ML22112A133

Kairos Power, LLC, Safety Evaluation for KP-TR-012 KP-FHR Mechanistic Source Term Methodology Topical Report, Revision 3 (EPID L-2020-TOP-0051/CAC 000431)

ML22112A133
Person / Time
Site:	99902069
Issue date:	04/22/2022
From:	William Kennedy NRC/NRR/DANU/UARL
To:	Hastings P Kairos Power
Cuadrado de Jesus S
Shared Package
ML22105A562	List: ML22112A133
References
CAC 000431, EPID L-2020-TOP-0051
Download: ML22112A133 (32)

Text

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION April 22, 2022 Mr. Peter Hastings Vice President, Regulatory Affairs and Quality Kairos Power LLC 707 W. Tower Ave.

Alameda, CA 94501

SUBJECT:

KAIROS POWER, LLC - SAFETY EVALUATION FOR KP-TR-012, KP-FHR MECHANISTIC SOURCE TERM METHODOLOGY TOPICAL REPORT, REVISION 3 (EPID NO: L-2020-TOP-0051/CAC NO. 000431)

Dear Mr. Hastings:

This letter provides the final safety evaluation for the Kairos Power LLC (Kairos) topical report KP-FHR Mechanistic Source Term Methodology Topical Report, Revision 3. By letter dated June 30, 2020 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML20182A785), Kairos submitted the topical report for U.S. Nuclear Regulatory Commission (NRC) staff review. On February 11, 2021 (ADAMS Accession No. ML21056A074), the NRC staff provided its review questions to Kairos. The review questions were discussed during closed meetings held between the NRC staff and Kairos on April 28, May 26, June 9, June 24, July 13, and July 28, of 2021. By letter dated August 19, 2021 (ADAMS Accession No. ML21231A289), Kairos submitted Revision 1 of the topical report to address the staff review questions. By letter dated February 16, 2022 (ADAMS Accession No. ML22047A325), Kairos submitted Revision 2 of the topical report to address concerns, discussed during a closed meeting held between the NRC staff and Kairos on February 7, 2022, regarding experimental data for vaporization of fission products. By letter dated March 28, 2022 (ADAMS Accession No. ML22088A228), Kairos submitted Revision 3 to address administrative errors in Revision 2 that redacted information that was previously made publicly available in Revision 1.

The NRC staffs final safety evaluation for KP-FHR Mechanistic Source Term Methodology Topical Report, Revision 3, is enclosed.

to this letter contains Proprietary Information. When separated from, this letter is DECONTROLLED.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION P. Hastings 2

The NRC staff provided Kairos a draft of the safety evaluation for the purpose of identifying proprietary information on October 29, 2021 (ADAMS Accession No. ML21307A043). On December 15, 2021, Kairos confirmed that the proprietary information in the draft safety evaluation was appropriately marked (ADAMS Accession No. ML21349B381).

The Advisory Committee on Reactor Safeguards (ACRS) was briefed on this topical report and the NRC staff draft safety evaluation on November 19, 2021, and November 30, 2021. The ACRS provided its recommendations for the publication of this safety evaluation in a letter dated December 20, 2021 (ADAMS Accession No. ML21342A179). By letter dated February 8, 2022, the NRC staff provided its response to the ACRS recommendations (ADAMS Accession No. ML22024A485). The enclosed safety evaluation is final, and a redacted version will be made publicly available.

The NRC staff requests that Kairos publish an accepted version of this topical report within 3 months of receipt of this letter. The accepted version shall incorporate this letter and the enclosed safety evaluation after the title page. The accepted version shall include an "-A" (designating accepted) following the topical report identification symbol.

If you have any questions, please contact Samuel Cuadrado at Samuel.CuadradodeJesus@nrc.gov.

Sincerely, William B. Kennedy, Acting Chief Advanced Reactor Licensing Branch Division of Advanced Reactors and Non-Power Production and Utilization Facilities Office of Nuclear Reactor Regulation Project No.: 99902069

Enclosures:

1. Proprietary (Non-Public) Safety Evaluation

2. Redacted (Public) Safety Evaluation cc:

Darrell Gardner gardner@kairospower.com Jim Tomkins tomkins@kairospowerwer.com

OFFICIAL USE ONLY PROPRIETARY INFORMATION Enclosure OFFICIAL USE ONLY PROPRIETARY INFORMATION KAIROS POWER, LLC - SAFETY EVALUATION REGARDING APPROVAL OF TOPICAL REPORT KP-TR-012, KP-FHR MECHANISTIC SOURCE TERM METHODOLOGY, REVISION NO. 3 (EPID NO. L-2020-TOP-0051)

SPONSOR AND TOPICAL REPORT INFORMATION Sponsor:

Kairos Power LLC (Kairos Power)

Sponsor Address:

Kairos Power LLC 707 W. Tower Ave.

Alameda, CA 94501 Docket/Project No(s).:

Project No. 99902069 Submittal Date: June 30, 2020 Submittal Agencywide Documents Access and Management System (ADAMS)

Accession No.: ML20182A785 Supplement ADAMS Accession No(s): KP-TR-012, Revision 3, March 28, 2022, ADAMS Accession No. ML22088A231 Brief Description of the Topical Report: The subject topical report (TR) provides a methodology to develop technology-specific mechanistic source terms (MSTs) for Kairos Power Fluoride Salt-Cooled, High-Temperature Reactor (KP-FHR) designs. The resulting MSTs are used in radiological consequence analyses that support applications for permits1, licenses, certifications, or approvals submitted under Title 10 of the Code of Federal Regulations (10 CFR) Part 50, Domestic Licensing of Production and Utilization Facilities, or Part 52, Licenses, Certifications, and Approvals for Nuclear Power Plants." Radiological atmospheric release source terms for design basis accidents (DBAs) developed using the TR methodology are for use in performing offsite radiological consequence analyses to show compliance with regulatory requirements for siting and safety analyses. Radiological atmospheric release source terms for anticipated operational occurrences (AOOs) and design basis events (DBEs) developed using the TR methodology are for use in implementing the methodology described in Nuclear Energy Institute (NEI) 18-04, Revision 1, Risk-Informed Performance-Based Guidance for Non-Light Water Reactor Licensing Basis Development, (Reference 1), as endorsed by the 1 As defined in 10 CFR 50.2, License means a license, including a construction permit or operating license under this part, an early site permit, combined license or manufacturing license under Part 52 of this chapter, or a renewed license issued by the Commission under this part, Part 52, or Part 54 of this chapter.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 2

NRC in Regulatory Guide (RG) 1.233, Guidance for a Technology-Inclusive, Risk-Informed, and Performance-Based Methodology to Inform the Licensing Basis and Content of Applications for Licenses, Certifications, and Approvals for Non-Light Water Reactors, (Reference 2).

For additional details on the submittal, please refer to the documents located at the ADAMS Accession No(s). identified above.

REGULATORY EVALUATION Regulatory Basis: 10 CFR 50.34(a)(1)(ii)(D), 10 CFR 50.34(b)(1), 10 CFR 52.47(a)(2)(iv),

10 FR 52.79(a)(1)(vi), 10 CFR 52.137(a)(2)(iv), or 10 CFR 52.157(d), as applicable The regulations cited above require that the reactor license application safety analysis report provides a description and safety assessment of the plant design features intended to mitigate the radiological consequences of accidents. Included in this assessment, is an evaluation of the safety features that are to be engineered into the facility and those barriers that must be breached as a result of an accident before a release of radioactive material to the environment can occur. The regulations also describe the main features of the evaluation, including postulation of a fission product release from the core (i.e., source term) and release to the environment, used together with applicable postulated site parameters, including site meteorology, to evaluate offsite radiological consequences. The safety assessment analyses are intended, in part, to show compliance with the radiological consequence evaluation factors for offsite doses at the exclusion area boundary (EAB) and outer boundary of the low population zone (LPZ). Applications for construction permits and operating licenses under 10 CFR 50.34, Contents of Applications; Technical Information, and combined licenses under 10 CFR 52.79, Contents of Applications; Technical Information in Final Safety Analysis Report, are also required to provide an analysis of the site using the same radiological consequence evaluation factors. Regardless of stationary power reactor application type, the radiological consequence evaluation factors for each of the cited regulations are the same:

An individual located at any point on the boundary of the exclusion area for any 2-hour period following the onset of the postulated fission product release, would not receive a radiation dose in excess of 25 rem total effective dose equivalent (TEDE).

An individual located at any point on the outer boundary of the LPZ, who is exposed to the radioactive cloud resulting from the postulated fission product release (during the entire period of its passage) would not receive a radiation dose in excess of 25 rem TEDE.

In addition, for stationary power reactor applications submitted after January 10, 1997, non-seismic siting criteria in 10 CFR 100.21(c)(2) require that the radiological dose consequences of postulated accidents shall meet the criteria set forth in 10 CFR 50.34(a)(1).

The TR MST methodology addresses the regulatory requirements for applications for power reactors cited above. The NRC staff acknowledges that portions of the methodology that describe modeling of physical phenomena related to development of the MST and modeling of atmospheric dispersion could be useful to inform analyses for non-power reactor applications, including testing facilities and research reactors. If referenced by an applicant for a facility other than a power reactor, the applicant should ensure that applicable regulatory requirements and any variations from the TR methodology are described in any application submitted.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 3

There is no NRC guidance on radiological source terms specifically applicable to non-light water reactor (non-LWR) designs. However, as a starting point for evaluation of non-LWR MST methodologies, the NRC staff refers to guidance about accident source terms that is of a generic nature (i.e., not dependent on the reactor technology) and positions on radiological consequence analysis and atmospheric dispersion in the following NRC documents:

NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR Edition, Section 15.0.3, Design Basis Accident Radiological Consequences of Analyses for Advanced Light Water Reactors, (Reference 3)

Regulatory Guide (RG) 1.183, Alternative Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors, (Reference 4), in particular Regulatory Position 2, Attributes of an Acceptable AST RG 1.145, Atmospheric Dispersion Models for Potential Accident Consequence Assessments at Nuclear Power Plants, Revision 1, (Reference 5)

RG 1.194, Atmospheric Relative Concentrations for Control Room Radiological Habitability Assessments at Nuclear Power Plants, (Reference 6)

In SECY-16-0012, Accident Source Terms and Siting for Small Modular Reactors and Non-Light Water Reactors, (Reference 7), the NRC staff stated that non-LWR applicants can use modern analysis tools to demonstrate quantitatively the safety features of those designs. In the NRC staff requirements memorandum (SRM) to SECY-93-092, Issues Pertaining to the Advanced Reactor (PRISM, MHTGR, and PIUS) and CANDU 3 Designs and Their Relationship to Current Regulatory Requirements, (Reference 8), the Commission approved the NRC staffs recommendation that source terms for non-LWRs be based upon a mechanistic analysis and that the acceptability of an applicants analysis will rely on the NRC staffs assurance that the following conditions are met:

The performance of the reactor and fuel under normal and off-normal conditions is sufficiently well understood to permit a mechanistic analysis. Sufficient data should exist on the reactor and fuel performance through the research, development, and testing programs to provide adequate confidence in the mechanistic approach.

