ML22101A223

From kanterella
Jump to navigation Jump to search
Final SE Kairos Fuel Performance Methodology Topical January 2022 - Redacted
ML22101A223
Person / Time
Site: 99902069
Issue date: 03/31/2022
From:
NRC/NRR/DANU/UARL
To: Hastings P
Kairos Power
Beasley B
Shared Package
ML21354A676 List:
References
doc log # 4942
Download: ML22101A223 (12)
v • d • e


Text

OFFICIAL USE ONLY - PROPRIETARY INFORMATION Enclosure OFFICIAL USE ONLY - PROPRIETARY INFORMATION KAIROS POWER LLC - FINAL SAFETY EVALUATION OF TOPICAL REPORT KP-FHR FUEL PERFORMANCE METHODOLOGY (REVISION 3) (EPID NO: L-2019-TOP-0056)

SPONSOR INFORMATION Sponsor: Kairos Power LLC Sponsor Address: 707 Tower Ave.

Alameda, CA 94501 Project No: 99902069 APPLICATION INFORMATION Submittal Date: June 10, 2021 Submittal Agencywide Documents Access and Management System (ADAMS)

Accession Number: ML21162A349.

Brief Description of the Topical Report: The topical report (TR), KP-FHR Fuel Performance Methodology (Revision 3), submitted by Kairos Power LLC (Kairos or the vendor) describes the methodology for analyzing a tristructural isotopic (TRISO)-coated fuel particle with a Uranium Oxycarbide (UCO) fuel kernel in a non-power or power Kairos Power Fluoride-Salt Cooled High Temperature Reactor (KP-FHR). The fuel performance methodology includes the manufacturing characteristics, material properties, physical models, and uncertainty analysis associated with modeling. The figures of merit (FOMs) are in-service failed fuel fraction and the fission product release from intact particles. Fission product releases from failed and intact particles, particles with manufactured defects, and dispersed uranium are considered in the analysis. The methodology uses the KP-BISON code, which is a modification of the Idaho National Laboratory BISON code, and is intended to cover normal operations, anticipated operational occurrences (AOOs) and design bases accidents (DBAs). The output from the fuel performance analysis serves as one of the inputs to the mechanistic source term analysis for either the non-power reactor or KP-FHR power reactor.

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

Implementation: An applicant who references a topical report in a licensing application must demonstrate that the application of the topical report to their specific facility is within the scope of the conditions in the topical report defining its application. The NRC staff verifies relevant criteria for accepted-for-use topical reports during each licensing action to ensure that the topical report's conclusions are both valid and applicable to the particular licensing action under review.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 2

OFFICIAL USE ONLY - PROPRIETARY INFORMATION Accordingly, upon implementation of this TR into a site-specific application, the staff will evaluate each topical area designated below to ensure that each topic appropriately interfaces with the proposed license application to ensure consistency. The staff will also make its regulatory determinations regarding the topics discussed below, as applicable, during its review of any future license application REGULATORY EVALUATION Regulatory Basis: For construction permit applications, Title 10 of the Code of Federal Regulations (10 CFR) Part 50, Domestic Licensing of Production and Utilization Facilities, Sections 50.34(a)(1)(ii)(D)(1) and (2) and 10 CFR 50.34(a)(4) apply because the TRISO particle is the primary means of fission product retention. For operating license applications, 10 CFR 50.34(b)(4) applies. Similar regulatory requirements exist for design certification applications, combined license applications, and standard design approvals in 10 CFR Part 52, Licenses, Certifications, and Approvals for Nuclear Power Plants (10 CFR 52.47(a)(2)(iv)(A) and (B), 10 CFR 52.47(a)(4), 10 CFR 52.79(a)(1)(vi)(A) and (B), 10 CFR 52.79(a)(5), 10 CFR 52.137(a)(2)(iv)(A) and (B), and 10 CFR 52.137(a)(4)). These regulations require an applicant to evaluate the off-site dose resulting from an accident considering the facility design features and barriers that must be breached before a release of radioactive material from the facility can occur.

In addition, 10 CFR 50.34(a)(3)(i) requires, in part, that an applicant for a construction permit to build a reactor provide principal design criteria (PDC) for the facility. Similar regulatory requirements exist in 10 CFR 52.47(a)(3)(i), 10 CFR 52.79(a)(4)(i), and 10 CFR 52.137(a)(3)(i) for design certification applications, combined license applications, and standard design approvals, respectively. The PDC establish criteria for structures, systems, and components that are important to safety.

