NUREG-1829, White Paper on Continued Applicability of NUREG-1829

White Paper on Continued Applicability of NUREG-1829

ML24205A015
Person / Time
Issue date:	11/13/2024
From:	David Rudland NRC/NRR/DNRL
To:	Michele Sampson NRC/NRR/DNRL
References
Download: ML24205A015 (1)

Text

MEMORANDUM TO:

Michele Sampson, Director Division of New and Renewed Licenses Office of Nuclear Reactor Regulation FROM:

David Rudland, Senior Technical Advisor Division of New and Renewed Licenses Office of Nuclear Reactor Regulation

SUBJECT:

WHITE PAPER ON CONTINUED APPLICABILITY OF NUREG-1829 The U.S. Nuclear Regulatory Commission (NRC) is considering amending its regulations (through the increased enrichment (IE) rulemaking) and updating guidance to facilitate the use of light-water reactor fuel containing uranium enriched to greater than 5.0 weight percent uranium-235. After receiving public comment on its rulemaking plan (Agencywide Documents Access and Management System Accession No. ML23032A504), the staff recommended Alternative 2, Rulemaking to Recategorize Large-Break Loss-of-Coolant Accidents as Beyond-Design-Basis Accidents. This alternative considers the transition break size (TBS) developed as part of a voluntary, risk-informed revision proposed for the emergency core cooling system requirements in Title 10 of the Code of Federal Regulations (10 CFR) 50.46a, Risk-informed changes to loss-of-coolant accident technical requirements, that was developed in the early 2000s and would have replaced 10 CFR 50.46a Acceptance criteria for reactor coolant system venting systems. The staff ultimately withdrew this proposed rulemaking in 2011. The TBS concept and the related IE rulemaking are based on the rationale that the probability of large-break loss-of-coolant accidents (LOCAs) associated with passive component failures larger than the TBS is extremely low and can be categorized as beyond design basis.

However, the NRC did not complete the original 10 CFR 50.46a rulemaking or otherwise implement the TBS concept in the regulations.

CONTACTS: David Rudland (NRR),

David Dijamco (NRR),

Seung Min (NRR),

301-415-1896 301-415-1502 301-415-2045 Eric Palmer (NRR),

Rob Tregoning (RES),

Matt Homiack (RES),

Eric.Palmer@nrc.gov 301-415-2324 301-415-2427 Chris Nellis (RES),

301-415-5973 November 13, 2024 Signed by Rudland, on 11/13/24

M. Sampson The original technical basis supporting the TBS concept comes from evaluating the impacts of degradation on piping and nonpiping passive components using data on operating experience up to approximately 2004, coupled with additional analysis. To incorporate the TBS methodology into the new proposed IE rulemaking, the staff needs to determine whether the TBS is still applicable, considering the operating history since the early 2000s. Therefore, the Office of Nuclear Reactor Regulation (NRR) Division of New and Renewed Licenses staff was tasked to determine whether this basis is still adequate.

The selection of the TBS was largely predicated on the results in NUREG-1829, Estimating Loss-of-Coolant Accident (LOCA) Frequencies Through the Elicitation Process, issued April 2008 (ML080630015), and NUREG-1903, Seismic Considerations For the Transition Break Size, issued February 2008 (ML080880140). The work discussed here focuses on the continued adequacy of NUREG-1829. The NRR Division of Engineering and External Hazards has conducted a similar assessment of NUREG-1903.

The staff undertook this effort by doing the following:

considering the impact of recent operational experience and the relevance to the TBS conducting a detailed probabilistic fracture mechanics study to verify the NUREG-1829 LOCA frequencies performing an internal and external elicitation to identify possible scenarios not considered, or underestimated, in NUREG-1829 that could result in primary pressure boundary breaches that are larger than the proposed TBS in both pressurized-and boiling-water reactors completing a database study to estimate whether worldwide operational experience would modify the LOCA frequencies The results of these analyses suggest that the probability of direct and indirect failures associated with the events considered does not challenge the proposed TBS. However, the staff identified a concern with the ongoing industry effort to optimize inspection requirements for components within the reactor coolant loop. In the original NUREG-1829 elicitation, the participants estimated the LOCA frequencies assuming that the typical component inspections would continue throughout the reactor life. The impact of the reduction in the number of future component inspections on the overall LOCA frequencies is unknown; therefore, the staff concluded that the elimination of inspections in the piping with diameters greater than the TBS would impact the NUREG-1829 conclusions. Therefore, the staff recommends that a risk-informed sample of 10 percent of these welds be inspected with qualified personnel and procedures before implementing this rule and in every subsequent inspection interval.

(Repeatedly inspecting the sample welds each interval is acceptable.) The attached white paper contains the details of the analyses, results, and recommendation.

Enclosure:

White Paper on the Continued Applicability of NUREG-1829

ML24205A015 E-Concurrence NRR-106 OFFICE NRR/DNRL QTE NRR/DNRL NAME DRudland JDougherty MSampson DATE 08/30/2024 09/19/2024 10/23/2024 OFFICE RES/DE NRR/DNRL NAME CAraguas DRudland DATE 11/13/2024

1 White Paper on the Continued Applicability of NUREG-1829 1.

Introduction The U.S. Nuclear Regulatory Commission (NRC) is considering amending its regulations (through the increased enrichment (IE) rulemaking) and updating guidance to facilitate the use of light-water reactor fuel containing uranium enriched to greater than 5.0 weight percent uranium-235. In SRM-SECY-21-0109, Staff RequirementsSECY-21-0109Rulemaking Plan on Use of Increased Enrichment of Conventional and Accident Tolerant Fuel Designs for Light-Water Reactors, dated March 16, 2022 (Agencywide Documents Access and Management System Accession No. ML22075A103), the Commission directed the staff to address and analyze fuel fragmentation, relocation, and dispersal issues relevant to fuels of higher enrichment and burnup levels. Fuel fragmentation and relocation are necessary precursors to ductile-failure-induced fuel dispersal and so are often discussed together.

However, from a regulatory perspective, dispersal poses a challenge that the other two phenomena do not, in that fuel fragmentation and relocation into the ballooned region of the fuel rod can be addressed within the existing regulatory framework. However, the staff believes that action may be needed to address and analyze fuel dispersal. As part of its regulatory basis document for this effort (ML23032A504), the staff proposed five alternatives to handle dispersal and requested stakeholder comment. After reviewing the public comments, the staff recommended Alternative 2, Rulemaking to Recategorize Large-Break Loss-of-Coolant Accidents as Beyond-Design-Basis Accidents. This alternative considers the transition break size (TBS) developed as part of a proposed voluntary, risk-informed revision of the emergency core cooling system requirements in Title 10 of the Code of Federal Regulations (10 CFR) 50.46a, Risk-informed changes to loss-of-coolant accident technical requirements, developed in the early 2000s. The TBS concept and the related IE rulemaking are based on the rationale that the probability of large-break loss-of-coolant accidents (LOCAs) associated with passive component failures larger than the TBS is extremely low. However, the NRC did not complete the original 10 CFR 50.46a rulemaking or otherwise implement the TBS concept in the regulations.

The selection of the TBS was largely predicated on the results contained in NUREG-1829, Estimating Loss-of-Coolant Accident (LOCA) Frequencies Through the Elicitation Process, issued April 2008 (ML080630015), and NUREG-1903, Seismic Considerations For the Transition Break Size, issued February 2008 (ML080880140). The original technical basis supporting the TBS concept came from evaluating the impacts of degradation on piping and nonpiping passive components using data on operating experience up to approximately 2004, coupled with additional analysis. To incorporate the TBS methodology into the new proposed IE rulemaking, the staff needs to determine whether the TBS is still applicable considering the operating history since the early 2000s. Therefore, the purpose of this white paper is to describe the staffs review of the piping and nonpiping degradation behaviors over the last 15 years and determine whether there are new events, operating experience, or failure mechanisms that may impact the proposed TBS by causing a significant increase in the likelihood of a break larger than the TBS than predicted in the NUREG-1829 analyses. The staff has considered the continued adequacy of the NUREG-1903 results elsewhere (ML24207A140).

1.2 Passive Component Failures/Degradation Assumed before 2008 and Continued Applicability to Transition Break Size NUREG-1829 contains the technical basis for TBS under normal operation and common transients. It includes LOCA frequencies estimated through expert elicitation, which considered

2 operating experience and insights from probabilistic fracture mechanics (PFM) studies related to plant design, operation, and material performance. Each expert on the elicitation panel qualitatively and quantitatively assessed important LOCA contributing factors and quantified their uncertainty. Each expert considered a variety of materials and their known loads and degradation mechanisms. Since the current goal is to determine whether anything new or different has occurred since this original elicitation effort, it is important to list the degradation assumed by the experts.

