largest component of thermochemical ECOPs, completely eliminating these costs would still leave ECOPs that are near or above selling price for many pathways. For these pathways to reach market parity, other components of ECOP, such as capital cost and fixed O&M, must decrease along with the major cost components.

A few cost trends are also apparent between pathways. By and large, C2+ products—molecules with two or more carbon atoms in a chain—have higher ECOPs than single-carbon (C1) products, which is to be expected. Electrochemical and thermochemical production of C2+ products, such as ethylene or jet fuel, often have a lower energy efficiency and lower selectivity compared to the synthesis of C1 products due to the high energy barrier to C-C bond formation and competition from the more facile C1 pathways.

The gravimetric energy density of the products (Table A1) also plays a major role in shaping ECOP trends since more energy-dense products require a larger input of energy to produce, driving up total electricity or feedstock hydrogen costs. Ethanol, for instance, has a gravimetric energy density over 40 percent lower than that of the other C2 products evaluated and accordingly has a lower ECOP than the other C2 pathways. The high and low gravimetric energy densities of hydrogen and carbon monoxide, respectively, similarly factor into these products’ outlier ECOP values.

The ratio of ECOP to product selling price (Figure 2) provides a clearer comparative view of how close each pathway is to reaching market parity. A ratio of less than 1 indicates that ECOP is lower than the selling price and the pathway has reached market parity, as we see for electrochemical CO production, ethanol from lignocellulosic biomass, and the two concrete production pathways. The methane and ethane production pathways have extremely high ratios, suggesting that it may not be feasible for these pathways to ever reach market parity. Most pathways have ECOP to selling price ratios of up to 7.5, highlighting a wide gap between ECOP and selling price that demands strong policy efforts to close this gap. As we will discuss in our policy recommendations, this information can be used to help prioritize the pathways with lower ECOP to selling price ratios that may more readily reach market parity.

Cost Sensitivity Analysis

To achieve market parity, the ECOP of most CO2 recycling pathways must decrease substantially. To assess which performance improvements or component cost reductions would be most effective, we analyzed the cost sensitivity to key inputs and functional parameters. The parameters evaluated include the electrical efficiency, electricity price, capacity factor, capital cost, various feedstock costs, and product selectivity. The product selectivity broadly describes the percentage of input feedstock and/or energy that goes toward making the desired product as opposed to undesired byproducts, as detailed in the appendix. Figure 3 shows the percent change in a pathway’s ECOP that results from increasing and decreasing an input value by 20 percent, with all else equal. The concrete production pathways are excluded since their costs are not defined by the metrics analyzed here, such as product selectivity and feedstock prices. This analysis demonstrates how sensitive ECOP is to an arbitrary but consistent percent change in each input, to help identify the main cost drivers for each pathway. Since this analysis focuses on sensitivity, it does not evaluate the extent or feasibility of cost reductions possible through improvements in each input and does not imply that a particular input value can or will change by 20 percent in the future. The extent of improvements available in the value of each input sets an important limit for ECOP reduction that should be considered alongside cost sensitivity.

For electrochemical pathways, the most significant techno-economic cost drivers are electrical energy conversion efficiency and product selectivity. This is because electricity represents the largest component of cost, and low efficiency and selectivity directly increase the overall electricity consumption of the pathways through energy conversion and faradaic efficiency losses, respectively. These two inputs are followed closely in importance by the electricity price itself.

In contrast, changes in capacity factor and electrolyzer capital cost have a relatively small impact on ECOP. These inputs directly affect CAPEX and fixed O&M costs, which comprise small portions of ECOP. Similarly, feedstock CO2 price is typically the weakest cost driver since CO2 input costs are a small portion of electrochemical ECOPs. This suggests an innovation agenda focused initially on electrochemical selectivity and conversion efficiency as opposed to capital cost reduction for electrolyzers and renewable electricity price reduction.

For thermochemical pathways, selectivity, electricity price, and green hydrogen price are the main cost drivers. In contrast with electrochemical pathways, most of the thermochemical processes receive the majority of their energy input from the hydrogen used as a reactant in the chemical reactions, with heat inputs and direct electricity use contributing relatively small input energy. As a result, ECOP is particularly sensitive to green hydrogen prices and to electricity prices that highly influence green hydrogen production costs. Since a lower selectivity directly increases hydrogen consumption for undesired reaction pathways, selectivity is the most prominent driver of ECOP. The capacity factor and capital cost for thermochemical pathways have a comparatively small influence on ECOP, since CAPEX makes a much smaller contribution to ECOP than hydrogen costs do. The electrical energy efficiency of plant equipment is the least significant ECOP driver for most pathways since the direct electricity consumption of the plant is dwarfed by the energy input from feedstock hydrogen.

For all pathways, the CO2 feedstock costs have a relatively small influence on ECOP. In real markets, the price of feedstock CO2 varies considerably based on the source of that CO2. This paper assumes feedstock CO2 is obtained through carbon capture from the flue gas streams of point-source emitters such as steel mills and purchased at a price of $50/tCO2.[16] A key alternative potential source of CO2 is direct air capture (DAC), which removes CO2 directly from the atmosphere. Since DAC does not rely on continued use of fossil fuels as a source of feedstock CO2, DAC could potentially avoid the upstream emissions associated with fossil fuel production that exist in most point-source carbon capture scenarios, which can be substantial. DAC would provide feedstock CO2 at higher prices, currently estimated between $250 and $600/tCO2,[17] though this cost could drop appreciably with broader deployment.[18] DAC might provide a fossil-free source of CO2 as DAC technology matures in the medium to long term. Since feedstock CO2 price has a small influence on ECOP, the higher price of feedstock CO2 from direct air capture would not greatly influence final product costs. For instance, increasing the CO2 feedstock price by 400 percent from $50/tCO2 to $250/tCO2, a price consistent with DAC, would only increase the ECOP of electrochemical CO2 reduction to ethylene by 19 percent.

Carbon Abatement Potential

Since reducing CO2 emissions is the key motivation of CO2 recycling, we calculated the carbon abatement potential of each pathway (Figure 4). This is based on a life cycle assessment to find the net decrease in emissions that would result from completely replacing conventional production pathways with CO2 recycling pathways. This analysis incorporated the feedstock CO2 consumption of the recycling pathway, the CO2 emissions associated with the recycling pathway, the emissions of the counterfactual incumbent production process that is displaced, and the global demand for the product.

The global carbon abatement potential of most CO2 recycling pathways lies between 150 MtCO2/yr and 1,700 MtCO2/yr (Figure 4a and 4b). These extensive carbon abatement potential values reflect both the product’s market size and the pathway’s intensive carbon abatement per ton of product. For example, urea has a low carbon content and therefore a low intensive carbon abatement per ton of product, leading to its relatively low carbon abatement potential of 138 MtCO2/yr. In contrast, the high carbon abatement potentials of the methane pathways largely reflect the product’s high global demand.

We estimated the sum carbon abatement potential for these CO2 recycling pathways together by adding the abatement potentials of each pathway, excluding the outlier methane pathways and redundant products as described in the methodology. The combined carbon abatement potential of the included pathways is 6.8 gigatonnes of CO2 per year (GtCO2/yr), or 16 percent of the 2019 global CO2 emissions of 43.1 GtCO2.[19] Adding in the abatement potential contribution from Sabatier process methane production, the more viable of the two methane pathways, the maximum abatement potential is 16 GtCO2/yr.

The intensive carbon abatement gives the tons of CO2 abated per ton of product made through each pathway based on the aforementioned life cycle emissions assessment (Figure 4c and 4d). Importantly, all pathways evaluated have a positive carbon abatement value under our assumption that the pathways use low-carbon inputs, indicating that using a CO2 recycling pathway to displace conventional production would indeed reduce CO2 emissions. Many pathways have carbon abatement values between 2 and 9 tCO2/metric ton of product, demonstrating an appreciable amount of carbon abatement for each unit of CO2 recycling production. The two concrete pathways have particularly low intensive carbon abatements. This is because, during CO2 curing of concrete, CO2 is incorporated at a relatively low weight percent into solely the cement component of concrete, and cement only accounts for an estimated 12.5 percent of the total mass of concrete.[20] However, the very large global demand for concrete leads to appreciable global carbon abatement potentials for concrete pathways. Urea also has a relatively low intensive carbon abatement, partly because urea has a low gravimetric carbon content, meaning a small amount of CO2 is consumed in the urea-producing reaction. Ethylene and jet fuel pathways have large intensive carbon abatements due to the high carbon content of these products. Hydrogen, on the other hand, has a large carbon abatement because the conventional pathway it displaces, steam methane reforming, has high emissions of roughly 12 tCO2/metric ton H2.

Marginal Abatement Cost and Revenue

To assess the economics of carbon abatement, the marginal cost of reducing carbon emissions, or marginal abatement cost (MAC), provides an initial economic metric for evaluating these pathways. The MAC gives the cost of each pathway in dollars per metric ton of CO2 abated and was calculated by dividing ECOP by intensive carbon abatement. Since the products of these pathways will be sold for revenue, we also calculated the revenue generated per metric ton of CO2 abated by recycling pathways, termed the marginal abatement revenue (MAR). The marginal abatement revenue was then subtracted from MAC to obtain the MAC after revenues. These three values are displayed in Figure 5 as a waterfall chart, in which MAC is shown as the initial value from which MAR is subtracted to yield MAC after revenues for each pathway.

MAC estimates lie in the vicinity of $200–$700/tCO2 for many of the pathways analyzed, with electrochemical pathways having generally higher MACs than thermochemical pathways. The MAC for electrochemical ethane production is the highest due to the pathway’s high ECOP. The MAC values for most pathways are greater than those of many other common carbon abatement levers, including renewable electricity and efficiency measures.[21] However, these CO2 recycling pathways fulfill a need for emissions reductions in hard-to-abate sectors like transportation that other levers on a MAC curve cannot readily satisfy. For this reason, it may be favorable to implement certain CO2 recycling pathways concurrently with other measures that have a lower MAC.

For most pathways, the MAR is less than the MAC, resulting in a net cost after revenues, which is represented as a positive value for MAC after revenues. However, for electrochemical CO production and the two concrete production pathways, the MAC after revenues is negative, indicating that a net profit can be made while abating emissions through these pathways.

Effective Carbon Price and Gross Subsidies

The value of MAC after revenues can also be taken to roughly represent the effective carbon price necessary for a pathway to reach market parity. We assume that an effective carbon price would be translated fully into an increase in the product selling price. In this case, applying an effective carbon price equal to the value of MAC after revenues for a CO2 recycling pathway would result in the pathway’s MAC being equal to its MAR (in other words, its ECOP being equal to the selling price), achieving market parity for the pathway.

Using this assumption, we calculated the cumulative carbon abatement potential of pathways that have reached market parity as a function of effective carbon price (Figure 6a). We excluded repeated pathways with higher breakeven carbon prices as described in the methodology. With no carbon price, 1.6 GtCO2/yr of cumulative abatement potential is available at market parity since electrochemical CO production, ethanol from lignocellulosic biomass, and the two concrete production pathways have ECOPs lower than product selling price. As the effective carbon price increases, the cumulative carbon abatement potential increases rapidly between $350/tCO2 and $425/tCO2 as more pathways reach market parity until cumulative carbon abatement potential reaches 7.3 GtCO2/yr at an effective carbon price of $425/tCO2. This sum is different from the previously cited abatement sum because we did not exclude electrochemical CO production from Figure 6a, as it has a much lower breakeven carbon price than syngas and the resolution between the two is important to illustrate. The cumulative potential jumps to 16.8 GtCO2/yr with the addition of Sabatier process methane production at an effective carbon price of $1,363/tCO2.

In a similar vein, we determined the gross value of subsidies needed, absent a carbon price, to bridge the gap between ECOP and product selling price for CO2 recycling pathways at the full scale of current global markets (Figure 6b). The needed subsidies vary widely between pathways but are in the range of hundreds of billions of dollars per year. For comparison, global fossil fuel subsidies were $318 billion in 2019.[22]

Critical Infrastructure Needs

For CO2 recycling to reach a global scale and contribute substantially toward reaching net-zero emissions, these pathways will need to draw on tremendous expansions of low-carbon electricity, CO2 transport, and electrolyzer infrastructure. Figure 7 shows the breakdown of the critical infrastructure capital cost and gigawatt capacity required to supply a product’s current global demand entirely using a given CO2 recycling pathway. All critical infrastructure requirements for methane production pathways are large outliers due to the enormous scale of global methane demand. The total infrastructure capital cost for CO2 recycling at global scale, summed across all pathways excluding methane production and duplicate products, is $27.5 trillion. This estimate is based on today’s capital costs for all forms of critical infrastructure, and as these capital costs are dropping rapidly (particularly for renewable generation capacity and electrolyzers),[23] the total infrastructure cost is expected to decrease significantly before CO2 recycling reaches large scale.

