Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2026 Jan 27;60(7):5505–5517. doi: 10.1021/acs.est.5c17460

Co-Location of Cellulosic Bioethanol and Alcohol-to-Jet (ATJ) Production Facilities for Targeted Scale-Up of Sustainable Aviation Fuel (SAF) Production

Maria F Bianco †,‡, Wenjun Guo †,‡, Madhu Khanna †,§,∥, Jeremy S Guest †,‡,§,*
PMCID: PMC12947612  PMID: 41592292

Abstract

Achieving aerospace industry net-zero emissions by 2050 requires rapid scaling of sustainable aviation fuel (SAF) production. Leveraging existing infrastructure, proven technologies like Alcohol-to-Jet (ATJ), and low carbon intensity (CI) feedstocks (e.g., switchgrass and miscanthus) can support this transition and help achieve near-term emissions reduction targets. This study evaluates the implications of lignocellulosic ethanol biorefinery siting and integration with petroleum refineries to produce SAF across 1000 sites randomly sampled from areas suitable for perennial grasses in the U.S. rainfed region. To better understand the logistics of material transport and handoffs, we integrated models of biomass harvest, transport, ethanol, and ATJ production in a stochastic framework based on Monte Carlo simulations to characterize SAF minimum selling price (MSP) and carbon intensity (CI), considering site-specific parameters (e.g., feedstock production, transportation, taxes, incentives). The results indicate trade-offs between MSP and CI across locations, with median MSP ranging from 7.9 to 12.8 USD·gal–1 and CI from −9.7 to 39.4 gCO2e·MJ–1. Despite high estimated decarbonization costs (580 USD·tonCO2e–1), our results indicate that site-specific deployment of ATJ with low-CI feedstocks can improve sustainability outcomes. The framework provides a systematic approach to assess cost and sustainability trade-offs across locations, considering the end-to-end supply chain and supporting an informed investment in SAF production.

Keywords: biorefinery siting, minimum selling price, global warming potential, location-specific analysis, supply chain

graphic file with name es5c17460_0008.jpg

graphic file with name es5c17460_0006.jpg

Introduction

The burning of fossil fuels is the largest contributor to global anthropogenic greenhouse gas (GHG) emissions, a major driver of climate change. To combat climate change and mitigate its effects, the United States (U.S.) aviation industry aims for net-zero emissions by 2050, − prioritizing sustainable biofuels in the near term as electrification remains challenging. The sustainable aviation fuel (SAF) Grand Challenge sets the target to meet 100% of aviation fuel demand by 2050, with a minimum of a 50% reduction in GHG emissions compared to fossil-derived fuel; this challenge is worthy of its name, given that the demand for jet fuel is expected to double to 35 billion gallons per year by 2050. In the near term, meeting the 2030 target of 3 billion gallons per year will require SAF production to rapidly scale up from early 2025 levels of 30,000 barrels per day (∼460 million gallons per year, less than 2% of current aviation fuel consumption of 1.7 million barrels per day). Leveraging existing assets at petroleum refineries and prioritizing commercially ready conversion technologies, such as alcohol-to-jet (ATJ), could be a stepping stone to achieving near-term targets and to scale up for future ones. While current biofuel producers are pursuing ATJ pathways primarily through new or dedicated facilities rather than retrofits of existing petroleum refineries, , this study highlights the potential to expand low-CI ethanol production across regions and leverage these new ethanol streams within existing petroleum refinery infrastructure to accelerate near-term SAF deployment.

Integrating ethanol upgrading to SAF at existing petroleum refineries creates significant opportunities to reduce the capital expenditures (CAPEX) and operating costs (OPEX), , which together represent one of the major barriers to SAF adoption. , While Tanzil et al. reported cost and GHG benefits from integrating lignocellulosic SAF pathways into petroleum refineries and Carlson et al. identified economic advantages under certain incentive structures, our study instead evaluates spatial and supply chain dynamics by analyzing the implications of deploying new, low-CI ethanol refineries to supply petroleum refineries producing SAF (via ATJ) on-site. Producing SAF at petroleum refineries allows it to be blended with Jet A fuel on-site, once ASTM-certified, and continue through the existing supply chain to terminals and airports via pipelines, − avoiding the need for separate transportation to blending terminals or dedicated pipelines. Additionally, current ethanol production relies primarily on corn, but the long-term goal is to convert it to more sustainable feedstocks. Jet A has an estimated CI of 84–90 gCO2e·MJ–1, ,,, while SAF derived from corn ethanol achieves an estimated 61 gCO2e·MJ–1. , However, to meet climate targets, policies have set target CIs of 47.5 gCO2e·MJ–1 or lower. Cellulosic ethanol has the potential to achieve more significant CI reductions, − with the challenge remaining to determine optimal locations for new biorefineries that take into consideration potential feedstock availability, feedstock CI, and the existing locations of petroleum refineries or fuel blending terminals.

In transitioning jet fuel production from fossil fuels to renewable carbon resources, a comprehensive view of the future supply chainincluding spatially explicit life cycle costs and emissions of future feedstock production scenariosis critical. Beginning with existing bioethanol production, for instance, even full redirection of current ethanol output to SAF production would meet only ∼43% of 2025 Jet A demand − (assuming an optimistic conversion factor of 0.56 gallons of jet per gallon of ethanol; Figure ). This underscores the need for additional bioethanol production capacity and strategic siting of new facilities if the ATJ pathway is to be scaled up, especially as jet fuel demand continues to grow. To meet the SAF Grand Challenge goals, the U.S. Department of Energy released a summary report from its 2023 Request for Information. The report underscores the need for models and simulations to assess supply chain scenarios, industry demand for biorefinery siting tools, and state-specific simulations. It also highlights the importance of identifying near-term opportunities to build the supply chain for large-scale SAF deployment, while expressing concern about the uncertainties and risks of early SAF facility investments. Some approaches for SAF supply chain modeling and location decisions include mixed integer linear programming (MILP) to minimize total SAF supply chain costs, , geographic information system (GIS)-based models (sometimes combined with MILP), − heuristics methods, or a combination of modeling tools. However, most existing analyses are restricted to a specific state or geographic region and do not integrate with existing jet fuel infrastructure. Compared to production at a greenfield facility, producing ATJ-SAF by partially repurposing a petroleum refinery is estimated to reduce the minimum selling price (MSP) by 14% and capital investment by 46% (described in detail in ). While Guo et al. focused on techno-economic and environmental benefits of three levels of refinery integration (stand-alone, colocated, or repurposed) in general, this study extends the analysis by incorporating spatially explicit feedstock production, transport logistics, and the optimization of product handoffs from ethanol refineries to regional jet fuel production facilities. Ultimately, charting long-term pathways to SAF production scale-up will require spatially explicit, robust modeling tools that, in addition to simulating the cultivation of more sustainable feedstocks, incorporate the logistics of feedstock and product handoffs among supply chain actors and the integration of new facilities with existing fuel production infrastructure.

1.

1

Open in a new tab

Location of jet fuel producers and existing bioethanol facilities. For petroleum refineries, the size of the circles represents installed capacity − in thousands of barrels of jet fuel per day (thousands of bpd). For ethanol facilities, the size of the circles represents the potential jet fuel production capacity they could support (via ATJ), considering a conversion factor of 0.56 gal jet per gal of ethanol, 42 gal of jet fuel per barrel, and 350 working days per year.

The objectives of this work were (1) to develop an open-source simulation platform to enable spatially explicit field-to-fuel simulation and analysis (including minimum selling price, MSP, and carbon intensity, CI) for crops to SAF (via ATJ), (2) to determine feasible locations for cellulosic ethanol biorefineries to supply petroleum refineries for SAF production in the U.S., and (3) to characterize the relative performance of this feedstock-to-ethanol-to-SAF pathway across candidate biorefinery locations using MSP and CI as key sustainability indicators. To this end, a feedstock and ethanol transport model was developed in Python to estimate unit feedstock and transportation costs for different candidate biorefinery locations, allocating farms to biorefineries and petroleum refineries to ethanol refineries based on distance. Results were used as input in the BioSTEAM open-source biorefinery simulation using a cellulosic biorefinery for feedstock-to-ethanol conversion and a SAF production simulation, both with location-specific parameters. , Techno-economic analysis (TEA) and a life cycle assessment (LCA) were performed with BioSTEAM to obtain financial and environmental indicators for each candidate location to present the decision space regarding the deployment of this integrated pathway. This analysis was conducted under uncertainty to provide a comprehensive view of site-specific performance, ranking locations by median MSP and CI while also capturing variability driven by spatially explicit parameters. This study advances siting analysis for cellulosic ethanol production and its integration with SAF production through regional supply chain simulation and the integration of logistics modeling with (bio)­refinery design and analysis. It further provides a flexible framework adaptable to diverse feedstocks and bioproducts, along with an open-source tool linking feedstock and ethanol transport/allocation with refinery integration for site-specific techno-economic and environmental evaluation.