The transport of fission products can be adequately modeled for all barriers and pathways to the environs, including the specific consideration of containment design. The calculations should be as realistic as possible so that the values and limitations of any mechanism or barrier are not obscured.

The events considered in the analyses to develop the set of source terms for each design are selected to bound severe accidents and design-dependent uncertainties.

The design-specific source terms for each accident category would constitute one component for evaluating the acceptability of the design.

In SRM-SECY-18-0096, Staff Requirements - SECY-18-0096 - Functional Containment Performance Criteria for Non-Light-Water-Reactors, (Reference 9), the Commission approved the NRC staffs proposed methodology for establishing functional containment performance criteria for non-LWRs. In SECY-18-0096, Functional Containment Performance Criteria for Non-Light Water Reactors, (Reference 10), the NRC staff described the functional containment concept as a barrier, or a set of barriers taken together, that effectively limits the physical

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 4

transport of radioactive material to the environment. Appendix C of RG 1.232, Guidance for Developing Principal Design Criteria for Non-Light-Water Reactors, (Reference 11), provides a rationale for containment design criteria for modular high temperature gas-cooled reactors. The rationale states that the term functional containment is applicable to advanced non-LWRs without a pressure retaining containment structure, and that a functional containment can be defined as a barrier, or set of barriers taken together, that effectively limit the physical transport and release of radionuclides to the environment across a full range of normal operating conditions, AOOs, and accident conditions.

Guidance in RG 1.233 describes the NRCs endorsement of NEI 18-04, Revision 1, which is a technology-inclusive, risk-informed, and performance-based methodology to inform the licensing basis and content of applications for non-LWR selection of licensing basis events (LBEs);

classification and special treatments of structures, systems, and components (SSCs); and assessment of defense in depth. The term licensing basis events is not defined in Part 50 or Part 52. As described in NEI 18-04, LBEs include AOOs, DBEs, DBAs, and beyond design basis events. Guidance in RG 1.233 provides high-level information on considerations for mechanistic source terms, including the relationship to probabilistic risk assessment.

TECHNICAL EVALUATION The subject TR describes a methodology an applicant can use to develop MSTs for a KP-FHR design for use in evaluating the offsite radiological consequences of DBAs, AOOs, and DBEs.

LBE identification for KP-FHR designs is not part of the MST methodology, and a list of KP-FHR LBEs is not provided in the TR. Therefore, this TR does not provide final values for event-specific MSTs. In the TR, Section 2.3.4, Additional Design Information Used in the Methodology, provides a list of assumptions related to KP-FHR design information used in the MST methodology, including some assumed LBE characteristics, based on the current state of design for the technology. As described in TR Section 8.2, Limitations, Limitation 8 on the use of the methodology requires that an applicant that uses the methodology verify that information in TR Sections 1.2.2, Key Design Features of the KP-FHR, and 2.3.4 is consistent with the specific KP-FHR design, or an applicant must justify deviations from the design assumptions listed in the TR. The NRC staff finds that Limitation 8 is reasonable and necessary, given that the design and LBE information used to develop the MST methodology consists of assumptions about plant conditions that account for the current lack of design finality. If the design features described in the TR are changed, the NRC staff determinations in this evaluation may not be applicable.

Mechanistic Source Term Approach Section 2, KP-FHR Mechanistic Source Term Evaluation Approach, of the TR describes the approach to the accident MST methodology, which is based upon evaluation of the accident source term phenomena and the concept of functional containment which are specific to the KP-FHR design. Kairos Power used a Phenomena Identification and Ranking Table (PIRT) process to inform the MST methodology. The NRC staff notes that the KP-FHR radiological source term PIRT covered phenomena associated with normal operations and transients that appear to be within the scope of the TR. The NRC staff did not review in detail or approve the PIRT as part of the review of this TR. However, based on the NRC staffs assessment of the PIRT tables in Section 2 of the TR, the identification and ranking of the relevant phenomena appear to be reasonable and no additional information was requested. In the TR methodology,

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 5

the development of event-specific radiological releases to the environment is predicated upon modeling the overall KP-FHR system as a series of sources of material at risk for release (MAR) and release fractions for each barrier that contains the MAR. The NRC staff notes that this system model is consistent with the description of functional containment in SECY-18-0096 and the discussion of MSTs in SECY-93-092 and RG 1.233. This evaluation of the barriers to release is also consistent with the safety analysis requirements in the regulations cited above.

Therefore, the NRC staff finds that the TR approach to developing MSTs by evaluating the sources of MAR and release fractions from each barrier is acceptable.

The TR Executive Summary, states that the source term evaluation methodology for the DBAs is deterministic and only credits the properties of the tri-structural isotropic (TRISO) particle fuel and the Flibe coolant for retention of radionuclides in the KP-FHR functional containment. The source term evaluation methodology for AOOs and DBEs uses a more realistic accounting of radionuclide barriers, which is consistent with the methodology for licensing basis events described in NEI 18-04. The NRC staff finds these aspects of the TR methodology acceptable because using a more realistic evaluation of barriers is consistent with the RG 1.233 discussion of LBE evaluation and related source terms. The NRC staff finds the evaluation of DBAs by only crediting retention in TRISO and Flibe is acceptable because it is consistent with the guidance in Standard Review Plan 15.0.3 and RG 1.183 on crediting safety-related SSCs and modeling design information in accident source terms.

To aid in the evaluation of offsite radiological consequences, the TR methodology develops event-specific radiological releases to the environment for LBEs (i.e., MSTs), and provides a method for estimating site-specific atmospheric dispersion factors for distances under 1,200 meters. However, with respect to the overall accident radiological consequence analysis methodology, the TR does not include specific positions on other factors in the consequence analysis such as dose coefficients and assumptions on breathing rate, other than referring to use of default parameters in the SNAP/RADTRAD computer code. The applicant using the TR methodology in a licensing application will address these additional considerations in the analyses that support the application.

Appendices A and B of the TR provide sample calculations to demonstrate the MST methodology. The NRC staff did not evaluate the calculations in Appendices A and B; therefore, the staff makes no finding on their acceptability for reference or use in licensing applications.

Sources of MAR in the KP-FHR In the TR, Sections 2.2, Sources of Material at Risk in the KP-FHR, and 2.3.1, Identification of MAR and affected barriers, describe the methodology step to identify sources of MAR, with more detailed discussion in subsequent sections of the identification of sources of MAR in the fuel and graphite pebbles, Flibe, graphite reflector structures, cover gas, and other sources such as graphite dust, filters, and cold traps. The distribution of steady-state MAR throughout the plant is an initial condition for the MST methodology. In the TR, Section 2.3.4 provides assumed design information (including assumptions on radionuclide activity and MAR) and relevant design bases. To address the initial conditions related to sources of MAR in the coolant, cover gas, and other sources, TR Section 8.2 provides Limitation 5, which states that an applicant may reference the TR for use only if the applicant establishes operating limitations on maximum circulating activity and concentrations relative to solubility limits in the reactor coolant, intermediate coolant, cover gas, and radwaste systems that are consistent with the

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 6

initial condition assumptions in the safety analysis report. The NRC staff finds that Limitation 5 is acceptable to ensure that the plant operating limits provide operating conditions consistent with the assumed initial conditions in the TR given that initial condition information is input to the MST methodology that depends on the specific design information. The NRC staff will evaluate the radionuclide inventories or initial activity concentrations in the coolant, cover gas, radwaste systems and other potential sources as an initial condition for the MST during the review of a licensing submittal which uses the MST TR methodology.

Section 2.5.1, SERPENT2, of the TR states that the radionuclide inventory in the core is calculated by the SERPENT2 code (Reference 12). Kairos Power plans to verify and validate SERPENT2 as part of the KP-FHR core design and analysis methodology, which will be provided in a future licensing submittal. The developer of SERPENT2 describes it as a multi-purpose three-dimensional continuous energy Monte Carlo particle transport code that has been developed for use in reactor physics applications, including isotope generation and depletion analysis. The guidance in RG 1.183, Section C.3.1, states that the core inventory should be determined using an appropriate isotope generation and depletion computer code.

The NRC staff finds that, because SERPENT2 was developed to perform reactor isotope generation and depletion analyses, the use of SERPENT2 to determine core inventory is appropriate, consistent with the guidance in RG 1.183, and is acceptable. The NRC staff will evaluate the development of the core inventory during the review of an applicant submittal that uses the MST TR methodology.

Barrier Evaluation Once the sources of MAR affected by an event sequence are identified, then the barriers to release to the environment for that sequence are identified. Section 2.3.2, Qualitative Evaluation, of the TR describes a qualitative evaluation of the barriers, while TR Section 2.3.3, Quantitative Evaluation, describes the quantitative evaluation. The barrier evaluations develop the modeling of the performance of the barriers (including performance of SSCs), phenomena affecting radionuclide retention and transport, and treatment of model uncertainty. The quantitative evaluation includes a step to screen sources of MAR based on a de minimis threshold. This screening determines the MAR which cannot affect the offsite dose figure of merit, and therefore can be treated conservatively. The de minimis screening for release pathways includes two types of thresholds: absolute and relative. The absolute threshold evaluates whether the release of the MAR ((

))

The relative threshold evaluates whether for the release of the MAR ((

)) For pathways that meet the de minimis thresholds, the MAR is not transferred to the next barrier.

The radionuclide release figures of merit for the DBAs are determined based upon the calculation of the worst 2-hour dose at the EAB and the dose for the duration of the plume at the LPZ, which are consistent with the siting and safety analysis requirements cited in the

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 7

Regulatory Evaluation section in this safety evaluation. The radionuclide release figures of merit for AOOs and DBEs are determined based upon the calculation of a 95th percentile representation of the 30-day doses at the site boundary, with consideration of uncertainty and modeling choices. This is consistent with the NEI 18-04 methodology. The methodology also has the user evaluate the effects of conservative inputs associated with screened MAR and assess the effects of model uncertainty, reasonable alternatives, and assumptions made due to lack of design detail on the radionuclide release figures of merit. The NRC staff finds that the de minimis screening described in TR Section 2.3.3 is acceptable because it assures that a comprehensive list of sources of MAR and release pathways (including those that are not likely to contribute more than a small fraction of the total offsite dose results) are included in the analysis with modeling assumptions consistent with their relative importance.