The critical infrastructure capital costs are dominated by new renewable electricity generation capacity due to the high electricity consumption of CO2 recycling pathways. The thermochemical pathways notably require similar quantities of renewable electricity investment to electrochemical pathways due to our assumption that the thermochemical feedstocks are produced electrochemically. The required renewables generation capital cost is trillions of dollars globally for each pathway, with a total renewables cost across pathways of $16.8 trillion. This total cost is based on a current average renewables capital cost of $2,000/kW across utility-scale solar, onshore wind, and offshore wind power and assumes complete displacement of current products with recycled products. With projected decreases in renewable power capital costs,[24] an average capital cost of $1,300/kW is reasonable to expect in 2030, which would require a total renewable generation capital cost of $10.9 trillion.

The electrochemical CO2 reduction pathways will require CO2 electrolyzer infrastructure, while the thermochemical pathways will largely require water electrolyzer build-out to supply their green hydrogen consumption. These electrolyzers represent the second largest critical infrastructure expense, though their capital cost is likely to decrease rapidly with broader deployment of this nascent technology. Transmission lines to deliver renewable electricity are a smaller but notable capital cost, ranging in the hundreds of billions of dollars for each pathway.

To estimate the capital cost of the CO2 transport network needed to supply CO2 for recycling markets, we used a CO2 pipeline capital cost of $42/tCO2/yr capacity, which was obtained from a model of a 20 MtCO2/yr network with a 500 km central spine.[25] This provides a minimal cost estimate, as smaller diameter or longer pipelines would cost more.[26] Overall, the electrochemical and thermochemical pathways show no systematic differences in CO2 transport infrastructure costs, which are driven by the global volumes of the process and the ratio of CO2 consumed to product generated in the reaction. The pathways directly consume a combined 5.3 GtCO2/yr that must be delivered by CO2 transport infrastructure. Most pathways require on the order of $50 billion of CO2 transport infrastructure, while electrochemical and thermochemical methane pathways would require hundreds of billions of dollars of CO2 transport infrastructure due to the large market for methane. These CO2 transport capital costs are nearly negligible compared to the large capital costs of the other forms of critical infrastructure. The total CO2 transport infrastructure capital cost of all pathways, excluding methane production and repeated products, is $222 billion, which is comparable to the combined annual revenues of the four largest chemical firms globally.[27]

In terms of the needed gigawatt capacity of renewable generation and electrolyzers, most pathways require between 200 and 2,000 GW of both renewable generation capacity and electrolyzers (Figure 7c and 7d). In total, excluding methane production and duplicate products, the pathways would collectively require 8,400 GW of new renewable capacity. Due to a global methane demand far exceeding the demand of all other processes, Sabatier process methane production alone would require 22,400 GW of renewable capacity. By comparison, the global installed renewable electricity capacity, including hydropower, was 2,500 GW at the end of 2019.[28] The installed capacity of wind power and solar PV is projected to grow at a rate of ≈180 GW/year through 2025, and this new capacity must serve other important needs for low-carbon energy, including electrification and displacement of fossil power generation.[29] Similarly, the total required electrolyzer capacity calculated using the same method is 8,000 GW. For context, the announced water electrolyzer projects that are expected to go online in 2021–2023 account for only 2.7 GW of capacity.[30] Without rapid and profound acceleration of both renewables and electrolyzer deployment, the demand of global-scale CO2 recycling will remain unmet.

Global Electricity and Hydrogen Consumption

To illustrate the volume of inputs that CO2 recycling at global scale will demand from this critical infrastructure, Figure 8 shows the electricity and hydrogen consumption of CO2 recycling pathways at full market scale. Most CO2 recycling pathways at global scale would require 1,100 to 9,200 TWh/yr of low-carbon electricity (Figure 8a and 8b). The sum across all pathways, excluding methane and duplicate products as described in the methodology, is 36,700 TWh/yr. Since the world consumes approximately 26,000 TWh/yr of electricity today,[31] this represents a monumental amount of electricity demand, which must be supplied entirely by low-carbon sources! At the same time, the global technical potential for electricity generation from solar and wind sources is hundreds of thousands of TWh/yr for each,[32] indicating that the required scale of renewable resource is available many times over if economics and other constraints allow.

The global feedstock hydrogen consumption of thermochemical CO2 recycling pathways is 30–100 Mt/yr for each pathway (Figure 8c), comparable to the current global hydrogen consumption of 70 Mt/yr.[33] Similar to electricity, this hydrogen demand must be supplied by low-carbon sources for CO2 recycling to abate emissions. With low-carbon hydrogen production reaching 8 Mt/yr in 2030 in the IEA’s Sustainable Development Scenario,[34] accelerating growth of low-carbon hydrogen will be needed to scale CO2 recycling.

Discussion

Our results lay the groundwork for a large-scale CO2 recycling deployment strategy. The analysis finds that four pathways have an ECOP lower than the selling price and may therefore currently be profitable: electrochemical CO production, ethanol production from lignocellulosic biomass, precast concrete carbonation, and the CarbonCure concrete process. This highlights opportunities for planning, investment, and policy actions. The ethanol from lignocellulosic biomass pathway is already commercial. Electrochemical CO production is not, though a large portion of CO is produced and consumed as syngas (CO + H2). Although there is a commercial CO market today, it is small and difficult to enter, limiting opportunities to scale. To accelerate commercialization of low-cost electrochemical CO production, produced CO could be mixed with low-carbon H2 and sold as low-carbon syngas. This would provide a market for electrochemical CO production while reducing emissions associated with conventional syngas formation from natural gas.

Reducing the Estimated Cost of Production of CO2 Recycling Pathways

The remaining CO2 recycling pathways have an ECOP above the selling price, making them unprofitable today. If the gap between ECOP and selling price is small enough to be bridged by subsidies designed to bring down ECOP, this can result in subsidy-supported profitability. Alternatively, as the result of carbon pricing or other market forces, the product selling price might increase to meet a higher ECOP. This price increase may reflect increased need for CO2 recycling in the future or demand for low-carbon products leading to a higher willingness to pay. Importantly, our ECOP calculations were designed to give globally representative production costs, and in certain locations or conditions, ECOP may be significantly closer to the current selling price.

Of course, price parity is not the sole criterion for adoption by companies and markets, and the net present value of existing assets, security of supply chains, ease of integration, and many other criteria are important in corporate decision making. However, most pathways assessed could cost substantially more to produce than today’s market-available alternatives, indicating a need for cost reduction and market-aligning policies.

Innovations in CO2 recycling technologies, particularly for catalysts, are likely to be the most effective way to reduce the ECOP of CO2 recycling pathways. Since direct electricity consumption is the largest contributor to ECOP for all electrochemical processes, improvements in electrical energy conversion efficiency are critical to reducing cost and can be achieved by lowering the required overpotentials on electrocatalysts for electrochemical CO2 reduction and water electrolysis. For both thermochemical and electrochemical pathways, product selectivity dramatically affects cost by influencing the required volume of material and energy inputs. By reducing the resource intensity of production, higher efficiencies and selectivities will also lessen the critical infrastructure needs to support CO2 recycling. Improved selectivities would also alleviate the need for intensive separations processes to isolate the desired product from byproducts, whose costs are not considered in this study. Since many of these technologies are relatively nascent, there are wide margins for improvement in catalyst performance that could yield significant cost reductions.

Improving selectivity will require developing a new generation of catalysts tailored to specific products and tuning reactor conditions for the desired product distribution. These two technical opportunities strongly recommend a targeted innovation agenda focused on developing thermo- and electrocatalysts with improved activity and selectivity, thereby reducing total power requirements and the need for separation and processing of products.

An alternative response to selectivity limitations is the commercialization of coproduced valuable molecules from the same reactor. CO2 recycling pathways often produce a wide range of hydrocarbons, alcohols, and oxygenates, all of which have market value. The conventional approach (applied in this study) is to design a CO2 recycling facility to produce one product with ≈30–70 percent selectivity and to discard the byproducts. This essentially wastes both the energy and feedstocks used to create these marketable byproducts. If facilities could instead be designed to simultaneously produce multiple products for sale, selectivity constraints could be repurposed as an asset. The cost of product separations could present a challenge to such an approach. In this context, one strategy is to coproduce easily separable products, such as a liquid and a gas, to avoid energy- and cost-intensive separation steps. For instance, carbon monoxide is often produced in large quantities as a byproduct of electrochemical and thermochemical direct CO2 conversion processes.[35] Instead, gaseous CO could be separated and sold as a coproduct of liquid products such as methanol or jet fuel.

An additional challenge of coproduction could be that the generation ratios of coproducts is mismatched to market demand, an issue that could become acute at large scale. Pathways with viable coproduction ratios should be identified for coproduction during planning. In addition, the selectivity of CO2 recycling catalysts could be tuned so that product ratios align closer with market demand. By making full use of the energy and feedstock inputs to the reactor, effective coproduction would allow us to meet global demand for CO2 recycling products with lower critical infrastructure needs and lower ECOP for each product.

A second strategy to decrease ECOP is lowering the costs of the key inputs and feedstocks for CO2 recycling pathways. These include the costs of electricity for the electrochemical processes and hydrogen or syngas for many thermochemical pathways. This cost reduction challenge is deepened by the requirement that the inputs have a low- to zero-carbon intensity to ensure a net emissions reduction through CO2 recycling. For this reason, low-carbon/renewable electricity, green hydrogen, and low-carbon syngas are the ideal forms of these inputs. As assumed in this study, feedstocks for thermochemical processes including hydrogen and syngas can potentially be made with electrochemical pathways, creating a hybrid process that has been found to be favorable for emissions reductions.[36]

Reducing the cost of low-carbon hydrogen production would be key to making thermochemical pathways less costly. For green H2 production by water electrolysis, cost reductions will be most sensitive to improvements in electrical energy conversion efficiency. Efficiency gains can be made by enhancing electrocatalyst activity to lower the required overpotential for water electrolysis. The aforementioned renewable electricity price reductions would also lower electricity input costs for the process, which make up the majority of ECOP for green H2. While the green H2 production cost is highly sensitive to product selectivity, there are no major competing reactions in water electrolysis, with the notable exception of seawater electrolysis in limited contexts.[37] Selectivities of nearly 100 percent for water electrolysis have been consistently achieved,[38] so improving catalyst selectivity in green H2 production should not be a focus.

An additional option is to use blue H2 instead of green H2, which is produced through steam methane reforming with carbon capture, making it a low-emissions hydrogen source. Costing around $1.5/kg–$3/kg,[39] blue H2 is currently less expensive than green H2 (estimated to cost $6.3/kg in this study), which could make it a useful near-term alternative input to CO2 recycling processes. However, blue H2 requires continued supply of natural gas, which entails substantial upstream/midstream methane leakage, CO2 emissions, and other detriments,[40] so green H2 would be the ideal hydrogen source in the longer term. In contrast, lowering the cost of CO2 is not a priority due to its relatively small effect on estimated cost of production.

Lowering renewable electricity costs can have a large impact on ECOP, but it represents a secondary measure for reducing ECOP and should be pursued accordingly. Innovation for renewable energy technologies would be helpful for decreasing renewable electricity costs. However, innovation funding to reduce ECOP is likely to be more effective through improving the performance of CO2 recycling catalysts, which are relatively underdeveloped and have larger margins for improvement than renewable electricity technologies. Additionally, improving catalyst selectivity and activity would reduce the energy intensity of CO2 recycling and thus bring down the large renewable energy infrastructure demands of these pathways.

Though lowering the price of renewable electricity should be a secondary priority from an ECOP reduction perspective, accelerated deployment of renewable energy capacity will be critical to meet the demanding low-carbon electricity consumption of global-scale CO2 recycling processes. This renewables deployment will have an additional benefit of lowering electricity costs through learning-by-doing, although deep cost reductions through this approach will be difficult to achieve due to the enormous global deployment required to achieve doublings of installed renewable capacity.