Methods

System Overview

To obtain MSP and CI indicators for SAF produced from cellulosic ethanol, we integrated several subsystem models to account for each step of its supply chain. The modeling objective was to find optimal locations for an ethanol facility with respect to jet producers’ locations (petroleum refineries with Jet A production process; Figure ). To explore siting opportunities beyond current ethanol infrastructure, this analysis is not limited to existing locations of bioethanol facilities but was extended to cover possibilities of colocated facilities across the full U.S. rainfed region (east of the 100th meridian).

The four submodels (Figure ) were sequentially simulated to obtain the final SAF indicators of unit costs (USD·gal–1) and GHG emissions (gCO2eq·MJ–1). Our simulations begin with spatially explicit modeling of feedstock production. Transportation is added to account for logistics in the supply of feedstock to the bioethanol refinery and to get feedstock delivered prices and CI across 1000 candidate biorefinery locations distributed across the entire U.S. rainfed region. Feedstock price and CI are then used as inputs in a cellulosic biorefinery model to obtain MSP and CI of ethanol at each location. Ethanol is assumed to be delivered to the nearest jet producer, incurring costs and GHG emissions during transport. Finally, the transformation of ethanol to SAF at the jet producer is simulated to obtain the final indicators of MSP and CI of SAF for each candidate biorefinery location.

2.

2

Open in a new tab

Submodels involved in the location analysis.

Feedstock Production and Conversion

Modeling In-Field Activities

Yields and CI of switchgrass and miscanthus were obtained as geospatial data from Fan et al., who used the biogeochemical process-based model DayCent, − which accounts for weather and spatial variation in growing conditions across the U.S., with a spatial resolution of 4 × 4 km2 (Figure S1). Only grid cells classified as cultivated land, grassland, or shrubland were included in this analysis, representing areas suitable for crop or perennial grass production. Nonarable lands such as urban areas, water bodies, and protected regions were excluded to ensure that only land available for biomass cultivation was considered. Feedstock CI considers soil organic carbon (SOC) change, in-field emissions, and upstream emissions from chemical inputs (nitrogen, phosphorus, and potassium fertilizers; herbicides; diesel fuel; Table S1). Breakeven prices of feedstock at farmgate were also obtained from the same source, which includes the costs of fertilizer, seeds, chemicals, planting and harvesting, on-farm storage, and the annual opportunity cost of land converted from its best alternative use (assuming a 10-year lifespan for switchgrass and 15 for miscanthus; Figure S2).

Biorefinery Models Description

The SAF production model consists of two submodels: (i) ethanol production via fermentation and (ii) subsequent upgrading of ethanol to SAF via the ATJ pathway. Ethanol production from switchgrass and miscanthus was modeled based on the NREL’s design for corn stover to ethanol conversion. For this study, an ethanol biorefinery size of 80 million gallons (MMgal) per year was selected, reflecting the median size of existing commercial ethanol facilities. This capacity was used to estimate the required feedstock collection radius based on regional biomass yields and land availability. The system’s heating and electricity demands were met by combusting process residues in the boiler turbogenerator, with excess electricity exported to the grid. All feedstock is converted to ethanol, with net electricity generation from process residues (stillage, biogas, and dewatered sludge) estimated at 3.7 kWh·gal–1 for switchgrass (fifth–95th percentile: 1.95–5.40 kWh·gal–1) and 1.0 kWh·gal–1 for miscanthus (fifth–95th percentile: 0–2.6 kWh·gal–1). Electricity excess was treated as a coproduct, and credited in the LCA using the displacement method (consistent with , ).

In the second stage, ethanol was upgraded to SAF via dehydration, oligomerization, hydrogenation, and separation. Ethanol upgrading facilities were assumed to be colocated with a petroleum refinery, modifying Guo et al.’s repurposing scenario to assume off-site ethanol production, to reduce capital costs by leveraging existing outside-battery-limit (OSBL) units (e.g., storage, wastewater treatment) and process units (i.e., steam methane reforming reactor for hydrogen production and hydrogenation reactors for subsequent oligomers processing). In addition to SAF, the ATJ process generates gasoline and diesel as coproducts, credited to CI via the displacement method and assumed to be sold at market rates. The biorefinery design was implemented in BioSTEAM, an open-source platform for biorefinery design, simulation, and evaluation (through TEA and LCA) under uncertainty. Unlike spreadsheet-based tools, BioSTEAM supports agile, flexible design by automating refinery scaling and uncertainty propagation, thereby streamlining the evaluation of alternative process configurations.

Linking Biorefinery Models

The bioethanol model takes the feedstock delivered price and CI obtained from the feedstock production and transport model and simulates the ethanol conversion system to obtain the minimum ethanol selling price and the CI of ethanol. It also incorporates location-specific parameters from Stewart et al., such as state-level financial incentives, income and property taxes, electricity unit price and CI, capital cost adjustment factor, and natural gas unit price (Table S2). The ethanol-to-SAF pathway was modeled by considering the results from the bioethanol refinery simulation and the ethanol transport model that accounts for transport prices and distances from candidate biorefinery locations and jet producers. The ethanol transport model determined the flow of ethanol (amount sent) to each jet producer, which would be an input to model the size of the ATJ process and how many jet producers were supplied by each biorefinery. This last factor will influence the final jet fuel cost, taken as a weighted average between the facilities involved.

Logistics Modeling

Transportation of Cellulosic Feedstocks and Ethanol

No intermediate storage or pretreatment was assumed to take place outside the biorefinery. Crops are left to dry in the field, transported as bales, and stored on the farm or outdoors at the ethanol refinery until they are processed. Distances were taken from the North American road network geospatial data set obtained from the Bureau of Transportation Statistics (BTS). In this study, feedstock transport distances were calculated using road network data for all farm-to-refinery pairs. To reduce computational burden and facilitate broader applicability, state-level average tortuosity factors are also included as scalable alternatives (for future work) in the accompanying open-source code (Figure S3). The model allocates feedstock supply by sequentially selecting the nearest farms to the biorefinery until the cumulative biomass input meets the requirement for producing 80 million gallons of ethanol per year, assuming a remanent moisture content of 20% in the transported biomass (see Section S1.4 of the SI for model formulation). Feedstock was assumed to be transported by a flatbed truck with 20 Mg (megagrams or metric tons) capacity in the form of square bales. Ethanol produced at the biorefinery was assumed to be transported by a tanker truck with an average capacity of 9000 gallons. Distances from bioethanol candidate locations to jet producers were calculated based on actual road networks and subjected to uncertainty later in the analysis to account for the incertitude in the exact location of the ethanol biorefineries (see Section S1.5 of the SI for model formulation). To ensure conservative cost estimates, all trucking distances were modeled as round trips, with no backhauls assumed. Compared to other transport models available online (e.g., FTOT), our framework enables high-dimensional scenarios, evaluates metrics across all candidate alternatives, explicitly incorporates uncertainty, and can be easily integrated with refinery design and simulation.

The trucking operational costs were obtained from the American Transportation Research Institute. According to this report, average marginal costs per mile by the U.S. region were considered to be 10% higher for tanker trucks; these costs include fuel, lease or purchase payments of the vehicle, maintenance, insurance, permits, tires, tolls, and driver wages and benefits. Fuel costs were replaced by the average diesel cost by state, also considering an average fuel economy of trucks by state. Operating margins and other costs (differentiated by type of truck) were taken from the ATRI report to obtain final costs per mile (Figure S4), which were divided by the considered capacity to obtain costs per km per Mg by state (Figure S5). All costs are in 2023 U.S. dollars (USD). To account for the GHG emissions from transportation, the average fuel economy of trucks by state was considered, as well as the energy content and the CO2 emissions coefficient of diesel fuel (see SI Section 1.7 and Figure S6 for more details).

Biorefinery Integration with Jet Fuel Production Facilities

Petroleum refineries currently producing jet fuel were filtered, and their jet fuel capacity was considered to determine the maximum possible ethanol supplied to each, as well as their locations for transport distances (Figure ). , Ethanol is transported from the biorefinery to the nearest jet producer until the biorefinery’s entire annual production is delivered or until the blending capacity of the jet producer is reached. In the latter case, the biorefinery would supply the remainder of its production to the next closest jet producer. The model allows multiple bioethanol locations to be linked to the same jet producer, enabling a comparison of alternative supply configurations to identify the most favorable option. For this work, the blending capacity at the jet producers’ facilities refers to the proportion of their current aviation fuel capacity that they are willing to divert to SAF production, leveraging their existing equipment and facilities; this was assumed to be 20% for the pioneers in the coproduction of both fuel types. Opportunity costs from diverting fossil jet fuel production to SAF were included, calculated as the profit loss of the replaced product and yield loss, consistent with the repurposing scenario of Guo et al.