Software The barrier evaluation determines whether detailed evaluation models are necessary to model the radionuclide behavior. Section 2.5.2, KP-Bison, of the TR states that the MST methodology uses KP-Bison, which is the Kairos Power version of the Bison code developed by Idaho National Laboratory (Reference 13), to assess the mechanical integrity of the TRISO-coated particles and the retention of fission products by intact or partially failed particles.

The TR states that Kairos Power plans verification and validation of the KP-Bison code as part of the KP-FHR fuel performance methodology, which is currently under review by the NRC staff (Reference 14). Section 8.2, Limitations, of the TR specifies that approval of the KP-FR fuel performance methodology TR is Limitation 1 on the use of the TR methodology. The NRC staff finds that this limitation is acceptable because the limitation ensures that the KP-Bison code will be appropriate for use to determine radionuclide releases from the fuel.

The MST term methodology also uses KP-SAM (which is the Kairos Power version of the System Analysis Module (SAM) code developed by Argonne National Laboratory (ANL)

(Reference 15) to provide event-specific thermal fluid conditions. Kairos Power states that it plans verification and validation of the KP-SAM code as part of KP-FHR transient methodology, which will be provided in a future licensing submittal. The NRC staff will evaluate the implementation of the KP-SAM code as input to the MST methodology during the review of a licensing submittal that uses the MST TR methodology.

Section 2.5.5, RADTRAD, of the TR identifies use of the Radionuclide, Transport, Removal, and Dose Estimation (RADTRAD) computer code within the Symbolic Nuclear Analysis Package (SNAP) (Reference 16) to model radionuclide transport within the gas space of the reactor building and calculate offsite doses. The TR states that non-applicable LWR models within SNAP/RADTRAD will not be used. The NRC staff has extensive experience with the SNAP/RADTRAD computer code, which was developed for the NRC for use in evaluation of DBA radiological consequences. The NRC staff uses the code in analyses performed to determine if the staff can confirm the results of licensee analyses submitted in license applications. The code allows for radionuclide source input and compartment modeling with user input on the source and transport between the compartments. Users of SNAP/RADTRAD can also provide input on retention or removal within and between compartments, or use correlation models included within the code, if appropriate. Based on its knowledge of and experience with the code, the NRC staff finds that the SNAP/RADTRAD computational framework to model transport and removal of radionuclides and estimate doses at selected receptors is flexible enough to support radiological consequence analysis for non-LWRs,

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 8

including the KP-FHR. Use of SNAP/RADTRAD, which is an NRC-developed computer code that includes correlations and models discussed in RG 1.183, is consistent with guidance in RG 1.183 for accident radiological consequence analysis.

The MST methodology atmospheric dispersion modeling assumes use of the Atmospheric Relative Concentrations in Building Wakes (ARCON) computer code (Reference 17), developed for the NRC to estimate short-term atmospheric dispersion factors for use in DBA radiological consequence analyses for receptors onsite, such as the control room. Although control room radiological habitability analysis is not within the scope of the TR, the MST methodology assumes the EAB and LPZ distances are less than 1,200 meters from the reactor building.

Because of the shorter distances, the TR describes use of the ARCON code in place of PAVAN (Reference 18), the NRC-developed computer code traditionally used for estimation of short-term atmospheric dispersion factors. The NRC staff has extensive experience with use of both ARCON and PAVAN through performance of confirmatory analyses of results submitted in license applications. The NRC staffs evaluation of the TR methodologys use of the ARCON code to estimate short-term atmospheric dispersion for the EAB and LPZ receptors is discussed in the Atmospheric Dispersion section of this safety evaluation.

Radionuclide Retention in Fuel Section 3, Evaluating Fuel Retention of Radionuclides, of the TR provides a description of the modeling of radionuclide transport from and retention in the fuel. Radionuclide retention in TRISO particles is considered by Kairos Power to be part of the functional containment. Kairos Power used information from the Advanced Gas Reactor (AGR) tests and performed a PIRT to identify fuel-related phenomena important to development of a radiological source term for transients in which the reactor coolant boundary integrity is maintained. Section 3.2, Sources of Radionuclides in Fuel, of the TR identifies the sources of radionuclides in the fuel and qualitatively describes potential release from the fuel due to manufacturing defects, heavy metal contamination, in-service fuel failures, and releases from intact TRISO fuel particles.

Section 3.2.1, Manufacturing Defects, of the TR states that the ((

)) The NRC staff is currently reviewing a separate Kairos Power TR on fuel qualification (Reference 19), which includes development of a KP-FHR fuel specification intended to ensure that the KP-FHR fuel pebble design can operate with the low failure fractions observed in the AGR testing program. Because these radionuclide release mechanisms from the fuel are consistent with the results of AGR program and known literature, the NRC staff finds that the radionuclide release mechanisms from the fuel are identified appropriately.

TR Section 3.3: Radionuclide Behavior and Retention Properties of Fuel The TR states that the TRISO particle is the primary medium for retention of fission product in the KP-FHR, where the MAR in the fuel particle is held up by the fuel kernel and the three coating layers. Table 3-1 of the TR provides a list of radionuclides that are assumed to be retained within a TRISO fuel particle under various coating layer failure conditions.

Section 3.3.1, Radionuclide Behavior in Fuel, of the TR describes how the chemical forms of the fission products were determined so that the movement of the radionuclides within the fuel and transport from the fuel can be appropriately modeled, and the chemical forms also support

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 9

the assumptions on radionuclide grouping in the fuel transport modeling. Section 3.3.2, Selection of Radionuclides, of the TR describes the production of radionuclides and the characteristics that affect their relative importance to support the selection of radioisotopes that are included in the MST. Based upon its review of the information in TR Section 3.3, the NRC staff finds that the TR provides an appropriate methodology to select radionuclides for inclusion in the MST based upon consideration of fission product production, generation of activation products, radioactive decay, and mobility of radionuclides in the fuel.

TR Section 3.4: Radionuclide Transport in Fuel The KP-FHR MST methodology uses the KP-Bison code described in the KP-FHR fuel performance methodology topical report, which is currently under NRC staff review. The KP-Bison code will be used to model the transport of fission products through a TRISO particle, based on assessing diffusion ((

))

through each coating layer and subsequent release to the carbon matrix in the fuel pebble. This modeling results in time-and position-dependent sources for fission product transport analysis in the pebble. The NRC staff finds that the TR modeling of radionuclide transport and release by assessing diffusion through and from the fuel is consistent with common practice for radiological source term analysis and the concept of developing an MST. In TR Section 8.2, Limitation 1 states that approval of KP-Bison for use in fuel performance analysis as captured in the KP-FHR fuel performance methodology topical report is a limitation on use of the MST topical report methodology. Given Limitation 1, the NRC staff finds the use of KP-Bison to be acceptable because it will mechanistically calculate the amount of failed fuel to use in development of the initial MAR.

The KP-Bison models transport of fission products as the four elements for which releases were historically detected and measured in post-irradiation testing on TRISO fuel in the AGR program: cesium (Cs), strontium (Sr), silver (Ag), and krypton (Kr). Based on this AGR data, these four elements have complete sets of diffusivities for the fuel pebbles. Table 3-3 of the TR provides the list of elements included in the MST, and their grouping. The NRC staff finds the grouping of radionuclides transported in the fuel acceptable because it is based on modeling transport of radionuclides with similar chemical and diffusion behavior in a consistent manner.

Section 8.2 of the TR, Limitation 4, states that confirmation of minimal ingress of Flibe into the pebble matrix carbon under normal and accident conditions, such that incremental damage to TRISO particles due to chemical interaction does not occur as captured in the KP-FHR fuel qualification methodology topical report, is a Limitation on the use of the TR methodology. The NRC staff finds that this limitation is acceptable because the limitation ensures that the conditions for radionuclide release from the fuel pebbles is bounded by the assumptions in the TR methodology.

Radionuclide Retention in Flibe Section 4, Evaluating Flibe Retention of Radionuclides, of the TR provides a description of the modeling of radionuclide transport and retention in Flibe. Because of tritiums unique qualities, the NRC staffs evaluation of the TR methodology for tritium transport and retention in Flibe and graphite is described in a separate section below. Section 8.2 of the TR, Limitation 7, states that the retention of radionuclides in solid Flibe is beyond the scope of the analysis described in the TR methodology. The NRC staff finds Limitation 7 acceptable because the methodology

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 10 described in Section 4 of the TR only covers phenomena related to molten Flibe and does not address the chemistry of radionuclide retention in solid Flibe.

TR Section 4.2: Sources of Radionuclides in Flibe The NRC staff evaluated the sources of radionuclides in Flibe to determine whether the methodology accounts for the appropriate sources of radionuclides. The NRC staff finds that the sources described by Kairos Power are appropriate as radionuclides born in Flibe and that escape from the TRISO fuel are considered. Additionally, it is appropriate because different pathways (e.g., transmutation, fission product release, etc.) are considered to determine the MAR in Flibe.

TR Section 4.3.1: Radionuclide Transport Groups in Flibe The NRC staff evaluated the part of the Kairos methodology to group radionuclides by their reduction-oxidation (redox) reactions with Flibe. The NRC staff finds it acceptable to group radionuclides by their chemical behavior in Flibe because this behavior will impact how the radionuclides are retained by Flibe. Additionally, this part of the methodology is acceptable because the grouping process ((

)) with the measurements taken as part of the Oak Ridge National Laboratory (ORNL) Molten Salt Reactor Experiment (MSRE) (Reference 20).

The method to determine whether a species forms a salt-soluble fluoride is acceptable because it uses fundamental principles of chemistry to determine whether a reaction proceeds.

Additionally, it is acceptable to ((

))

The NRC staff finds the sorting of insoluble metallic phases acceptable because species in the reduced metallic phase are not likely to form salt-soluble compounds and this grouping is consistent with the MSRE behavior (References 20 and 21). Additionally, phase separated radionuclides will not be inhibited by the Flibe barrier and are available to be released based on representative vapor properties for the groupings.

For the radionuclides grouped as gases, the NRC staff finds the grouping acceptable because

((

)) Additionally, MSRE experience shows that low amounts of noble gases were found in the salt and high amounts were found in the off-gas system. Therefore, the NRC staff finds it acceptable for Kairos Power to [

].

The NRC staff finds it acceptable to group radionuclides into groups for the reduced phase (noble metals), salt-soluble fluorides, redox dependent species, and noble gases as shown in Table 4-2 of the TR. This is consistent with the data generated during the MSRE. Additionally, the NRC staff finds the radionuclide transport groupings acceptable because the methodology

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 11 determines the thermodynamically favored chemical states for radionuclides in the Flibe and groups them accordingly.