1.3 Description of Tables 3.4 and 3.5 from NUREG-1829 The panelists developed a structure to allow them to better answer the elicitation questions.

Under that structure, they considered the aging factors for each material in a particular system being investigated and aging mechanisms that have been experienced or could surface in the future. They then categorized these by system and reactor type. Tables 3.4 and 3.5 in NUREG-1829, reproduced below, show the degradation mechanics considered for piping components.

Appendix B to NUREG-1829 contains similar tables for the reactor pressure vessel (RPV),

steam generators, pressurizers, pump, and valves. This information represents the state of knowledge in aging mechanisms and operation experience at the time of the elicitation. The remainder of this document investigates operational events or understood degradation mechanisms that have occurred since the early 2000s.

3

4

5 2.

Operational Experience and Degradation Not Included in NUREG-1829 that May Impact the Transition Break Size This section of the white paper considers items such as degradation, operating experience, and a change to American Society of Mechanical Engineers (ASME) requirements that have occurred since the development of the original basis in NUREG-1829 and discusses their impacts on the proposed TBS. These items were identified through the staffs experience, research, and elicitations from subject matter experts both within and external to the NRC (ML24193A120). The items include direct failures of both piping and non-piping components and draw qualitative conclusions on the impacts to TBS. The staff considers indirect failures in the next section.

2.1 Piping Operational Experience or Degradation Not Included in NUREG-1829 2.1.1 Thermal Embrittlement 2.1.1.1 Cast Austenitic Stainless Steel Thermal Embrittlement In 2019, the NRC staff approved a reactor coolant pump casing flaw tolerance analysis for subsequent license renewal (SLR), as described in Pressurized Water Reactor Owners Group (PWROG)-17033-NP-A, Revision 1, Update for Subsequent License Renewal: WCAP-13045, Compliance to ASME Code Case N-481 of the Primary Loop Pump Casings of Westinghouse Type Nuclear Steam Supply Systems, issued November 2019 (ML19319A188). The flaw tolerance analysis for pressurized-water reactor (PWR) reactor coolant pump casings fabricated with cast austenitic stainless steel (CASS) materials is an update to the previous industry analyses that were based on ASME Code Case N-481. As part of the evaluation, this updated analysis used the latest guidance on the estimation of saturated fracture toughness for CASS materials described in NUREG/CR-4513, Revision 2, Estimation of Fracture Toughness of Cast Stainless Steels during Thermal Aging in LWR Systems, issued May 2016 (ML16145A082).

The updated flaw tolerance analysis also demonstrates that the potential thermal embrittlement of CASS materials does not affect the integrity of reactor coolant pump casings for 80 years of operation.

In addition, the NRC describes aging management programs (AMPs) for SLR in NUREG-2191, Generic Aging Lessons Learned for Subsequent License Renewal (GALL-SLR) Report, issued July 2017. Specifically, AMP XI.M12 describes a program to manage the effects of thermal aging embrittlement for the Class 1 piping and components fabricated with CASS. SLR applicants may use an inspection approach (e.g., using enhanced visual, or qualified ultrasonic examination) or a component-specific flaw tolerance analysis consistent with the flaw evaluation provisions in the ASME Boiler and Pressure Vessel Code (ASME Code),Section XI, Appendix C. The NRC staff periodically updates the guidance on AMPs considering the latest research results and plant operating experience. The use of AMP XI.M12 also provides assurance that the structural integrity of the Class 1 piping and components fabricated with CASS materials will be adequately maintained for 80 years of operation.

Finally, the operating experience indicates that no significant cracking has been observed in the Class 1 piping and components fabricated with CASS materials, consistent with the discussion in section 6.3.1 of NUREG-1829. The absence of service-induced cracking in the CASS piping and components also supports the finding that the contribution of the thermal aging embrittlement of CASS materials to LOCA frequencies is not significantly higher now than it was when evaluated in NUREG-1829.

6 The TBS concept and the related increased enrichment rulemaking are based on the rationale that the probability of large-break LOCAs associated with the main coolant line, which is larger than the TBS, is extremely low. Therefore, the large-break LOCA of the main coolant line may not be treated as a design-basis accident for the purpose of evaluating the fuel dispersal. The staff conducts a more rigorous evaluation of the potential effect of thermal embrittlement of CASS materials on the rupture frequency of the main coolant line in section 4.3.7 and appendix F, table F.7, of NUREG-1829. Accordingly, the discussion below considers the continued applicability of this evaluation based on associated knowledge gained since the completion of NUREG-1829.

NUREG-1829 evaluates the thermal embrittlement effect using the PRAISE code with embrittled properties consistent with a highly aged CASS material (CF8M). The analysis only assumes fatigue crack growth, which is reasonable because the CASS materials susceptible to thermal aging are highly resistant to environmentally assisted cracking such as stress corrosion cracking (SCC) due to the relatively high ferrite content of the CASS materials. The reactor coolant pressure boundary (Class 1) piping and components fabricated with CASS have not shown service-induced cracking or leakage, as the industry operating experience to date confirms.

Appendix F, table F.7, of NUREG-1829 describes the piping break probability for 60 years of operation for the PWR hot-leg weld location based on the degraded fracture toughness of CASS materials. The table indicates that the break probabilities for the unaged CF8M material and the extremely aged CF8M material are on the order of 1x10-18 and 1x10-14, respectively. These failure probabilities are cumulative probabilities. Table F.7 of NUREG-1829 also indicates that the 60-year probabilities for a Category 1 LOCA (i.e., smallest LOCA with a flow rate greater than 100 gallons per minute (gpm)) for the unaged and extremely aged CASS materials are comparable to the break probabilities discussed above (i.e., on the order of 1x10-18 and 1x10-14 for the unaged and extremely aged CASS materials, respectively). Even though relatively large uncertainty may be associated with these extremely low break probability values, the analysis results support the finding that cracking in CASS materials is a negligible contributor to the frequencies of main coolant loop piping LOCAs.

Table F.7 of NUREG-1829 indicates that the assumed total number of the heatup, cooldown transient cycles in the fatigue crack growth analysis is 180 cycles for 60 years of operation (i.e., 3 cycles per year). If these cycles are assumed to occur for 80 years of operation, the annual number of heatup, cooldown cycles is 2.25 cycles per year. This annual number of cycles is reasonable for the 80-year fatigue crack growth analysis because it is greater than 2 cycles per year and reasonably represents annual heatup, cooldown cycles. Therefore, the total number of the heatup, cooldown cycles used in the NUREG-1829 analysis can be also used in the 80-year analysis.

The analysis in table F.7 also considers a leak detection capability of 5 gpm and inspections at 0, 20, and 40 years, as discussed in NUREG-1829, section F.2. However, a flaw in actual CASS piping may not be detected due to the grain structure of CASS that can redirect and attenuate the ultrasonic beams during the inspections. Therefore, this paper provides additional discussion below on eliminating the effect of the inspections on the failure probabilities discussed above (i.e., the effect of the inspections decreasing the failure probabilities).

Conservatively assuming that the inspections probability of detecting a flaw is 99 percent, the probability of not doing so during the three inspections assumed in the analysis is approximately 1x10-6 (i.e., (1-0.99)3). Therefore, to eliminate the effect of the inspections on the failure probabilities, the break and LOCA probabilities with inspections need to be increased by

7 approximately 6 orders of magnitude (i.e., multiplication by 1x106). This adjustment is conservative because the flaw depth in the initial phase of fatigue crack growth does not involve the very high probability of detection (close to 0.99) that is considered in the foregoing discussion. With this adjustment to the break and LOCA probabilities, the break probability for the unaged CF8M material without inspection is on the order of 1x10-12, and the break probability for the extremely aged CF8M material without inspection is on the order of 1x10-8.

For the extremely aged CF8M material, the probability of a Category 1 LOCA (the smallest LOCA) without inspection is also on the order of 1x10-8.

The adjusted break and LOCA probabilities considering no inspection are extremely small (on the order of 1x10-8 or lower). Therefore, these failure probabilities based on the NUREG-1829 analysis support the conclusion that CASS thermal aging embrittlement has a negligible effect on the LOCA frequencies for the main coolant line.

The 60-year failure probabilities of the NUREG-1829 analysis are based on the extremely aged CASS material and, therefore, the applicability of the extremely aged fracture toughness evaluated in NUREG-1829 to 80 years of operation is further discussed below. As a matter of fact, the NUREG-1829 analysis assumes that the CASS materials have the extremely aged fracture toughness throughout the evaluation time period without any reduction in fracture toughness during the operating time.

