45Q - Frequently Asked Questions and Additional Resources

For any of the LCA software, we want to ensure that we’re able to review the models exactly as the applicant or third-party analysts have constructed and used them for LCA results. To this end, an entire database export is preferred, as that reduces the chances for missing data or processing steps required by NETL. For example if the LCA model is prepared in openLCA, we would prefer the model be submitted as a database backup (i.e., a .zolca file). Some other examples are: sphere GaBi - .gcd and SimaPro .SPBackup

Carbon oxides used as a tertiary injected in enhanced oil or natural gas recovery should be treated as a co-product and follow the appropriate co-product management methods described in the CO2U Guidance and 45Q Addendum, e.g., system expansion. The expanded system can be truncated at CO₂ production with the use phase assumed to be the same in the proposed and comparison product systems. The comparison CO₂ can be CO₂ from another source such as natural dome CO₂ which is frequently used in EOR operations.

Carbon oxides sent to be disposed of in secure geological storage should be treated as a waste with the waste management operations (e.g., energy used for CO₂ storage) included in the system boundary.

However, the amount of carbon oxides sent to a recovery project or disposal should not be included in the quantity of carbon oxide utilized (i.e., the denominator in the DF calculation, as seen in the 45Q Addendum Section 2.1.9.3).

Note: The applicant should see TR 1-45Q regarding treatment and credit calculation for the non-utilization fractions of the total captured CO₂ stream.

Federal employees are obligated to not disclose any confidential information under 18 U.S.C. 1905. As a Federal agency, DOE must comply with requests made under the Freedom of Information Act (FOIA) (5 U.S.C. 552). Under FOIA, Federal agencies are required to disclose agency records requested by a member of the public, including information received from outside parties, unless FOIA specifically exempts the information’s disclosure. FOIA Exemption 4 (5 U.S.C. 552(b)(4)) exempts from FOIA disclosure confidential commercial and financial information obtained from a person outside the agency.

Whenever a document submitted to DOE by a private business contains information which may be exempt from public disclosure, it will be handled in accordance with the procedures in 10 CFR 1004.11. While DOE is responsible for making the final determination with regard to the disclosure or nondisclosure of information contained in any requested documents, DOE will consider the submitter's views in making its determination. Therefore, if the submitter believes the information submitted would constitute trade secrets or commercial/financial information that is privileged and confidential, it is recommended that the submitter label the documents as such.

An adequate critical review requires a thorough evaluation of the Life Cycle Inventory (LCI) data used as part of the study. 26 CFR 1.45Q 4(c)(5) states that in addition to a report, the model must be submitted if the LCA has not been verified by an independent third-party review. 26 CFR 1.45Q 4(c)(4) indicates that if conducted, an independent third-party review must include an assessment of the model and supporting data.

If the applicant does not want to disclose this information to DOE, an alternative would be for the applicant to seek an independent third-party review of their LCA. However, the third-party reviewer is expected to thoroughly evaluate the model and LCI as noted above. If DOE, in its review, determines that this level of review did not occur or was insufficient, resulting in major errors, DOE may request additional information from the applicant. Additional information regarding this option is noted in Section 2.4.5 of the 45Q Addendum.

DOE may need to share nontax information with the IRS, whose employees are also subject to confidentiality laws (Privacy Act (PA), Freedom of Information Act (FOIA), and IRC section 6103 and other USC code sections).

As noted in the previous FAQ, a complete critical review requires that all relevant portions of the model are reviewed. Data from third-party providers must be verified in some way to complete a review.

As noted in the 45Q Addendum Section 1.5, “use of any third-party licensed data shall include permission from the data providers for DOE review access.” If this permission cannot be obtained, summary information regarding the third-party processes must be submitted and DOE may contact you to verify its accuracy.

From the CO2U Guidance Section 2.1.9.1: “The LCA shall include at least 99% of the carbon and energy inputs and outputs to the system boundary. Completeness shall be demonstrated based on a carbon and energy balance. Completeness shall be tested by determining if the inclusion of secondary or tertiary processes to the system boundary would change the environmental impact result for the Product System by 1%. If more than one process is excluded the sum of the exlusions shall change the result less than 1%. Completeness shall be determined based on the expected or median value for data representativeness.