((

))

((

]

Grouping of Radionuclide Elements by Exchange Reactions with Flibe The NRC staff reviewed the part of the methodology to determine the transport groupings for radionuclide oxides released from the TRISO fuel. The NRC staff finds it acceptable to assess these exchange reactions because this method accounts for the chemical state of radionuclides released from the TRISO to determine whether these radionuclides form salt-soluble compounds or remain as insoluble oxides. Additionally, radionuclide species formed from these reactions will be grouped in the same manner as those formed from redox reactions. As described above in the discussion of TR Section 4.3.1, the NRC staff found the groupings acceptable.

Quantification of Release Fractions for Flibe This TR section generally describes the process through which radionuclides may escape from the Flibe barrier. The details for the process are covered in TR Sections 4.3.1.1 through 4.3.1.5.

The NRC staff evaluation of the details in these sections are covered in the following sections of this safety evaluation. Additionally, Kairos Power states that aerosolization is covered in TR Section 7, Evaluating Radionuclide Transport in the Gas Space and Atmospheric Transport, and therefore in its review of TR Sections 4.3.1.2 through 4.3.1.5, provided below, the NRC staff only evaluates vaporization of radionuclides from the Flibe.

Liquid Phase Equilibria in Molten Flibe Solutions The NRC staff evaluated the part of the methodology that assess reactions between salt-soluble radionuclides to determine the chemical state of the radionuclides. The NRC staff finds it necessary, and acceptable, to consider reactions between radionuclides in the Flibe because these reactions may impact the transport behavior of the radionuclides. Additionally, this section of the methodology accounts for reactive vaporization of radionuclides that may occur in certain scenarios (e.g., Flibe spill). The NRC staff finds this acceptable because all potential radionuclide release mechanisms from the Flibe should be evaluated. Additionally, the NRC

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 12 staff finds the method to use the Gibbs Free Energy of Reaction (Gibbs Energy) to determine equilibrium quantities of radionuclides acceptable because the use of Gibbs Energy is consistent with methods described by ANL in its development of a mechanistic source term approach for liquid-fueled molten salt reactors (Reference 22).

This part of the methodology allows Kairos Power to determine the chemical activities of radionuclide species to be analyzed in the transport analysis. The NRC staff reviewed the method to determine the total quantity of each species that is formed. The NRC staff finds the method acceptable because the TR states that activity coefficients are determined experimentally and an activity coefficient of 1 is assumed for species in solution unless a smaller value is justified. The NRC staff finds the use of pure compound vapor pressures acceptable, as it will likely result in a higher calculated vapor pressure of radionuclide species when compared to scenarios accounting for nonidealities (Reference 23), which is conservative and will be supported by data as described in Limitation and Condition 11. Additionally, as described below, Kairos Power provided certain measures to ensure a dilute solution of radionuclides in the Flibe. This will help reduce the chance of nonidealities in the solution causing increases in vapor pressures of radionuclides, and subsequent higher releases of radionuclides from the Flibe.

The NRC staff also reviewed the rationale for assuming dilute solutions. A dilute solution is necessary to minimize certain chemical interactions that could increase the vaporization of Flibe and radionuclide species. The dilute solution assumption is supported by use of TRISO fuel, which will only release a small quantity of radionuclides relative to the Flibe volume, and Limitation 5 in TR Section 8.2, which states Kairos Power will establish a limitation on maximum circulating activity in the salt. Having a maximum initial concentration coupled with no radionuclides added to the salt above a de minimis level during a transient will allow the NRC staff to verify that there is a dilute solution of radionuclides in Flibe during the source term implementation. A dilute solution is also important to ensure that radionuclides do not have significant impacts on the physical properties of the Flibe as described in Section 4.3.1 of the TR. Additionally, the NRC staff finds this acceptable because the TR states that the MST methodology is restricted to scenarios where radionuclides are in a dilute solution and concentrations are well below solubility limits. The NRC staff finds Limitation 5 acceptable because these controls on circulating activity in the salt are necessary to support the dilute solution assumption used in the TR methodology.

TR Section 4.3.1.2: Equilibrium Vapor Pressure of Pure Radionuclide Species Over Flibe Solutions The NRC staff evaluated the part of the methodology that determines the equilibrium vapor pressure of radionuclides over the Flibe. The NRC staff finds this part of the methodology acceptable because Kairos Power will determine the quantity of radionuclides that are present in the vapor phase above the salt at equilibrium between the salt and vapor phase based on fundamental chemistry and physics. Using equilibrium vapor pressures is conservative because it assumes that these pressures are reached instantaneously and not slowed by factors such as mass transfer to the Flibe-gas interface or reaction kinetics for formation of volatile chemical species. The methodology is also acceptable because the methodology conservatively considers the vapor pressure of pure compounds, which will likely result in a higher calculated vapor pressure than when accounting for non-idealities in the solution.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 13 TR Section 4.3.1.3: Grouping Radionuclide Vapor Behavior The NRC staff reviewed the part of the methodology to choose a representative species for the salt-soluble fluoride radionuclide transport group. The NRC staff finds it acceptable because the methodology compares the vapor pressures of different potential species in the salt to determine a representative species for the transport analysis. The NRC staff also finds the use of cesium fluoride (CsF) as the representative species appropriate because CsF has a high vapor pressure compared to radionuclide species that have insignificant vapor pressure at operating conditions.

TR Section 4.3.1.4: Vapor Pressure of Radionuclide Phases Over Flibe Solutions The NRC staff evaluated how the MST methodology calculates the vapor pressure of radionuclides over the Flibe. The NRC staff finds the methodology acceptable because it uses an appropriate thermodynamic relationship to determine the partial pressure of radionuclide species above Flibe. The NRC staff finds this part of the methodology acceptable because it uses representative species that appropriately models transport for each of the radionuclide groups. Additionally, the NRC staff finds the use of an equilibrium relationship acceptable because the use of TRISO fuel, which will only release a small quantity of radionuclides relative to the Flibe volume, coupled with Kairos Power Limitation 5, which will set circulating activity operational limits consistent with design basis accident initial conditions, allows for quantification of maximum equilibrium concentrations of MAR available for release. Therefore, it is conservative to use an equilibrium relationship because under these assumptions no radionuclides are added to the coolant during a postulated event. Also, this relationship assumes that equilibrium vapor pressures are reached instantaneously. This assumption means that the equilibrium concentration of radionuclides in the Flibe will be highest at the start of an event and use of this concentration will result in a higher vapor pressure of radionuclides over the Flibe.

Section 8 of the TR, Limitation 2, states that Kairos Power will provide justification for use of thermodynamic data and vapor pressure correlations for representative species in safety analysis reports for licensing application submittals. This is necessary for the methodology to determine vapor pressures of radionuclides over Flibe to be acceptable. This is because Limitation 2 will ensure that the data used to calculate vaporization of radionuclides (i.e., vapor pressures) is applicable to the KP-FHR operating conditions and that appropriate uncertainties in the data are considered. The NRC staff finds Limitation 2 acceptable because it requires an applicant that references this TR to analyze thermochemical data for representative species used in the methodology to ensure the data is applicable to the Kairos Power design and that an applicant can account for uncertainties and errors in the data. The NRC staff finds that it is acceptable to analyze the specific radionuclides referenced because these are the species that are used in the transport methodology described in Section 4 of the TR.

TR Section 4.3.1.5: General Vaporization Rate Law The NRC staff reviewed the portion of the MST methodology used to calculate the rate of mass transfer of radionuclide species from the Flibe to the cover gas. The NRC staff finds this part of the MST methodology acceptable because it calculates the rate of mass transfer from the surface of the Flibe to the bulk gas space atmosphere and uses the partial pressure of the radionuclide species as the driving force for mass transfer. Use of the partial pressure over the

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 14 Flibe is acceptable because it does not consider the partial pressure of the radionuclide in the cover gas, which maximizes the driving force out of the Flibe, and is therefore conservative.

The method to determine the applicable mass transfer coefficient is acceptable because the coefficient will be empirically derived and will represent the expected KP-FHR conditions.

Therefore, the NRC staff finds that the vaporization rate equation is acceptable to model the rate of radionuclide vaporization from the Flibe.

TR Section 4.3.1.6: Experimental Justification of Vaporization Rates The NRC staff reviewed the proposed experimental justification of vaporization rates described in this section of the MST methodology. The staff finds this part of the MST methodology acceptable because these experiments will provide data to demonstrate the portion of the MST methodology used to calculate radionuclide vaporization is conservative. The staff has added Limitation and Condition 11 to ensure these data are provided to the staff at the licensing stage for review.

Radionuclide Retention in Flibe - Additional Limitations on Use of TR The NRC staff reviewed Section 4 of the TR as it relates to the methodology to model retention of radionuclides by the KP-FHR reactor coolant. As described above, the NRC staff finds this portion of the methodology acceptable, subject to the limitations identified by Kairos Power in TR Section 8.2. Several KP-FHR design features provide the NRC staff assurance that the methodology is reasonable. These include limits on circulating activity in the Flibe, the margin between postulated accident temperatures and the boiling point of Flibe, the near-atmospheric operating pressure, and the ability of TRISO to retain radionuclides at postulated accident temperatures. Based upon these findings and the influence of the unique features of the KP-FHR design on the retention of radionuclides in the Flibe, the NRC staff imposes the following additional limitation and condition on the use of the TR:

Limitation and Condition 9 - The use of this topical report is limited to the KP-FHR design and is not applicable to other molten salt reactor designs because the KP-FHR utilizes TRISO fuel which is stated to retain most radionuclides. As described in the topical report and the NRC staff safety evaluation, this allows for the use of certain simplifying assumptions related to retention of radionuclides in the molten salt.

The NRC staff also imposed another limitation and condition on the use of the TR to ensure that the proposed method to determine vaporization of radionuclides from the Flibe is supported by experimental data.

Limitation and Condition 11 - An applicant that uses this methodology must provide experimental data as described in Section 4.3.1.6 of this topical report.

Tritium Transport and Retention In this TR, Kairos Power presents the tritium source term methodology that includes tritium formation, and its transport and holdup in fuel pebbles and core moderator, graphite structures, vessel, primary piping, and intermediate heat exchangers. Kairos Power properly treated tritium as a unique contributor to the KP-FHR MST and recognized it in the MST methodology with

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 15 appropriate attention to its unique attributes and phenomena involved in its potential release. In the TR, Section 3.2.5 presents tritium formation, Section 3.5 presents retention of tritium in the graphite pebble, Section 5.3.1 presents tritium speciation, and Section 5.3.2 presents tritium retention and permeation. The NRC staff review of these sections follows.