The following paragraphs discuss the fracture toughness of the CASS materials evaluated in NUREG-1829, section 4.3.2 and appendix F (table F.7), in comparison with the limiting fracture toughness based on the latest NRC guidance on the estimation of fracture toughness for CASS materials (i.e., NUREG/CR-4513, Revision 2). The analysis in NUREG-1829 uses a fracture toughness (JIc) value of 1.11 kilopounds per inch (195 kilojoules per square meter (kJ/m2)) for the unaged CASS material, as described in table F.7. The fracture toughness (JIc) of the extremely aged CASS in the analysis is 0.2 kilopound per inch (35 kJ/m2), which is lower than that of the unaged CASS by a factor of more than 5.

As described in table 4 of NUREG/CR-4513, Revision 2, the lower bound fracture toughness at 290-320 degrees Celsius for the static-cast CF8M material with a ferrite content of 30 to 40 percent and a nickel content equal to or greater than 10 weight percent can be calculated by using J = 46(a)0.26 in kJ/m2, where a is the crack extension in millimeters. In NUREG/CR-4513, Revision 2, this static-cast CF8M material group with the high ferrite and nickel content discussed above involves the most limiting (lowest) fracture toughness.

NUREG/CR-4513, Revision 2, recommends using the lower bound fracture toughness method to estimate the saturated fracture toughness for CASS materials for which ferrite content but not specific composition is known. The ferrite content range of 30-40 percent used in this discussion is the highest ferrite content range that is expected for the piping systems of domestic nuclear power plants.

Based on the 0.2 millimeter offset line in the fracture toughness curve (J = 46(a)0.26), the staff notes that the JIC value of the most limiting CASS material group is approximately 33 kJ/m2 in accordance with NUREG/CR-4513, Revision 2, table 4. This fracture toughness value is comparable to the JIC value (35 kJ/m2) of the extremely aged CASS used in the analysis of NUREG-1829, section 4.3.7 and appendix F, table F.7. This comparison indicates that the fracture toughness values are comparable between the limiting saturated CASS material in accordance with NUREG/CR-4513, Revision 2, and the extremely aged CASS material evaluated in NUREG-1829. Therefore, the fracture toughness level of the extremely aged CASS material used in the NUREG-1829 analysis is regarded to reasonably represent the limiting

8 fracture toughness of thermally aged CASS materials for 80 years of operation, consistent with NUREG/CR-4513, Revision 2.

As discussed previously, the break and Category 1 LOCA (smallest LOCA) probabilities of the main coolant line, which are based on the hot-leg weld location and limiting CF8M fracture toughness, are on the order of 1x10-8 without considering the beneficial effect of inspections.

For 80 years of operation, the corresponding annual frequencies of failure are on the order of 1x10-9 per year or lower. These extremely low failure frequencies support the finding that the potential thermal embrittlement of CASS materials has a negligible effect on the integrity of the main coolant line.

2.1.1.2 Austenitic Stainless Welds Thermal Embrittlement NUREG-1829 does not specifically discuss potential impacts associated with thermal embrittlement of austenitic stainless-steel welds (ASSWs) because the expert elicitation process did not identify thermal embrittlement of ASSWs as a significant contributor to LOCA frequencies.

In comparison, NUREG/CR-6428, Revision 1, Effects of Thermal Aging on Fracture Toughness and Charpy-Impact Strength of Stainless Steel Pipe Welds, issued August 2018 (ML18222A161), indicates that ASSWs are susceptible to thermal embrittlement. One observation is that variations in fracture toughness among unaged ASSWs are very large (e.g., by a factor of 5 or greater). This observation suggests a potential need to better understand the effects of welding processes and weld material compositions (before and after welding) on the resultant variations in weld fracture toughness.

In a similar manner to the thermal aging embrittlement of CASS, the ferrite phase in the ASSWs may increase the tendency of thermal aging embrittlement of ASSWs. However, the higher ferrite content in the ASSWs can also increase the resistance to service-induced cracking such as SCC and, therefore, can minimize the potential effect of thermal aging on the structural integrity of ASSWs due to the absence of service-induced cracking.

The operating experience with BWRs indicates that boiling-water reactor (BWR) piping is susceptible to intergranular stress corrosion cracking (IGSCC) in the heat-affected zone of welded austenitic stainless-steel piping. However, the operating experience indicates that no IGSCC has been observed in the weld itself of BWR piping. Similarly, Alloy 600/82/182 materials are susceptible to primary water SCC in the main loop piping of PWRs, but the operating experience with the PWR main loop piping indicates that no cracking has been observed in the ASSWs. Therefore, the current operating experience suggests that the potential effect of thermal aging of ASSWs on LOCA frequencies is not a significant consideration based on the high resistance of ASSWs to service-induced cracking.

ASSWs are periodically inspected, and any relevant indications such as stress corrosion cracks are repaired, replaced, or monitored in accordance with the requirements in ASME Code,Section XI, and augmented inspection programs such as the BWR SCC program. Therefore, the ASSWs that may be susceptible to thermal embrittlement are already under the existing inspection and condition monitoring activities.

In relation to this topic, the NRC staff is currently evaluating a need to update the Z-factors (elastic-plastic correction factors) that are used in the analytical evaluation of flaws in accordance with ASME Code,Section XI, Appendix C, in consideration of the potential effect of

9 thermal embrittlement. The NRC staff is also discussing this topic with the ASME Code Committee members and industry representatives. Based on these discussions, the NRC staff will evaluate whether the current approaches for the ASSW analytical flaw evaluations may need to be enhanced. In addition, the NRC staff is proposing additional inspections of similar metal welds to ensure that plants implementing this rule have acceptable performance monitoring. Section 3.1.4 discusses this concept further.

2.1.2 Stress Corrosion Cracking of Stainless Steel in Pressurized-Water Reactors 2.1.2.1 Background on Issue During a typical inservice inspection (ISI) on October 21, 2021, at Civaux Unit 1 (four loop, 1,450 megawatt PWR) in France, Électricité de France (EDF) found circumferential cracking at several locations near an elbow in the emergency core cooling system. The maximum depth of the cracking was 5.6 millimeters and, in one case, extended all the way around the circumference of the pipe. Similar cracking was also found at Civaux Unit 2 and Chooz Unit 2.

On November 13, 2021, it was reported that cracking, less severe than the others, was found in EDFs Penly Unit 1 (four loop, 1,300 megawatt PWR) during a routine inspection in the safety injection system piping near the cold leg. These lines are typically stagnant, but some circulation occurs in these sections of the safety injection system due to the proximity of the piping to both the hot-and cold-leg reactor coolant piping. The flaws were in a nonisolable section of the piping system that is susceptible to thermal fatigue. In most cases, the cracks identified were long but very shallow. Through destructive examination, EDF determined the root cause of the cracking was IGSCC and hypothesized that cracking is driven by thermal stratification loading and weld residual stress (WRS). Upon expanding its inspection program, EDF inspected over 400 welds either by destructive testing or by updated nondestructive evaluation procedures that were better suited for both detecting and sizing cracks caused by IGSCC. This expanded program found over 100 flaws at a variety of locations in the residual heat removal and emergency core cooling system piping. In most cases, the cracks were long but very shallow. In addition to the shallow flaw, EDF also found an 85 percent deep,152-millimeter (6-inch)-long circumferential crack in a nonisolable location in the safety injection system piping of Penly Unit 1 near the hot leg at a place where the thermal stratification loads found in the other locations is not expected.

2.1.2.2 Safety Concern This type of SCC was not expected in the heat-affected zone of a stainless-steel weld due to the PWR water chemistry. In fact, the only operational experience in the United States for this type of cracking occurred in conditions with elevated residual stresses, high-carbon weld chemistry, or stagnant water chemistry. The staff have not identified any IGSCC operational experience under typical PWR conditions. However, while these locations are part of plants risk-informed piping inspection programs due to the susceptibility for thermal fatigue, many of these nonisolable sections are not inspected with the proper nondestructive examination technique to identify this type of cracking.

Also, the large crack found at Penly Unit 1 was near a weld that was repaired twice during fabrication. The component associated with the weld also underwent significant cold work during installation to address misalignment between the mating surfaces. It is suspected that the WRS due to these repairs, in conjunction with the bulk cold work, is the root cause of this deep cracking. While there may be some requirements for recording repairs conducted during

10 fabrication, history has shown that many weld repairs went unrecorded, leading to uncertainty as to whether U.S. plants could have this type of weld repair.