Biorefinery Siting

Siting and Evaluation of Candidate Biorefineries

For this study, 1000 candidate locations were randomly selected from the grid of suitable land for perennial grass production, with selection probability weighted by yields (Figure S7). These sites represent approximate locations for exploring the U.S. rainfed region; actual final siting would require detailed, site-specific assessments that were outside the scope of this study (e.g., transportation access, land ownership, permitting). To determine the optimal locations for bioethanol refineries supplying petroleum refineries, several assumptions and simplifications were made. It was assumed that individual farms would be the size of the DayCent modeling grid resolution (i.e., 4 × 4 km2) and that only 20% of the area (320 ha) would be planted with the chosen feedstock (switchgrass or miscanthus). This assumption yields a regional land utilization factor with a median of 5.5% (ranging from 0% at the fifth percentile to 17% at the 95th percentile; Figure S8). , An exhaustive search-based method was used for this work, since the objective was to analyze trade-offs between locations. The method consists of four subsystem models in a stochastic framework that incorporate inputs represented by probability distributions (Monte Carlo simulations). These models are applied sequentially to narrow the decision space from the initial 1000 candidate locations and highlight locations with better performance for the chosen metrics. The components included in each model are those related to farm activities and crop, transport from the farm to the biorefinery, processing at the biorefinery, transport to the jet fuel producer, and final processing to SAF (Figure ). To assess the sensitivity of results to the selection of candidate biorefinery locations, the MSP and CI of delivered feedstock were recalculated for 30 independent sets of 1000 candidate biorefinery locations; ultimately, results were similar across all simulation sets (Figure S9).

Locality-Specific Analysis of Candidate Sites

Location-specific parameters were considered in the bioethanol refinery to simulate over 37 states (Table S2): these parameters include income, property and state tax rates, electricity and natural gas prices, a location capital cost factor, available incentives for ethanol production, and CI of electricity (by eGRID subregion/balancing regions). State sales tax rate is not included in the analysis since ethanol will be sold as an intermediate product in the supply chain. Location-specific taxes and utility prices were also included in the SAF production process for 44 states.

Uncertainty and Sensitivity Analyses

A total of 78 independent economic, LCA, and technological performance parameters were varied across 1000 Monte Carlo simulations using Latin hypercube sampling (12 parameters for the feedstock and ethanol transport models, 27 for the ethanol refinery, and 39 for the SAF refinery), with distributions assigned based on assumptions from literature (Tables S3–S5). Outputs from Monte Carlo simulations were sequentially passed to the downstream models. A detailed overview of the steps in the uncertainty and sensitivity analyses (Steps 1 through 9), including data handoffs, is provided in Figure . Additional details are also provided below.

3.

3

Open in a new tab

Schematic representation of the integrated modeling framework used to assess the SAF production uncertainty. The diagram shows the interaction between four modules and how uncertainty in input parameters is propagated to evaluate ethanol and SAF MSP and CI.

Uncertainty in Feedstock Production and Transport

Feedstock yields, costs, and logistics parameters like distances and transport capacities were varied in the feedstock production and transport model’s uncertainty analysis, generating 1000 samples for each of the 1000 candidate locations (Table S3; Figure – Step 1). These simulations generated distributions of delivered feedstock price and CI, denoted as x . To limit the computational burden, these values were not directly used as inputs for the ethanol biorefinery model; they were later integrated into the analysis through a linear relationship between x and ethanol MSP and CI (denoted as y ), which is derived separately (Figure – Steps 2 to 3.1).

Uncertainty in Biorefinery Performance

Efficiencies of reactions and equipment, material prices and CIs, and electricity prices and CIs (with baseline values for each state) were all treated as uncertain following defined distributions (Table S4). Additional state-dependent factors, including taxes, financial incentives, and the capital cost factor (Table S2), were not subjected to uncertainty. For each state, we disaggregated the ethanol MSP and CI ( y ) into two components: the fixed contribution of the biorefinery (stemming from capital, operations, and maintenance; represented as b 0 ) and the variable contribution from the feedstock (the slope of the linear relationship between x and y , represented as a 0 ; Figure , Tables S6–S9). To derive these components, we performed 1000 Monte Carlo simulations for each state under uncertainty: once excluding feedstock price to derive b 0 (Figure – Step 2) and once including it to estimate a 0 (Step 3). The resulting state-level distributions (1000 simulations per state) of fixed (Step 2.1) and variable (Step 3.1) contributions were then integrated with feedstock production and transport model outputs for 1000 candidate locations (Step 4). This enabled full uncertainty characterization at each site without requiring separate biorefinery simulations for all locations (Sections S3.4 and S3.5).

Uncertainty in Ethanol Transport

Distances, transport capacities, costs, and CIs were treated as uncertain parameters in the ethanol transport model (Table S3) to account for spatial variability within the area represented by each randomly selected candidate site. For each biorefinery candidate location and jet fuel producer, unit ethanol transport costs were computed under uncertainty across 1000 samples (Figure , Step 5). The model also determined the ethanol allocation from each candidate biorefinery to its nearest jet producersincluding the number of producers supplied and the amount sent to eachbased on deterministic optimization (i.e., without uncertainty; Step 5.1). Ethanol delivered MSP and CI (denoted as y ′) were then calculated by summing the ethanol production MSP/CI ( y ) with the corresponding transport costs to each supplied jet producer (Step 5.2).

Uncertainty in SAF Production via ATJ

Efficiencies of processes, prices and CI of materials, and electricity prices and CIs (with baseline values for each state) were also varied across 1000 Monte Carlo simulations following defined distributions for SAF production (Table S5). A similar approach was applied to the SAF ATJ model as was used for the ethanol biorefinery, with the added complexity of incorporating nine distinct jet fuel producer capacities. These capacities were defined based on the volumes of ethanol allocated to each jet producer facility via the ethanol transport model. Simulations were performed for each of 44 states and each jet producer capacity, under uncertainty, to derive: (i) the fixed contribution ( b 1 ) to SAF MSP/CI ( z ) with ethanol delivered price/CI ( y ′) set to zero (Figure – Steps 6, 6.1), and (ii) the variable contribution, represented by the slope ( a 1 ) of the linear relationship between y ′ and z, using nonzero y ′ values (Steps 7, 7.1; Figure S11; Tables S10 and S11). Using the state- and capacity-specific linear coefficient distributions ( a 1 and b 1 ), the SAF MSP and CI were computed for each supplied jet producer across 1000 samples at each candidate location (Step 8). Final SAF values for each candidate location were calculated as weighted averages across the relevant ATJ facilities using the ethanol supply volumes (from the transport model) as weights (Step 9).

Assumptions of Variable Independence

The uncertainty analysis assumes independence among uncertain variables across the four modeling components. While the electricity price and CI appear as uncertain inputs in both the bioethanol and SAF models, they are treated as independent. This is justified by the possibility that the jet fuel producer may be in a different state than the biorefinery or that temporal or regional variations in electricity markets could lead to uncorrelated fluctuations. Ultimately, electricity unit price and CI are not relevant to the SAF refinery, as it does not rely on external electricity. Instead, it generates its own electricity by burning natural gas in a boiler-turbogenerator system and is designed to operate as electricity-neutral.

Results and Discussion

Across all 1000 candidate locations, SAF derived from switchgrass or miscanthus had CIs that were consistently lower than conventional, petroleum-derived aviation fuel (CAF) and generally below reported values for SAF from corn (via ATJ; estimated at ∼61 gCO2eq·MJ–1) , and corn stover (estimated at 28 gCO2eq·MJ–1; , Figure A). In the case of switchgrass, site-specific CI scores ranged from −9.7 [−28.1 to 7.7] gCO2e·MJ–1 to 39.4 [30.5 to 48.9] gCO2e·MJ–1 (median; fifth and 95th percentiles in brackets). Candidate sites with the highest CIs were concentrated in the northeast (including Maine and New York) and southeast (including Florida and Georgia), while the lowest CIs were found in South Dakota and Illinois (Figure B). Similarly, for miscanthus, the locations with the highest CI scores were found in Florida, Georgia, Maine, and New York, while the best-performing sites were located in South Dakota, Nebraska, and Illinois. Miscanthus-derived SAF CI scores ranged from −1.3 [−17 to 12.1] gCO2e·MJ–1 to 33.1 [25.4 to 41.7] gCO2e·MJ–1 (Figure S12). Despite spatial variability, all candidate sites for both switchgrass and miscanthus have the potential to achieve adequate CI reduction (i.e., achieve a CI below the maximum threshold of 50 kgCO2e·mmBTU–1, approximately 47.39 gCO2e·MJ fuel–1) to qualify for SAF tax credits under the Clean Fuel Production Credit , (also known as the 45Z credit), provided that prevailing wage and apprenticeship requirements are met (Figure A). Moreover, several locations, primarily in Illinois, Nebraska, and South Dakota, and to a lesser extent in Iowa and Arkansas, could achieve the maximum 45Z credit for SAF derived from switchgrass by reaching CI values below 0 gCO2e·MJ fuel–1 (Figure S13). Other instruments are available for credit, like the Low Carbon Fuel Standard (LCFS), only applicable for fuels marketed in California, which provides credit for any SAF with a CI lower than the CAF benchmark of 87.89 gCO2e·MJ fuel–1. , In addition, under the Renewable Fuel Standard (RFS), SAF derived from cellulosic feedstocks becomes eligible for credit once it achieves at least a 60% CI reduction relative to the petroleum baseline (Figure A). , We calculated the MSP without including policy incentives, as eligibility and allocation vary by policy, jurisdiction, and where the fuel is marketed rather than produced, placing them outside the scope of this study. Nevertheless, such incentives can substantially reduce the effective MSP of SAF. For example, the federal 45Z credit provides up to 1.75 USD·gal–1, the California LCFS can contribute up to about 1.50 USD·gal–1 depending on CI reduction, and the state of Illinois recently enacted a 1.50 USD·gal–1 SAF credit. Together, these mechanisms illustrate how federal and state policies can narrow the cost gap between SAF and CAF, and future work may explicitly evaluate how individual policies influence these reductions.