TR Section 3.2.5: Tritium Tritium (3H or T) is mainly produced in the KP-FHR via a set of neutron reactions in the Flibe salt. Little tritium is generated in the fuel and Kairos Power assumed that tritium formation in the fuel can be ignored based on the de minimis principle; hence Kairos Power did not consider tritium formation in the fuel as relevant to modeling of the tritium contribution to the MST. The NRC staff considers this assumption acceptable because the amount of tritium generated in fuel is negligible compared to the tritium generated in the salt (Reference 24).

The neutron reactions in the Flibe salt that contribute to the majority of tritium production are presented via Equations 7 through 12 of the TR. Equations 7 and 8 can be used to describe the tritium generation through neutron capture by lithium (6Li and 7Li), a component of Flibe, and the major source of the tritium in the KP-FHR. A much smaller fraction of tritium is produced indirectly due to neutron capture by beryllium (9Be) and production of 7Li, which is expressed in Equation 9. Likewise, the neutron capture by 9Be in Equation 11 contributes indirectly to the production of tritium by the production of 6Li in Equation 12. The NRC staff considers this tritium source characterization in Equations 7 through 12 to be appropriate as it is consistent with the available literature (References 25 and 26).

TR Section 3.5: Retention of Tritium in the Graphite Pebble This TR section states that there are two types of graphite materials in the KP-FHR core:

nuclear graphite reflector blocks and graphite in fuel pebbles and moderator pebbles. Since graphite can retain tritium through absorption, it can be considered as a tritium sink. Pebbles have a relatively high surface area and thus collectively store the most tritium MAR in the primary system. The tritium source term methodology indicates how the tritium stored in graphite pebbles and structures can be desorbed.

Kairos Power has modeled the tritium mass transfer between the pebbles and either the Flibe or cover gas using Henrys and Sieverts Laws. These laws can be applied (with appropriate uncertainty) to characterize how the tritium concentration in the salt increases via generation and decreases via tritium retention in the graphite, as described on Section 3.5 of the TR. The quantity of absorbed tritium in fuel pebbles is a function of the (i) Flibe tritium concentration, (ii) pebble removal rate (pebbles will float in the KP-FHR core, and will be removed from the top of the core, partially desorbed, and re-inserted to the core bottom), (iii) temperature, and (iv) operating history as pebbles are predicted to saturate quickly.

The volatile tritium species released in the Flibe are T2 and TF (tritium fluoride). Tritium uptake into graphite in the core accounts for pebble recirculation and is only calculated for T2 in the Flibe. The TR assumes that any TF present does not interact with graphite. The TR based this approach on the tritium analysis results from graphite samples tested in the Massachusetts Institute of Technology fluoride salt irradiation experiments (Reference 27). Based on experimental verification provided in Reference 27, the NRC staff finds the assumption that TF has no interaction with graphite to be acceptable for the MST methodology.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 16 In Equations 18 to 26, Kairos Power included a detailed set of mass transfer correlations that describe the tritium retention models in the TR including tritium transit from the bulk salt to the graphite pebble surface. Equation 19 shows the dependence of Reynolds number (Re) for fixed and fluidized beds of spheres on the pebble diameter, as well as the density and viscosity of the salt and the velocity of salt flow in the core. The correlation selected by Kairos Power for Equation 19 corresponds to Item E in Table 5-23 of TR Reference 60. Based on a comparison with the public literature, the NRC staff finds that Kairos correctly selected and applied the correlation in Equation 19 in the TR. Equation 22 is a mass transfer correlation that was taken from TR Reference 59 but is also listed in the TR Reference 60. The correlation is Item E in the Correlations column in Table 5-23 of the TR Reference 60, and the Re range from 55 to 1,500 of Item E is used in the TR. Based on its review of public literature, the NRC staff finds that the Kairos MST methodology correctly selected and applied the correlation in Equation 22.

TR Section 5.3.1: Tritium Speciation This TR section states that tritium is instantly formed and remains at the equilibrium balance of TF and T2 predicted by the redox potential as shown in Equation 63 of the TR. The TF and T2 can both exist as dissolved gases in Flibe. The TF and T2 have different diffusivities and solubilities in Flibe and have different transport paths in the reactor system. For example, T2 can permeate through metals whereas TF will not (Reference 26). However, both can evolve to a gas phase above the salt, as discussed in Reference 28. The NRC staff finds that the tritium speciation characterization presented in the TR is appropriate and consistent with the available literature referenced.

TR Section 5.3.2: Tritium Retention and Permeation As shown in TR Figure 7-8 and Section 5.3.1, respectively, tritium permeates through the reactor vessel wall and other metals. Section 5.3.2 of the TR recognizes that tritium permeation through piping, vessels, and the intermediate heat exchanger as a unique contributor to sources of MAR for transients in the primary system is properly recognized in the methodology. For a given tritium flux, there will be faster diffusion of tritium away from the salt/metal interface and a much lower surface tritium concentration compared to the graphite case. The methodology uses Sieverts Law for the salt/metal interface. As illustrated in TR Figure 7-8, the tritium retention in the graphite reflector is typically limited to regions near the outer surface of the reflector, which interface with the downcomer. The NRC staff notes that the mechanism for tritium retention in the reflector is similar to that in the fuel pebbles because, as noted in TR Section 3.5, the MST methodology assumes that both contain graphite IG-110.

In TR Section 5.3.2, Kairos Power models the reactor vessel downcomer as an annular flow region between the graphite reflector as the inner diameter and metallic reactor vessel wall as the outer diameter and uses Equations 67 to 71 to evaluate the tritium mass transport. Equation 67 was modified so that the Sherwood number (Sh) for mass transport in the region is obtained from an empirical correlation measured for Reynolds numbers (Re) between 10,000 and 100,000 and Schmidt numbers (Sc) between 430 and 100,000. Kairos Power provided a reference for the source of this correlation (see TR Reference 63). The NRC staff reviewed TR Reference 63 and located the correlation that is used by Kairos Power as the modified Equation 67 in TR. In evaluating Kaiross choice among the several correlations presented in TR Reference 63, the NRC staff found that the correlation selected by Kairos Power represents

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 17 the data from all the runs in the experiment, with an average deviation of 5.4 percent and maximum deviation of 13.2 percent. This correlation is also listed in Table 5-19 of the TR Reference 60 as Item R. Based on this review, the NRC staff confirms that Kairos Power made the correct choice of correlation for the modified Equation 67 in the TR by comparing with public literature.

Kairos Power used an alternate correlation for the Equation 68 in the TR. Here, TR Reference 64 is used to justify the use of the Chilton-Colburn analogy to construct an empirical mass transfer correlation along with applicable Nusselt number (Nu) based heat transfer correlation for both laminar and turbulent conditions. The Chilton-Colburn analogy is generally applicable for laminar and turbulent conditions when Prandtl numbers (Pr) are between 0.6 and 60, and Schmidt numbers (Sc) are between 0.6 and 3,000, as discussed in TR Reference 64. The NRC staff reviewed TR Reference 64 and finds the correlation used by Kairos Power for Equation 68 to be acceptable because this equation is consistent with the equation in TR Reference 64, which is a peer reviewed publication.

Tritium Source Term Limitations Section 8 of the TR, Limitation 3, states that Kairos Power will address validation of the tritium transport model in future submittals which use the MST methodology. The NRC staff finds this limitation to be acceptable because Kairos Power recognizes the lack of validation needs to be addressed in future submittals that use this methodology.

The TR methodology calculates tritium absorption into graphite using a bulk diffusivity as modeling assumption and neglects graphite heterogeneous features like pores and grains.

The NRC staff notes that the TR methodology does not describe the validation of the assumption on tritium diffusivity and solubility in Flibe and its effect on the calculations of tritium absorption into graphite. Therefore, the NRC staff imposes the following additional limitation and condition on the use of the TR:

Limitation and Condition 10 - In any future license application submittal that references this TR, an applicant needs to provide information to justify that the calculation of tritium absorption into graphite is not sensitive to the assumptions on tritium diffusivity and solubility in Flibe.

Section 8 of the TR, Limitation 6, states that quantification of the transport of tritium in nitrate salt and between the nitrate salt and the cover gas will be addressed in future submittals which implement the MST methodology. The NRC staff finds the tritium source term method to be acceptable given Limitation 3, which specifies that an applicant address validation concerns in any future license application submitted for NRC approval, and Limitation and Condition 10, which specifies that an applicant provide information to address the sensitivity of the assumptions regarding tritium diffusivity and solubility in Flibe.

Radionuclide Retention for Other Sources of MAR Section 6 of the TR provides a description of the modeling of radionuclide transport and retention from sources of MAR other than the fuel, coolant, and graphite reflector. The TR

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 18 identifies the potential sources of MAR and the phenomena associated with release from the radioactive waste systems, chemistry control system, intermediate heat transfer system, and the pebble handling and storage system (PHSS). The TR points out that the sources of MAR described in TR Section 6 are radionuclides that were transferred from a previous barrier (e.g.,

coolant) and do not generate additional MAR. Unique sources of MAR and release for the KP-FHR design (e.g., release from graphite dust) are described using information applicable to accident and normal operating conditions and degraded barriers. The TR states that any other phenomena that bear on dose criteria will be evaluated in subsequent licensing submittals.

Based on its assessment of the information provided, the NRC staff finds that the identification of other sources of MAR and the potential release and transport phenomena are comprehensive.

Section 6.3, Model Interfaces, of the TR states that the source term analysis for the identified other sources of MAR will use the same methodology as used of the sources of MAR in the fuel, coolant, and graphite reflector. The NRC staff finds this appropriate because the radionuclide transport phenomena are not unique to the location of the MAR. The NRC staff finds the discussion in TR Section 6, Other Sources of MAR, acceptable because the TR identifies relevant sources of MAR, release mechanisms, and transport phenomena based on the current state of knowledge of the KP-FHR design details, as being subject to the limitations in Section 8.2 of the TR.

Radionuclide Transport and Retention in Gas Space Section 7, Evaluating Radionuclide Transport in the Gas Space and Atmospheric Transport, of the TR provides a description of the modeling of radionuclide transport and retention in the gas space, which includes releases to the reactor building, aerosol characterization, transport in the reactor building, and use of the RADTRAD computer code for transport modeling. The methodology identifies phenomena relevant to radionuclide transport methodology, which was separated into two sets related to the condition of the reactor coolant boundary for the event.

Events with intact reactor coolant boundary included evaluation of ((

)), while events with a compromised reactor coolant boundary evaluate

((

)).

For the LBE evaluation, the methodology treats modeling of MAR transport and retention in the gas space for AOOs and DBEs differently based on whether the event is risk-significant or not.