Improvements in the capital cost and capacity factor—which determine total CAPEX from an intensive and extensive standpoint, respectively—will not be as effective in reducing ECOP compared to the measures already discussed. However, as efficiencies and selectivities improve and the cost of inputs goes down, variable costs will become a smaller portion of total cost and ECOP will become increasingly sensitive to CAPEX. This shift will signal an important milestone for CO2 recycling pathways, where efforts to reduce capital cost will become crucial and policy levers such as capital cost loan programs will become financially salient. Importantly, Figure 1 demonstrates that many pathway ECOPs would remain higher than the selling price even if the costs of electricity and hydrogen went to zero. Although variable costs dominate ECOP currently and should be the main focus for cost reductions at the outset, concurrently lowering CAPEX and fixed O&M cannot be neglected if cost parity is to be achieved in the longer term.

Our analysis finds it exceedingly difficult and often impossible to achieve profitability for a pathway by improving only one of the techno-economic inputs alone. Decreasing ECOP sufficiently requires cost reductions across the board, from improvements in technological performance to cheaper feedstocks. This calls for redoubled efforts across sectors. It also suggests a coordinated approach involving both innovation investment and market-aligning policies.

Carbon Abatement

Our findings demonstrate that replacing conventional pathways with CO2 recycling could significantly abate emissions. Excluding methane pathways, completely producing the products investigated using CO2 recycling would abate 6.8 GtCO2/yr. On an intensive basis, most pathways abate 2–11 tCO2 per metric ton of product when replacing conventional production, representing substantial abatement for each unit of CO2 recycling production. However, these carbon abatement potentials are heavily contingent on our assumption that the CO2 recycling pathways consume renewable electricity and low-carbon feedstocks. If these pathways instead consume fossil-based grid electricity and feedstocks generated using conventional petrochemical processes like SMR-derived H2, CO2 recycling is likely to emit more CO2 than the current production methods.[41] Therefore, it is critically important that CO2 recycling pathways use low-carbon inputs to ensure that deploying these pathways does not ultimately increase emissions. In this context, developing the critical infrastructure to generate and deliver these low-carbon inputs at low cost is a priority.

The MAC and breakeven effective carbon price of most pathways are relatively high, in the hundreds of dollars per tCO2. We note that these estimates do not incorporate the costs or savings of displacing conventional production, such as the avoided fuel costs of operating a chemical plant at reduced capacity due to partial displacement by a CO2 recycling pathway. If displacing conventional production results in net savings, the MAC and breakeven effective carbon price would be lower.

Critical Infrastructure Needs

Our analysis demonstrates that CO2 recycling at scale will require large quantities of new infrastructure for renewable electricity generation, electricity transmission, electrolyzer capacity, and CO2 transport. We estimate this infrastructure will cost trillions of dollars in capital investment for each pathway at global scale. These findings indicate that, in addition to reducing the cost of production for CO2 recycling pathways, building these supportive infrastructures at an aggressive pace and scale will be key to enabling CO2 recycling.

In order to maximize emissions abatement, CO2 recycling must be supplied with low-carbon electricity. Due to the high electricity demand of each CO2 recycling pathway, new low-carbon electricity generation capacity constitutes the large majority of critical infrastructure capital costs for CO2 recycling. Since CO2 recycling will join a range of growing demands for clean electricity in a sustainable energy future, including electrified transport and heating, it is likely that even more low-carbon generation, mostly renewable generation, will be required to meet the needs of CO2 recycling along with these other demands. This makes accelerated deployment of low-carbon/renewable electricity capacity crucial to enabling CO2 recycling at scale.

A build-out of long-range electricity transmission lines will likely be needed to deliver renewable electricity to CO2 recycling centers far from renewable energy resources. Our estimates show that new transmission infrastructure for CO2 recycling globally would account for a small portion of the overall critical infrastructure capital investment needs but would still require hundreds of billions of dollars in capital investment. In this analysis, we attribute new transmission costs entirely to new renewable generation installations at a transmission capital cost of $300 per kW of renewable capacity.[42] However, the transmission cost may ultimately be lower since the new transmission lines will probably be shared by multiple sources and not all renewable projects will require entirely new transmission build-out.

The second most capital-intensive type of critical infrastructure for most pathways is electrolyzer capacity. CO2 electrolyzers will be needed to perform electrochemical CO2 reduction for the electrochemical pathways, and water electrolyzers will supply green hydrogen for the thermochemical pathways. The capital investment needed for electrolyzers can be expected to drop substantially in coming years through innovation and rapid doublings of installed capacity.

CO2 transport pipelines appear to be the least expensive form of critical infrastructure, with most pathways requiring less than $50 billion in CO2 transport capital costs. These CO2 transport pipelines will be needed to deliver feedstock CO2 from its source, either a DAC or point-source carbon capture facility, to its end use in a CO2 recycling pathway. Our CO2 transport costs represent minimal cost estimates, since the CO2 transport networks modeled are very high capacity and smaller pipelines would require more capital. Nevertheless, for most pathways CO2 transport accounts for less than 3 percent of the total infrastructure cost, suggesting that CO2 transport will likely remain the least expensive piece of critical infrastructure even with higher cost estimates.

As an important caveat, these estimates of critical infrastructure capital cost do not consider the additional difficulties of building the infrastructure. In many nations, especially OECD countries, new energy infrastructure is contentious and is not guaranteed on the basis of cost competitiveness alone.

Greater efficiencies in infrastructure build-out can be found by “piggybacking” on existing industrial and chemical infrastructure. For instance, oil and gas pipelines can potentially be repurposed to transport hydrogen or CO2 feedstocks to the CO2 recycling plant. The thermochemical CO2 recycling pathways in particular will use many of the same process units conventionally used in chemical plants, which could be repurposed for CO2 recycling processes. Some CO2 recycling pathways would even directly use existing infrastructure. For example, both CO2 recycling urea production and the CarbonCure concrete process simply require an external CO2 input stream retrofit to a conventional urea or concrete plant.

Market Size

One important feature to these CO2 recycling pathways is that their products represent existing markets on the order of 100 Mt/yr globally. Although current cost barriers are substantial, the market for these fuels, chemicals, and materials already exists and is robust, as these products are essential to our global society. For this reason, CO2 recycling products have a low risk of saturating existing markets in early deployment or driving down selling prices. The reliable demand and largely stable prices for the products provide a high degree of near-term certainty that should encourage actors making entry into a nascent CO2 recycling industry.

Since our analysis of carbon abatement potential, infrastructure needs, global electricity/hydrogen consumption, and needed subsidies assume CO2 recycling pathways will supply current global demand for their products, these estimates represent a maximum scale case. The true size of these extensive metrics is likely to be lower than our estimates, particularly in the nearer term. Additionally, as the per-unit capital cost of renewable capacity and electrolyzers continue to drop rapidly, the total infrastructure cost at the time CO2 recycling reaches global scale will likely be significantly lower. Nevertheless, these estimates help bound the possibility space for the scale of CO2 recycling.

Policy Recommendations

Our findings indicate that, while CO2 recycling is a potentially pivotal carbon abatement tool, a measured approach must be taken to improving the techno-economics of and deployment of CO2 recycling pathways. Policies to support CO2 recycling pathways should prioritize the pathways with the most promise for reaching commercial maturity to avoid expending limited resources on pathways that have limited prospects of becoming techno-economically viable in the foreseeable future. For these prioritized pathways, policymakers should seek to both create demand for CO2 recycling products and to bring down the costs of CO2 recycling pathways.

Prioritizing Pathways

For prioritizing certain CO2 recycling pathways, our analysis suggests pathways that are closest to market and might benefit most from policy to help support market entry. We define these as pathways with an ECOP to selling price ratio of less than five. Under this definition, the following pathways should be prioritized for early market entry: green hydrogen production, electrochemical CO production, CO2 hydrogenation to light olefins, CO2 hydrogenation to methanol, CO2 hydrogenation to jet fuel, ethanol production from lignocellulosic biomass, CO2 recycling urea production, precast concrete carbonation curing, and the CarbonCure concrete process. Because these CO2 recycling processes are relatively close to reaching market parity (or are profitable today), demand pull policies such as public procurement could prove cost effective at bringing these pathways to scale quickly, achieving substantial emissions reductions. In addition, these pathways, along with pathways that have ECOP to selling price ratios from 5 to 7.5, remain excellent candidates for investment in innovation and technical development.

Though many of the remaining pathways have large carbon abatement potentials, pursuing these less promising pathways could hinder the progress of CO2 recycling, and policy support on these pathways would be ineffective, absent a breakthrough innovation. These pathways, namely electrochemical methane and ethane production and Sabatier process methane production, would require meeting a highly demanding set of techno-economic targets to reach market parity. The greater than 25:1 ECOP to selling price ratios for these processes present a difficult gap to bridge through subsidies, and demand-side policies like portfolio standards could force producers to adopt exceedingly expensive production processes and pass on undue price increases to consumers. Large subsidies or innovation spending for processes that are unlikely to become profitable could be better spent supporting more viable CO2 recycling pathways. As a result, methane and ethane markets may require ongoing supplies of conventional natural gas even as measures to minimize natural gas demand and emissions continue in parallel. Given the high ECOP to selling price ratios, it may prove necessary to balance these emissions with CO2 removal practice elsewhere.[43]

Pursuing the more promising pathways first could have spillover benefits for the less promising pathways. Under a targeted innovation agenda, technological advances for the prioritized CO2 recycling pathways could be applied broadly. For instance, a breakthrough in our understanding of how to tailor the selectivity of electrochemical CO2 reduction catalysts for C-C coupling may provide insights for targeting a range of C2+ products through electrochemical CO2 reduction. Additionally, the renewable energy and CO2 transport infrastructure deployed for scaling near-term CO2 recycling pathways could be leveraged by all remaining pathways later on. With these spillover benefits, it may become more viable in the medium to long term to pursue currently unfavorable pathways and unlock the sizable carbon abatement potential associated with them. This logic may also apply to the deployment of CO2 recycling pathways for low-volume, high-market price products that are outside the scope of this study. These pathways include some listed in our broad summary of CO2 recycling pathways in Table A5.

Policies for Cost Reductions and Market Deployment

In addition to identifying the most promising CO2 recycling pathways, our findings highlight how policies can most effectively reduce the ECOP of these pathways. The primary lever we propose for lowering ECOP is technological innovation policy focused on improving the activity and selectivity of CO2 recycling catalysts. Innovation policy has been found to be an important contributor and accelerant of cost reduction and performance increase.[44] Because the efficiency and yield of CO2 recycling pathways were found to have a large influence on ECOP in our sensitivity analysis, a targeted RD&D program to enhance catalytic activity and product selectivity of electro- and thermocatalysts would improve the efficiency and yield of CO2 recycling pathways, respectively, unlocking cost reductions. Since many of these technologies, particularly electrochemical CO2 reduction, are relatively nascent, an innovation program could achieve substantial improvements in catalyst performance and drive deep cost reductions. The costs of renewable electricity, green H2, and/or low-carbon syngas inputs to the pathways also have a high impact on ECOP, and innovation in renewable electricity technologies as well as catalysts to produce green H2 and low-carbon syngas will be key to lowering these input costs. Finally, production pathways for these three inputs would see deployment-related cost reductions if scaled rapidly to supply the growing CO2 recycling pathways that depend on them.

In tandem with cost reductions of CO2 recycling pathways, policies should include market-aligning measures for the pathways ready for early market entry, to lower barriers to entry and create demand. However, the differential between ECOP and selling prices suggests care and consideration in prioritizing market-aligning policies. Some pathways, such as methane production or electrochemical methanol production, are so expensive that even large expenditures might prove unable to stimulate market adoption today. In contrast, supporting the purchase and market adoption of products that are relatively close to market prices would create supply chains for some products, provide contracting experience and quality standards, and accelerate cost reduction for enabling technology for the other products and pathways.