Therefore, given the limited inspection sampling, the possibility of weld repairs, and the French operating experience, the formation of a large, undetected, circumferential crack in these pipe systems in the United States is a reasonable possibility. Propagation of a circumferential crack to a pipe rupture will impact plant safety. Even though the stainless-steel piping is highly tolerant of flaws, the loss of a nonisolable piping system within the safety injection system can limit the ability to cool the core in the case of a LOCA.

2.1.2.3 LIC-504 Summary To address these concerns, the staff analyzed this issue following the guidance for risk-informed decision-making (RIDM) in LIC-504, Revision 5, Integrated Risk-Informed Decision-Making Process for Emergent Issues, dated March 4, 2020 (ML19253D401). For these analyses, the staff used PFM to bound the annual frequencies at which SCC in the safety injection system could initiate small-break, medium-break, or large-break LOCAs. These LOCA values1 were then used with probabilistic risk assessment techniques to estimate the change in core damage frequency (CDF) that would occur with these LOCAs for all operating PWRs in the U.S. fleet. The results of these analyses suggest that with the change in the LOCA initiating event frequency, the CDF ranged from 1.3x10-8 per year for the plant with the lowest risk increase and 1.4x10-6 per year for the plant with the highest risk increase. Overall, the risk results from the analysis show that all plants would fall into the very low safety significance or low to moderate safety significance ranges that are used as guidance in LIC-504.

However, as with any risk-informed analysis, safety margins and performance monitoring were also considered. With the ongoing inspection procedure in place as part of the risk-informed program, the staff felt that both safety margins and performance monitoring would be impacted due to the inspection sampling and the fatigue qualified nondestructive examination method.

However, as part of its safety assessment, the industry implemented a needed recommendation from the Nuclear Energy Institute document NEI 03-08, Revision 3, Guideline for the Management of Materials Issues, issued February 2017, to modify the inspection technique at those locations where the IGSCC observed in the French plants occurred, and the safety margins and performance monitoring will be maintained, thus allowing reasonable assurance of structural integrity. From this, the staff recommended, and management approved, taking no direct action but continuing to monitor the industrys response. The staffs summary report (ML23236A052) contains further details.

2.1.3 Main Loop Piping Inspection Frequency The piping welds in the main reactor coolant loop of a PWR or piping larger than the feedwater and residual heat removal lines in a BWR are inspected either through a standard ASME inspection program, an augmented inspection program (e.g., Code Case N-770 for dissimilar metal weld inspections), or risk-informed ISI (e.g., Code Case N-716-2). The ASME code development community has made an effort to reduce the number of inspections in the welds that have been shown to have low safety significance or no active degradation. PFM and other bases are used to justify the reduction in the number of inspections and, in some cases, the possible elimination of inspections. While the bases may be sufficient to support the reduction in 1

Higher LOCA values were calculated as part of the sensitivity studies conducted but were considered conservative for use in calculating CDF.

11 inspections, the conclusions in NUREG-1829 were based on the ongoing inspections for the welds in consideration. In addition, the inspection results for these welds provide essential performance monitoring insights into the ongoing reliability of the current inspection approach and the associated technical bases of the inspections (e.g., deterministic or probabilistic fracture mechanics analysis assumptions and results). The ongoing inspection results also provide direct evidence of emerging degradation that may not have been considered in the analyses supporting the NUREG-1829 conclusions. Therefore, the continuing inspection of these welds is essential to the basis supporting the TBS.

2.1.4 Impacts on Transition Break Size From the LIC-504 effort related to SCC at the French plants, the pipe size considered was a 10.75-inch outer diameter, which was categorized as larger than a large-break LOCA in the LIC-504 work. From these analyses, the large-break LOCA frequency, which equated to a 5,000 gpm leak rate, was estimated as less than 7.5x10-6 per year at 80 years. Note that the PFM runs recorded no large-break LOCAs, and this frequency was estimated from the approach described in NUREG/CR-7278, Technical Basis for the use of Probabilistic Fracture Mechanics in Regulatory Applications, issued January 2022 (ML22014A406), with 95 percent confidence. From NUREG-1829, the LOCA frequencies for a 12-inch diameter ranged between 3x10-7 per year (at the mean) and 5x10-6 per year (at the 95th percentile), while at a 3-inch diameter, the LOCA frequencies ranged from 2x10-5 per year (at the mean) and 2x10-4 per year (at the 95th percentile). Clearly, the results from NUREG-1829 bound the LOCA frequency calculated in the LIC-504 effort. However, since many cracks were found in different piping in the same plant in France, it is possible that multiple failures may cause a LOCA greater than the TBS. However, multiple common-cause failures could only occur in conjunction with a large transient (e.g., seismic), which has a low frequency of occurrence. While this could lead to a scenario in which the TBS chosen may be nonconservative, the operating history in France has not been observed in the United States. In fact, through its safety assessment, the PWROG surveyed locations in these systems that were inspected and found no similar SCC. As discussed in the LIC-504 French SCC assessment (ML23236A052), future inspections will be conducted at these critical locations using an IGSCC qualified approach. If similar cracking is found to occur in the U.S. fleet, the conclusion made in this document would need to be reconsidered.

For the thermal embrittlement issue, the treatment of embrittlement of CASS in NUREG-1829 (i.e., the use of fracture toughness equivalent to 80 years of operation in the analyses) and the current aging management requirements for these weld locations provide reasonable assurance that the current understanding of thermal embrittlement of CASS does not challenge the LOCA frequencies presented in NUREG-1829. However, the impacts of thermal embrittlement of stainless-steel welds were not identified as principal attributes affecting LOCA frequencies in NUREG-1829. Based on the current understanding and ongoing research, it is not likely that this embrittlement will impact the LOCA frequencies due to their treatment in the ASME Code,Section XI, programs and aging management requirements. However, the staff is still investigating the flaw evaluation procedures for these welds; therefore, revisiting their impact on the LOCA frequencies may be prudent when the research, code development, and staff review are complete.

The reduction in piping ISIs may become a technical concern when the reduction would impact the results in NUREG-1829 and the selected TBS. The elimination of those inspections in the piping with diameters greater than the TBS will impact the NUREG-1829 conclusions.

Therefore, it is prudent to continue with sampling-based inspection programs for these welds to

12 ensure the NUREG-1829 LOCA frequencies remain valid and to provide appropriate performance monitoring of these welds. Since some of the welds in the piping with diameters greater than the TBS may be inspected continually as part of augmented and risk-informed inspection programs (e.g., ASME Code Case N-770), the main concern is the inspections of similar metal welds with no expected degradation mechanisms. ASME Code Case N-716-2 covers these welds under Category R.1.20. However, the wording of the code case does not allow for a minimum number of these welds to be inspected in the overall risk-informed ISI program (see section 4(b)(2) of Code Case N-716-2).

Therefore, for reactor coolant pressure boundary piping with a diameter greater than the TBS, the staff recommends inspecting a risk-informed sampling (i.e., those welds with the highest failure potential) of at least 10 percent of the similar metal circumferential piping welds in a PWR and the Category A welds (in accordance with Boiling Water Reactor Vessel and Internals Program (BWRVIP)-75-A, BWR Vessel and Internals Project, Technical Basis for Revisions to Generic Letter 88-01 Inspection Schedules, EPRI Technical Report 1012621, October 2005) in a BWR before implementation of the rule and in every subsequent ISI interval using ASME Code,Section XI, qualified personnel and procedures. These inspections will ensure that assumptions used in NUREG-1829 remain valid over the course of the reactor life, and any novel degradation will be identified before it threatens the integrity of the piping.

The staff discussed whether the welds chosen for the 10 percent sample should be the same welds every inspection interval (in accordance with ASME Code,Section XI) or whether different welds should be examined each inspection interval in order to inspect most of the welds throughout the reactor life. The main concern is that some welds may have an undocumented weld repair that may create a flaw that would challenge the integrity of the pipe, and inspecting different welds each interval could locate any flaw growing from this repair. First, as most weld repairs are local (i.e., they only extend a short distance around the circumference), this promotes the conditions that will lead to a short, deep-surface crack that would leak before rupture and not challenge the integrity. Second, since these pipes are in general extremely flaw tolerant, the circumferential critical flaw size for typical loads is very long

(~180 degrees) and very deep (>50 percent). If there is a mechanism like SCC active in these pipes, the loads and WRSs are expected to promote this type of crack development. From the work related to Wolf Creek Generating Station (ML072400199), natural flaw growth analyses suggest that even if a long surface crack develops under these conditions, it is expected to leak before becoming critical due to the flaw shape and the flaw tolerance of the material. In addition, recent work with the Extremely Low Probability of Rupture (xLPR) code also suggests the extremely low probability of a situation that threatens the integrity of the piping (ML22088A006).