4.

4

Open in a new tab

(A, C) SAF CI/MSP data from switchgrass for all candidate locations sorted by median values. Eligibility for tax credits depends on other requirements as well (e.g., meeting wage and apprenticeship requirements for 45Z, selling the fuel in California for LCFS). (B, D) Maps of SAF CI/MSP by candidate location.

Across locations, the major factors governing CI variations are SOC sequestration and the GHG benefits of grid electricity offsets, the latter of which is influenced by regional balancing regions. While switchgrass-derived SAF achieves most of its CI reductions from SOC sequestration of the feedstock, lower-CI sites tend to prioritize locations with lower feedstock CI rather than higher electricity injection. Electricity injection represents between 36 and 43.5 gCO2e·MJ fuel–1 (average values for Pareto Frontier locations; Figure ); this accounts for a major percentage (55 to 66%) of the total positive contributions to the total SAF CI (Figure S14) if the feedstock is switchgrass. For its part, SAF derived from miscanthus achieves the greatest CI reduction from SOC sequestration, ranging between 24.9 and 59.2 gCO2e·MJ fuel–1, equivalent to 46 to 84% of total positive contributions (Figure S15). SOC sequestration from switchgrass varies between 11.2 gCO2e·MJ fuel–1 and 41.4 gCO2e·MJ fuel–1 across Pareto sites, equivalent to 19 to 58% of the total positive contributions to SAF CI (Figure S14). In terms of electricity injection for miscanthus-derived SAF, it can vary between 9.2 and 13 gCO2e·MJ fuel–1 for Pareto frontier locations (16 to 22% of total positive contributions; Figure S15), a significantly lower amount than for switchgrass. The latter could be explained by the fact that miscanthus has a slightly lower lignin content (16 vs 17%, which is burned to produce energy) and a higher feedstock-to-ethanol conversion rate (mean values of 114 gallons of ethanol per dry Mg of miscanthus vs 99.5 gallons of ethanol per dry Mg of switchgrass). Thus, less residual biomass is available for electricity production in miscanthus biorefineries.

5.

5

Open in a new tab

(A) MSP vs CI values of SAF derived from switchgrass for 1000 candidate locations (gray circles) with Pareto front (blue circles). MSP and CI values of SAF represent median values from the uncertainty analysis. (B) Trade-offs between SAF and CAF and the minimum cost of decarbonization at the current state of technology, with CAF price set at 2.4 USD·Gal–1 and its CI at 84 gCO2e·MJ–1. (C) Specific hypothetical candidate locations on the Pareto front (state maps) with corresponding MSPs and CIs of SAF.

In contrast to the environmental benefits of SAF relative to CAF, the former is often considerably more expensive (Figure B). The spot price of CAF (distribution terminal or refinery gate equivalent price) for June 2024 was 2.4 USD·gal–1. However, this price has been subject to considerable volatility: over the past 10 years (2014–2024), U.S. jet fuel spot prices ranged from 0.61 USD·gal–1 (April 2020) to 4.12 USD·gal–1 (June 2022), with an average of 2.03 USD·gal–1. In comparison, we estimate that median MSPs for switchgrass-derived SAF would range from 7.9 to 12.8 USD·gal–1 (Figure C,D) and from 7.9 to 12.3 for miscanthus (Figure S12C,D) across candidate locations. Ethanol conversion represents the highest contributor to SAF costs (between 40 and 50%; Figures S16 and S17), largely due to material costs from hydrolysis and fermentation, material costs from wastewater treatment, and installed equipment costs from utilities and storage: these costs represent ∼26%, 22%, and 26% of the contributions to the total annualized cost of biomass-to-ethanol conversion, respectively, at baseline values (Figure S18). As expected, locations with lower feedstock costs (<120 USD·Mg–1 for switchgrass, <140 USD·Mg–1 for miscanthus; Figure S2) and higher yields (>13 Mg·ha–1 for both crops; Figure S1) are the sites with lower SAF production costs. These are mainly located in southern states such as Oklahoma, Louisiana, and Alabama for switchgrass and Oklahoma, Kentucky, Ohio, and Arkansas for miscanthus. This pattern is primarily driven by differences in land rent (used as a proxy for the opportunity cost of land), with southern states having some of the lowest average rents and Midwestern states among the highest. State incentives for ethanol production, where applicable, have little effect on SAF MSP. For example, they reduce switchgrass-derived SAF price by only 0.3% in Alabama, 0.5% in Kentucky, and 1.9% in Kansas (states outside the Pareto frontier), with a slightly larger reduction in Nebraska of 2.7% (Figure S16; 2.8% for miscanthus, Figure S17). Compared to values reported in literature, our calculated MSPs are similar to the reported SAF levelized production cost (which does not include a discount rate and, therefore, is expected to be lower than MSP) from cellulosic feedstocks, which can range from 6.83 to 11.41 USD·gal–1. Throughout the literature, however, SAF production costs can vary significantly depending on technology, feedstock, and plant scale.

Across candidate sites, locations with high CI tend to have lower SAF costs and vice versa, revealing apparent trade-offs between economic and environmental goals for sustainability (Figures and S19). In the Midwest, this pattern is primarily driven by assumptions embedded in the feedstock breakeven price at farmgate, which accounts for the opportunity cost of converting productive cropland to energy crops. As a result, states in the Corn Belt (e.g., Illinois, Iowa, and Indiana) exhibit strong agronomic and environmental performance, with switchgrass yields exceeding 12 Mg·ha–1 and miscanthus yields surpassing 16.5 Mg·ha–1. These yields are accompanied by considerable SOC sequestration, leading to net CI values of −111.3 [−173.7 to −28.5] gCO2e·Mg–1 for switchgrass and −169.8 [−216.7 to −103.0] gCO2e·Mg–1 for miscanthus. However, these same sites also exhibit some of the highest feedstock breakeven prices (>140 USD·Mg–1 for both crops). In contrast, southeastern statesincluding Florida, Georgia, North Carolina, South Carolina, and Virginiaexhibit a different type of trade-off, this time between yield and CI. From Florida to North Carolina, switchgrass yields are relatively high (>15 Mg·ha–1), but over 40% of candidate sites in each state have net CI values above zero (Florida 85%, Georgia 55.8%, North Carolina 42%, South Carolina 49.4%, and Virginia 31.2%). Miscanthus displays the opposite trend: lower yields in southern states (<13 Mg·ha–1 from Texas to Florida), but with net CI values below zero for most of the rainfed region (−139 [−250.9, −53.6] gCO2e·Mg–1).

Our simulations reveal that SAF-level CI and MSPs follow similar trends as their respective feedstock-level counterparts, underscoring the dominant role of feedstock characteristics. The trade-off between SAF CI and cost suggests that, in the case of switchgrass-derived SAF, investors must choose between siting ethanol production facilities in regions with low feedstock costs and limited SOC benefits or in regions with high SOC sequestration potential and higher feedstock costs, with the added benefit of the possibility to inject electricity to the grid. To support decisions that balance economic and environmental objectives, a Pareto front can be constructed to identify deployment sites that optimize both cost and CI (Figure A). For switchgrass, the Pareto front includes 12 sites across 5 states (Figure C). Sites in Louisiana and Oklahoma are associated with the lowest costs but the highest CIs. Sites in Nebraska and South Dakota offer the lowest CIs but the highest costs. Notably, several low-CI candidate sites on the Pareto front are located near existing ethanol biorefineries, offering opportunities to reduce capital investment through repurposing. If these facilities were adapted to process switchgrass (or other low-CI crops) instead of corn, ethanol conversion costs, the largest contributor to SAF price, could be reduced. State-level differences among Pareto locations reflect variation in feedstock CI and electricity emission factors, which drive CI differences, while MSP variations are primarily influenced by electricity prices, state taxes, and feedstock costs.