Regarding specific policies, incentives for CO2-based products could narrow the cost gap between a CO2 recycling pathway and existing market alternatives, as was done with tax credits and feed-in tariffs for renewable power. A comparable policy strategy could include production tax credits for CO2 recycled products. Similarly, mandates could serve to create early market pull and establish supply chains and methodologies for standards. For example, governments could require that a certain percentage of a product under direct procurement is supplied by CO2 recycling pathways (i.e., “buy clean” requirements). This strategy has proven important to many clean energy technologies, notably in biofuel procurement mandates in defense-related applications. This was tried successfully in California’s Buy Clean Act,[45] which included structural steel and rebar but did not include CO2 recycled materials. New York State is considering a clean concrete legislation that would provide modest price support for novel low-carbon cement formulations.[46] Importantly, the development, testing, and codification of standards and life cycle methodologies from early procurements provided an important basis for future regulation and drove investment in qualification and performance enhancement. In addition, prize competitions or milestone payments for CO2 recycling technologies upon reaching certain stages of maturity can be effective at creating early markets for these pathways and spur further innovation as technologies transition from the RD&D stage toward market deployment.[47]

Critical Infrastructure Build-out

Innovation and market policy alone will not deliver CO2 recycled products to markets due to infrastructure limits. The analysis shows that critical electrical and CO2 transmission infrastructure investments are required to serve manufacturers and supplies. In most markets, governments provide critical funds or critical authorization of rate base to build and maintain critical energy infrastructure. Infrastructure policies should include funding for construction, permitting and regulatory support, public outreach and engagement, and, in rare cases, condemnation or exercise of eminent domain—this should apply to both transmission and renewable generation infrastructure. Similar policy measures are important components of the Biden “Build Back Better” policy drive, Canada’s Hydrogen Roadmap, and the EU’s Green Deal plan.[48]

Conclusion

CO2 recycling pathways show real promise for displacing existing fossil-intensive production pathways of high volume products. If implemented on a global scale, these CO2 recycling pathways (excluding costly methane) could collectively result in up to 6.8 GtCO2/yr of carbon abatement, although the pathways that are currently profitable account for 1.6 GtCO2/yr of abatement potential. The estimated cost of production (ECOP) varies significantly between pathways but is typically significantly higher (2.5–7.5 times more expensive) than the product selling price. A few key pathways may prove the most attractive for early market entry due to their ECOP to selling price ratios of less than five, and many others are suitable for further RD&D.

To provide relevant climate and economic benefits through commercial markets, ECOP reductions of key CO2 recycling pathways are essential and urgently needed. A variety of cost reduction levers can and should be used to achieve profitability, including improving a pathway’s electrical efficiency and selectivity, as well as reducing the costs of key inputs such as renewable electricity, green H2, electrochemically produced CO, and low-carbon syngas inputs. Well-crafted innovation policy combined with market-alignment policies can contribute substantially to all these approaches. On a global scale, CO2 recycling pathways would require major investment in critical infrastructure. Enormous growth of both renewable electricity (generation and transmission), electrolyzer capacity, and CO2 transport networks are critical pathways for global deployment.

Appendix

Appendix Tables

[49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67]

[68] [69] [70] [71]

Overview of Selected CO2 Recycling Pathways

The CO2 recycling pathways analyzed in this study are listed in Table 1. Each pathway is defined by the pairing of a chemical process (e.g., CO2 hydrogenation) and its desired product (e.g., methanol). Several of the chemical processes can be adapted to target various products, and in these cases, we evaluate each process-product pairing as a distinct pathway.

Eight of the pathways are electrochemical, meaning the driving force for the reactions comes from electricity. They include the following pathways:

  • Water electrolysis to produce hydrogen: In water electrolysis, electricity is used to split water molecules into hydrogen (H2) and oxygen (O2). Though not strictly a CO2 recycling pathway, the generated H2 is a key input for many thermochemical CO2 recycling pathways.
  • Electrochemical CO2 reduction to produce carbon monoxide, methane, ethylene, or ethanol: The electrochemical CO2 reduction pathways entail reacting CO2 and water in a CO2-saturated aqueous solution to produce a wide range of carbon-based or hydrocarbon products.[72] By tailoring the selectivity of the electrocatalyst, the reaction can be directed to produce different molecules as the major product.
  • Electrochemical CO2 reduction to produce methanol: Methanol is produced with low selectivities in typical electrochemical CO2 reduction in aqueous electrolyte, but several studies have achieved significantly higher methanol selectivities by conducting electrochemical CO2 reduction in an electrolyte consisting of a mixture of water and an ionic liquid such as [Bmim]BF4.[73] We model our electrochemical CO2 reduction to methanol pathway based on these ionic liquid electrolyte studies. The need to use an ionic liquid electrolyte may present additional challenges to scale that are not considered here.
  • Electrochemical CO2 reduction to produce ethane: We model electrochemical CO2 reduction to ethane based on catalysis studies that use PdCl2 in the aqueous electrolyte to enhance performance. The need to use this additive, which is consumed in the reaction process, may entail additional challenges to scale that are not considered here.
  • Electrochemical CO2 reduction to produce syngas: A common byproduct of electrochemical CO2 reduction is H2 generated through the hydrogen evolution side reaction. In the electrochemical syngas pathway, electrochemical CO2 reduction is used to create carbon monoxide (CO) with H2 as a byproduct to yield syngas, a mixture of CO and H2. We design this pathway to produce syngas with a CO/H2 mole ratio of 0.8.

The remaining nine pathways are thermochemical pathways for which the driving force is thermal, though often the majority of the energy input for these processes comes from feedstock hydrogen as discussed in the results section. The thermochemical pathways are:

  • Fischer-Tropsch (F-T) synthesis to produce light olefins or jet fuel: F-T synthesis involves the reaction of syngas (CO and H2) to form various hydrocarbons with adjustable product selectivity.[74] We model the F-T pathways as using separately generated green hydrogen from renewables-powered electrolysis as the feedstock H2 source and using feedstock CO generated from renewables-powered electrochemical CO2 reduction. The “light olefins” product mixture includes ethylene, propylene, and butylene.
  • CO2 hydrogenation to produce light olefins, methanol, or jet fuel: In CO2 hydrogenation, CO2 and H2 are reacted to form various hydrocarbons, and catalyst selectivity can be used to tailor the product mix.[75] In this paper, we do not distinguish between hydrogenation mechanisms, such as the methanol reaction involving a methanol intermediate or F-T based CO2 hydrogenation, in which CO2 is reduced to CO in situ before reacting with H2 through the F-T process.[76] We assume CO2 hydrogenation pathways use green hydrogen feedstock.
  • Sabatier process to produce methane: The Sabatier process, also known as CO2 methanation, combines CO2 and H2 (assumed to be green H2 in this study) to form methane thermochemically.[77]
  • Reverse water-gas shift (RWGS) reaction to produce syngas: The RWGS reaction is a reversible process by which CO2 and H2 (assumed to be green H2) are converted to CO and H2O.[78] The RWGS reaction is typically taken only partially to completion in order to yield syngas.
  • Bosch-Meiser process to produce urea: The Bosch-Meiser process is a large-scale industrial pathway that reacts ammonia and CO2 to produce urea.[79] Conventionally, the feedstock ammonia is often produced onsite by combining steam methane reforming (SMR)-derived hydrogen with nitrogen (N2) in the Haber-Bosch process. The CO2 produced from SMR is then combined with the produced ammonia to create urea through Bosch-Meiser.[80] This study uses an alternative CO2 recycling Bosch-Meiser process in which green ammonia is produced using green hydrogen and N2 through the Haber-Bosch process, and externally sourced feedstock CO2 is then reacted with this green ammonia to produce urea.
  • Lignocellulosic biomass fermentation to produce ethanol: Ethanol production from lignocellulosic biomass, perhaps more specifically categorized as a biochemical process, involves the breakdown of lignin and polysaccharides into simple sugars, which are then fermented to produce ethanol.[81] Feedstock lignocellulosic biomass can be sourced from agricultural residues such as corn stover, forestry residue, and other wastes and nonfood crops.

The remaining two pathways are different CO2 recycling-based modifications to a single step in conventional concrete production. We model the concrete pathways by assuming that the pathways are identical to conventional concrete production with identical costs and emissions except for the single modified step.

  • Precast concrete carbonation curing: In precast concrete carbonation curing, newly formed precast concrete is exposed to CO2-enriched air in a closed container for 4–48 hours, allowing CO2 to be taken up into the concrete in the form of calcium carbonates and calcium silicates.[82] The carbonation process improves the strength and durability of the concrete and replaces the conventional steam curing step in precast concrete production.
  • CarbonCure process for all concrete types: In addition to the carbonation curing process, we also evaluate a similar pathway developed by CarbonCure Technologies, a start-up company. The CarbonCure process injects CO2 into precast, ready-mix, or masonry concrete during the mixing stage, similarly incorporating the CO2 into the concrete as minerals.[83] Since this process has different costs and CO2 uptake than the precast concrete carbonation curing pathway and is also applicable to forms of concrete other than precast, we evaluate the CarbonCure process as a separate pathway.

Detailed Methodology

Our assumptions and calculations are described in detail below. All ton units are in metric tons.

Figures of Merit

In our calculations we use several related but distinct figures of merit to describe the performance of CO2 recycling catalysts. These figures of merit are defined here:

  • Carbon selectivity is defined as the percentage of input moles of carbon that go toward the desired product. This is used to calculate the CO2 consumption of electrochemical and thermochemical pathways as well as the CO consumption of Fischer-Tropsch pathways.

Several figures of merit apply exclusively to thermochemical pathways:

  • Hydrogen selectivity is the percentage of input moles of H2 in a thermochemical process that is converted to the desired product. This is used to calculate the hydrogen consumption of thermochemical pathways.
  • Thermochemical plant electrical efficiency is the percentage of input electricity that is converted to usable forms of energy by thermochemical plant equipment. Thus, the energy efficiency value for thermochemical pathways applies only to the direct electricity consumption of the plant, which is a very small portion of total energy input to the thermochemical process.

Other figures of merit apply exclusively to electrochemical pathways:

  • Electrochemical electrical efficiency is calculated as the ideal cell voltage for the desired reaction divided by the real cell voltage for that reaction. Disregarding current that goes to undesired reactions, this metric represents the percentage of input electrical energy that is converted into chemical energy in the desired product. This metric is used along with faradaic efficiency to calculate the total electricity consumption of electrochemical pathways.
  • Faradaic efficiency gives the percentage of overall input electrical current that goes toward the desired product in an electrochemical pathway. This metric represents a form of selectivity of the pathway. It is used along with electrical efficiency to calculate total electricity consumption.

When evaluating the effect of selectivity in our cost sensitivity analysis, we varied the faradaic efficiency, carbon selectivity, and hydrogen selectivity simultaneously for each pathway and described the resulting percent change in ECOP with the umbrella term of “selectivity” for the sake of simplicity. When the general terms “product selectivity” or “selectivity” are used throughout the report, they are referring to faradaic efficiency, carbon selectivity, and hydrogen selectivity collectively.

Estimated Cost of Production (ECOP) and Sensitivity Analysis

The estimated cost of production (ECOP, in $/metric ton product) for each pathway was composed of the following cost estimates: feedstock cost (low-carbon hydrogen, CO, biomass, CO2, and/or ammonia feedstock), electricity cost, CAPEX, and fixed operation and maintenance (O&M) cost. The input value assumptions for the calculations are summarized in Table A1, A2 and A3. The calculation methodology differed slightly between the electrochemical and thermochemical processes.

To determine the per-unit chemical feedstock consumption of electrochemical and thermochemical pathways (in tons of feedstock consumed per ton of product), we divided the feedstock consumption ratio from the reaction’s stoichiometry by the appropriate selectivity metric of the pathway. Dividing by product selectivity accounts for the additional feedstock that is consumed to produce reaction byproducts. To find the H2 consumption, we used the hydrogen selectivity, which gives the percentage of input hydrogen that goes to the desired product. This allows us to exclude CO2 conversion to CO in many thermochemical pathways, which does not consume H2.

To calculate CO2 consumption through this method, we used the carbon selectivity of the pathway—the selectivity to the desired product among all carbon-containing products. We multiplied carbon selectivity by the CO2 conversion (the percent of feedstock CO2 that is reacted) to obtain the carbon yield, and divided the stoichiometric consumption ratio by carbon yield. This allowed us to exclude the hydrogen evolution side reaction in electrochemical CO2 reduction, which does not consume CO2. This method accounts for the consumption of CO2 to produce the desired product and byproducts, as well as the unconverted CO2 that remains and is assumed to be wasted. The carbon selectivity and hydrogen selectivity used in our calculation are listed in Table A1.