In general, barrier transport and retention for non-risk-significant AOOs and DBEs are modeled in a more deterministic or conservative manner, similar to DBAs. Risk-significant AOOs and DBEs have more detailed mechanistic modeling. The NRC staff finds this is consistent with the guidance in RG 1.233 and is therefore acceptable.

The methodology states that the SNAP/RADTRAD code will be used to model the radionuclide transport and retention in the reactor building and releases to the environment. As discussed earlier in this safety evaluation, the SNAP/RADTRAD code was developed to model radionuclide transport and retention in compartments (e.g., reactor building); therefore, the NRC staff finds SNAP/RADTRAD is an appropriate tool for this type of assessment. Table 7-1 in the TR provides the radionuclide groups for transport in the gas space. This grouping is based on similar chemical and transport phenomena in the gas space. Although the grouping is a modification of the default grouping in SNAP/RADTRAD, the KP-FHR MST-specific grouping can be input by the user into the SNAP/RADTRAD code. The KP-FHR MST methodology only

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 19 models two processes for decontamination in the reactor building: gravitational settling of aerosol and radioactive decay. Therefore, the physical form of the radionuclides in the reactor building and environment is modeled in SNAP/RADTRAD as either a gas or an aerosol. The NRC staff finds the grouping and treatment of the physical form of radionuclides to be consistent with the state of practice for source term analysis, and will evaluate the specific use of the model (including computer code inputs) in subsequent license applications that reference the KP-FHR MST TR.

Aerosol Characterization Section 7.3, Characterizing Aerosols, of the TR describes the aerosol generation methodology for two types of event scenarios in the KP-FHR design basis: pipe breaks and PHSS transients.

Unlike in LWRs, the aerosols generated in the KP-FHR are not a consequence of core overheating and fuel melting, but instead are a result of coolant release and aerosolization of the salt. Sections 7.3.1, Material at Risk During a Pipe Break, and 7.3.2, Pipe Break Airborne Release Fraction Approach, of the TR describe the characterization of aerosols and airborne release fractions for pipe breaks. For events with a compromised reactor coolant boundary involving a pipe break, the MAR is the circulating activity in the molten salt within the pipe (either Flibe or nitrate salt depending on the event scenario). Section 7.3.1 of the TR describes the methodology to calculate an airborne release fraction for this MAR, considering the assumptions in TR Table 7-2. Airborne release fractions are based on mechanical aerosolization of the salt leaving the pipe or salt splashing on a surface, with the assumption all radionuclides dissolved in the salt aerosolize at an equivalent fraction. Any potential retention in insulation materials around the pipe is conservatively neglected. The assumed location of the postulated pipe break is that which produces the highest activity in the gas space. The NRC staff finds that these assumptions on retention in insulation and the location of the pipe break are conservative and acceptable.

Section 7.3.1.2 of the TR describes aerosol formation from jet breakup as the salt is released from the pipe by modeling the effects of gas entrainment in liquid jets based on theoretical correlations. The methodology derives a formula based on theoretical correlations ((

)) to calculate a representative diameter for the distribution of particles formed by the jet breakup and the airborne release fraction as a function of driving pressure. The NRC staff evaluated the information in TR Section 7.3.1.2 and finds that the assumptions are reasonably based on theoretical correlations to physical phenomena and the supporting references are relevant to the modeling of aerosol formation from molten salt by jet breakup.

Section 7.3.1.3 of the TR describes aerosolization of salt splashing on a surface occurring as a result of entrained air in the developing pool causing bubble burst aerosol formation. The TR model is based on correlations for continuous spills into already-formed pools, which are bounding for splashing in the developing pool. The TR notes that the correlations used show good agreement with experiments. The TR assumes a conservative bubble burst entrainment coefficient to represent Flibe with impurities to result in a salt splash aerosol airborne release fraction as a function of the volumetric flow rate of the spilled liquid. The NRC staff evaluated the information in TR Section 7.3.1.3 and finds that the assumptions are reasonably based on empirical correlations and the supporting references are relevant to the modeling of aerosol formation from molten salt by spilling or splashing onto a surface.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 20 Section 7.3.3, Material at Risk in the Pebble Handling and Storage System, and Section 7.3.4, Airborne Releases for Material at Risk in the Pebble Handling and Storage System, of the TR describe the aerosol and gaseous airborne release fractions for releases from the PHSS. The MAR in the PHSS includes the radionuclides in TRISO fuel, aerosols that are plated out on graphite dust or Flibe dust, and in the cover gas. Mechanisms for release include diffusion of MAR from fuel pebbles due to temperature transients in pebble handling or storage, mechanical damage to the fuel pebbles, or cover gas leakage events. The NRC staff finds that assuming equilibrium cycle PHSS loading to determine the MAR in the TRISO fuel is reasonable because this provides the likely conditions during operation and is appropriate for modeling the releases from fuel in PHSS events. The MAR associated with graphite dust in the PHSS will be determined ((

)) The estimation of MAR associated with graphite dust is based on the radionuclide loading and the graphite dust generation rate, which will be evaluated as inputs to the MST in a licensing application which uses the TR methodology. The TR describes that the MAR in Flibe dust formed from frozen Flibe is assumed to have the same mass normalized activity levels as the molten reactor coolant, and the airborne release fraction will be calculated taking into consideration bounding respirable particle diameters less than 10 micrometers (µm). Cover gas leakage events model the release of MAR in the cover gas and settled graphite dust to the reactor building ((

] Radionuclide release from mechanical damage to fuel pebbles is based on assuming the loss of the functional containment for the fuel, ((

)). The NRC staff finds that the modeling assumptions related to releases from the PHSS are based upon conservative assumptions or detailed calculations considering uncertainty, and therefore are acceptable.

Transport and Retention in Buildings Section 7.4, Building Transport Models, of the TR describes the modeling of radionuclide and retention in the buildings, including use of the SNAP/RADTRAD computer code. The TR methodology describes how the code will be used to model the gas space in the reactor building as a single volume. The specific reactor building volume used as input to the model is based on site-specific design information, which will be evaluated by the NRC staff in a licensing application that uses the TR methodology. The methodology provides for different levels of mechanistic detail or conservatism in the modeling of transport in the reactor building depending on whether it is modeling DBAs, non-risk-significant AOOs and DBEs, or risk-significant DBEs.

Radionuclides are sourced into the gas space, with the input based on the MAR transport and release models for the previous barriers of fuel, coolant, graphite, and other sources of MAR.

The KP-FHR reactor building is not assumed to be a safety-related structure, will not be leak-tight, and appropriately is not considered to be part of the functional containment. Although dependent upon the event evaluated, releases from the KP-FHR reactor coolant boundary, PHSS, or radwaste systems are not assumed to be at high pressure to force the release into the reactor building. Therefore, the NRC staff considers it reasonable to conclude that the release

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 21 can be modeled with a passive release rate, subject to Limitation 8 on the use of the TR.

Releases into the reactor building are located within the beryllium confinement, which is assumed to have a passive leakage rate equivalent to releasing all radionuclides in the building within a 2-hour window. The NRC staff finds the 2-hour release window to be acceptable because it is consistent with assumptions used in RG 1.183 for passive release from an open containment or other building for a LWR fuel handling accident, which also is not a high-pressure release. The TR methodology assumes that during this holdup time, the radionuclides within the reactor building gas space undergo radioactive decay and, depending on the event, aerosol natural deposition. The KP-FHR MST models activity reduction through radioactive decay only within the reactor building.

Section 7.5.2, Volumetric Flow Rates, of the TR describes the modeling of volumetric flow rates in the reactor building. The TR methodology accounts for the operation of heating, ventilation, and air conditioning systems, the effect of the temperature gradient between the beryllium confinement and the rest of the building, and wind loadings on the exterior of the building, depending on the event. Volumetric flow rates for DBA calculations will be prescribed to give a conservative result, while AOOs and DBEs are modeled differently. The NRC staff has evaluated the information in TR Section 7.5.2 and finds the modeling of volumetric flow rates for DBAs acceptable because it generally is consistent with guidance on DBA radiological consequence analyses and results in a bounding potential dose. The NRC staff finds the modeling of AOOs and DBEs to be acceptable because it is consistent with the RG 1.233 guidance on using realistic assumptions in evaluation of LBEs for comparison to the frequency-consequence target.

Aerosol Transport and Retention Section 7.4.2, Release Models, Section 7.5.3, Aerosol Formation Heights, and Section 7.5.4, Aerosol Particle Density, of the TR provide information on the transport and retention of aerosols in the building. The KP-FHR methodology uses the Henry correlation (Reference 29) implemented in the SNAP/RADTRAD code to model gravitational settling of aerosols. This model correlates aerosol removal rate to the aerosol density within the volume, with the removal rate proportional to the aerosol density and inversely proportional to the effective settling height.

It does not explicitly model gravitational settling rates differently for different particle sizes within the applicable range of particles. As described in Reference 29, the Henry correlation is based on aerosol removal experiments using sodium oxide aerosols and the correlation was compared to other reactor aerosol experiments that included a range of particle sizes. The Henry correlation takes user input on the settling height and theoretical particle density to adjust the experimentally determined aerosol settling rates for the specific use. Kairos Power describes the use of the Henry correlation as conservative when applied to Flibe, carbon, and nitrate aerosols and TR Section 7.5.4 provides the theoretical densities for the particles for input to adjust the correlation, as implemented in SNAP/RADTRAD.

The NRC staff finds implementation of the Henry correlation in SNAP/RADTRAD is conservative because RADTRAD only accounts for radioactive aerosols in determining the aerosol density within the volume, whereas the presence of non-radioactive aerosols would increase the aerosol density and subsequently increase the aerosol settling rate. The TR methodology only applies aerosol settling in the reactor building where temperatures assure that vapors have condensed to aerosols (i.e., aerosol deposition is not modeled in the cover gas region), and it is also conservative ((

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 22

)) Based on its assessment of Reference 29 and the information in TR Section-7.4.2, the NRC staff finds the correlation is acceptable for calculating gravitational settling for aerosol concentrations under dry conditions such as those expected within the KP-FHR buildings. The NRC staff notes that it found modeling of aerosol natural deposition acceptable within the safety-related primary containment for several LWRs. Although the KP-FHR, as described in TR Section 1.2.2, does not have a safety-related containment or a similar structure that is leak-tight and pressure retaining, the Henry correlation modeling of aerosol gravitational settling is applicable because it is dependent only on the settling height and aerosol density in the volume of the building. Because settling height and the aerosol density are not dependent on whether the structure is leak-tight and pressure retaining, the NRC staff finds that both can be determined by the applicant based on the physical dimensions of the final design. Therefore, based on the inputs to the Henry correlation, along with the conservative deterministic assumptions in the TR modeling of the reactor building for DBAs, the NRC staff finds the TR modeling of aerosol transport and retention in the reactor building to be acceptable.