The Nuclear Regulatory Commission (NRC) staff approved the PDC for the KP-FHR in topical report KP-TR-003 (ADAMS Accession No. ML20167A174) on May 22, 2020. KP-FHR PDC 10, Reactor design states that the reactor core and associated heat removal, control, and protection systems shall be designed with appropriate margin to ensure that specified acceptable system radionuclide release design limits are not exceeded during any condition of normal operation, including the effects of anticipated operational occurrences. Establishing fuel design limits and ensuring these limits are not exceeded represent a fundamental underpinning of the safety assessment of a nuclear power plant required by 10 CFR 50.34(a)(1).

Part 100, Reactor Site Criteria, of 10 CFR establishes approval requirements for proposed sites for stationary power and testing reactors subject to 10 CFR Part 50 or Part 52. In particular, 10 CFR 100.11, Determination of exclusion area, low population zone, and population center distance, requires an applicant to assume a fission product release for the determination of an exclusion area, a low population zone, and a population center distance.

The vendor plans to credit the retention of the TRISO particle to inform its dose analysis, and this TR will, therefore, play a role in the staff review of the siting criteria.

TECHNICAL EVALUATION In the following safety evaluation, a section, table, or figure number without additional description refers to the vendors topical report. If section, table, or figure numbers are

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 3

OFFICIAL USE ONLY - PROPRIETARY INFORMATION referenced from other reports or sections for this safety evaluation, additional descriptive information will be provided.

Portions of the Topical Report Not Addressed by This Safety Evaluation The focus of this safety evaluation is on the calculational framework and associated uncertainty analysis for the two FOMs. As such, this safety evaluation does not make findings associated with the following portions of the topical report:

1. Section 2.3, Phenomena Identification and Ranking Tables.
2. Section 2.4, Fission Product Transport and all the Section 2.4 subsections.
3. Section 3, Fuel Modeling - Material Properties and Physical Models, and all Section 3 subsections.
4. Sections 4.1.1; Verification; 4.1.2, Validation (and all 4.1.2 subsections); and 4.1.5, Validation, Verification and Uncertainty Quantification Results.
5. Section 5, KP-BISON Code, and all Section 5 subsections.
6. Topical report Table 3-8, Diffusion coefficients for Key Fission Products Modeled in KP-BISON.
7. Section 6.4.2, TRISO and pebble models.
8. Section 6.3, Fission Product Release, as it pertains to fuel pebble mechanical and chemical interactions with the salt environment, and possible wear which are stated by the vendor to be outside the scope of this tropical report.

The staff will make findings on the above sections in a subsequent revision to the TR after code validation and uncertainty quantification has been performed. Item 8 is outside the scope of the KP-FHR Fuel Performance TR and will be addressed in a separate TR covering fuel qualification. Collectively, the exclusion of the eight numbered items constitutes Limitation and Condition 1.

Methodology Overview and Review of Calculational Framework The staffs review is limited to the calculational framework and the associated uncertainty analysis used to determine the upper tolerance limit associated with intact and failed fuel particle release when using an approved fuel performance code. Using a fuel performance code (e.g., KP-BISON) to calculate the in-service failure fraction and intact release fraction is referred to as a coupled model. As described in TR Section 6, means other than a coupled model (i.e., other than a deterministic failure model) can be used to determine in-service fuel failures such as using applicable post irradiation examination (PIE) data. While the staff agrees that other methods to determine in-service fuel failure could be conceptually sound, the details regarding how these would be implemented have not been provided and the staff makes no findings regarding the use of other in-service failure determinations. This constitutes Limitation and Condition 2: use of this TR in the submitted form requires use of an approved fuel performance code to determine the in-service failed fuel fraction, the resulting failed and intact particle fission product release, and the manufactured defect fraction fission product release.