For the electrochemical pathways, the electricity cost ($/metric ton product) was determined by multiplying the electricity price ($/kWh) by the gravimetric energy density of the product (kWh/metric ton) and dividing by both the electrical efficiency and the faradaic efficiency. The feedstock costs were calculated by multiplying the $/ton feedstock purchase price by the per-unit feedstock consumption of the pathway. The CAPEX ($/metric ton product) was calculated by dividing the required energy input including efficiency and selectivity losses (kWh/metric ton product) by the lifetime hours of operation as determined by the capacity factor and equipment lifetime, which was then multiplied by the intensive capital cost ($/kW). The capital cost was multiplied by 1.05 to reflect a weighted average cost of capital (WACC) of 5 percent. The annual fixed O&M cost ($/yr) was estimated as 10 percent of the total capital investment for thermochemical plants and 4 percent of total capital costs for electrolyzers, so fixed O&M cost ($/metric ton product) was calculated by taking this percentage of the capital cost ($/metric ton product) and multiplying by the equipment lifetime (years). The ECOP was found by summing the cost components in $/metric ton product: electricity cost, feedstock costs, CAPEX, and fixed O&M cost.

For each thermochemical pathway, we selected a techno-economic analysis (TEA) from the literature that designed a full plant in a chemical process simulator such as Aspen Plus to execute the pathway. The literature TEAs used, along with all other sources for input values, are cited in Table A1. From the mass and energy balances in the stream summary of each analysis, we extracted values for the direct electricity consumption of plant equipment (kWh/metric ton product) and plant capital cost ($/metric ton/yr of capacity). Using these values, the feedstock cost ($/metric ton product) was calculated by multiplying feedstock consumption (ton feedstock/ton product) by the feedstock price ($/metric ton feedstock). For most pathways, the feedstocks were hydrogen and CO2. The feedstock price used was equivalent to our calculated ECOP for the electrochemical pathway that produces the feedstock (e.g., water electrolysis to produce feedstock H2). The CAPEX ($/metric ton product) was determined by dividing total capital cost ($/metric ton/yr of capacity) by the capacity factor and then dividing by the equipment lifetime (years). Direct electricity costs were calculated by multiplying the direct electricity consumption (kWh/ton product) by the electricity price ($/kWh). Fixed O&M costs ($/metric ton product) were calculated in a manner identical to the electrochemical pathway methodology. The ECOP was calculated by summing the electricity cost, feedstock costs, CAPEX, and O&M cost. We did not include in our calculations the cost of process heat for thermochemical pathways. CO2 recycling pathways require relatively low operating temperatures of 200–350 ºC,[84] compared to temperatures well above 1,000 ºC needed for steelmaking or concrete production, for which process heat becomes a predominant cost.[85] As such, process heat accounts for a minor portion of CO2 recycling production costs[86] and can be reasonably neglected.

For thermochemical pathways, we make a distinction between direct electricity consumption and overall electricity consumption. Direct electricity consumption is the amount of electricity consumed by plant equipment during operation. This value is used to calculate the direct electricity costs as part of ECOP. The overall electricity consumption includes direct electricity consumption plus the electricity consumed to produce the chemical feedstocks for the pathway, such as green hydrogen. The amount of electricity consumed to produce feedstocks is calculated by multiplying the feedstock consumption of a thermochemical pathway (ton feedstock/ton product) by the previously calculated electricity consumption of the electrochemical pathway that produces that feedstock (kWh/ton feedstock). Overall electricity consumption is significantly larger than direct electricity consumption and is used for calculating carbon abatement and infrastructure needs as detailed in subsequent sections.

We determined the sensitivity of ECOP to each variable techno-economic input by individually increasing and decreasing each input value by 20 percent in our model, with all else equal, and noting the resulting percent change in ECOP. When calculating the sensitivity of ECOP to selectivity, we varied the faradaic efficiency, carbon selectivity, and hydrogen selectivity up or down 20 percent simultaneously.

Carbon Abatement Metrics and Effect of Subsidies

The intensive carbon abatement (tCO2 abated/metric ton product) of each pathway was defined as the net decrease in CO2 emissions that would result from displacing a conventional production process with a CO2 recycling pathway that produces the same product. Since a CO2 recycling product and its corresponding conventional product are functionally identical, we assume that the emissions associated with the combustion or use of the products (gate to grave emissions) are equivalent. Under this assumption, the change in emissions from displacing conventional processes with CO2 recycling arises entirely within the cradle to gate life cycle scope. Thus, carbon abatement was calculated through a cradle to gate life cycle assessment of emissions. The intensive carbon abatement of each pathway (tCO2 abated/ton product) was calculated by adding the per-unit CO2 consumption of the pathway (tCO2 consumed/ton product) to the counterfactual cradle to gate CO2 emissions of producing the product using a conventional pathway (tCO2 emitted/ton product) and subtracting the emissions of the CO2 recycling pathway associated with carbon capture, renewable electricity generation, process heat, and reaction byproduct CO2. The intensive carbon abatement was then multiplied by the global demand for the product (Mt/yr) to obtain the global carbon abatement potential of the pathway (MtCO2/yr).

The per-unit CO2 consumption of each pathway is identical to the per-unit CO2 consumption metric described in the ECOP calculation. The counterfactual emissions of the conventional pathway represent the gross emissions reduction of displacing conventional production processes. These counterfactual emissions values were found using the ecoinvent version 3 life cycle inventory database[87] and literature studies cited in Table A1. We assumed that the process of point-source carbon capture created CO2 emissions equivalent to 10 percent of the amount of CO2 captured. Thus, we included emissions equivalent to 10 percent of the per-unit CO2 consumption of the pathway in the abatement calculation. We also incorporated life cycle emissions from renewable electricity generation of 25 gCO2/kWh[88] by multiplying this value by the overall electricity consumption of the pathway (kWh/metric ton product). The emissions associated with process heat were found by assuming process heat is provided by natural gas-derived steam and calculating the emissions associated with providing the required heat load from literature TEAs using this steam.[89] The byproduct CO2 generation was found for the Fischer-Tropsch pathways by converting the molar CO to CO2 selectivity (obtained from literature catalysis studies) to a mass ratio of CO2 to desired product using the selectivity to desired product and the molar mass of each species. For the ethanol from lignocellulosic biomass pathway, we did not incorporate emissions associated with land-use change.

Additionally, the marginal carbon abatement cost (MAC) for each pathway was obtained by dividing ECOP ($/metric ton product) by the intensive CO2 abatement (tCO2/ton product). The marginal abatement revenue was found by dividing the product selling price ($/metric ton product) by the CO2 abatement (tCO2/ton product). The MAC after revenues was calculated by subtracting MAR from MAC.

The MAC after revenues for each pathway was taken as the effective carbon price needed for the pathway to reach market parity, where costs equal revenues. We used this framing to find the cumulative carbon abatement potential that has reached market parity as a function of effective carbon price. At each effective carbon price value, the carbon abatement potential of all pathways with a MAC after revenues less than or equal to the effective carbon price were summed together. In the case of repeated products, we excluded the cost with the higher MAC after revenues. We included both concrete production pathways since they apply to different types of concrete. Electrochemical carbon monoxide production was included along with one syngas pathway because of these two pathways’ very different MAC after revenues values. Separately, we calculated the gross subsidies needed to close the cost-price gap for each pathway by multiplying the difference between selling price and ECOP by the global demand for the product.

Global Critical Infrastructure Needs and Resource Consumption

To estimate the annual electricity consumption (TWh/yr) of each pathway at global scale, the overall electricity consumption of the pathway (TWh/ton product) was multiplied by the global demand for the product (Mt product/yr). For thermochemical pathways, the overall electricity consumption included the electricity needed to produce the pathway’s feedstocks electrochemically. Since the vast majority of conventional chemical production and industrial energy use does not come from electricity[90] and current electricity mixes are mostly fossil-based, we assume that the conventional processes being displaced by CO2 recycling use no renewable electricity. Therefore, all renewable electricity consumption from deploying CO2 recycling has full additionality. The global electricity consumption was then divided by the hours per year of operation based on a 50 percent capacity factor to obtain the capacity of renewable electricity (GW) required to power the pathway at global scale. The required electrolyzer capacity (GW) was similarly found using a 50 percent capacity factor along with the global electricity consumption for electrochemical pathways and the global electricity consumption minus the direct electricity consumption for the thermochemical pathways. The 50 percent capacity factor was chosen to represent a combination of wind and solar with complementary output profiles.[91] These GW capacity values were converted to global capital costs ($B) using the electrolyzer capital costs ($/kW) listed in Table A2 and using a $2,000/kW capital cost for renewable energy, representative of an even mix of solar and wind power at today’s costs.[92]

To estimate the transmission infrastructure capital cost associated with the required global renewable electricity capacity, we multiplied the GW renewable capacities by a $300/kW transmission line capital cost.[93] This transmission capital cost is based on a review of existing renewable energy projects and models and assumes transmission costs are fully allocated to the associated renewable installation. If certain renewable generation projects do not require new transmission infrastructure or the use of new transmission infrastructure is shared between various actors, our estimated costs would be lower. Transmission capital costs vary widely, as reported in the cited study, so $300/kW was chosen as a representative value.

For estimates of CO2 transport pipeline capital costs, the global CO2 consumption of each pathway (tCO2/yr) was first determined by multiplying the intensive CO2 consumption of the pathway (tCO2/ton product) by the global volume of product demand (ton product/yr). Then, the global CO2 consumption was multiplied by a CO2 transport pipeline capital cost of $42/tCO2/yr of capacity to obtain the global CO2 transport capital cost ($B). This capital cost was drawn from a previously analyzed CO2 transport network with 20 MtCO2/yr capacity consisting of a 500 km spine pipeline and two 10 km distribution pipelines on each end.[94] In this model, the 500 km pipeline has a diameter of 32” and the 10 km distribution pipelines have a diameter of 20”. This provides a minimal cost estimate of CO2 transport costs, as lower capacity, smaller diameter pipelines would cost more.

The global hydrogen consumption of the hydrogen-consuming thermochemical pathways was determined using the per-unit H2 consumption of the pathway (tH2/ton product), which was found with the same method used for all per-unit feedstock consumption values. We multiplied the per-unit H2 consumption by the global demand for the product (Mt/yr) to obtain the global hydrogen consumption (MtH2/yr).

Inputs and Assumptions

We designed both the electrochemical and thermochemical CO2 recycling pathways to consume low-carbon electricity, here modeled as renewable energy, and assumed the H2, CO, and/or ammonia consumed as a reactant in thermochemical pathways are supplied by onsite electrochemical processes powered by renewable energy. All numerical assumptions are detailed in Table A1, A2, and A3.

We assume the feedstock CO2 is sourced from carbon capture at point-source emitters and delivered to the CO2 recycling plant via a CO2 pipeline network at a total cost of $50 per metric ton of CO2 (tCO2).[95] While lower carbon capture costs may be available in certain contexts, these lower costs will not be accessible on average for the global scope and scale of this study. For the thermochemical pathways that consume electrochemically generated feedstocks, we use our own calculated ECOP of the electrochemical feedstock production pathway (e.g., water electrolysis for feedstock green H2) as the feedstock price.

Though similar studies often assume renewable electricity prices based on the levelized cost of energy (LCOE) for renewable generators or average power purchase agreement (PPA) prices for renewables, these methods do not reflect the duty cycles of industrial use, contributions of network costs (costs of transmission and distribution), and electricity taxes to the end-use industrial electricity price paid by producers. While producers may have access to PPA-range renewable electricity prices at certain times, in limited contexts, and with low capacity factors, these low prices will not be available on average for the global production scale evaluated in this study. Therefore, for our estimate of the renewable electricity price paid for global-scale CO2 recycling processes, we must incorporate the duty cycle requirements and contributions of network costs and taxes beyond renewable PPA prices.

To estimate the renewable electricity price used in this study, we take renewable power PPA prices in a region to be equivalent to the wholesale price of renewable electricity, and calculate the corresponding industrial price of renewable electricity by increasing the PPA price to reflect the additional contributions of network costs and taxes to industrial prices. Wholesale electricity prices account for 10–70 percent of industrial electricity prices in different countries.[96] We assume that renewable PPA prices on average account for the same percentage of the final renewable electricity industrial price in a particular country. To estimate the renewable electricity price used in this study, we assume that CO2 recycling producers have access to industrial electricity markets, and that wholesale electricity prices represent 50 percent of the price of industrial electricity on a global basis. We use an average global renewables PPA value of 4.75 ¢/kWh[97] and divide it by 0.5 to obtain an industrial electricity price for renewables of 9.5 ¢/kWh. In keeping with this framing, we assume a capacity factor of 50 percent for electrochemical pathways supplied by renewable electricity, which was chosen to represent a combination of wind and solar with complementary output profiles.[98] Since thermochemical pathways have very small direct electricity consumption, we assume a higher capacity factor of 90 percent.