Also, the ASME Code,Section XI, philosophy for inspection sampling is based on the probability of failure and impact to the plant (e.g.,Section XI requires inspection of 25 percent of Class 1 piping but only 7.5 percent of Class 2 piping.) The sample comprises welds with the highest failure potential in that group, and, therefore, unless operational experience suggests otherwise, continuing to inspect these welds each interval is justified as follows. With the low probability of a long, deep, undocumented weld repair and an active cracking mechanism in these similar metal welds, inspecting the same welds with the highest failure potential (based on the ASME Code,Section XI, philosophy) each interval is an appropriate risk-informed choice since it allows for monitoring needed to identify novel degradation and to ensure the NUREG-1829 assumptions remain valid, while eliminating the extra resources needed to inspect different welds each interval (e.g., different scaffolding arrangement, different fixturing, weld preparations).

13 Therefore, the staff consensus was that the ASME Code-based reinspection of the initial inspection sample during subsequent inspections was technically sufficient to provide reasonable assurance that the NUREG-1829 assumptions could be validated, and any novel degradation would be identified before challenging the integrity of this piping. In addition, the staff believes that the inspection approach should be continually informed by the most recent operating experience, inspection results, and research data regarding the susceptibility of the primary coolant loop piping to degradation mechanisms. Appendix A describes the dissenting opinion and its basis.

2.2 Reactor Pressure Vessel Operational Experience or Degradation Not Included in NUREG-1829 2.2.1 Carbon Macrosegregation 2.2.1.1 Background on the Issue In early 2015, regions of carbon macrosegregation (CMAC) were discovered in European Pressurized Reactor RPV heads manufactured for a plant in Flamanville, Manche, France, by the manufacturer, AREVA Creusot Forge (ACF), after chemical and mechanical tests on a similar vessel head conducted in late 2014. The French Nuclear Safety Authority (ASN; lAutorité de sûreté nucléaire) instructed the French licensee, Electricity of France (EDF;

Électricité de France S.A.), to perform extensive reviews where it identified steam generator channel heads produced by ACF and Japanese Casting and Forging Corporation (JCFC) to be the most likely to contain a region of CMAC. The ASN asked for nondestructive testing of these steam generator channel heads, and, eventually, higher than expected carbon content was measured in a JCFC steam generator channel head. The result was an accelerated shutdown of EDF plants to perform required nondestructive testing.

2.2.1.2 Safety Concern Positive carbon hot-top macrosegregation, generically referred to as CMAC, is a phenomenon that occurs when convective currents carry concentrated carbon, as an alloying element, to the top of an ingot during solidification. Caused by differences in solubility in the liquid and solid phases, it results in regions of heterogeneity both near the surface and within the body of the ingot, with higher strength and lower toughness relative to the bulk material. The beltline region of the RPV, which surrounds the active fuel, experiences the most neutron damage and thus is the region most likely to fail during normal operations or under postulated accident conditions.

Reasonable assurance that likelihood of failure within the RPV beltline is acceptably low is provided by regulations related to design requirements, preservice fabrication and inspection, ISI and monitoring, and operating restrictions. However, the possibility of heterogeneous material properties due to CMAC within the RPV beltline could make it more susceptible to neutron embrittlement.

To address these concerns, the NRC staff conducted a preliminary safety assessment to determine the potential safety significance posed to the U.S. fleet by CMAC as observed in components overseas. The most deleterious CMAC effects have only been observed in components made by ACF using a specific processing route. Subsequently, the NRC inspection team performed a vender inspection to review component-specific information for U.S.

components produced by ACF. The staff concluded that ACF produced only a very small population of U.S. steam generator, pressurizer, and RPV heads using this specific processing route.

14 CMAC is also only expected to occur in localized regions of a forged component and not to create a through-thickness band of inhomogeneous material. Open literature reviews identified bounds for the effects of CMAC on fracture toughness, resulting in the NRC developing a conservative estimate of the impact of CMAC. Finally, semiquantitative analysis concluded that the probability of failure of an RPV head or steam generator channel head is extremely low and bound by the failure probability for the RPV shell. The RPV shell failure probability is regulated under 10 CFR 50.61, Fracture toughness requirements for protection against pressurized thermal shock events; 10 CFR 50.61a, Alternate fracture toughness requirements for protection against pressurized thermal shock events; and Appendix G, Fracture Toughness Requirements, to 10 CFR Part 50, Domestic Licensing of Production and Utilization Facilities, to provide assurance that it remains acceptably low. Therefore, the NRC staffs preliminary safety assessment concluded that the safety significance of CMAC to the U.S. fleet is negligible.

In parallel with the NRCs activities, the ASN and the French Institute for Radiological Protection and Nuclear Safety (IRSN: Institut de radioprotection et de sûreté nucléaire) issued a report to describe their experimental efforts and summarize the results of a portion of their activities to address CMAC in France. Additionally, the Electric Power Research Institute (EPRI) performed a probabilistic analysis to assess the potential safety significance of postulated regions of CMAC in U.S. components. The results from the NRCs independent analysis are consistent with those from both the deterministic analysis conducted by the ASN/IRSN and the PFM assessment performed by EPRI.

2.2.1.3 LIC-504 Summary The final staff evaluation of the safety significance used the five key principles of RIDM, in accordance with LIC-504, Revision 5, March 4, 2020 (ML19253D401): (1) compliance with existing regulations, (2) consistency with the defense-in-depth philosophy, (3) maintenance of adequate safety margins, (4) demonstration of acceptable levels of risk, and (5) implementation of defined performance measurement strategies. The staff considered four options for responding to this issue:

(1)

Monitor and Evaluate: The NRC staff would continue monitoring all domestic and international information associated with the CMAC topic. The staff would evaluate all new information as it becomes available, to ensure that conservatism in the staffs final safety determination is maintained. New information may prompt additional actions if warranted.

(2)

Issue a Generic Communication: The NRC staff would issue a generic letter to licensees operating with forged components produced by ACF requesting that they confirm that their quality assurance program under Appendix B, Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants, to 10 CFR Part 50 verified compliance with applicable NRC regulations and ASME Code requirements. Licensees are required to submit a written response to a generic letter in accordance with 10 CFR 50.54(f).

(3)

Issue Orders Requiring Inspections: The NRC staff would order the licensees operating with components produced by ACF to conduct nondestructive examination during the next scheduled outage. Examinations would verify both the condition of the components and carbon levels.

15 (4)

Issue Orders Suspending Operation: The NRC staff would issue an order for all plants operating with components produced by ACF to immediately shut down and perform inspections. In this option, plants would not restart until the issue is addressed and the NRC grants approval.

Upon reviewing the options and applying the RIDM process in accordance with LIC-504, the staff selected option 1 as the most appropriate based on the material and processing information the staff reviewed during the vendor inspection of AREVA Inc.; experimental data and evaluation reported by the ASN/IRSN; PFM analyses conducted by EPRI; staff review of the open literature on CMAC in steel ingots and performance impacts; and the staffs semiquantitative analysis of the failure likelihood of potentially affected components. This assessment confirmed (1) the NRC staffs preliminary safety assessment that the safety significance of CMAC to the U.S. fleet is negligible and (2) no immediate action is warranted and the NRC staff will continue to monitor and evaluate CMAC effects on industry. The staff documented its assessment in a memorandum dated February 22, 2018 (ML18017A441).

2.2.2 Quasi-Laminar Indications In July 2012, ultrasonic inspections of RPV ring forgings at the Doel 3 and Tihange 2 nuclear power plants in Belgium revealed thousands of subsurface, nearly axial indications. After extensive investigation, the Belgian licensee, Electrabel, concluded that the indications consisted of hydrogen flakes that originated from fabrication. These flakes, termed quasi-laminar indications, were approximately circular, disc-shaped cracks and oriented approximately parallel to the RPV wall. The NRC issued Information Notice 2013-19, Quasi-Laminar Indications in Reactor Pressure Vessel Forgings, dated September 22, 2013 (ML13242A263), to inform industry of the quasi-laminar indications discovered in the Belgian RPV forgings. Electrabel concluded that every indication in Doel 3 and Tihange 2 met the acceptance criteria of ASME Code,Section XI, IWB-3612 (in conjunction with the flaw characterization rules of an early version of the approved ASME Code Case N-848-1), for continued operation of the plants.