Manufacturing defects are another source of fission-product release and these are addressed in the methodology by either a specification or characterization of as-manufactured fuel. Table 2-3 provides the advanced gas reactor (AGR) 5/6/7 manufacturing defect by type (e.g., Silicon Carbide (SiC) coating defect fraction) which the vendor states is representative of their expected defect fractions. The staff agrees the calculational framework is sufficient to determine the associated fission product release of manufacturing defects using either a specification or

OFFICIAL USE ONLY-PROPRIETARY INFORMATION characterization as both are direct code inputs based on previous experience or measurement of the as-built fuel. In the context of this topical report, this approach is acceptable because particle batches which do not meet the specification will be rejected. To determine the fission product release for manufacturing defects, an approved fuel performance code is necessary as kernel retention at a minimum is assumed. The use of an approved fuel performance code is addressed by Limitation and Condition 2.

a e - prov, es e

spec, Ica,on or Isperse uranium which the ven or sta es is representative for their particles. The staff agrees the calculational framework is sufficient to determine the associated fission product release of dispersed uranium using either a specification or characterization as both are code inputs. These inputs are based on either measurements of the as-manufactured particle or the particles that are determined to meet a provided specification. To determine any mitigation of potential fission roduct release from dis ersed uranium, an approved fuel performance code is necessary ((

)) The use of an approved fuel performance co 10n 2.

TRISO Failure Mechanisms To calculate the failure fraction and resulting release, the calculation framework must define the failure mechanisms and how each coating's fission product retention characteristics are modified. Table 2-8, 'TRISO Fuel Failure Mechanism Modeled in the Fuel Performance Analysis Model," provides a list of UCO TRISO failure modes and a rationale for considering the failure mode in the calculational framework. The NRC staff agrees that the provided list of failure mechanisms is complete and concurs with the provided rationale based on literature reviews and historical experience examining irradiated TRISO particles. Table 2-9 provides the vendor's determination of the failure modes which are not considered in the failed fraction calculational framework. The failure mechanisms not considered are based on a combination of observed UCO fuel kernel behaviors (e.g., carbon monoxide chemical attack of the SiC and kernel migration) and expected operating conditions as given, in part, by Table 2-6, "KP-FHR Expected Normal Operating Conditions." Based on review of the staff approved Electric Power Research Institute (EPRI) topical report, "Uranium Oxycarbide (UCO) Tristructural Isotropic (TRISO) Coated Particle Fuel Performance," (EPRI-AR-1-A) (ADAMS Accession No. ML20336A052) the staff agrees with the vendor's failure modes which have been eliminated from the methodology if the operating conditions for the proposed non-power or power KP-FHR stay within the range given by Figure 4-6 in EPRI-AR-1-A. For conditions which lie outside the operating range of EPRI-AR-1-A, the failure modes in Sections 2.2.2 and 6.2 must be reevaluated and topical report updated as necessary. This restriction is reflected in Limitation and Condition 3.

In the Section 6.2, "Particle Failure," of the TR, the vendor lists the resultin mechanisms that are considered:

OFFICIAL USE ONLY-PROPRIETARY INFORMATION

OFFICIAL USE ONLY-PROPRIETARY INFORMATION for the in-service failure fraction calculation for the ((

failure modes and the associated modifications to the coating layer diffusivities. The [

)) failure mode compares the calculated maximum stress of the SiC coating layer o e rac ure strength of the coatin la er as sam led from its Weibull distribution as a function of time fuel du rmmation of the a equac dressed by the performance code validation as re ec ed in Limitation and Condition 2.

The failed fuel fraction calculational scheme in Fi 5

OFFICIAL USE ONLY-PROPRIETARY INFORMATION

OFFICIAL USE ONLY-PROPRIETARY INFORMATION Uncertainty Parameters An important aspect in determining a conservative fission product release from particles with compromised coatings layers is the handling of manufacturing properties, physical models, material properties, and operational (irradiation conditions) uncertainties. Topical report Section 4.1.3 provides the methodology used for uncertainty quantification which propagates individual uncertainties to yield an upper tolerance limit for the particle failure probability. This methodology is evaluated in the following safety evaluation subsections.

Particle Manufacturing Uncertainties Particle manufacturing uncertainties are specified in Table 4-1. The range of nominal values (Mean values) in Table 4-1 is consistent with values based on Reference 42 of EPRI-AR-1-A.

The u er limit on the standard deviation is based either on the d

the range o ould yield selected fuel properties outside the range of the AGR 1 and 2 95/98 tolerance limit extrema values given in RI-AR-1-A, Table 5

  • n va I atIon o e aI ure mec anisms eyon e expenmen a range o e GR 1 an

, or the AGR 5/6/7 data, if available, is uncertain. Therefore, while the staff agrees with the fuel manufacturing range of nominal values based on EPRI-AR-1-A, Table 5-5, and that the sampling variation range covers a sufficiently high percentage of the TRISO particle population, additional justification is needed to demonstrate that the fuel performance code either reasonably predicts particle behavior, or the methodology assumes particle failure, for particle properties sampled beyond the code validation range (e.g., AGR 1 and 2 95/98 tolerance limit extrema values). The code validation range relative to the Table 4-1 sampling range is addressed by Limitation and Condition 5. This condition is only relevant for particles at the extreme tails of the distribution. While it may be possible to justify conservative failure fractions when sampling outside the code validation range, the fuel specification or the as-manufactured particles must stay within the code validation range (or failure should be assumed for particles at the likely failure extremes).