In the TR, Section 7.5.3 states that the settling height input for the Henry correlation is determined by the aerosol formation height, with different modeling for LBEs as compared to DBAs. The NRC staff finds the determination of aerosol formation height for input to the Henry correlation to be consistent with the guidance in RG 1.233 for LBEs because the TR methodology to determine the aerosol formation height uses realistic assumptions with consideration of uncertainty. The NRC staff also finds that determination of aerosol formation heights for DBA calculations is consistent with the conservative evaluation of DBAs because it results in a conservative estimation of aerosol removal by the Henry correlation. Based on its experience with the SNAP/RADTRAD code, the NRC staff finds that SNAP/RADTRAD can estimate radioactive aerosol density for use in the Henry correlation to model aerosol removal within the reactor building. Because user input on the volume and release height is important to the calculation, the NRC staff will evaluate the use of the aerosol retention modeling when more design detail is available in a future licensing submittal that uses the TR.

Section 7.6, Building Transport Outputs, of the TR describes the dose analysis outputs for comparison to regulatory acceptance criteria for the DBAs or to the frequency-consequence target in NEI 18-04 for the LBEs. For DBAs, SNAP/RADTRAD calculates and reports TEDE at the EAB for the most limiting 2-hour period, which the NRC staff finds acceptable for comparison to the offsite dose criteria for siting and safety analyses required by the regulations cited above in the Regulatory Evaluation section of this safety evaluation. For DBAs, one TR methodology result is that the TEDE at the outer boundary of the LPZ will be reported for an exposure period after the start of the release of 30 days or for the duration of the passage of the plume, whichever is longer. In the TR, Section 7.6.2, Cumulative 30-Day Dose, describes how the user will adjust the SNAP/RADTRAD simulation time to ensure that if doses are still rising after 30 days, the simulation will be extended until there is no further effective increase in accumulated dose. The NRC staff finds this simulation time adjustment to enable the evaluation of effectively the entire exposure to the radiological release is acceptable because the adjustment is consistent with the requirements for evaluating the dose at the outer boundary of the LPZ for the duration of the passage of the plume, as given in the regulations cited above in the Regulatory Evaluation section of this safety evaluation. The NRC staff also finds the 30-day TEDE at the site boundary, which is reported for the LBE evaluation acceptable because the period is consistent with the metric in the frequency-consequence target in NEI 18-04 and endorsed by RG 1.233.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 23 Not all information to complete the consequence analysis, such as dose factors and specific design values related to the physical features of the plant, is described in the subject MST methodology TR. The NRC staff will review the radiological consequence analyses which use the MST methodology in future licensing submittals which reference the TR.

Atmospheric Dispersion Section 2.5.4, ARCON96, of the TR states the MST methodology uses the ARCON96 computer code to calculate offsite atmospheric dispersion values (also known as relative concentrations or /Q values) at the EAB and outer boundary of the LPZ rather than the PAVAN computer code. Both PAVAN and ARCON96 are NRC codes approved for calculating relative concentrations. The PAVAN code implements the guidance in RG 1.145, Atmospheric Dispersion Models for Potential Accident Consequence Assessments at Nuclear Power Plants, Revision 1, for determining offsite /Q values at the EAB and outer boundary of the LPZ, whereas ARCON96 implements the guidance in RG 1.194, Atmospheric Relative Concentrations for Control Room Radiological Habitability Assessments at Nuclear Power Plants, for determining onsite /Q values for the control room.

Large LWR nuclear power plants typically have EAB and LPZ distances that range from 800 to 6,000 meters. The ARCON96 computer code was developed to model shorter distances in the vicinity of buildings typical of control room habitability dose evaluations. The ARCON96 dispersion algorithms are based on field measurements taken out to distances of 1200 meters.

The TR states that, because the scope of the KP-FHR source term methodology is limited to the evaluation of EAB and LPZ dispersion distances less than 1,200 meters from the reactor building, Kairos Power considers the ARCON96 computer code methodology for calculating offsite atmospheric dispersion values appropriate for KP-FHR applications. Kairos implementation of the methodology is discussed below.

The NRC staff notes that the methodologies for calculating /Q values for the control room, technical support center, and routine releases are not covered in the scope of this TR. Those methodologies would need to be evaluated for applications using this MST methodology if applicable. Also, certain locations are affected by atmospheric transport and diffusion conditions that may be more restrictive than assumed in the contiguous 48 states, including effects caused by variations in the duration of daylight and darkness (e.g., limited inversion depths and extended persistence of various conditions). If the ARCON96 computer code methodology for calculating offsite atmospheric dispersion values is to be used in these locations (e.g., Alaska), the NRC staff notes that the applicability of the dispersion algorithms in ARCON96 may not apply or may require further modification.

Atmospheric Dispersion Models In the TR, Section 2.5.4 states that Kairos application of ARCON96 does not modify the source code. Section 7.7, Atmospheric Dispersion Models, of the TR states that the KP-FHR source term methodology includes the governing equations used by ARCON96. The TR states that when the equations are coupled with the methodology described in TR Sections 7.7 through 7.9, Dispersion Outputs, for selecting inputs and processing outputs, it provides the technical

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 24 basis and methodology for implementing ARCON96 to compute near-field EAB and LPZ /Q values for a KP-FHR design. The TR states that this methodology is consistent with the intent of RG 1.145 regarding directional dependence.

Section 7.7.1, Selecting the Conservative Release Distance, of the TR outlines the methodology for selecting the conservative release distance for the analysis. Computer code ARCON96 will calculate the directionally dependent /Q values from the side of building from which radioactive m aterial is being released to the EAB and LPZ. ((

))

Section 7.7.2, Calculation of the Time Averaged Percentile /Q Values, of the TR outlines the methodology for the calculation of the time averaged percentile /Q values. The output of the ARCON96 computer code model provides the time averaged directional 95th percentile /Q values. To calculate directionally independent 95th percentile values, Kairos Power will set the sector window to 360 degrees when the model is run. This will effectively include all the hours in the dataset and will therefore represent each wind direction, regardless of sector.

Kairos Power also outlined how it will calculate the 99.5th percentile directionally dependent /Q values. The ARCON96 computer code produces a complementary cumulative distribution function provided as one of its standard outputs. The ARCON96 computer code provides these complementary cumulative distribution functions in terms of the total number of 1-hour periods that would produce a /Q value greater than a given value. The TR states that Kairos Power plans to divide this frequency distribution by the total number of 1-hour inputs to produce the probability that the /Q value would be greater than a given value. The TR states that, from this set of complementary cumulative distribution functions, a /Q value greater than 99.5 percent of the data can be determined for each time interval calculated by ARCON96. Kairos Power also provides Equation 105 in the TR as a method to calculate a /Q value for a given time interval that does not start from zero. ((

]

Section 7.7.3, Selecting the Conservative Release Wind Direction, of the TR outlines how Kairos Power will select the conservative release wind direction used in the analysis. For each release location, Kairos Power will analyze all 16 wind directions consistent with the guidance outlined in RG 1.145. For DBAs and non-risk-significant AOOs and DBEs, Kairos Power will use the most conservative wind direction to determine dose at the receptor using the 99.5th percentile /Q value. For risk-significant AOOs and DBEs, all wind sectors are sampled uniformly.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 25 The NRC staff reviewed the methodologies for selecting the conservative release distance and wind direction used in the analysis and finds them to be consistent with NRC guidance. The NRC staff used ARCON96 and its outputs to verify the methodology used for identifying the directional 95 h percentile provided by ARCON96 and calculating the directionally independent 95th percentile /Q values using ARCON96. The NRC staff also performed an analysis of the methodology used to calculate the 99.5 h percentile directionally dependent /Q values and confirmed that the approach would produce results consistent with the guidance in RG 1.145, which outlines a methodology to select 99.5th percentile values.

Atmospheric Dispersion Inputs Section 7.8, Atmospheric Dispersion Inputs, of the TR, describes the ARCON96 computer code input parameters that will be used in the analysis. The release height will be set to zero meters and the applicant states that ground level releases are bounding for all release types in this methodology. The TR discusses how to identify the receptor data. The direction to source is the direction from the receptor to the source in degrees. The distance to the receptor is the distance to the EAB and LPZ and the release locations will be defined as the building walls closest to the EAB and LPZ. The TR states that the height to the intake (i.e., receptor) is the height of the intake above ground level at the EAB and LPZ. The intake height will be set to a ground level intake of 0.0 m for conservatism. The terrain elevation difference is the difference in elevation between the base of the reactor building and the EAB and LPZ. The TR states that the meteorological data will be consistent with the guidance of RG 1.23, Meteorological Monitoring Programs for Nuclear Power Plants, Revision 1 (Reference 30). The TR states that wind direction sector width for direction dependent calculations will be set to 45 degrees, consistent with guidance in RG 1.145. For direction independent calculations, the wind direction sector width will be set to 360 degrees. The minimum wind speed used for all calculations is 0.5 m/s. Surface roughness and averaging sector width constant will be set to default unless it is shown that it is more conservative to set a site-specific value. The TR also outlines the use of the initial horizontal and vertical diffusion coefficients.

The NRC staff reviewed the atmospheric dispersion inputs listed in the TR. The NRC staff finds the inputs acceptable for use because they are consistent with information outlined in RG 1.194.

Dispersion Outputs In Section 7.9, Dispersion Outputs, of the TR, Kairos Power again addressed the pre-calculated 95th percentile /Q values and complementary cumulative distribution functions produced by ARCON96. The TR notes that ARCON96 reports 95 h percentile /Q values for five different time intervals. The TR states that while ARCON96 does not directly report 99.5th percentile directionally dependent, time averaged /Q values, ARCON96 does report the cumulative distribution functions for directionally dependent /Q values. The methodology for using the cumulative distribution functions to calculate the 99.5th percentile /Q values is outlined in TR Section 7.7.2. The NRC staff discussed the methodology above and found it acceptable.

Example Figures and Calculations The ARCON96 computer code is distributed with a sample problem as an example to demonstrate its functions. Kairos Power used the outputs from this sample problem to

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 26 demonstrate its methodology for determining the 95th percentile and 99.5th percentile /Q values and plotted the results in Figures 7.9 and 7.10 of the TR. The methodology used is outlined in Section 7.7.2 of the TR. The NRC staff focused its confirmatory analysis of this methodology on the process and equations outlined in TR Section 7.7.2 as discussed by NRC staff above in this document.

In the example calculations presented in the appendices of the TR, Kairos Power used /Q values from an application previously approved by the NRC and input parameters that do not represent the atmospheric methodology set forth in the TR. Kairos Power is not seeking approval of these calculations and the NRC staff did not use the information in the appendices in its analysis of the proposed atmospheric dispersion methodology.