Miscanthus follows trends similar to those of switchgrass in our simulations (Figure S19), with 16 locations on the Pareto front. The lowest costs for miscanthus-derived SAF are found in Oklahoma and Ohio (7.8 to 8.6 USD·gal–1), but these sites also exhibit high CIs (19 to 15.7 gCO2e·MJ–1). Sites in Arkansas, Illinois, and Nebraska display intermediate performance for both metrics, with MSPs ranging from 8.7 to 10 USD·gal–1 and CI values from 14.6 to 6.2 gCO2e·MJ–1. In contrast, eastern South Dakota shows the lowest CI values (0.7 to −1.3 gCO2e·MJ–1) for miscanthus-derived SAF, but at a higher MSP of approximately 10.8 USD·gal–1. Although miscanthus outperforms switchgrass in terms of feedstock CI, its breakeven price at the farm is similar or higher, despite higher biomass yields. Our analysis shows that switchgrass can lead to a lower overall CI for the final fuel product when ethanol biorefineries can export electricity to the grid, thereby offsetting emissions. In these scenarios, grid injection offsets 55–66% of the total emissions contributions (i.e., the sum of positive emission sources) for switchgrass-derived SAF and 16–22% for miscanthus-derived SAF (Figures S14 and S15). However, as the power grid continues to decarbonize, these emissions offset benefits may diminish, potentially making miscanthus the more favorable feedstock.

For both feedstocks, our cost estimates reflect refinery gate prices (i.e., wholesale prices). In contrast, jet fuel prices paid at airports (retail prices) are considerably higher due to the inclusion of transportation, storage, and distribution fees, along with taxes, fixed-based operator (FBO) markups, and concession or flowage fees charged by airports to FBOs, often constituting a substantial portion of overall fuel costs. Given the current CAF wholesale prices and associated CI values, the minimum cost of carbon abatement using switchgrass is approximately 580 USD·Mg CO2eq–1 at an Arkansas site (Figure B), and an estimated 660 USD·Mg CO2eq–1 for miscanthus-derived SAF at a site in Oklahoma (Figure S19B). These values are at the upper end of cost estimates for direct air carbon capture and storage (DACCS). , For SAF to be competitive with DACCS, abatement costs would need to decline to roughly 250 USD·Mg–1. , While DACCS costs are also high today, they are expected to decrease over timemirroring the trajectory anticipated for SAF. However, relying exclusively on DACCS to offset emissions from petroleum-based jet fuel would perpetuate dependence on fossil fuels while significantly increasing demand for CO2 storage capacitya resource already constrained by scalability and competing demands from other hard-to-decarbonize sectors. Technological alternatives such as electrification and hydrogen-based propulsion remain technically and economically infeasible for long-haul flights in the near term. Consequently, biofuels represent the most viable strategy for achieving short- and medium-term decarbonization targets in aviation.

To meet the U.S. SAF Grand Challenge goals by 2030, SAF production costs must decrease rapidly. Our analysis indicates that particular attention should be given to ethanol conversion, which remains the largest contributor to overall SAF costs (Figures S16 and S17). Sugar cane offers an interesting feedstock alternative because it does not require pretreatment, enabling lower ethanol costs. However, its cultivation is geographically limited, reinforcing the short-term viability of switchgrass and miscanthus as near-term options for domestic feedstock production. Short-rotation forestry crops such as poplar or willow have also been considered in literature, though limited TEA/LCA studies suggest they deliver smaller CI reductions and less favorable economics than switchgrass or miscanthus. , Future work could extend this analysis to additional feedstockssuch as corn stover or other agricultural residues.

Beyond feedstock selection, the literature on supply chain modeling for biofuels highlights additional system-level considerations. Examples include integrating social objectives such as local job creation or social asset factors, optimizing facility numbers and capacities to meet demand, , and incorporating depots or preprocessing facilities with biomass storage. , Most studies agree that conversion is the largest cost contributor and key barrier to commercializing cellulosic ethanol , and often assume dedicated SAF facilities supplying airports, , which may overlook blending requirements. While some studies use transportation cost minimization as the primary objective, our results indicate that feedstock yields and feedstock costs may play a more decisive role in siting decisions, and future work that allows irregular rather than circular collection areas may enhance siting optimization.

The challenges ahead extend beyond reducing costs and CI, but also include scaling up production capacity. For example, to fully substitute jet fuel production with SAF for a median size jet fuel producer (∼19.3 thousand bpd), a 500 million gallon per year ethanol refinery would be required. This capacity exceeds the size of the largest existing ethanol refinery by about 100 million gallons per year and would necessitate a feedstock collection radius of approximately 82 [58 to 170] km, assuming 20% of arable land within that area is devoted to energy crops, a figure that may not be realistic, compared to 32 [22.5 to 67] km used for the biorefinery capacity of this study (80 million gallon per year); or 270 [186 to 550] km for 2% land devoted to energy crops. While the model assigns ethanol to the nearest petroleum refinery, the ethanol transport distance has a relatively minor influence on SAF MSP and CI; consequently, Pareto optimal biorefinery sites are not necessarily located near petroleum refineries. Ethanol transport distances for MSP-optimal sites range from approximately 15 to 250 km, while CI-optimal sites involve longer distances, between 230 and 550 km (Figure S20). Ultimately, ethanol transport costs account for less than 1.5% of the SAF cost for MSP-optimal sites and less than 5.5% for CI-optimal sites (Figure S16), indicating that leveraging existing petroleum refineries may be a cost-effective alternative to constructing new facilities. Although ethanol-derived SAF is only one of several potential pathwaysand current aircraft are not yet certified for 100% SAF useour results demonstrate the potential of the ATJ pathway, especially when ethanol production facilities are integrated within the supply chain of petroleum refineries to leverage existing infrastructure and use feedstocks capable of sequestering carbon for greater CI reductions. Our findings also highlight the potential of cellulosic feedstocks like switchgrass and miscanthus, which are adaptable to various soil types, require fewer inputs, offer higher biomass yields, and provide carbon sequestration benefits. With sustained investment, these feedstocks could help catalyze the growth of a circular bioeconomy across the U.S. Heartland and beyond.

Supplementary Material

es5c17460_si_001.pdf (15.3MB, pdf)

Acknowledgments

This work was funded by the DOE Center for Advanced Bioenergy and Bioproducts Innovation (U.S. Department of Energy, Office of Science, Biological and Environmental Research Program under Award Number DE-SC0018420). Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the U.S. Department of Energy.

All data and code used in this study are available in the BioSTEAM Bioindustrial Park GitHub repository, under the SAF_location branch. Please note that this branch is currently separate from the main branch but may be merged into the main repository in the future. (https://github.com/BioSTEAMDevelopmentGroup/Bioindustrial-Park/tree/SAF_location/biorefineries/Location_SAF).

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.5c17460.

  • Assumptions for feedstock CI calculation based on SOC sequestration results from DayCent simulation; feedstock yield, GHG emissions and breakeven price at farmgate maps; breakdown of transportation costs by U.S. region and average state transportation prices maps for US; assumptions used to calculate transportation GHG emissions and average state values map; variation of average tortuosity factors for farm to ethanol refinery transportation by state and crop; map of 1000 candidate locations used for each crop; state-specific parameters used for ethanol biorefinery and SAF refinery simulations; detailed TEA and LCA assumptions and data for uncertainty; results for linear relation between feedstock price/CI and ethanol price/CI per state for each crop, and linear relation between ethanol price/CI and SAF price/CI per state for a 80 MMgal of ethanol per year capacity; results of disaggregated cost and CI contributors for all Pareto front locations for SAF derived from switchgrass and miscanthus; results for SAF derived from miscanthus, for all candidate locations and for Pareto frontier locations; and breakdown of costs for ethanol conversion when feedstock price is zero (PDF)

The authors declare no competing financial interest.