Thus, the DOE review may find errors or data which are not of sufficient quality or representativeness but changing these items do not or cannot change the result (the Displacement Factor) by greater than 1%. Such errors would be considered “immaterial” and would not affect the acceptance of the submitted LCA.

Section 2.1.9.2 of the 45Q addendum makes clear that a sensitivity analysis is required for 45Q LCAs, while an uncertainty analysis is not. The results of the sensitivity analysis should lead to a break-even analysis on highly sensitive parameters.

Section 2.2.3.2 of the CO2U Guidance, which is not superseded by a corresponding 45Q Addendum section, discusses a common approach for a sensitivity analysis. Exhibit 2-27 in the same section (recreated below) shows an example of an expected results graph from this analysis. Changing the percentage of the parameters “all-at-once” does not satisfy the requirements for the sensitivity analysis, which instead should test the individual contribution of uncertain parameters to the overall impacts. To do this, the parameters should be changed by +/- 10% “one-at-a-time” so the individual parameter impacts can be seen and potentially displayed in a “tornado graph.” Please refer to the previously referenced CO2U Guidance requirements for a sensitivity analysis. Highly sensitive parameters must be determined to review and show how their values affect the LCA results and if their values are within expected or verified ranges. In the example below, this would mean calculating the value at which Parameters A and B cause the total lifecycle greenhouse gases (GHGs) of the Proposed Product System to equal the lifecycle GHGs of the Comparison Product System.

According to Section 2.2.2.3 of the 45Q Addendum, the expectation is that “the applicants will develop custom, parameterized unit process or unit processes of all operations within the operational control of the utilization project.” However, if black box unit processes are used as background data, the input quantities of the black box process may be varied during a sensitivity analysis. For example, the quantity of material inputs can be adjusted by a set percentage point individually the same way a parameter would change.

Additionally, Section 2.2.2.6 of the CO2U Guidance provides information on when the use of proxy data requires a sensitivity analysis. This section states that documentation and a sensitivity analysis are required to evaluate data gaps. “For example, if a unit process for a material does not exist, identify at least two proxy data sets equivalent to the mass of the required material input that would theoretically bound a low and high environmental profile for a similar type of material. Double the mass of the requirement for each profile and test the sensitivity of the study results to the importance and relative changes in study-level results. If the study results are determined to be sensitive (materially changes the interpretation of the study findings) then this deficiency and its impact on results must be prominently noted with the study conclusion; see Section 2.2.3.2 on sensitivity analysis. Alternatively, if time and resources permit, conduct primary research and modeling to fill the data gap.”

Sufficient information must be provided to allow for confirmation of the applicant's mass, energy, and carbon balances, as stated in Section 2.4.5 of the 45Q Addendum. Generally, this should include information regarding the mass, energy content, and carbon content of each input and output. For example, the energy balance should essentially be a comprehensive accounting of all energy demands and any energy products/co-products – types, amounts, sources, properties (e.g., heating value). This could be presented in tabular format to satisfy the completeness check requirements. It is not necessary to close the energy balance so long as the source of discrepancies can be described and justified qualitatively. However, non-closure of the engineering balances due to data gaps may be subject to proxy data requirements, as discussed in Section 2.2.2.6 of the CO2U Guidance.

An energy balance table similar to the carbon balance table may be required for unit processes that generate energy flows, consume significant energy flows, or include sensitive parameters for the model. The NETL guidance requires that 99% of mass and energy of inputs and outputs are included in the system boundary (Section 2.1.9.1 of the CO2U Guidance) unless the applicant can justify otherwise.

Section 2.1.3.2 of the 45Q Addendum provides a hierarchy for selecting the relevant technologies for the Comparison Product System. In order, that section requires the representation of the U.S. average greenhouse gas technology or the industry standard practice if the U.S. average is unavailable.