An obvious condition from this cost framing is that the true costs of recycling CO2 will vary substantially across regions and jurisdictions and that some locations may provide high-capacity electricity at much lower costs. These sensitivities and ranges will be published separately.

In our techno-economic analysis, we choose values for faradaic efficiency, carbon selectivity, and hydrogen selectivity that are 20 percent lower than those of the highest performance catalysts reported in the literature. This 20 percent decrease from the highest performance values was applied to reflect that many of the cited catalysis studies at the bench scale do not demonstrate industrially relevant production rates (measured as current density or space-time yield), and industrial demonstrations of these pathways today would not achieve bench-scale performance.

Faradaic efficiency values were found directly reported in the literature. For thermochemical pathways, carbon selectivities on a molar basis were directly reported in the literature. For each electrochemical pathway, the carbon selectivity was found by converting the faradaic efficiencies to a molar basis. First, we listed out the faradaic efficiency toward each product/byproduct as reported in the cited catalysis study, excluding the hydrogen evolution reaction, which does not involve carbon atoms. Then, each faradaic efficiency was divided by the moles of electrons consumed per mole of CO2 consumed in the associated reaction. Finally, the resulting quotient for the desired product was divided by the sum of all quotients to obtain the carbon selectivity. For the hydrogen selectivity of thermochemical pathways, we directly used the hydrocarbon selectivity values directly reported in the literature, which give the molar carbon selectivity toward the desired product among all hydrocarbon products (excluding non-hydrocarbon byproducts such as CO). The hydrocarbon selectivity is a close approximation of hydrogen selectivity because, as dictated by the stoichiometry, all side reactions in CO2 hydrogenation consume between three and four moles of H2 per mole of CO2 consumed, and F-T consumes between two and three moles H2 per mole of CO.[99] This narrow range of H2 to carbon consumption ratios means that a conversion from hydrocarbon selectivity to hydrogen selectivity would result in a very similar value. All faradaic efficiency, carbon selectivity, and hydrogen selectivity values obtained or derived from the literature were decreased by 20 percent to obtain the values used in this study, as described in the previous paragraph.

To estimate the electrical energy efficiency of the electrochemical pathways, we divided the ideal cell voltage by the real cell voltage indicated in the studies cited in Table A1. The ideal cell voltage is the difference between the reversible potentials of the electrochemical CO2 reduction half-reaction and its paired oxygen evolution reaction. When the real cell voltage was not directly reported by a study, we used the cathode potential used for electrochemical CO2 reduction in the study and assumed an oxygen evolution reaction overpotential of 400 mV.[100] We found the difference between the study’s cathode potential and the assumed oxygen evolution reaction potential to obtain the real cell voltage. For the thermochemical pathways, we assumed a plant equipment electrical efficiency of 75 percent, which applied only for our calculation of direct electricity consumption for thermochemical pathways. Our efficiency and selectivity assumptions along with the associated references are summarized in Table A1.

The conversion values listed in Table A1 give the molar percentage of feedstock CO2 or CO that is ultimately consumed in the reactor. We assume high conversions of 90–100 percent. Though single-pass conversions have not been demonstrated at these levels, we assert these conversion values based on the assumption of using recycle streams with multiple passes to increase overall conversion.

We assumed an electrolyzer capital cost of $1,000/kW for water electrolysis.[101] For electrochemical CO2 reduction pathways, since these technologies have not reached commercial scale and CO2 electrolyzer capital cost data is therefore not available, we modeled the electrolyzer based on water electrolyzers with a capital cost of $1,000/kW. For thermochemical pathways, we obtain capital cost estimates from plant simulations in the literature (Table A2). The annual fixed O&M costs were estimated as a small percentage of the total capital cost (Table A3).

Notes

[1] Joeri Rogelj et al., “Zero Emission Targets as Long-Term Global Goals for Climate Protection,” Environmental Research Letters 10, no. 10 (October 1, 2015): 105007, https://doi.org/10.1088/1748-9326/10/10/105007; V. Masson-Delmotte et al., “Global Warming of 1.5°C. An IPCC Special Report on the Impacts of Global Warming of 1.5°C above Pre-Industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate Change, Sustainable Development, and Efforts to Eradicate Poverty” (Intergovernmental Panel on Climate Change, 2018).

[2] Gunnar Luderer et al., “Residual Fossil CO2 Emissions in 1.5–2 °C Pathways,” Nature Climate Change 8, no. 7 (July 2018): 626–33, https://doi.org/10.1038/s41558-018-0198-6.

[3] P. Psarras, C. M. Woodall, and J. Wilcox, “The Role of Carbon Utilization,” in CDR Primer, ed. B. Kolosz, J. Freeman, and Wilcox, J., n.d.; Committee on Developing a Research Agenda for Utilization of Gaseous Carbon Waste Streams et al., Gaseous Carbon Waste Streams Utilization: Status and Research Needs (Washington, D.C.: National Academies Press, 2019), https://doi.org/10.17226/25232; D. Sandalow et al., “Carbon Dioxide Utilization (CO2U) ICEF Roadmap 2.0” (Innovation for Cool Earth Forum, 2017).

[4] KAPSARC, “CCE Guide Overview,” in Guide to the Circular Carbon Economy, 2020.

[5] Arun Majumdar and John Deutch, “Research Opportunities for CO2 Utilization and Negative Emissions at the Gigatonne Scale,” Joule 2, no. 5 (May 2018): 805–9, https://doi.org/10.1016/j.joule.2018.04.018.

[6] Cameron Hepburn et al., “The Technological and Economic Prospects for CO2 Utilization and Removal,” Nature 575, no. 7781 (November 7, 2019): 87–97, https://doi.org/10.1038/s41586-019-1681-6; Phil De Luna et al., “What Would It Take for Renewably Powered Electrosynthesis to Displace Petrochemical Processes?,” Science 364, no. 6438 (April 26, 2019): eaav3506, https://doi.org/10.1126/science.aav3506; Oleksandr S. Bushuyev et al., “What Should We Make with CO2 and How Can We Make It?,” Joule 2, no. 5 (May 2018): 825–32, https://doi.org/10.1016/j.joule.2017.09.003; Antonio José Martín and Javier Pérez-Ramírez, “Heading to Distributed Electrocatalytic Conversion of Small Abundant Molecules into Fuels, Chemicals, and Fertilizers,” Joule 3, no. 11 (November 2019): 2602–21, https://doi.org/10.1016/j.joule.2019.09.007; IEA, “CCUS in Clean Energy Transitions,” 2020.

[7] Ernst Worrell et al., “Energy Use and Energy Intensity of the U.S. Chemical Industry” (Lawrence Berkeley National Laboratory, 2000).

[8] European Technology Platform for Zero Emission Fossil Fuel Power Plants, “The Costs of CO2 Transport” (European Technology Platform for Zero Emission Fossil Fuel Power Plants, 2010).

[9] European Technology Platform for Zero Emission Fossil Fuel Power Plants.

[10] Sara Budinis et al., “An Assessment of CCS Costs, Barriers and Potential,” Energy Strategy Reviews 22 (November 2018): 61–81, https://doi.org/10.1016/j.esr.2018.08.003.

[11] IEA, “Electricity Market Report – December 2020” (International Energy Agency, 2020).

[12] IRENA, “Renewable Power Generation Costs in 2019,” 2020, 144.

[13] Mahdi Fasihi, Dmitrii Bogdanov, and Christian Breyer, “Techno-Economic Assessment of Power-to-Liquids (PtL) Fuels Production and Global Trading Based on Hybrid PV-Wind Power Plants,” Energy Procedia 99 (November 2016): 243–68, https://doi.org/10.1016/j.egypro.2016.10.115.

[14] Adam Christensen, “Assessment of Hydrogen Production Costs from Electrolysis: United States and Europe,” June 1, 2020, 72.

[15] S. Schlömer et al., Annex III: Technology-Specific Cost and Performance Parameters. In: Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, ed. O. Edenhofer et al. (Cambridge: Cambridge University Press, 2014), https://doi.org/10.1017/CBO9781107415416.

[16] Budinis et al., “An Assessment of CCS Costs, Barriers and Potential.”

[17] Jeff Tollefson, “Sucking Carbon Dioxide from Air Is Cheaper than Scientists Thought,” Nature 558, no. 7709 (June 7, 2018): 173–173, https://doi.org/10.1038/d41586-018-05357-w; David W. Keith et al., “A Process for Capturing CO2 from the Atmosphere,” Joule 2, no. 8 (August 2018): 1573–94, https://doi.org/10.1016/j.joule.2018.05.006.

[18] Keith et al., “A Process for Capturing CO2 from the Atmosphere”; Mahdi Fasihi, Olga Efimova, and Christian Breyer, “Techno-Economic Assessment of CO2 Direct Air Capture Plants,” Journal of Cleaner Production 224 (July 2019): 957–80, https://doi.org/10.1016/j.jclepro.2019.03.086.

[19] Pierre Friedlingstein et al., “Global Carbon Budget 2019,” Earth System Science Data 11, no. 4 (December 4, 2019): 1783–1838, https://doi.org/10.5194/essd-11-1783-2019.

[20] Portland Cement Association, “How Concrete Is Made,” 2019, https://www.cement.org/cement-concrete/how-cement-is-made.

[21] Kenneth Gillingham and James H. Stock, “The Cost of Reducing Greenhouse Gas Emissions,” Journal of Economic Perspectives 32, no. 4 (November 1, 2018): 53–72, https://doi.org/10.1257/jep.32.4.53; S Julio Friedmann et al., “Levelized Cost Of Carbon Abatement: An Improved Cost-Assessment Methodology For A Net-Zero Emissions World” (Center on Global Energy Policy, 2020).

[22] IEA, “Energy Subsidies,” 2020.

[23] IRENA, “Renewable Power Generation Costs in 2019.”

[24] IRENA, “Future of Wind” (International Renewable Energy Agency, 2019); NREL, “Electricity Annual Technology Baseline (ATB)” (National Renewable Energy Laboratory, 2020).

[25] European Technology Platform for Zero Emission Fossil Fuel Power Plants, “The Costs of CO2 Transport.”

[26] European Technology Platform for Zero Emission Fossil Fuel Power Plants.

[27] Alex Tullo, “C&en’s Global Top 50,” C&en, n.d., 7.

[28] IRENA, “Renewable Capacity Statistics 2020” (International Renewable Energy Agency, 2020).

[29] IEA, “Renewables 2020 - Analysis and Forecast to 2025” (International Energy Agency, 2020).

[30] IEA, “IEA Hydrogen Project Database” (International Energy Agency, 2020).

[31] Enerdata, “Global Energy Statistical Yearbook 2020,” 2020.

[32] IEA, “Offshore Wind Outlook 2019: World Energy Outlook Special Report,” Offshore Wind, 2019, 98; Xi Lu, Michael B McElroy, and Juha Kiviluoma, “Global Potential for Wind-Generated Electricity,” Proceedings of the National Academy of Sciences 106, no. 27 (2009): 10933–38; Athina Korfiati et al., “Estimation of the Global Solar Energy Potential and Photovoltaic Cost with the Use of Open Data,” International Journal of Sustainable Energy Planning and Management, March 29, 2016, 17-30 Pages, https://doi.org/10.5278/IJSEPM.2016.9.3.

[33] IEA, “The Future of Hydrogen,” 2019, https://www.iea.org/reports/the-future-of-hydrogen.

[34] IEA, “Low-Carbon Hydrogen Production, 2010-2030, Historical, Announced and in the Sustainable Development Scenario, 2030” (International Energy Agency, 2020).

[35] Haiyan Yang et al., “A Review of the Catalytic Hydrogenation of Carbon Dioxide into Value-Added Hydrocarbons,” Catalysis Science & Technology 7, no. 20 (2017): 4580–98, https://doi.org/10.1039/C7CY01403A; Wenchao Ma et al., “Electrocatalytic Reduction of CO2 to Ethylene and Ethanol through Hydrogen-Assisted C–C Coupling over Fluorine-Modified Copper,” Nature Catalysis 3, no. 6 (June 2020): 478–87, https://doi.org/10.1038/s41929-020-0450-0.

[36] Brian M. Tackett, Elaine Gomez, and Jingguang G. Chen, “Net Reduction of CO2 via Its Thermocatalytic and Electrocatalytic Transformation Reactions in Standard and Hybrid Processes,” Nature Catalysis 2, no. 5 (May 2019): 381–86, https://doi.org/10.1038/s41929-019-0266-y.