Therefore, dings reg eliminated from the methodology, as there is insufficient validation basis for doing so. This is reflected in Limitation and Condition 6.

6 OFFICIAL USE ONLY-PROPRIETARY INFORMATION

OFFICIAL USE ONLY-PROPRIETARY INFORMATION Uncertainty Parameters - Material Properties and Physical Models Other sources of uncertainty are associated with material properties and physical models as given in Tables 4-2 and 4-3. The material properties and physical models are either modeled by physical laws or by correlations which are a function of input conditions such as fuel properties or operating conditions. Often how a property changes with operation conditions or fuel duty is represented by a correlation(s) based on experimental data. Typically, these correlations are best fits of the data and hence do not re resent an fit uncertaint. To ca ture this uncertainty, the vendor defines IS conserva Ive as ong as a c ec or nonconserva Ive p ysIca e avIor Is pe ormed. A review of the sensitivity analysis results for cases outside the established model or property ranges should be performed to identify any unexpected nonconservative behavior. The staff agrees that a sensitivity analysis can be used to determine which items in Table 4-2 significantly affect the failure probability and intact release but makes no finding on the sensitivity methodology, including what constitutes a significant impact. The final items in Tables 4-2 and 4-3 which will be included in the uncertainty analysis, and the sensitivity methodology used to determine the significant parameters, is addressed by Limitation and Condition 6.

Uncertainty Parameters -Irradiation (Operating) Conditions The final source of uncertainty is the irradiation conditions or operating conditions the TRISO particles may experience during their lifetime. The expected normal operating conditions are a function of pebble trajectories, allowed operating bands, and the code uncertainties used to predict these conditions. As described in Section 4.1.3.1.3, "Irradiation Conditions," the vendor has chosen to a I OFFICIAL USE ONLY-PROPRIETARY INFORMATION

assumption.

OFFICIAL USE ONLY-PROPRIETARY INFORMATION a,on an ndition 8, regar uence Upper Tolerance In-service Failed Particle Fraction Methodology With the failure mechanisms and uncertainty parameters defined, a statistical method to determine the overall uncertainty in failure fraction and fission product release must be constructed. In Section 4.1.3.2, "Uncertain Calculational Scheme," the vendor defines the method chosen for use in this TR OFFICIAL USE ONLY-PROPRIETARY INFORMATION

OFFICIAL USE ONLY-PROPRIETARY INFORMATION Because the expected failure probability of particles is low, selecting a on ses dology to ensure e

oes not dominate the calculated failure probability.

Upper Tolerance Fission Product Release Methodology The in-service fission product release fraction follows a similar process to the failed fraction described above but the contribution of intact articles is included. An 9

OFFICIAL USE ONLY-PROPRIETARY INFORMATION

OFFICIAL USE ONLY-PROPRIETARY INFORMATION v

een provI e.

subsequent revision of the topica a

The uncertainty method used to determine the overall in-service fission product release is acceptable, as it accounts for the failed particle probability discussion in the previous section of this safety evaluation related to uncertainty parameters. The staff also agrees the weighting method of the ((--)) is appropriate to determine the in-service release from intact and failed coatinQs as it accounts for the predicted distribution of the particle states. The overall methodology-calculated fission product release is the in-service upper tolerance limit release combined with conservative inputs for manufacturing defects and dispersed uranium fractions. These fractions are finally applied to the total core TRISO particle inventory to determine a conservative prediction of the total reactor fuel fission product release.

LIMITATIONS AND CONDITIONS An applicant for a KP-FHR design may reference the TR for use as applied to the applicant's facility subject to the following limitations and conditions:

1.

This safety evaluation does not approve the portions of the TR described in the section of this safety evaluation titled, "Portions of the Topical Report Not Addressed by This Safety Evaluation."

2.

An applicant referencing this TR must use an NRC staff-approved fuel performance code to determine the in-service failed fuel fraction, the resulting failed and intact particle fission product release, manufactured defect fraction fission product release, and other sources of fission product release (e.g., dispersed uranium).