To supplement the evaluation under ASME Code,Section XI, Electrabel performed conservative PFM evaluations to show that the frequency of crack initiation is less than the probability of failure criterion of 1x10-6 per year. The U.S. industry, through EPRI, also performed PFM evaluations, assuming a postulated quasi-laminar indication population that has 10 times more indications than those observed in the Doel 3 RPV forging and also determined that the through-wall cracking frequency (TWCF) was less than the probability of failure acceptance criterion of 1x10-6 per year.

The staff documented its assessment of the potential for quasi-laminar indications to impact the structural integrity of U.S. RPVs in Technical Assessment of Potential Quasi-Laminar Indications in Reactor Pressure Vessel Forgings, September 8, 2015 (ML15282A218). The staff based its assessment on the five principles of RIDM. The staff considered three options in this RIDM evaluation: (1) evaluate, communicate, and follow developments with no other actions, (2) initiate actions to require ultrasonic examination for quasi-laminar indications, and (3) impose immediate shutdown of potentially affected plants. The staff recommended option 1 as the most appropriate action. The staff determined that if quasi-laminar indications similar to those discovered at Doel 3 and Tihange 2 existed in U.S. plants, the indications are not expected to significantly affect RPV integrity under normal or accident conditions. The basis for this conclusion is the EPRI PFM analysis, in which a vessel with 10 times as many indications as observed in the worst forging at Doel 3 would have a TWCF less than the acceptance

16 criterion of 1x10-6 per year. This comparison of the TWCF frequency with the acceptance criterion of 1x10-6 per year addresses the risk principle of RIDM with the conservative assumption that TWCF is equivalent to large early release frequency. The staff also evaluated the other four principles of RIDMmeeting current regulations, maintaining defense in depth, maintaining safety margins, and performance monitoringand determined that option 1 adequately addressed these principles based on existing ASME Code design and inspection requirements, the conservatism of the PFM analysis, and the minimal quantitative risk increases. This analysis supports the conclusion that the potential existence of quasi-laminar indications is not expected to significantly affect the structural integrity of U.S. RPVs under normal or accident conditions.

2.2.3 Small Surface Breaking Flaws A 2016 study by the Oak Ridge National Laboratory (ORNL), documented in ORNL/TM-2015/59531/REV-01, The Effect of Shallow Inside-Surface-Breaking Flaws on the Probability of Brittle Fracture of Reactors Subjected to Postulated and Actual Operational Cool-Down Transients: A Status Report, issued February 2016 (ML16043A170), analyzed several RPVs subjected to plant cooldown from operating temperature and pressure to cold shutdown using the FAVOR PFM computer code. This ORNL analysis showed that the conditional probability of crack initiation (CPI) and conditional probability of vessel failure (CPF) values for shallow, circumferential, inner small surface-breaking flaws (SSBFs) that extend just into the ferritic base metal of the RPV are significantly greater than those for the quarter thickness reference flaw in ASME Code,Section XI, Appendix G. NRC technical letter report (TLR)-RES/DE/REB-2021-13, Summary of Investigations into Addressing the Shallow Surface-Breaking Flaw Issue, issued December 2021 (ML21260A245), describes sensitivity studies performed using FAVOR to determine whether the higher CPI and CPF values could result in TWCF values higher than 1x10-6 per year. The 2021 NRC TLR determined the impact of thermomechanical cladding modeling, warm prestress, and loading path on the SSBF CPI and CPF values and concluded that these three parameters did not significantly change the CPI and CPF values. The report also concluded that the small impact of these parameters on the SSBF CPI and CPF values should be interpreted with consideration of realistic cooldown transient frequencies and the likelihood of exceeding the calculated pressure-temperature limit curve. The TLR noted additional factors that should be considered, such as conservatisms in the FAVOR modeling assumptions. It stated that by taking into account the realistic cooldown transient frequencies and the additional factors, the TWCF values for SSBFs are expected to be far below 1x10-6 per year.

2.2.4 Reactor Pressure Vessel Embrittlement 2.2.4.1 Background on Issue The NRCs regulations, and associated codes and standards, applicable to RPV integrity are designed to function synergistically to provide reasonable assurance that RPV integrity will be maintained over the operating lifetime of each plant. Within these regulations, the material toughness predicted by the embrittlement trend curve (ETC) model in Regulatory Guide (RG) 1.99, Revision 2, Radiation Embrittlement of Reactor Vessel Materials, issued May 1988 (ML003740284), is used to demonstrate that acceptable margin is maintained in both normal operation (10 CFR Part 50, Appendix G) and during pressurized thermal shock events (10 CFR 50.61). In conjunction, requirements for performance monitoring through surveillance programs (10 CFR Part 50, Appendix H, Reactor Vessel Material Surveillance Program Requirements) exist to demonstrate that the generic ETC model predictions adequately

17 describe the properties of critical plant-specific RPV materials over the entire reactor operating lifetime.

However, the existing RG 1.99 (and 10 CFR 50.61) ETC model, developed in the mid-1980s, has characteristics that manifest as underprediction of RPV material neutron embrittlement under the high fluences that will be reached at many PWR plants with operation beyond 60 years. Further, the amount of the underprediction rises with increasing fluence. In parallel, licensees are allowed to defer, and many have deferred, surveillance capsule testing that is intended to confirm embrittlement predictions from the ETC model. These two issues compound the uncertainty associated with understanding the toughness behavior and impact the staffs long-term confidence in the integrity of the RPV for certain plants.

2.2.4.2 Safety Concern To determine the impacts of these underpredictions, the staff conducted a risk-informed analysis (ML21314A228). Leveraging deterministic and probabilistic fracture mechanics analyses, the staff conducted a scoping study on a variety of BWR and PWR conditions in attempts to bound the fleet behavior and estimate the risk impact of these underpredictions. For the normal operating condition, FAVOR runs were conducted on a variety of heatup and cooldown scenarios, including both hypothetical (following the pressure-temperature limit curve) and actual transients. The results suggest that for most plants, the TWCF due to the underprediction is small (< 1x10-6 per year), but if that underprediction becomes larger than 100 degrees Fahrenheit, the risk increases. For pressurized thermal shock, the staff used the methodology in 10 CFR 50.61a and calculated the mean nil-ductility reference temperature known as RTmax using the ETCs from RG 1.99, the ETC embedded in 10 CFR 50.61a known as EONY, and the ASTM International standard ASTM E900-15, Standard Guide for Predicting Radiation-Induced Transition Temperature Shift in Reactor Vessel Materials. When the TWCFs for these different RTmax values were evaluated, all remained below 1x10-6 per year.

Even though the risk appeared low for most plants, the many uncertainties associated with the analyses (e.g., plant-specific details related to transients, surveillance) make drawing general conclusions about the plant-specific impacts very difficult. These uncertainties are compounded by the apparent lack of performance monitoring data from many plants in the license renewal period. That is, some plants are continually delaying capsule withdrawal and testing, thus limiting the high-fluence, plant-specific data needed to verify ETC accuracy. Licensees can delay their final capsule from their original 40-year program because ASTM E182-82, Standard Practice for Conducting Surveillance Tests for Light-Water Cooled Nuclear Power Reactor Vessels (incorporated by reference in 10 CFR Part 50, Appendix H), was not developed to explicitly account for design lives beyond 40 years. ASTM E185-82 states that the last capsule can be withdrawn when the capsule fluence is not less than once or greater than twice the peak end-of-life vessel fluence. Therefore, the practice of delaying the final capsule from the original 40-year program to the initial period of extended operation, and then again to the subsequent period of extended operation, is still in conformance with the ASTM standard. In accordance with Administrative Letter 97-04, NRC Staff Approval for Changes to 10 CFR Part 50, Appendix H, Reactor Vessel Surveillance Specimen Withdrawal Schedules, September 30, 1997 (9709290106Legacy Library), the NRC currently only requires licensees to ensure that they are in conformance with the ASTM E185-82 schedule. Therefore, plants can continually delay the withdrawal of capsules as currently allowed under ASTM E185-82. In addition, the industry is conducting the PWR Supplemental Surveillance Program to provide high-fluence data for benchmarking ETC models up to 1x1020 neutrons per centimeter squared (n/cm2).

While these new data will help in the validation and development of the ETC, the first specimens

18 will not be available until after 2025 and may not provide sufficient performance monitoring for all the vessel materials in the fleet.