Conclusions for Atmospheric Dispersion The Kairos Power MST TR describes the applicants methods for using ARCON96 to calculate offsite atmospheric dispersion values for the EAB and outer boundary of the LPZ. Because the methodology differs from the NRCs guidance, the applicant outlined why the use of the ARCON96 computer code methodology is appropriate for its KP-FHR source term methodology and outlined how its use would be implemented. The applicant outlined the inputs that would be used for the model and explained how the outputs from the model would be used to meet the requirements in RG 1.145. The NRC staff reviewed the inputs and assumptions outlined in the TR. The NRC staff performed an independent analysis on the offsite atmospheric dispersion methodology and found that the approach would produce results consistent with the guidance in RG 1.145. For these reasons, the NRC staff finds the use of the ARCON96 computer code methodology for calculating offsite /Q values at the EAB and outer boundary of the LPZ acceptable for use in the KP-FHR mechanistic source term methodology subject to the limitations and conditions listed below in this safety evaluation.

LIMITATIONS AND CONDITIONS The NRC staffs review includes the limitations and conditions identified by the applicant in the TR and the staffs acceptance is partially based on the limitations and conditions as presented.

In addition, the NRC staff imposes three additional limitations and conditions on use of the TR (Limitations and Conditions 9, 10, and 11). An applicant may reference the TR for use as applied to the applicants facility only if the applicant demonstrates compliance with the following limitations and conditions:

1.

Approval of KP-Bison for use in fuel performance analysis as captured in KP-TR-010, KP-FHR Fuel Performance Methodology.

2.

Justification of thermodynamic data and associated vapor pressure correlations of representative species.

3.

Validation of tritium transport modeling methodology.

4.

Confirmation of minimal ingress of Flibe into pebble matrix carbon under normal and accident conditions, such that incremental damage to TRISO particles due to chemical interaction does not occur as captured in KP-TR-011, Fuel Qualification Methodology for the Kairos Power Fluoride Salt-Cooled High Temperature Reactor (KP-FHR).

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 27

5.

Establishment of operating limitations on maximum circulating activity and concentrations relative to solubility limits in the reactor coolant, intermediate coolant, cover gas, and radwaste systems that are consistent with the initial condition assumptions in the safety analysis report.

6.

Quantification of the transport of tritium in nitrate salt and between nitrate salt and the cover gas.

7.

The phenomena associated with radionuclide retention discussed in this topical report is restricted to molten Flibe. The retention of radionuclides in solid Flibe is beyond the scope of the current analysis.

8.

The methodology presented in this topical report (KP-TR-012) is based on design features of a KP-FHR provided in Section 1.2.2 and Section 2.3.4. Deviations from these design features will be justified by an applicant in safety analysis reports associated with license application submittals.

9.

The use of this TR is limited to the KP-FHR design and is not applicable to other molten salt reactor designs because the KP-FHR utilizes TRISO fuel, which is stated to retain most radionuclides. As described in the topical report and the related NRC staff safety evaluation, this allows for the use of certain simplifying assumptions related to retention of radionuclides in the molten salt.

10.

In any future license application submittal that references this TR, an applicant needs to provide information to justify that the calculation of tritium absorption into graphite is not sensitive to the assumptions on tritium diffusivity and solubility in Flibe.

11.

An applicant that uses this methodology must provide experimental data as described in Section 4.3.1.6 of this topical report.

CONCLUSION The NRC staff concludes that Kairos Powers topical report KP-TR-012, KP-FHR Mechanistic Source Term Methodology, Revision 3, provides an acceptable methodology for development of event-specific mechanistic source terms for use by Kairos Power Fluoride Salt Cooled High Temperature Reactor designs in offsite radiological consequence analyses for AOOs, design basis events, and design basis accidents based on: (1) the methodology being based on models of the chemical and physical phenomena that are supported by empirical data; (2) appropriate discussion of the consideration of model and design uncertainty; and (3) consistency with the guidance on performing radiological consequence analyses, subject to the limitations and conditions discussed above. Accordingly, the NRC staff concludes that K-TR-012, KP-FHR Mechanistic Source Term Methodology, Revision 3, can be used for development of MSTs for KP-FHR designs to support reactor licensing applications for permits, licenses, certifications, or approvals under 10 CFR Parts 50 or 52.

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 28 REFERENCES

1.

Nuclear Energy Institute, NEI 18-04, Revision 1, Risk-Informed Performance-Based Technology Inclusive Guidance for Non-Light Water Reactor Licensing Basis Development, August 2019 (ADAMS Accession No. ML19241A472)

2.

RG 1.233, Guidance for a Technology-Inclusive, Risk-Informed, and Performance-Based Methodology to Inform the Licensing Basis and Content of Applications for Licenses, Certifications, and Approvals for Non-Light Water Reactors, June 2020 (ADAMS Accession No. ML20091L698)

3.

NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR Edition, Section 15.0.3, Design Basis Accident Radiological Consequence Analyses for Advanced Light Water Reactors, March 2007 (ADAMS Accession No. ML070230012)

4.

Regulatory Guide (RG) 1.183, Alternative Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors, July 2000 (ADAMS Accession No. ML003716792)

5.

RG 1.145, Atmospheric Dispersion Models for Potential Accident Consequence Assessments at Nuclear Power Plants, Revision 1, February 1983 (ADAMS Accession No. ML003740205)

6.

RG 1.194, Atmospheric Relative Concentrations for Control Room Radiological Habitability Assessments at Nuclear Power Plants, June 2003 (ADAMS Accession No. ML031530505)

7.

SECY-16-0012, Accident Source Terms and Siting for Small Modular Reactors and Non-Light Water Reactors, February 7, 2016 (ADAMS Accession No. ML15309A319)

8.

SRM-SECY-93-092, SECY-93-0092 - Issues Pertaining to the Advanced Reactor (PRISM, MHTGR, and PIUS) and CANDU 3 Designs and Their Relationship to Current Regulatory Requirements, July 30, 1993 (ADAMS Accession No. ML003760774)

9.

SRM-SECY-18-0096, Staff Requirements - SECY-18-0096 - Functional Containment Performance Criteria for Non-Light-Water-Reactors, December 4, 2018 (ADAMS Accession No. ML18338A502)

10.

SECY-18-0096, Functional Containment Performance Criteria for Non-Light Water Reactors September 28, 2018 (ADAMS Accession No. ML18114A546)

11.

RG 1.232, Guidance for Developing Principal Design Criteria for Non-Light-Water Reactors, April 2018 (ADAMS Accession No. ML17325A611)

12.

Leppnen, J., et al. (2015) "The Serpent Monte Carlo code: Status, development and applications in 2013." Ann. Nucl. Energy, 82 (2015) 142-150.

13.

R. L. Williamson, J. D. Hales, S. R. Novascone, M. R. Tonks, D. R. Gaston, C. J.

Permann, D. Andrs, and R. C. Martineau. Multidimensional multiphysics simulation of nuclear fuel behavior. Journal of Nuclear Materials, 423:149-163, 2012.

14.

Kairos Power, KP-TR-010, KP-FHR Fuel Performance Methodology, Revision 3, June 2021 (ADAMS Accession No. ML21162A351)

15.

ANL/NE-17/4, SAM Theory Manual, Argonne National Laboratory, March 2017

16.

W. Arcieri, D. Mlynarczyk, and L. Larsen, SNAP/RADTRAD 4.0: Description of Models and Methods, U.S. NRC, NUREG/CR-7220, June 2016. (ADAMS Accession No. ML16160A019)

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY PROPRIETARY INFORMATION 29

17.

Atmospheric Relative Concentrations in Building Wakes, Pacific Northwest National Laboratory, NUREG/CR-6331, Revision 1, May 1997 (ADAMS Accession No. ML17213A190)

18.

PAVAN: An Atmospheric-Dispersion Program for Evaluating Design-Basis Accidental Releases of Radioactive Materials from Nuclear Power Stations, NUREG/CR-2858, PNL-4413, November 1982 (ADAMS Accession No. ML12045A149)

19.

Kairos Power, KP-TR-011, Fuel Qualification Methodology for the Kairos Power Fluoride Salt-Cooled High Temperature Reactor (KP-FHR), Revision 0 (ADAMS Accession No. ML20182A830)

20.

Compere, E.L., et. al., ORNL-4865, Fission Product Behavior in the Molten Salt Reactor Experiment, Oak Ridge National Laboratory, October 1975.

21.

Kedl, R.J., ORNL-TM-3884, The Migration of a Class of Fission Products (Noble Metals) in the Molten-Salt Reactor Experiment, Oak Ridge National Laboratory, December 1972.

22.

Jerden, J., Grabaskas, D., Bucknor, M., ANL/CFCT-19/4, Development of a Mechanistic Source Term Approach for Liquid-Fueled Molten Salt Reactors, Argonne National Laboratory, July 2019.

23.

Kalilainen, J., Nichenko, S., Krepel, J., Evaporation of materials from the molten salt reactor fuel under elevated temperatures, Journal of Nuclear Materials, Volume 533, May 2020

24.

TRISO particles: The Most Robust Nuclear Fuel on Earth, Office of Nuclear Energy, July 9, 2019. https://www.energy.gov/ne/articles/triso-particles-most-robust-nuclear-fuel-earth

25.

D. Grabaskas, T. Fei, J. Jerden, ANL/NSE-20-15, Technical Letter Report on the Assessment of Tritium Detection and Control in Molten Salt Reactors: Final Report, May 2020. (ADAMS Accession No. ML20157A155)

26.

M. Young, Wu Huali, and Scarlat Raluca, Characterization of Tritium Transport in the Flibe-Graphite System, for In-Situ Absorption by the Fuel Elements of the FHR, NURETH-16, August 30-September 4, 2015.

27.

K. Dolan, G. Zheng, D. Carpenter, S. Hwang and L. Hu, Tritium content and chemical form in nuclear graphite from molten fluoride salt irradiations, Fusion Science and Technology, February 21, 2020

28.

A. Suzuki, T. Terai and S. Tanaka, Tritium release behavior from Li2BeF4 molten salt by permeation through structural materials, Fusion Engineering and Design, Vols.

51-52, pp. 863-868, November 2000.

29.

Fauske & Associates, Inc., FAI Aerosol Correlation, An Industry Degraded Core Rulemaking (IDCOR) Program Report, 1984.

30.

RG 1.23, Meteorological Monitoring Programs for Nuclear Power Plants, Revision 1, March 2007 (ADAMS Accession No. ML070350028)

Principal Contributor(s): Michelle Hart Alexander Chereskin Imtiaz Madni Jason White Date: April 22, 2022