References

  1. Calvin, K. ; Dasgupta, D. ; Krinner, G. ; Mukherji, A. ; Thorne, P. W. ; Trisos, C. ; Romero, J. ; Aldunce, P. ; Barrett, K. ; Blanco, G. ; Cheung, W. W. L. ; Connors, S. ; Denton, F. ; Diongue-Niang, A. ; Dodman, D. ; Garschagen, M. ; Geden, O. ; Hayward, B. ; Jones, C. ; Jotzo, F. ; Krug, T. ; Lasco, R. ; Lee, Y.-Y. ; Masson-Delmotte, V. ; Meinshausen, M. ; Mintenbeck, K. ; Mokssit, A. ; Otto, F. E. L. ; Pathak, M. ; Pirani, A. ; Poloczanska, E. ; Pörtner, H.-O. ; Revi, A. ; Roberts, D. C. ; Roy, J. ; Ruane, A. C. ; Skea, J. ; Shukla, P. R. ; Slade, R. ; Slangen, A. ; Sokona, Y. ; Sörensson, A. A. ; Tignor, M. ; Van Vuuren, D. ; Wei, Y.-M. ; Winkler, H. ; Zhai, P. ; Zommers, Z. ; Hourcade, J.-C. ; Johnson, F. X. ; Pachauri, S. ; Simpson, N. P. ; Singh, C. ; Thomas, A. ; Totin, E. ; Arias, P. ; Bustamante, M. ; Elgizouli, I. ; Flato, G. ; Howden, M. ; Méndez-Vallejo, C. ; Pereira, J. J. ; Pichs-Madruga, R. ; Rose, S. K. ; Saheb, Y. ; Sánchez Rodríguez, R. ; Ürge-Vorsatz, D. ; Xiao, C. ; Yassaa, N. ; Alegría, A. ; Armour, K. ; Bednar-Friedl, B. ; Blok, K. ; Cissé, G. ; Dentener, F. ; Eriksen, S. ; Fischer, E. ; Garner, G. ; Guivarch, C. ; Haasnoot, M. ; Hansen, G. ; Hauser, M. ; Hawkins, E. ; Hermans, T. ; Kopp, R. ; Leprince-Ringuet, N. ; Lewis, J. ; Ley, D. ; Ludden, C. ; Niamir, L. ; Nicholls, Z. ; Some, S. ; Szopa, S. ; Trewin, B. ; Van Der Wijst, K.-I. ; Winter, G. ; Witting, M. ; Birt, A. ; Ha, M. ; Romero, J. ; Kim, J. ; Haites, E. F. ; Jung, Y. ; Stavins, R. ; Birt, A. ; Ha, M. ; Orendain, D. J. A. ; Ignon, L. ; Park, S. ; Park, Y. ; Reisinger, A. ; Cammaramo, D. ; Fischlin, A. ; Fuglestvedt, J. S. ; Hansen, G. ; Ludden, C. ; Masson-Delmotte, V. ; Matthews, J. B. R. ; Mintenbeck, K. ; Pirani, A. ; Poloczanska, E. ; Leprince-Ringuet, N. ; Péan, C. . IPCC, 2023: Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland; IPCC: Geneva, Switzerland, 2023. [Google Scholar]
  2. U.S. Department of Agriculture . Department of Energy, and Department of Transportation. SAF Grand Challenge Roadmap, 2022. https://www.energy.gov/sites/default/files/2022-09/beto-saf-gc-roadmap-report-sept-2022.pdf (accessed September 24, 2022).
  3. The National Academies Press . Getting to Net-Zero Emissions by 2050. https://nap.nationalacademies.org/resource/other/dels/net-zero-emissions-by-2050/?utm_source=google_grants&utm_medium=cpc&utm_campaign=15107095915&utm_content=145823965111&utm_term=carbon%20emissions&gad_source=1&gclid=CjwKCAjwzIK1BhAuEiwAHQmU3o1vPPVbp7r7Dxgwr5eQuNmEzLrC9fj1_XeoR3IgIllZCbX0M4DS_hoCRpcQAvD_BwE#tab-top (accessed March 21, 2025).
  4. Honeywell Introduces UOP eFiningTM Technology for New Class of Sustainable Aviation Fuel. https://www.honeywell.com/us/en/press/2023/05/honeywell-introduces-uop-efining-technology-for-new-class-of-sustainable-aviation-fuel (accessed August 03, 2024).
  5. Grim, R. G. ; Tao, L. ; Abdullah, Z. ; Cortright, R. ; Oakleaf, B. . The Challenge Ahead: A Critical Perspective on Meeting U.S. Growth Targets for Sustainable Aviation Fuel; National Renewable Energy Laboratory: Golden, CO, 2024. [Google Scholar]
  6. Kerry, J. ; United States Department of State; United States Executive Office of the President. The Long-Term Strategy of the United States, Pathways to Net-Zero Greenhouse Gas Emissions by 2050. 2021.
  7. Energy Information Administration (EIA) . Jimmy Troderman U.S. sustainable aviation fuel production takes off as new capacity comes online. https://www.eia.gov/todayinenergy/detail.php?id=65204 (accessed May 14, 2025).
  8. Huffstutter, P. J. ADM Sees Its Sustainable Aviation Fuel Coming Online by 2026 Reuters, 2022. https://www.reuters.com/article/business/adm-sees-its-sustainable-aviation-fuel-coming-online-by-2026-idUSL1N32V187/ (accessed August 03, 2024).
  9. Summit Next Gen to Use Honeywell Ethanol to Jet Fuel Technology for Production of Sustainable Aviation Fuel. https://www.honeywell.com/us/en/press/2023/05/summit-next-gen-to-use-honeywell-ethanol-to-jet-fuel-technology-for-production-of-sustainable-aviation-fuel (accessed August 03, 2024).
  10. Tanzil A. H., Brandt K., Zhang X., Wolcott M., Stockle C., Garcia-Perez M.. Production of Sustainable Aviation Fuels in Petroleum Refineries: Evaluation of New Bio-Refinery Concepts. Front. Energy Res. 2021;9:735661. doi: 10.3389/fenrg.2021.735661. [DOI] [Google Scholar]
  11. Carlson N. A., Talmadge M. S., Singh A., Tao L., Davis R.. Economic Impact and Risk Analysis of Integrating Sustainable Aviation Fuels into Refineries. Front. Energy Res. 2023;11:1223874. doi: 10.3389/fenrg.2023.1223874. [DOI] [Google Scholar]
  12. Tanzil A. H., Zhang X., Wolcott M., Brandt K., Stöckle C., Murthy G., Garcia-Perez M.. Evaluation of Dry Corn Ethanol Bio-Refinery Concepts for the Production of Sustainable Aviation Fuel. Biomass Bioenergy. 2021;146:105937. doi: 10.1016/j.biombioe.2020.105937. [DOI] [Google Scholar]
  13. Shahriar M. F., Khanal A.. The Current Techno-Economic, Environmental, Policy Status and Perspectives of Sustainable Aviation Fuel (SAF) Fuel. 2022;325:124905. doi: 10.1016/j.fuel.2022.124905. [DOI] [Google Scholar]
  14. U.S. Department of Energy . Alternative Fuels Data Center: Sustainable Aviation Fuel. https://afdc.energy.gov/fuels/sustainable-aviation-fuel (accessed September 24, 2024).
  15. Moriarty, K. ; Kvien, A. . U.S. Airport Infrastructure and Sustainable Aviation Fue; National Renewable Energy Laboratory: Golden, CO; 2021. [Google Scholar]
  16. Smart Freight Centre and MIT Center for Transportation & Logistics. Sustainable Aviation Fuel Greenhouse Gas Emission Accounting and Insetting Guidelines, 2021. https://sustainable.mit.edu/sustainablefuel/ (accessed September 24, 2024).
  17. Jing L., El-Houjeiri H. M., Monfort J.-C., Littlefield J., Al-Qahtani A., Dixit Y., Speth R. L., Brandt A. R., Masnadi M. S., MacLean H. L., Peltier W., Gordon D., Bergerson J. A.. Understanding Variability in Petroleum Jet Fuel Life Cycle Greenhouse Gas Emissions to Inform Aviation Decarbonization. Nat. Commun. 2022;13(1):7853. doi: 10.1038/s41467-022-35392-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Bhatt A. H., Zhang Y., Milbrandt A., Newes E., Moriarty K., Klein B., Tao L.. Evaluation of Performance Variables to Accelerate the Deployment of Sustainable Aviation Fuels at a Regional Scale. Energy Convers. Manage. 2023;275:116441. doi: 10.1016/j.enconman.2022.116441. [DOI] [Google Scholar]
  19. Wang, M. ; Cai, H. ; Lee, U. ; Kar, S. ; Sykora, T. ; Liu, X. . Development of R&D GREET 2023 Rev1 to Estimate Greenhouse Gas Emissions of Sustainable Aviation Fuels for 40B Provision of the Inflation Reduction Act; Argonne National Laboratory, 2024. [Google Scholar]
  20. Staples M. D., Malina R., Olcay H., Pearlson M. N., Hileman J. I., Boies A., Barrett S. R. H.. Lifecycle Greenhouse Gas Footprint and Minimum Selling Price of Renewable Diesel and Jet Fuel from Fermentation and Advanced Fermentation Production Technologies. Energy Environ. Sci. 2014;7(5):1545–1554. doi: 10.1039/C3EE43655A. [DOI] [Google Scholar]
  21. United States Code . Sec. 45Z - Clean Fuel Production Credit. INTERNAL REVENUE CODE, 2018. https://www.govinfo.gov/app/details/USCODE-2023-title26/USCODE-2023-title26-subtitleA-chap1-subchapA-partIV-subpartD-sec45Z/summary.
  22. Scown C. D., Nazaroff W. W., Mishra U., Strogen B., Lobscheid A. B., Masanet E., Santero N. J., Horvath A., McKone T. E.. Lifecycle Greenhouse Gas Implications of US National Scenarios for Cellulosic Ethanol Production. Environ. Res. Lett. 2012;7(1):014011. doi: 10.1088/1748-9326/7/1/014011. [DOI] [Google Scholar]