Through this evaluation, the staff demonstrated (ML21314A228) that underpredicting the embrittlement of the RPV, coupled with a lack of surveillance testing in long-term operation, may eventually impact the staffs confidence in RPV integrity for certain plants. These analyses demonstrate that the current regulations may not be adequate to ensure the long-term safety margins and performance monitoring necessary to provide reasonable assurance that RPV integrity will be maintained over the extended operating lifetime of each plant. The staff views this issue as a long-term concern that does not compromise the current integrity of the RPVs.

2.2.4.3 Path Forward In SECY-22-0019, Rulemaking Plan for the Revision of Embrittlement and Surveillance Requirements for High-Fluence Nuclear Power Plants in Long-Term Operation, dated March 8, 2022 (ML21314A215), the staff requested Commission approval to conduct rulemaking to amend the RPV embrittlement and surveillance regulatory requirements. This plan presented the technical basis and need for the change and suggested several options for Commission consideration. The staffs recommended option called for revising Appendix H to 10 CFR Part 50 to include additional surveillance testing requirements for long-term operation and to revise applicable regulations (10 CFR 50.61) and guidance (RG 1.99) to include a revised fluence function fit for all materials that will experience high fluence levels (e.g., a new ETC or an update to existing ETCs). The staff estimated it would take 4 to 5 years to complete the rulemaking after receiving approval from the Commission. As of September 2024, the Commission still has the rulemaking plan for review. If the Commission does not approve this rulemaking, it is suggested that the staff re-evaluate the TBS based on the known underprediction of the ETC.

2.2.5 Change in Reactor Pressure Vessel Inspection Frequencies Licensees have been submitting to the NRC for review and approval alternatives to the ISI frequencies required by ASME Code,Section XI, for certain RPV welds. Therefore, the staff investigated the impacts of the change in the inspection frequencies on the LOCA frequencies in NUREG-1829. For BWR RPV circumferential welds, the alternative ISI frequency is based on the PFM evaluations in NRC-approved proprietary topical reports BWRVIP-05, BWR Vessel and Internals Project: BWR Reactor Pressure Vessel Weld Inspection Recommendations, issued September 1995, and BWRVIP-329-A, BWR Vessel and Internals Program: Updated Probabilistic Fracture Mechanics Analyses for BWR RPV Welds to Address Extended Operation, issued December 2021. For BWR RPV nozzle-to-vessel welds and BWR nozzle inner radii, reduced inspections are based on meeting the provisions of ASME Code,Section XI, Code Case N-702 (which has been update to N-702-1). For PWRs, the alternative ISI frequency is based on the PFM evaluations and performance monitoring in NRC-approved nonproprietary topical report WCAP-16168-NP-A, Revision 3, Risk-Informed Extension of the Reactor Vessel In-Service Inspection Interval, issued October 2011 (ML11306A084).

The BWRVIP-05 report contains the original BWR PFM evaluations, and the BWRVIP-329-A report updates those evaluations. The PFM evaluations included the effects of embrittlement of the RPV beltline welds, a postulated distribution of fabrication flaws within the RPV welds, consideration of service-induced flaw growth, and a limiting low-temperature overpressure event to determine the RPV probability of failure in terms of TWCF. The probability of failure acceptance criterion used in the PFM evaluations is 1x10-6 per year (originally 5x10-6 per year in

19 the BWRVIP-05 work). The PFM evaluations justify inspections of only the BWR RPV beltline axial welds and elimination of the inspections for BWR RPV beltline circumferential welds.

Licensees alternative requests must include plant-specific information to show that the plant is bounded by the PFM evaluations in BWRVIP-05 or BWRVIP-329-A.

ASME Code,Section XI, Code Case N-702, is incorporated in RG 1.147, Revision 19, Inservice Inspection Code Case Acceptability, ASME Section XI, Division 1, issued October 2019 (ML19128A244), with conditions, and is applicable to nozzle-to-vessel welds and nozzle inner radii of BWR nozzles (typically recirculation inlet and outlet nozzles) that are not feedwater nozzles or control rod drive return nozzles. This has been updated to ASME Code Case N-702-1 (approved without conditions in RG 1.147, Revision 20, issued December 2021 (ML21181A222)) that incorporated the conditions in RG 1.147, Revision 19. ASME Code,Section XI, Code Case N-702, and the conditions in RG 1.147, Revision 19, define criteria that must be met to allow a reduction of inspections to 25 percent of the nozzle population of the nozzle types covered by the code case. Fluence effects are included for periods of extended operation (i.e., more than 40 years of operation). Code Cases N-702 and N-702-1 are based on a probability of failure acceptance criterion for the PFM evaluations of 5x10-6 per year.

The WCAP-16168-NP-A report describes risk-informed evaluations based for the most part on the studies performed for PWR pilot plants during the development of the alternate pressurize thermal shock rule in 10 CFR 50.61a. The evaluation methodology in WCAP-16168-NP-A is applicable to ASME Code,Section XI, Examination Category B-A and Examination Category B-D welds in PWRs. The PFM evaluation portion of the methodology included effects of embrittlement of the RPV beltline welds, a postulated flaw distribution within the RPV welds consisting of embedded and inner-surface breaking fabrication flaws, consideration of service-induced flaw growth, and limiting design-basis transients to determine the RPV probability of failure in terms of TWCF. The PFM evaluations are supplemented by performance monitoring that consists of a fleetwide inspection schedule (ML11153A033) of the welds in the scope of the report. The PFM evaluations used probability of failure acceptance criteria of 1.76x10-8 per year, 3.16x10-7 per year, and 4.42x10-7 per year for Westinghouse, Combustion Engineering, and Babcock & Wilcox-designed plants, respectively. The PFM evaluations and fleetwide inspections justify extension of the ISI interval required by ASME Code,Section XI.

Licensees alternative requests must include plant-specific information, specified in appendix A to WCAP-16168-NP-A and the NRC safety evaluation of the report, to show that the plant is bounded by the PFM evaluation for the corresponding PWR pilot plant analyzed in WCAP-16168-NP-A.

2.2.6 Impacts on Transition Break Size For the SSBF, CMAC, and quasi-laminar indications, the results of the NRC analyses suggest that the TWCFs in all cases were well below the acceptance criteria of 1x10-6 per year.

However, these TWCFs were conditional on the existence of the defects in the vessel wall.

While the actual probability of occurrence for these defects is unknown, it can be safely assumed to be less than the assumed probability of occurrence in the analyses, due to other inspections and research done in the past. Therefore, the realistic TWCFs for these defects can be considered extremely low and will not impact the proposed TBS.

For the embrittlement issue, as described above, there may be a potential safety issue due to the nonconservative predictions of embrittlement and uncertainty in surveillance testing schedules in both the NRC guidance and regulations. The staff proposed to modify both the regulations and guidance to address these issues, as described in SECY-22-0019. Even though

20 the analyses suggest that the TWCF in most cases is less than 1x10-6 per year, the safety margins and performance monitoring are impacted by these issues and raise unknown uncertainty in the resulting TWCF. However, as noted above, the staff does not believe this is a current issue, and it only becomes a concern in the future for higher fluence plants. Therefore, the staff recommends revisiting the impact of these embrittlement concerns following Commission action on the rulemaking plan.

Concerns arise when inspection frequencies are decreased and analyses are used to justify a low-risk situation (e.g., quasi-laminar indications), because performance monitoring is needed to ensure that the analyses remain valid and no unexpected degradation is occurring. However, in the cases discussed, performance monitoring was covered by either limiting the inspection relief to the circumferential welds only (BWR), reducing the required 10-year inspections to 25 percent of the population (vessel-to-nozzle welds and inner radii) rather than eliminating inspections, or presenting a fleet-wide performance monitoring plan (PWR) that demonstrated that inspection data would continue to be generated over the period of the relief. With that added confidence, the impact on the TBS of the change in the inspection frequencies for RPV welds is insignificant because the criteria for the probability of RPV failure that must be met for the licensing and regulatory actions described above are within the range of the LOCA frequencies developed in NUREG-1829. Also, the prescribed performance monitoring plan for these welds continues to generate data throughout the service life of these reactors that allow the staff to verify the continued adequacy of the analyses used in the basis and should identify any emergent degradation not considered in the analyses.