3. The approval of this TR is limited to operation of the non-power or power KP-FHR reactors, within the bounds supported by the AGR program, as reflected in Figure 4-6 in EPRI-AR-1-A.
4. An applicant referencing this TR must provide detailed methodologies to account for SiC layer stress due to aspherical particles and a cracked IPyC layer that are reviewed and approved by the staff as part of a future submittal. The applicant proposed methodologies in topical report Section 6.2, "Particle Failure," but sufficient details were not provided for the staff to make a reasonable assurance finding on the proposed methods.

10 OFFICIAL USE ONLY-PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 11 OFFICIAL USE ONLY - PROPRIETARY INFORMATION

5. An applicant referencing this TR must provide acceptable justification, approved by the NRC staff, that sampling of fuel properties outside those used for the code validation range predicts conservative failure fractions and release rates. For particles that are sampled outside the code validation range, the assumption of complete coating failure (i.e., a TRISO failure) is acceptable.
6. The MC fuel properties and parameters sampled must continue to include all the parameters in Tables 4-1, 4-2, and 4-3, unless otherwise justified and approved by the NRC staff.
7. The calculation framework may not be used to evaluate fuel performance during anticipated AOOs, DBAs, and BDBEs. In the present form of the methodology presented in the TR, justification for the temperature and irradiation conditions is provided only for quasi steady-state conditions, and additional information would need to be provided for the NRC staff to approve use of the methodology for AOOs, DBAs, and BDBEs.
8. The applicants assumption regarding the accumulation of burnup and fast fluence that yields conservative failure fractions and fission product release requires further justification by the vendor and approval by the NRC staff. An applicant referencing this TR must justify that the actual limiting fuel performance parameters (such as an expected limiting particle/pebble history in the reactor) are bounded by the burnup and fluence profile used in this TR.
9. The number of fuel property parameters must be large enough that the standard error in equation 83 is approximately an order of magnitude or less than the failure probability, p, also in equation 83.
10. An applicant referencing this TR must provide the sampling range associated with the physical models and material properties that is acceptable to the NRC staff.
11. An applicant referencing this TR must provide a means of addressing the diffusivity uncertainties in Table 3-8 of the topical report that is acceptable to the NRC staff.

CONCLUSION Subject to the limitations and conditions stated above, the NRC staff has determined that the KP-FHR Fuel Performance Methodology provides an acceptable methodology for determining a conservative UCO TRISO particle fission product release from in-service failed and intact particles, manufacturing defects, and dispersed uranium. This conclusion considers that in-service failures and releases are derived from the vendors coupled methodology based on:

(1) incorporation of the relevant UCO TRISO failure mechanisms; (2) incorporation of the relevant uncertainties which affect the in-service failure fractions and the overall particle releases; (3) the methodology used to determine the sampled individual uncertainties over a range of variation; and (4) a statistical framework which combines individual uncertainties into conservative upper tolerance limits for the failed fuel fractions and overall fuel fission product release, subject to the limitations and conditions discussed above. Accordingly, the NRC staff concludes that KP-FHR Fuel Performance Methodology can be used as a calculational framework to determine UCO TRISO particle fission product release in designs referencing this TR and can be used to support a demonstration of compliance with NRC regulations in 10 CFR 50.34(a)(1)(ii)(D)(1) and (2) and 10 CFR 50.34(a)(4) for construction permit applications, 10 CFR 50.34(b)(4) for operating license applications, 10 CFR 52.47(a)(2)(iv)(A) and (B) and 10 CFR 52.47(a)(4) for design certification applications, 10 CFR 52.79(a)(1)(vi)(A) and (B) and 10 CFR 52.79(a)(5) for combined license applications, 10 CFR 52.137(a)(2)(iv)(A) and (B) and 10 CFR 52.137(a)(4)) for standard design approvals, and 10 CFR 100.11 for all applications.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 12 OFFICIAL USE ONLY - PROPRIETARY INFORMATION REFERENCES 1 Owen, D., 1958. Tables of Factors for One-Sided Tolerance Limits for a Normal Distribution.

Sandia Corporation Systems Analysis Department Statistical Division, p.8.

Principal Contributor(s): Jeffrey Schmidt, NRR Antonio Barrett, NRR Boyce Travis, NRR Date: March 31, 2022

Retrieved from "https://kanterella.com/w/index.php?title=ML22101A223&oldid=5530323"