  23. Dwivedi P., Wang W., Hudiburg T., Jaiswal D., Parton W., Long S., DeLucia E., Khanna M.. Cost of Abating Greenhouse Gas Emissions with Cellulosic Ethanol. Environ. Sci. Technol. 2015;49(4):2512–2522. doi: 10.1021/es5052588. [DOI] [PubMed] [Google Scholar]
  24. Lee U., Kwon H., Wu M., Wang M.. Retrospective Analysis of the U.S. Corn Ethanol Industry for 2005–2019: Implications for Greenhouse Gas Emission Reductions. Biofuels, Bioprod. Biorefin. 2021;15(5):1318–1331. doi: 10.1002/bbb.2225. [DOI] [Google Scholar]
  25. Scully M. J., Norris G. A., Falconi T. M. A., MacIntosh D. L.. Carbon Intensity of Corn Ethanol in the United States: State of the Science. Environ. Res. Lett. 2021;16(4):043001. doi: 10.1088/1748-9326/abde08. [DOI] [Google Scholar]
  26. Langholtz, M. H. ; U.S. Department of Energy . 2023 Billion-Ton Report: An Assessment of U.S. Renewable Carbon Sources; Oak Ridge National Laboratory, 2024. [Google Scholar]
  27. U.S. Energy Information Administration . Petroleum Refineries |U.S. Energy Atlas, 2023. https://atlas.eia.gov/datasets/6547eda91ef84cc386e23397cf834524_22/explore?location=50.183026%2C-136.924637%2C8.90 (accessed July 09, 2024).
  28. U.S. Energy Information Administration . Ethanol Plants | U.S. Energy Atlas 2023. https://atlas.eia.gov/datasets/3f984029aadc4647ac4025675799af90_112/explore?location=34.816139%2C-99.641106%2C5.00 (accessed July 09, 2024).
  29. U.S. Energy Information Administration . Table 3. Capacity of Operable Petroleum Refineries by State; From EIA-820, “Annual Refinery Report”; Refinery Capacity, 2024. https://www.eia.gov/petroleum/refinerycapacity/table3.pdf.
  30. U.S. Department of Energy . Sustainable Aviation Fuel (SAF) Grand Challenge: Building Supply ChainsRequest for Information (RFI) DE-FOA-0003157 Summary Report. Renewable Energy 2024.
  31. Masum F. H., Coppola E., Field J. L., Geller D., George S., Miller J. L., Mulvaney M. J., Nana S., Seepaul R., Small I. M., Wright D., Dwivedi P.. Supply Chain Optimization of Sustainable Aviation Fuel from Carinata in the Southeastern United States. Renewable Sustainable Energy Rev. 2023;171:113032. doi: 10.1016/j.rser.2022.113032. [DOI] [Google Scholar]
  32. Akter, H. A. ; Huang, Y.-K. ; Dwivedi, P. . Developing a Supply Chain Model for Sustainable Aviation Fuel Using Logging Residues in Georgia, United States; SSRN: Rochester, NY,2024. 10.2139/ssrn.4739842. [DOI] [Google Scholar]
  33. Mousavi-Avval S. H., Khanal S., Shah A.. Assessment of Potential Pennycress Availability and Suitable Sites for Sustainable Aviation Fuel Refineries in Ohio. Sustainability. 2023;15(13):10589. doi: 10.3390/su151310589. [DOI] [Google Scholar]
  34. Martinkus N., Shi W., Lovrich N., Pierce J., Smith P., Wolcott M.. Integrating Biogeophysical and Social Assets into Biomass-to-Biofuel Supply Chain Siting Decisions. Biomass Bioenergy. 2014;66:410–418. doi: 10.1016/j.biombioe.2014.04.014. [DOI] [Google Scholar]
  35. Hu H., Lin T., Wang S., Rodriguez L. F.. A cyberGIS Approach to Uncertainty and Sensitivity Analysis in Biomass Supply Chain Optimization. Appl. Energy. 2017;203:26–40. doi: 10.1016/j.apenergy.2017.03.107. [DOI] [Google Scholar]
  36. Ullah K. M., Masum F. H., Field J. L., Dwivedi P.. Designing a GIS-Based Supply Chain for Producing Carinata-Based Sustainable Aviation Fuel in Georgia, USA. Biofuels, Bioprod Biorefin. 2023;17(4):786–802. doi: 10.1002/bbb.2483. [DOI] [Google Scholar]
  37. Antony P. S., Vanderghem C., MacLean H. L., Saville B. A., Posen I. D.. A Framework to Estimate National Biofuel Potential by Siting Production Facilities: A Case Study for Canola Sustainable Aviation Fuel in Canada. Energy Adv. 2024;3(7):1612–1631. doi: 10.1039/D3YA00579H. [DOI] [Google Scholar]
  38. Davis C. D., Sreekumar S., Altman R., Clarens A. F., Lambert J. H., Colosi L. M.. Geospatially Explicit Technoeconomic Assessment of Sustainable Aviation Fuel Production: A Regional Case Study in Virginia. Fuel Commun. 2024;19:100114. doi: 10.1016/j.jfueco.2024.100114. [DOI] [Google Scholar]
  39. Guo, W. ; Kudli, L. ; Bhagwat, S. ; Guest, J. . Co-Processing Sustainable Aviation Fuel at Petroleum Refineries to Reduce Costs and Carbon Intensity ChemRxiv 2025. 10.26434/chemrxiv-2025-n5tkz. [DOI] [PMC free article] [PubMed]
  40. Cortes-Peña Y., Kumar D., Singh V., Guest J. S.. BioSTEAM: A Fast and Flexible Platform for the Design, Simulation, and Techno-Economic Analysis of Biorefineries under Uncertainty. ACS Sustainable Chem. Eng. 2020;8(8):3302–3310. doi: 10.1021/acssuschemeng.9b07040. [DOI] [Google Scholar]
  41. Stewart D. W., Cortés-Peña Y. R., Li Y., Stillwell A. S., Khanna M., Guest J. S.. Implications of Biorefinery Policy Incentives and Location-Specific Economic Parameters for the Financial Viability of Biofuels. Environ. Sci. Technol. 2023;57(6):2262–2271. doi: 10.1021/acs.est.2c07936. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Glover, E. Aviation Biofuels, 2010. http://large.stanford.edu/courses/2010/ph240/glover2/ (accessed May 25, 2024).
  43. U.S. Department of Agriculture . The Environmental Opportunities and Challenges of Emerging Biofuels Production, 2007. https://www.nrcs.usda.gov/sites/default/files/2022-10/Internal-MM-Issues.pdf.
  44. Fan X., Khanna M., Lee Y., Kent J., Shi R., Guest J. S., Lee D.. Spatially Varying Costs of GHG Abatement with Alternative Cellulosic Feedstocks for Sustainable Aviation Fuels. Environ. Sci. Technol. 2024;58(26):11352–11362. doi: 10.1021/acs.est.4c01949. [DOI] [PubMed] [Google Scholar]
  45. Kantola I. B., Masters M. D., Blanc-Betes E., Gomez-Casanovas N., DeLucia E. H.. Long-Term Yields in Annual and Perennial Bioenergy Crops in the Midwestern United States. GCB Bioenergy. 2022;14(6):694–706. doi: 10.1111/gcbb.12940. [DOI] [Google Scholar]
  46. Dhaliwal J. K., Grosso S. J. D., Saha D.. Simulating Soil Carbon Sequestration, Yield, and N2O Fluxes with DayCent under Long-Term No-till and Cover Crop-Based Cotton Cropping System. Agric., Ecosyst. Environ. 2026;395:109926. doi: 10.1016/j.agee.2025.109926. [DOI] [Google Scholar]
  47. Parton, W. J. ; Hartman, M. D. . Evolution of Century and DayCent Ecosystem Models During the Last 40 Years AGU Fall Meeting Abstracts 2024; Vol. 2024. [Google Scholar]
  48. Takeda N., Rowlings D., Parton W., Grace L., Day K., Nguyen T., Grace P.. Soil Carbon Sequestration Potential in Subtropical Grasslands Estimated by DayCent-CABBI. Soil Sci. Soc. Am. J. 2025;89(1):e70003. doi: 10.1002/saj2.70003. [DOI] [Google Scholar]
  49. Hartman, M. D. ; Parton, W. J. ; Blanc-Betes, E. ; Berardi, D. . DayCent-CABBI: An Ecosystem Model with Microbial-Explicit and Deep Soil Organic Matter Dynamics for Evaluating the Sustainability of Bioenergy Crop Production AGU Fall Meeting Abstracts 2024; Vol. 2024. [Google Scholar]
  50. Jarecki M., Kariyapperuma K., Deen B., Graham J., Bazrgar A. B., Vijayakumar S., Thimmanagari M., Gordon A., Voroney P., Thevathasan N.. The Potential of Switchgrass and Miscanthus to Enhance Soil Organic Carbon SequestrationPredicted by DayCent Model. Land. 2020;9(12):509. doi: 10.3390/land9120509. [DOI] [Google Scholar]
  51. Necpálová M., Anex R. P., Fienen M. N., Del Grosso S. J., Castellano M. J., Sawyer J. E., Iqbal J., Pantoja J. L., Barker D. W.. Understanding the DayCent Model: Calibration, Sensitivity, and Identifiability through Inverse Modeling. Environ. Modell. Software. 2015;66:110–130. doi: 10.1016/j.envsoft.2014.12.011. [DOI] [Google Scholar]