[37] Sören Dresp et al., “Direct Electrolytic Splitting of Seawater: Opportunities and Challenges,” ACS Energy Letters 4, no. 4 (April 12, 2019): 933–42, https://doi.org/10.1021/acsenergylett.9b00220; Amar A. Bhardwaj et al., “Ultrathin Silicon Oxide Overlayers Enable Selective Oxygen Evolution from Acidic and Unbuffered PH-Neutral Seawater,” ACS Catalysis 11, no. 3 (February 5, 2021): 1316–30, https://doi.org/10.1021/acscatal.0c04343.

[38] Charles C. L. McCrory et al., “Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for Solar Water Splitting Devices,” Journal of the American Chemical Society 137, no. 13 (April 8, 2015): 4347–57, https://doi.org/10.1021/ja510442p.

[39] IEA, “The Future of Hydrogen.”

[40] Zachary D. Weller, Steven P. Hamburg, and Joseph C. von Fischer, “A National Estimate of Methane Leakage from Pipeline Mains in Natural Gas Local Distribution Systems,” Environmental Science & Technology 54, no. 14 (July 21, 2020): 8958–67, https://doi.org/10.1021/acs.est.0c00437.

[41] Tackett, Gomez, and Chen, “Net Reduction of CO2 via Its Thermocatalytic and Electrocatalytic Transformation Reactions in Standard and Hybrid Processes”; B. Parkinson et al., “Levelized Cost of CO2 Mitigation from Hydrogen Production Routes,” Energy & Environmental Science 12, no. 1 (2019): 19–40, https://doi.org/10.1039/C8EE02079E; Viola Becattini, Paolo Gabrielli, and Marco Mazzotti, “Role of Carbon Capture, Storage, and Utilization to Enable a Net-Zero-CO2 -Emissions Aviation Sector,” Industrial & Engineering Chemistry Research, January 5, 2021, https://doi.org/10.1021/acs.iecr.0c05392.

[42] Will Gorman, Andrew Mills, and Ryan Wiser, “Improving Estimates of Transmission Capital Costs for Utility-Scale Wind and Solar Projects to Inform Renewable Energy Policy,” Energy Policy 135 (December 2019): 110994, https://doi.org/10.1016/j.enpol.2019.110994.

[43] S Julio Friedmann et al., “Net-Zero and Geospheric Return: Actions for 2030 and Beyond” (Center on Global Energy Policy and Global CCS Institute, 2020).

[44] Varun Sivaram et al., Energizing America: A Roadmap to Launch a National Energy Innovation Mission, 2020.

[45] “Buy Clean California Act,” Pub. L. No. AB 262 (2017).

[46] Todd Kaminsky, “The New York State Low Embodied Carbon Concrete Leadership Act,” Pub. L. No. S8965 (2020).

[47] Varun Sivaram et al., “To Bring Emissions-Slashing Technologies to Market, the United States Needs Targeted Demand-Pull Innovation Policies” (Center on Global Energy Policy, 2021).

[48] Biden for President, “The Biden Plan to Build a Modern, Sustainable Infrastructure and an Equitable Clean Energy Future,” 2020; European Commission, “The European Green Deal,” 2019; Government of Canada, “Hydrogen Strategy for Canada,” 2020, 141.

[49] De Luna et al., “What Would It Take for Renewably Powered Electrosynthesis to Displace Petrochemical Processes?”; Christensen, “Assessment of Hydrogen Production Costs from Electrolysis: United States and Europe”; Parkinson et al., “Levelized Cost of CO2 Mitigation from Hydrogen Production Routes.”

[50] Rainer Küngas, “Review—Electrochemical CO2 Reduction for CO Production: Comparison of Low- and High-Temperature Electrolysis Technologies,” Journal of The Electrochemical Society 167, no. 4 (February 14, 2020): 044508, https://doi.org/10.1149/1945-7111/ab7099; Régis Anghilante et al., “Bottom-up Cost Evaluation of SOEC Systems in the Range of 10–100 MW,” International Journal of Hydrogen Energy 43, no. 45 (November 2018): 20309–22, https://doi.org/10.1016/j.ijhydene.2018.08.161; Xiaogang Li et al., “Exclusive Ni–N4 Sites Realize Near-Unity CO Selectivity for Electrochemical CO 2 Reduction,” Journal of the American Chemical Society 139, no. 42 (October 25, 2017): 14889–92, https://doi.org/10.1021/jacs.7b09074; Rahman Daiyan et al., “Antipoisoning Nickel–Carbon Electrocatalyst for Practical Electrochemical CO2 Reduction to CO,” 2019, 8.

[51] Stephanie Nitopi et al., “Progress and Perspectives of Electrochemical CO2 Reduction on Copper in Aqueous Electrolyte,” Chemical Reviews 119, no. 12 (June 26, 2019): 7610–72, https://doi.org/10.1021/acs.chemrev.8b00705; IEA, “Methane Tracker 2020” (International Energy Agency, 2020); Hanqing Pan and Christopher J. Barile, “Electrochemical CO2 Reduction to Methane with Remarkably High Faradaic Efficiency in the Presence of a Proton Permeable Membrane,” Energy & Environmental Science 13, no. 10 (2020): 3567–78, https://doi.org/10.1039/D0EE02189J.

[52] Tackett, Gomez, and Chen, “Net Reduction of CO2 via Its Thermocatalytic and Electrocatalytic Transformation Reactions in Standard and Hybrid Processes”; Wilson A. Smith et al., “Pathways to Industrial-Scale Fuel Out of Thin Air from CO2 Electrolysis,” Joule 3, no. 8 (August 2019): 1822–34, https://doi.org/10.1016/j.joule.2019.07.009; Lu Lu et al., “Highly Efficient Electroreduction of CO2 to Methanol on Palladium-Copper Bimetallic Aerogels,” Angewandte Chemie International Edition 57, no. 43 (October 22, 2018): 14149–53, https://doi.org/10.1002/anie.201808964; Dexin Yang et al., “Selective Electroreduction of Carbon Dioxide to Methanol on Copper Selenide Nanocatalysts,” Nature Communications 10, no. 1 (December 2019): 677, https://doi.org/10.1038/s41467-019-08653-9; Xiaofu Sun et al., “Molybdenum-Bismuth Bimetallic Chalcogenide Nanosheets for Highly Efficient Electrocatalytic Reduction of Carbon Dioxide to Methanol,” Angewandte Chemie International Edition 55, no. 23 (June 1, 2016): 6771–75, https://doi.org/10.1002/anie.201603034.

[53] Ma et al., “Electrocatalytic Reduction of CO2 to Ethylene and Ethanol through Hydrogen-Assisted C–C Coupling over Fluorine-Modified Copper”; Madhav Ghanta, Darryl Fahey, and Bala Subramaniam, “Environmental Impacts of Ethylene Production from Diverse Feedstocks and Energy Sources,” Applied Petrochemical Research 4, no. 2 (June 2014): 167–79, https://doi.org/10.1007/s13203-013-0029-7; Ye Chen et al., “Ethylene Selectivity in Electrocatalytic CO2 Reduction on Cu Nanomaterials: A Crystal Phase-Dependent Study,” Journal of the American Chemical Society 142, no. 29 (July 22, 2020): 12760–66, https://doi.org/10.1021/jacs.0c04981; Cao-Thang Dinh et al., “CO2 Electroreduction to Ethylene via Hydroxide-Mediated Copper Catalysis at an Abrupt Interface,” Science 360, no. 6390 (May 18, 2018): 783–87, https://doi.org/10.1126/science.aas9100.

[54] Chung Shou Chen, Jane Hui Wan, and Boon Siang Yeo, “Electrochemical Reduction of Carbon Dioxide to Ethane Using Nanostructured Cu2O-Derived Copper Catalyst and Palladium(II) Chloride,” The Journal of Physical Chemistry C 119, no. 48 (December 3, 2015): 26875–82, https://doi.org/10.1021/acs.jpcc.5b09144; Anthony Vasileff et al., “Electrochemical Reduction of CO2 to Ethane through Stabilization of an Ethoxy Intermediate,” Angewandte Chemie 132, no. 44 (October 26, 2020): 19817–21, https://doi.org/10.1002/ange.202004846.

[55] Jing Yuan et al., “Efficient Electrochemical Reduction of CO2 to Ethanol on Cu Nanoparticles Decorated on N-Doped Graphene Oxide Catalysts,” Journal of CO2 Utilization 33 (October 2019): 452–60, https://doi.org/10.1016/j.jcou.2019.07.014; Dilan Karapinar et al., “Electrochemical CO2 Reduction to Ethanol with Copper-Based Catalysts,” ACS Energy Letters 6, no. 2 (February 12, 2021): 694–706, https://doi.org/10.1021/acsenergylett.0c02610; Yangyang Zhang et al., “Cu/Cu2O Nanoparticles Supported on Vertically ZIF-L-Coated Nitrogen-Doped Graphene Nanosheets for Electroreduction of CO2 to Ethanol,” ACS Applied Nano Materials 3, no. 1 (January 24, 2020): 257–63, https://doi.org/10.1021/acsanm.9b01935.

[56] Ji Hoon Lee et al., “Tuning the Activity and Selectivity of Electroreduction of CO2 to Synthesis Gas Using Bimetallic Catalysts,” Nature Communications 10, no. 1 (December 2019): 3724, https://doi.org/10.1038/s41467-019-11352-0; Qun He et al., “Electrochemical Conversion of CO2 to Syngas with Controllable CO/H2 Ratios over Co and Ni Single‐Atom Catalysts,” Angewandte Chemie International Edition 59, no. 8 (February 17, 2020): 3033–37, https://doi.org/10.1002/anie.201912719.

[57] Run-Ping Ye et al., “CO2 Hydrogenation to High-Value Products via Heterogeneous Catalysis,” Nature Communications 10, no. 1 (December 2019): 5698, https://doi.org/10.1038/s41467-019-13638-9; Xiaoliang Liu et al., “Selective Transformation of Carbon Dioxide into Lower Olefins with a Bifunctional Catalyst Composed of ZnGa2O4 and SAPO-34,” Chemical Communications 54, no. 2 (2018): 140–43, https://doi.org/10.1039/C7CC08642C; Jian Li et al., “Carbon Dioxide Hydrogenation to Light Olefins over ZnO-Y2O3 and SAPO-34 Bifunctional Catalysts,” Catalysis Communications 129 (September 2019): 105711, https://doi.org/10.1016/j.catcom.2019.105711; Mehdi Sedighi and Majid Mohammadi, “CO2 Hydrogenation to Light Olefins over Cu-CeO2/SAPO-34 Catalysts: Product Distribution and Optimization,” Journal of CO2 Utilization 35 (January 2020): 236–44, https://doi.org/10.1016/j.jcou.2019.10.002; Thai Ngan Do and Jiyong Kim, “Green C2-C4 Hydrocarbon Production through Direct CO2 Hydrogenation with Renewable Hydrogen: Process Development and Techno-Economic Analysis,” Energy Conversion and Management 214 (June 2020): 112866, https://doi.org/10.1016/j.enconman.2020.112866.

[58] Liangshu Zhong et al., “Cobalt Carbide Nanoprisms for Direct Production of Lower Olefins from Syngas,” Nature 538, no. 7623 (October 2016): 84–87, https://doi.org/10.1038/nature19786; Bingxu Chen et al., “Tailoring of Fe/MnK-CNTs Composite Catalysts for the Fischer–Tropsch Synthesis of Lower Olefins from Syngas,” Industrial & Engineering Chemistry Research 57, no. 34 (August 29, 2018): 11554–60, https://doi.org/10.1021/acs.iecr.8b01795; H. M. Torres Galvis et al., “Supported Iron Nanoparticles as Catalysts for Sustainable Production of Lower Olefins,” Science 335, no. 6070 (February 17, 2012): 835–38, https://doi.org/10.1126/science.1215614; Xiaoliang Liu et al., “Design of Efficient Bifunctional Catalysts for Direct Conversion of Syngas into Lower Olefins via Methanol/Dimethyl Ether Intermediates,” Chemical Science 9, no. 20 (2018): 4708–18, https://doi.org/10.1039/C8SC01597J; Yuping Liu et al., “Low-Olefin Production Process Based on Fischer–Tropsch Synthesis: Process Synthesis, Optimization, and Techno-Economic Analysis,” Industrial & Engineering Chemistry Research 59, no. 18 (May 6, 2020): 8728–39, https://doi.org/10.1021/acs.iecr.0c00542; F. Jiao et al., “Selective Conversion of Syngas to Light Olefins,” Science 351, no. 6277 (March 4, 2016): 1065–68, https://doi.org/10.1126/science.aaf1835.