As noted in Section 2.1.3.2 of the CO2U guidance, the U.S. average GHG performance technology is calculated as the production weighted average emissions intensity of all facilities producing the product. This can be calculated as the total GHG emissions across all facilities divided by the total product across all facilities. This approach accurately represents the full range of emissions intensity for the product of interest in the comparison system. Also, it must include facilities that have carbon capture equipment installed in the industry of interest. EPA reporting databases, such as the Greenhouse Gas Reporting Program (GHGRP), can be used in combination with industry production data to characterize a sector’s facility-level GHG emission intensity performance. This information should be coupled with appropriate supply chain emissions to account for purchased raw materials and energy.

According to the EPA GHGRP 2022 data, there were 18.1 million metric tons of CO₂ captured from industrial sources (as reported by subpart PP), comprising 117 different facilities.¹ This does not include CO₂ that is captured and utilized on-site (e.g., CO₂ capture from ammonia production and utilization at integrated facilities to make urea). According to EPA, the top sectors in terms of number of facilities capturing CO₂ are ethanol, natural gas processing, and ammonia.

If CO₂ is sourced from one of the top sectors that are conducting carbon capture already, the existence of carbon capture should be reflected in the Comparison Product System technology. It is not valid to compare a facility with carbon capture as the source of CO₂ in the Proposed Product System to the same facility without carbon capture in the Comparison product system because the latter would not reflect the relevant technologies in the hierarchy established in Section 2.1.3.2.

Section 2.1.3.1 of the CO2U guidance also describes in detail how to determine the functional unit for the proposed system, which in turn dictates what products must also be included in the comparison system.

The CO2U GuidanceSection 2.1.9.1 advises applicants to include any process within the system boundary that may contribute >1% to the global warming potential. However, the 45Q Addendum notes that if the downstream impacts are equivalent between the two systems, then the system boundary can exclude identical processes after the point of functional equivalence.

The NETL guidance does not specifically allow truncation of data upstream of the utilization process even if it is a consistent input for both the proposed and comparison systems. However, from Section 2.1.9.1: “The LCA shall include at least 99% of the carbon and energy inputs and outputs to the system boundary. Completeness shall be demonstrated based on a carbon and energy balance. Completeness shall be tested by determining if the inclusion of secondary or tertiary processes to the system boundary would change the environmental impact result for the Product System by 1%. If more than one process is excluded the sum of the exclusions shall change the result less than 1%. Completeness shall be determined based on the expected or median value for data representativeness.”

If a data limitation exists, we recommend that you identify a proxy and note the use of the proxy as part of your completeness check. Note, the NETL openLCA LCI Database does not have to be the sole source of life cycle inventory data. You are welcome to use additional public and commercial sources to supplement any missing data.

NETL does not have specific recommendations or requirements for any externally-produced tools or commercial databases or software. Per NETL 45Q guidance Section 2.1.3.2, the applicant should use the provided NETL U.S. average GHG technology baselines for products and/or functions in their Comparison Product System wherever possible. If the applicant identifies data that better represent the U.S. average GHG technology for the products and/or functions in their Comparison Product System, justification and documentation (including references to peer-reviewed, industry, or government sources) must be provided to verify and validate the decision to not use NETL unit processes.

The 45Q toolkit provides optional tools (e.g., the NETL 45Q U.S. Average (openLCA) LCI Database) meant for ease of use and flexibility to suit a range of taxpayer needs and be compatible with numerous external (e.g., commercially available) LCA software and databases. However, externally produced tools and data may not be readily compatible with the methods or data quality necessary for a 45Q LCA. These external databases and software must be evaluated by the applicant for appropriateness of use within the context of their LCA based on the required methods. Please note that within a given database, data quality varies process by process and sector by sector. For example, data for niche products might receive less focus by developers because they are assumed to be insignificant to most LCAs. This could make those data sets of insufficient quality for a specific 45Q LCA if the products are significant contributors to that LCA. For this reason, while NETL recognizes that using external LCA databases for background data may be convenient and can increase modeling consistency, NETL cannot make blanket recommendations for all taxpayers, sectors, or products given the range and complexity of use cases.