  52. BioSTEAM Development Group . Bioindustrial-Park: BioSTEAM’s Premier Repository for Biorefinery Models and Results.
  53. Stewart D. W., Guo W., Li Y., Fan X., Coppess J. W., Khanna M., Guest J.. Greenhouse Gas Accounting Procedures in Low Carbon Fuel Policies Overlook the Spatial Variability of Miscanthus-Derived Sustainable Aviation Fuel. ACS Sustainable Resour. Manage. 2025;2:1185–1194. doi: 10.1021/acssusresmgt.4c00486. [DOI] [Google Scholar]
  54. Tao L., N Markham J., Haq Z., J Biddy M.. Techno-Economic Analysis for Upgrading the Biomass-Derived Ethanol-to-Jet Blendstocks. Green Chem. 2017;19(4):1082–1101. doi: 10.1039/C6GC02800D. [DOI] [Google Scholar]
  55. Bureau of Transportation Statistics (BTS) . North American Roads; National Transportation Atlas Database, 2024. https://geodata.bts.gov/datasets/0b6c2fd2e3ac40a7929cdff1d4cf604a/explore (accessed September 24, 2024). [Google Scholar]
  56. Duffy, M. Estimated Costs for Production, Storage and Transportation of Switchgrass; Iowa State University, 2008. https://www.extension.iastate.edu/agdm/crops/pdf/a1-22.pdf. [Google Scholar]
  57. U.S. Department of Energy . Ethanol Production and Distribution. https://afdc.energy.gov/fuels/ethanol-production (accessed September 24, 2024).
  58. U.S. Department of Transportation . Freight and Fuel Transportation Optimization Tool (FTOT). https://volpeusdot.github.io/FTOT-Public/ (accessed November 19, 2025).
  59. Leslie, A. ; Murray, D. . An Analysis of the Operational Costs of Trucking: 2024 Update; American Transportation Research Institute, 2024; p 62. [Google Scholar]
  60. The Currency editors . Average gas prices by state in 2023 Empower. https://www.empower.com/the-currency/life/gas-prices-by-state (accessed September 25, 2024).
  61. Geotab . State of Fuel Economy in Trucking. Geotab. https://www.geotab.com/truck-mpg-benchmark/ (accessed September 24, 2024).
  62. U.S. Department of Energy . Alternative Fuels Data Center: Fuel Properties Comparison. https://afdc.energy.gov/fuels/properties (accessed September 24, 2024).
  63. U.S. Energy Information Administration - EIA . Independent Statistics and Analysis Carbon Dioxide Emissions Coefficients. https://www.eia.gov/environment/emissions/co2_vol_mass.php (accessed September 24, 2024).
  64. Baral N. R., Davis R., Bradley T. H.. Supply and Value Chain Analysis of Mixed Biomass Feedstock Supply System for Lignocellulosic Sugar Production. Biofuels, Bioprod. Biorefin. 2019;13(3):635–659. doi: 10.1002/bbb.1975. [DOI] [Google Scholar]
  65. Kumar A., Sokhansanj S.. Switchgrass (Panicum Vigratum, L.) Delivery to a Biorefinery Using Integrated Biomass Supply Analysis and Logistics (IBSAL) Model. Bioresour. Technol. 2007;98(5):1033–1044. doi: 10.1016/j.biortech.2006.04.027. [DOI] [PubMed] [Google Scholar]
  66. Boutwell, M. Inflation Reduction Act Sustainable Aviation Fuel Credit. Stillwater Associates. https://stillwaterassociates.com/inflation-reduction-act-sustainable-aviation-fuel-credit-carbon-intensity-matters/ (accessed October 16, 2024).
  67. California Air Resources Board . Low Carbon Fuel Standard. https://ww2.arb.ca.gov/our-work/programs/low-carbon-fuel-standard (accessed Jun 04, 2024).
  68. E. P. A. U.S. . Renewable Fuel Standard Program. https://www.epa.gov/renewable-fuel-standard-program ((accessed 2024–06–04) (accessed Jun 04, 2024).
  69. U.S. Energy Information Administration . U.S. Gulf Coast Kerosene-Type Jet Fuel Spot Price FOB (Dollars per Gallon). https://www.eia.gov/dnav/pet/hist/eer_epjk_pf4_rgc_dpgD.htm (accessed October 15, 2024).
  70. NASS (National Agricultural Statistics Service) . USDA (United States Department of Agriculture). Quick Stats. Rent, Cash, Cropland, Non-Irrigated - Expense, 2016. https://quickstats.nass.usda.gov/.
  71. O’Malley, J. ; Pavlenko, N. ; Kim, Y. H. . Meeting the SAF grand challenge: Current and future measures to increase US Sustainable aviation fuel production capacity International Council on Clean Transportation 2023.
  72. Jones, N. As Carbon Air Capture Ramps Up, Major Hurdles Remain. Yale Environment 360. https://e360.yale.edu/features/direct-air-capture (accessed October 16, 2024).
  73. Höglund, R. Direct air capture is cheaper than you think. https://www.milkywire.com/articles/direct-air-capture-is-cheaper-than-you-think (accessed October 16, 2024).
  74. Farbes, J. ; Haley, B. ; Jones, R. . Marginal Abatement Cost Curves for U.S. Net-Zero Energy Systems; Evolved Energy Research LLC: San Francisco, CA, 2021. https://www.edf.org/sites/default/files/documents/MACC_2.0%20report_Evolved_EDF.pdf. [Google Scholar]
  75. IEA . Direct Air Capture, 2022. https://www.iea.org/reports/direct-air-capture-2022/executive-summary (accessed October 16, 2024).
  76. Sacchi R., Becattini V., Gabrielli P., Cox B., Dirnaichner A., Bauer C., Mazzotti M.. How to Make Climate-Neutral Aviation Fly. Nat. Commun. 2023;14(1):3989. doi: 10.1038/s41467-023-39749-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Speizer S., Fuhrman J., Lopez L. A., George M., Kyle P., Monteith S., McJeon H.. Integrated Assessment Modeling of a Zero-Emissions Global Transportation Sector. Nat. Commun. 2024;15(1):4439. doi: 10.1038/s41467-024-48424-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Limb B. J., Smith J. P., Simske S., Quinn J. C.. Addressing the Sustainable Aviation Fuel Grand Challenge: Economic and Environmental Insights from Feedstock Optimization. Environ. Sci. Technol. 2025;59(35):18628–18640. doi: 10.1021/acs.est.5c05842. [DOI] [PubMed] [Google Scholar]
  79. Budsberg E., Crawford J. T., Morgan H., Chin W. S., Bura R., Gustafson R.. Hydrocarbon Bio-Jet Fuel from Bioconversion of Poplar Biomass: Life Cycle Assessment. Biotechnol. Biofuels. 2016;9(1):170. doi: 10.1186/s13068-016-0582-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. You F., Tao L., Graziano D. J., Snyder S. W.. Optimal Design of Sustainable Cellulosic Biofuel Supply Chains: Multiobjective Optimization Coupled with Life Cycle Assessment and Input–Output Analysis. AIChE J. 2012;58(4):1157–1180. doi: 10.1002/aic.12637. [DOI] [Google Scholar]
  81. Marvin W. A., Schmidt L. D., Benjaafar S., Tiffany D. G., Daoutidis P.. Economic Optimization of a Lignocellulosic Biomass-to-Ethanol Supply Chain. Chem. Eng. Sci. 2012;67(1):68–79. doi: 10.1016/j.ces.2011.05.055. [DOI] [Google Scholar]
  82. O’Neill E. G., Martinez-Feria R. A., Basso B., Maravelias C. T.. Integrated Spatially Explicit Landscape and Cellulosic Biofuel Supply Chain Optimization Under Biomass Yield Uncertainty. Comput. Chem. Eng. 2022;160:107724. doi: 10.1016/j.compchemeng.2022.107724. [DOI] [Google Scholar]
  83. Trejo-Pech C. O., Larson J. A., English B. C., Yu T. E.. Cost and Profitability Analysis of a Prospective Pennycress to Sustainable Aviation Fuel Supply Chain in Southern USA. Energies. 2019;12(16):3055. doi: 10.3390/en12163055. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

es5c17460_si_001.pdf (15.3MB, pdf)

Data Availability Statement

All data and code used in this study are available in the BioSTEAM Bioindustrial Park GitHub repository, under the SAF_location branch. Please note that this branch is currently separate from the main branch but may be merged into the main repository in the future. (https://github.com/BioSTEAMDevelopmentGroup/Bioindustrial-Park/tree/SAF_location/biorefineries/Location_SAF).

Articles from Environmental Science & Technology are provided here courtesy of American Chemical Society

RESOURCES