[59] Juan Navarro et al., “Policies and Motivations for the CO2 Valorization through the Sabatier Reaction Using Structured Catalysts. A Review of the Most Recent Advances,” Catalysts 8, no. 12 (November 22, 2018): 578, https://doi.org/10.3390/catal8120578; M. A. A. Aziz et al., “CO2 Methanation over Heterogeneous Catalysts: Recent Progress and Future Prospects,” Green Chemistry 17, no. 5 (2015): 2647–63, https://doi.org/10.1039/C5GC00119F; William L. Becker, Michael Penev, and Robert J. Braun, “Production of Synthetic Natural Gas From Carbon Dioxide and Renewably Generated Hydrogen: A Techno-Economic Analysis of a Power-to-Gas Strategy,” Journal of Energy Resources Technology 141, no. 2 (February 1, 2019): 021901, https://doi.org/10.1115/1.4041381.

[60] Tackett, Gomez, and Chen, “Net Reduction of CO2 via Its Thermocatalytic and Electrocatalytic Transformation Reactions in Standard and Hybrid Processes”; Oliver Martin et al., “Indium Oxide as a Superior Catalyst for Methanol Synthesis by CO2 Hydrogenation,” Angewandte Chemie International Edition 55, no. 21 (May 17, 2016): 6261–65, https://doi.org/10.1002/anie.201600943; Mar Pérez-Fortes et al., “Methanol Synthesis Using Captured CO2 as Raw Material: Techno-Economic and Environmental Assessment,” Applied Energy 161 (January 2016): 718–32, https://doi.org/10.1016/j.apenergy.2015.07.067.

[61] Ivan Muñoz et al., “Life Cycle Assessment of Bio-Based Ethanol Produced from Different Agricultural Feedstocks,” The International Journal of Life Cycle Assessment 19, no. 1 (January 2014): 109–19, https://doi.org/10.1007/s11367-013-0613-1; D Humbird et al., “Process Design and Economics for Biochemical Conversion of Lignocellulosic Biomass to Ethanol: Dilute-Acid Pretreatment and Enzymatic Hydrolysis of Corn Stover,” March 1, 2011, https://doi.org/10.2172/1013269.

[62] Shaik Afzal et al., “Minimizing CO2 Emissions for Syngas Production Units Using Dry Reforming of Methane,” in Computer Aided Chemical Engineering, vol. 40 (Elsevier, 2017), 2617–22, https://doi.org/10.1016/B978-0-444-63965-3.50438-4; Ebrahim Rezaei and Stephen Dzuryk, “Techno-Economic Comparison of Reverse Water Gas Shift Reaction to Steam and Dry Methane Reforming Reactions for Syngas Production,” Chemical Engineering Research and Design 144 (April 2019): 354–69, https://doi.org/10.1016/j.cherd.2019.02.005.

[63] Becattini, Gabrielli, and Mazzotti, “Role of Carbon Capture, Storage, and Utilization to Enable a Net-Zero-CO2 -Emissions Aviation Sector”; Jie Li et al., “Integrated Tuneable Synthesis of Liquid Fuels via Fischer–Tropsch Technology,” Nature Catalysis 1, no. 10 (October 2018): 787–93, https://doi.org/10.1038/s41929-018-0144-z; Timothy Skone, “NETL Fischer-Tropsch Black Box Model,” 2015; Yuefeng Liu et al., “Titania-Decorated Silicon Carbide-Containing Cobalt Catalyst for Fischer–Tropsch Synthesis,” ACS Catalysis 3, no. 3 (March 2013): 393–404, https://doi.org/10.1021/cs300729p; Meng Yang et al., “Selective Fischer–Tropsch Synthesis for Jet Fuel Production over Y3+ Modified Co/H-β Catalysts,” Sustainable Energy & Fuels 4, no. 7 (2020): 3528–36, https://doi.org/10.1039/D0SE00468E.

[64] Fasihi, Bogdanov, and Breyer, “Techno-Economic Assessment of Power-to-Liquids (PtL) Fuels Production and Global Trading Based on Hybrid PV-Wind Power Plants”; Ye et al., “CO2 Hydrogenation to High-Value Products via Heterogeneous Catalysis”; Peng Gao et al., “Direct Conversion of CO2 into Liquid Fuels with High Selectivity over a Bifunctional Catalyst,” Nature Chemistry 9, no. 10 (October 2017): 1019–24, https://doi.org/10.1038/nchem.2794; Chundong Zhang et al., “Direct Conversion of Carbon Dioxide to Liquid Fuels and Synthetic Natural Gas Using Renewable Power: Techno-Economic Analysis,” Journal of CO2 Utilization 34 (December 2019): 293–302, https://doi.org/10.1016/j.jcou.2019.07.005; Benzhen Yao et al., “Transforming Carbon Dioxide into Jet Fuel Using an Organic Combustion-Synthesized Fe-Mn-K Catalyst,” Nature Communications 11, no. 1 (December 2020): 6395, https://doi.org/10.1038/s41467-020-20214-z.

[65] Jiranart Jeenchay and Kitipat Siemanond, “Ammonia/Urea Production Process Simulation/Optimization with Techno-Economic Analysis,” in Computer Aided Chemical Engineering, vol. 43 (Elsevier, 2018), 385–90, https://doi.org/10.1016/B978-0-444-64235-6.50070-X; Hanfei Zhang et al., “Techno-Economic Comparison of 100% Renewable Urea Production Processes,” Applied Energy 284 (February 2021): 116401, https://doi.org/10.1016/j.apenergy.2020.116401.

[66] Hilal El-Hassan and Yixin Shao, “Carbon Storage through Concrete Block Carbonation,” Journal of Clean Energy Technologies 2, no. 3 (2014): 287–91, https://doi.org/10.7763/JOCET.2014.V2.141; Duo Zhang, Zaid Ghouleh, and Yixin Shao, “Review on Carbonation Curing of Cement-Based Materials,” Journal of CO2 Utilization 21 (October 2017): 119–31, https://doi.org/10.1016/j.jcou.2017.07.003; Yixin Shao, “Beneficial Use of Carbon Dioxide in Precast Concrete Production,” June 26, 2014, https://doi.org/10.2172/1155035.

[67] CarbonCure Technologies Inc., “CarbonCure’s 500 Mt CO2 Reduction Technical Roadmap” (CarbonCure Technologies Inc., 2020).

[68] Sara Budinis et al., “An Assessment of CCS Costs, Barriers and Potential,” Energy Strategy Reviews 22 (November 2018): 61–81, https://doi.org/10.1016/j.esr.2018.08.003.

[69] European Technology Platform for Zero Emission Fossil Fuel Power Plants, “The Costs of CO2 Transport” (European Technology Platform for Zero Emission Fossil Fuel Power Plants, 2010).

[70] Will Gorman, Andrew Mills, and Ryan Wiser, “Improving Estimates of Transmission Capital Costs for Utility-Scale Wind and Solar Projects to Inform Renewable Energy Policy,” Energy Policy 135 (December 2019): 110994, https://doi.org/10.1016/j.enpol.2019.110994.

[71] S. Schlömer et al., Annex III: Technology-Specific Cost and Performance Parameters. In: Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, ed. O. Edenhofer et al. (Cambridge: Cambridge University Press, 2014), https://doi.org/10.1017/CBO9781107415416.

[72] Nitopi et al., “Progress and Perspectives of Electrochemical CO2 Reduction on Copper in Aqueous Electrolyte.”

[73] Lu et al., “Highly Efficient Electroreduction of CO2 to Methanol on Palladium-Copper Bimetallic Aerogels”; Yang et al., “Selective Electroreduction of Carbon Dioxide to Methanol on Copper Selenide Nanocatalysts”; Sun et al., “Molybdenum-Bismuth Bimetallic Chalcogenide Nanosheets for Highly Efficient Electrocatalytic Reduction of Carbon Dioxide to Methanol.”

[74] Hessam Jahangiri et al., “A Review of Advanced Catalyst Development for Fischer–Tropsch Synthesis of Hydrocarbons from Biomass Derived Syn-Gas,” Catal. Sci. Technol. 4, no. 8 (2014): 2210–29, https://doi.org/10.1039/C4CY00327F.

[75] Ye et al., “CO2 Hydrogenation to High-Value Products via Heterogeneous Catalysis.”

[76] Ye et al.

[77] Charlotte Vogt et al., “The Renaissance of the Sabatier Reaction and Its Applications on Earth and in Space,” Nature Catalysis 2, no. 3 (March 2019): 188–97, https://doi.org/10.1038/s41929-019-0244-4.

[78] L. Pastor-Pérez et al., “CO2 Valorisation via Reverse Water-Gas Shift Reaction Using Advanced Cs Doped Fe-Cu/Al2O3 Catalysts,” Journal of CO2 Utilization 21 (October 2017), https://doi.org/10.1016/j.jcou.2017.08.009.

[79] Carl Bosch and Wilhelm Meiser, Process of Manufacturing Urea, United States Patent Office US1429483A, filed 1920, and issued 1922.

[80] IEA, “The Future of Hydrogen.”

[81] H. Zabed et al., “Fuel Ethanol Production from Lignocellulosic Biomass: An Overview on Feedstocks and Technological Approaches,” Renewable and Sustainable Energy Reviews 66 (December 2016): 751–74, https://doi.org/10.1016/j.rser.2016.08.038.

[82] Shao, “Beneficial Use of Carbon Dioxide in Precast Concrete Production.”

[83] CarbonCure Technologies Inc., “CarbonCure’s 500 Mt CO2 Reduction Technical Roadmap.”

[84] Ye et al., “CO2 Hydrogenation to High-Value Products via Heterogeneous Catalysis.”

[85] S Julio Friedmann, Zhiyuan Fan, and Ke Tang, “Low-Carbon Heat Solutions for Heavy Industry: Sources, Options and Costs Today” (Center on Global Energy Policy, 2019).

[86] Liu et al., “Low-Olefin Production Process Based on Fischer–Tropsch Synthesis”; Do and Kim, “Green C2-C4 Hydrocarbon Production through Direct CO2 Hydrogenation with Renewable Hydrogen.”

[87] Gregor Wernet et al., “The Ecoinvent Database Version 3 (Part I): Overview and Methodology,” The International Journal of Life Cycle Assessment 21, no. 9 (September 2016): 1218–30, https://doi.org/10.1007/s11367-016-1087-8.

[88] S. Schlömer et al., Climate Change 2014 Mitigation of Climate Change.

[89] State of Oregon Department of Environmental Quality, “Calculating Greenhouse Gas Emissions from Steam Production,” 2019.

[90] Worrell et al., “Energy Use and Energy Intensity of the U.S. Chemical Industry.”

[91] Fasihi, Bogdanov, and Breyer, “Techno-Economic Assessment of Power-to-Liquids (PtL) Fuels Production and Global Trading Based on Hybrid PV-Wind Power Plants.”

[92] IRENA, “Renewable Power Generation Costs in 2019.”

[93] Gorman, Mills, and Wiser, “Improving Estimates of Transmission Capital Costs for Utility-Scale Wind and Solar Projects to Inform Renewable Energy Policy.”

[94] European Technology Platform for Zero Emission Fossil Fuel Power Plants, “The Costs of CO2 Transport.”

[95] Budinis et al., “An Assessment of CCS Costs, Barriers and Potential.”

[96] IEA, “Electricity Market Report – December 2020.”

[97] IRENA, “Renewable Power Generation Costs in 2019.”

[98] Fasihi, Bogdanov, and Breyer, “Techno-Economic Assessment of Power-to-Liquids (PtL) Fuels Production and Global Trading Based on Hybrid PV-Wind Power Plants.”

[99] Ye et al., “CO2 Hydrogenation to High-Value Products via Heterogeneous Catalysis.”

[100] McCrory et al., “Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for Solar Water Splitting Devices.”

[101] Christensen, “Assessment of Hydrogen Production Costs from Electrolysis: United States and Europe”; Zhiyuan Fan et al., “Green Hydrogen in a Circular Carbon Economy: Opportunities and Limits” (Center on Global Energy Policy, 2021).