In addition to the NETL toolkit, taxpayers may consult other resources for their analyses, while independently evaluating whether the selected data satisfy the applicable section 45Q and NETL 45Q LCA requirements. For example, in addition to the NETL 45Q openLCA database, which was developed to support section 45Q LCA work, NETL has released a variety of other datasets across several platforms. These include contributions to the Federal LCA Commons (FLCAC), the Energy Data eXchange (EDX), and the Open Hydrogen Initiative (OHI) LCA Toolkit. NETL strives to use internally consistent LCA methods across its products, including a preference for system expansion over allocation where appropriate, but data-quality decisions are made process-by-process based on the context of the product and expected use. In addition to NETL data sources, other resources are available that could potentially be useful in calculations needed for a 45Q LCA; for example, the GREET platform and commercially available databases and tools.

GREET is not required to be used for a section 45Q LCA, and neither section 45Q nor the regulations designate a particular LCA database or software platform. Treas. Reg. section 1.45Q-4(c) requires the lifecycle analysis to conform with ISO 14040:2006 and ISO 14044:2006. Notice 2024-60, Section 3.02, further requires the LCA Report to be completed in accordance with the most current applicable NETL guidance as of the beginning of the year in which the LCA Report is submitted. GREET is a modeling platform containing numerous datasets some of which may be useful in a 45Q LCA, but use of GREET does not, by itself, establish compliance with section 45Q. Applicants must evaluate whether individual GREET datasets, assumptions, parameters, system boundaries, and modeling choices are appropriate for their section 45Q LCA and are consistent with the applicable ISO standards and NETL guidance.

Applicants should consider the following:

• GREET does not have a version specifically designated for section 45Q for utilization. Some GREET models include carbon capture, sequestration and utilization scenarios but those datasets do not cover the whole range of products or functions potentially applicable under Section 45Q (26 CFR 1.45Q-4(a)). It should be noted that the presence of a pathway or dataset in GREET does not by itself establish compliance with section 45Q requirements.
• Because the statutory language ties the tax credit amount to the GHG emissions being displaced, the 45Q LCA guidance prioritizes system expansion (NETL 45Q Guidance Section 2.1.10) and accounts for it using a multi-product functional unit. In contrast, GREET employs different co-product management strategies, such as system expansion with displacement and allocation to arrive at a single product, which are not compatible without proper justification.
• Many GREET datasets use the final product as the reference flow without information needed to easily truncate the system boundary at an intermediate step, as is preferred under the NETL 45Q LCA Guidance Section 2.1.4.
• ISO 14044:2006 defines “waste” as “substances or objects which the holder intends or is required to dispose of.” It should be noted that “emit” is a separate definition from “dispose,” which means that carbon dioxide does not meet the definition of waste because it is frequently emitted. Therefore, the “zero-burden assumption” cannot be applied merely because another LCA framework or dataset characterizes the stream as a waste. The treatment of a carbon oxide stream in a section 45Q LCA must be consistent with the study goal, applicable system boundaries, and co-product management requirements outlined in the NETL 45Q guidance. Treatment of captured carbon dioxide specifically is further explained in FAQ 2. GREET datasets may apply different assumptions to carbon dioxide or carbon monoxide streams, so applicants should review and, where necessary, modify those assumptions before incorporating the data into a section 45Q LCA.
• Treas. Reg. section 1.45Q-4(c)(1) requires greenhouse gas mass values to be adjusted for their relative global warming potential according to Table A-1 of 40 CFR part 98, subpart A. Different GREET models and versions may use different characterization factors or conventions. Applicants should therefore transfer inventory data in a manner that permits the section 45Q-required characterization factors to be applied and should verify that the factors used are consistent with Treas. Reg. section 1.45Q-4(c)(1). Applicants must account for the uptake and emissions of biogenic carbon separately and should not import a GREET biogenic-carbon accounting convention without determining whether it would change the result of the NETL 45Q LCA. The system boundaries required of a 45Q LCA often make the use of the “0/0” biogenic accounting shortcut inaccurate. Any uptake, release, co-product treatment, and system-boundary assumptions should be transparent, consistently applied, and documented in the LCA Report.
• The NETL 45Q LCA Guidance (Section 2.2.3.1) informs which materials and energy flows may be considered negligible, which may differ from GREET assumptions of negligibility. For example, GREET defaults to excluding construction impacts from the system boundary. Whether exclusions are appropriate based on negligibility under the NETL Guidance depends on the applicant’s product system.
• NETL 45Q LCA Guidance Section 2.1.3.2 requires that the Comparison Product System represent the U.S. Average GHG Technology. GREET data come from a variety of types of studies, not all of which intend to represent a U.S. Average. The profiles sometimes represent industry standard technologies, sometimes include market mixes of industry standard technologies to approximate regional or national averages (e.g. electricity grid mix), and sometimes represent novel or low-TRL (technology readiness level) technologies. Thus, when using GREET or other data for the Comparison Product System especially, it is important to read the documentation behind the data to determine if it meets the 45Q requirements or if modifications may need to be made.
• Unit-process-level data quality indicators (DQIs) required under ISO 14044 and NETL guidance are not always explicitly embedded in tabular exports from GREET. Furthermore, because upstream emissions in some worksheets are aggregated across life cycle stages, applicants extracting GREET data would need to manually trace intermediate stages to support the stage-by-stage contribution and sensitivity analyses. Lastly, because linked datasets may be modeled in separate worksheets (e.g., R&D GREET1 and GREET2), performing sensitivity analysis is also not straightforward. The 45Q LCA expectations are further explained in the NETL 45Q LCA Guidance (Section 2.1.8).

The following are common mistakes observed in 45Q LCAs:

• Using GREET data as a stand-in for taxpayer operations – GREET data can be used for CPS data where appropriate and potentially for background data within the PPS, but Notice 2024-60 requires direct or primary data to address the actual operational performance of the taxpayer's system for the taxable year covered by the LCA Report.
• Extracting GREET datasets without reviewing or modifying the parameters – The GREET models include default parameters, which may not be representative of the scenario being modeled in a section 45Q LCA. GREET intends the users to change these parameters to meet the goals of the study. Applicants should review the parameters and assumptions relevant to the selected dataset and modify them where needed to satisfy the applicable section 45Q methodology and data-quality requirements. Background datasets, including electricity and other upstream datasets, should be evaluated for temporal, geographic, technological, and other relevant representativeness; this does not necessarily mean that every background dataset must be recreated for the taxable year.
• Using the wrong units – Common errors are confusing short tons and metric tons, and lower and higher heating values. GREET uses a combination of U.S. customary engineering units and SI units. Mass units may include grams (g), pounds (lb), tons (short ton, 2,000 lb), and tonnes (metric ton, 1,000 kg), depending on the worksheet and data source. Within GREET, units are identified within the applicable tables or worksheets and should be verified when transferring data. Users should not assume that the same unit convention applies across all worksheets.
• Excluding upstream impacts unintentionally – Because GREET datasets sometimes present emissions under different column formats, emissions from certain stages may be unintentionally excluded. When transferring GREET process inventories, users should preserve the intended system boundary and ensure that upstream contributions are represented where needed.
• Using the wrong emission factors – The default emission factors used by GREET may differ from those in Table A-1. GREET inventory data should be transferred as individual emission species rather than as pre-characterized GHG impacts so that the appropriate 45Q characterization factors can be applied in the applicant’s chosen modeling tool.

There is no general section 45Q requirement to use the most recent GREET release. The GREET version and dataset selected should be appropriate and representative for the modeled process, and the applicant should identify and document the version used, relevant settings and modifications, and the basis for its selection. Newer GREET releases may incorporate corrections or updated assumptions and therefore should be considered when selecting a dataset, but annual releases do not necessarily update every underlying dataset to the year of release. If an interim version is used (for example, R&D GREET 2025 Rev. 1), that version should be identified and documented. If an applicant determines that a dataset from a different GREET version is more representative of the modeled flow, the applicant should explain and document that determination in the LCA Report.

In general, applicants should review which parameters are relevant to the dataset of interest before using a GREET dataset. In the worksheet version of GREET, input parameters are typically highlighted in yellow, and certain scenario selections are available through drop-down menus shown in peach-colored cells. While major input parameters can be modified in the “Input” tab of GREET (worksheet version), there are also other input parameters in other tabs that may also warrant review. Examples include the “Target Year of Simulation” parameter in the “Input” tab, which selects the year for analysis, “NG Production Region” and “NG Delivery Region,” and electricity generation scenarios. For more specific datasets, users may also consider reviewing plant efficiencies and power losses, as appropriate for their analysis.