Whenever multiple issues are observed in the same component, their cumulative effects on the overall failure frequency must be considered. Given that the TWCF is conditional on the existence of the indications considered, which is extremely low, it can be inferred that the cumulative effects of the actual TWCF would be also extremely low. However, this may not be the case if the impacts occur simultaneously, such as if SSBF occurs at the same location as CMAC. Given that the probability of occurrence of these individual effects is extremely low, it can be surmised that the probability of both occurring at the same time and location is even lower. In addition, the interaction effects are not a large concern since the degradation typically occur in different locations (CMAC typically occurs in the middle of the wall thickness, while SSBF occurs near the inner surface at the cladding interface). In addition, with the provided performance monitoring efforts for RPV weld inspection relief, the TWCFs were also extremely low, such as 1x10-8 per year, thus not adding significantly to the cumulative effects. The largest potential impact considered was from the nonconservative embrittlement predictions; however, as described above, the main safety concern was not the risk numbers (<1x10-8 per year when actual transients are assumed) but the performance monitoring issues. Therefore, when the cumulative effects of all these topics are considered, the staff believes the overall TWCF would still be extremely low. Still, the staff recommends revisiting the impact of these embrittlement concerns on the TBS following Commission action on the rulemaking plan.

2.3 Steam Generator Operational Experience or Degradation Not Included in NUREG-1829 2.3.1 Change in Shell Weld Inspection Frequencies for Steam Generators Licensees have been submitting to the NRC for review and approval plant-specific alternatives to extending the ISI interval required by ASME Code,Section XI, for steam generator welds and nozzle inside radii. The technical basis for the alternative is the PFM evaluations in EPRI Technical Report 3002014590, Technical Bases for Inspection Requirements for PWR Steam

21 Generator Feedwater and Main Steam Nozzle-to-Shell Welds and Nozzle Inside Radius Sections, issued April 2019 (ML19347B107), and EPRI Technical Report 3002015906, Technical Bases for Inspection Requirements for PWR Steam Generator Class 1 Nozzle-to-Vessel Welds and Class 1 and Class 2 Vessel Head, Shell, Tubesheet-to-Head and Tubesheet-to-Shell Welds, issued in 2019 (ML20225A141), supplemented by plant-specific performance monitoring during the approved extended interval.

The PFM evaluation portion of the methodology included effects of a postulated fabrication flaw distribution within the welds and nozzle inside radii, consideration of service-induced flaw growth (fatigue), various ISI scenarios, and design-basis transients to determine the probability of failure in terms of probability of rupture, with rupture defined as the case when the crack driving force exceeds the material resistance to cracking in the linear elastic regime. The probability of failure acceptance criterion used in the PFM evaluations is 1x10-6 per year. The PFM evaluations justify ISI interval extension for steam generator shell welds, nozzle-to-vessel welds, and nozzle inside radii. Licensees alternative requests must include plant-specific information to show that the analyses in the EPRI reports are applicable to the plant. In addition, the plant-specific performance monitoring plan proposed as part of the requests would demonstrate that these welds will continue to perform as they have in the past, and that emergent degradation will be identified before it affects the weld integrity.

Based on this rationale, the impact on the TBS of the ISI interval extension for steam generator welds and nozzle inside radii is insignificant. This is because the probability of steam generator failure criterion that must be met for the licensing action described above is within the range of the LOCA frequencies developed in NUREG-1829, and it is expected to be less than the 1x10-5 per year frequency that was considered as an upper limit when selecting the TBS. Finally, the proposed performance monitoring plan for these welds continues to generate data throughout the service life of these steam generators, which allows the staff to verify the continued adequacy of the analyses used in the technical basis. This plan should also identify any emergent degradation not considered in the analyses.

2.4 Pressurizer Operational Experience or Degradation Not Included in NUREG-1829 2.4.1 Change in Shell Weld Inspection Frequencies for Pressurizers Licensees have been submitting to the NRC for review and approval plant-specific alternatives to extending the ISI interval required by ASME Code,Section XI, for pressurizer welds. The technical basis for the alternative is the PFM evaluations in EPRI Technical Report 3002015905, Technical Bases for Inspection Requirements for PWR Pressurizer Head, Shell-to-Head, and Nozzle-to-Vessel Welds, issued December 2019 (ML21021A271),

supplemented by plant-specific performance monitoring during the approved extended interval.

The PFM evaluation portion of the methodology included the effects of a postulated fabrication flaw distribution within the welds, consideration of service-induced flaw growth (fatigue), various ISI scenarios, and design-basis transients (including insurge-outsurge transients) to determine the probability of failure in terms of the probability of rupture, with rupture defined as the case when the crack driving force exceeds the material resistance to cracking in the linear elastic regime. The probability of failure acceptance criterion used in the PFM evaluations is 1x10-6 per year. The PFM evaluations justify ISI interval extension for pressurizer shell welds and nozzle-to-vessel welds. Licensees alternative requests must include plant-specific information to show that the analyses in the EPRI report are applicable to the plant. In addition, the plant-specific performance monitoring plan proposed as part of the requests would demonstrate

22 that these welds will continue to perform as they have in the past, and that emergent degradation will be identified before it affects the weld integrity.

Based on this rationale, the impact on the TBS of extending the ISI interval for pressurizer welds is insignificant because the probability of pressurizer failure criterion that must be met for the licensing action described above is within the range of the LOCA frequencies developed in NUREG-1829, and it is expected to be less than the 1x10-5 per year frequency that was considered as an upper limit when selecting the TBS. Finally, the proposed performance monitoring plan for these welds continues to generate data throughout the service life of these pressurizers, which allows the staff to verify the continued adequacy of the analyses used in the technical basis. This plan should also identify any emergent degradation not considered in the analyses.

3.

Secondary-Side Piping Failure that May Cause Indirect Failures Greater than the Transition Break Size This effort hypothesized that the secondary piping system, like the safety injection system, fails due to degradation and impacts the larger reactor coolant system piping due to pipe whip or other dynamic effects increasing the LOCA frequencies. While this is a possibility, regulations and guidance are in place to prevent this behavior in the design stage. General Design Criterion 4, Environmental and dynamic effects design bases, in Appendix A, General Design Criteria for Nuclear Power Plants, to 10 CFR Part 50 states that structures, systems, and components shall be appropriately protected against dynamic effects, including the effects of missiles, pipe whipping, and discharging fluids, that may result from equipment failures and from events and conditions outside the nuclear power unit. In addition, NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR Edition, Section 3.6.2, Determination of Rupture Locations and Dynamic Effects Associated with the Postulated Rupture of Piping, contains guidance on where piping breaks must be postulated to design the hardware and equipment needed to properly mitigate the dynamic effects of pipe ruptures. Adherence to this regulation and guidance ensures that reasonable assurance of safety is maintained.

With adherence to the regulation and guidance, the flaw tolerance of nuclear piping, and the limited operating experience in piping failure, the probability of secondary piping failure causing enough damage to the larger reactor coolant system piping to increase the LOCA frequencies is very low. However, because of the many plant systems in the variety of different plant designs in the United States, there is uncertainty in the type, material, and proximity of high-energy secondary piping systems relative to the larger reactor coolant system piping. Therefore, it is prudent for licensees to cover any possible impacts of secondary-side failure causing indirect failure of piping greater than the TBS when demonstrating plant-specific applicability for the TBS. The NRC will provide guidance to aid the licensee in making this determination.

4 Summary of the Supporting Efforts In addition to the activities described above, the staff conducted in-house and external elicitations, additional operational experience database analyses, and PFM analyses to support the development of the basis for continued adequacy of NUREG-1829.

4.1 Summary of the Elicitations

23 The staff conducted both internal and external elicitations to separately evaluate the continued adequacy of the technical basis supporting the TBS concept for use in the proposed IE rulemaking. The external experts expect LOCA frequencies associated with TBS ruptures to be less than those estimated in NUREG-1829, which served as the starting point for the TBS selection. Neither the internal nor external elicitations identified any generic issues or scenarios that either were not considered in TBS development or have changed since TBS development that could undermine its technical basis. Therefore, the results from these elicitations support the continued adequacy of the TBS concept for the IE rulemaking.

However, both the internal and external elicitations did identify topics that should be addressed within either the proposed rule, the statement of considerations, rule implementation guidance, or guidance to demonstrate plant applicability. Topics identified during the elicitation for consideration within these various documents include probabilistic risk assessment requirements, impacts of plant changes, SCC in main loop and recirculation piping, indirect piping failures, direct and indirect seismic failure evaluations, maintenance of mitigative capabilities, grid stability, common-cause maintenance errors, NUREG-1829 uncertainties, plant security impacts, and attributes that could increase plant-specific LOCA frequencies. This document and the rulemaking package have already addressed many of these topics, and the elicitations served to refine their proposed treatment as well as identify some novel issues that were not initially considered. Details of the issues and dispositions can be found at ML24193A128.

4.2 Operating Experience Analysis Summary As part of this effort, the staff updated the NUREG-1829 LOCA frequency estimates based on an analysis of operating experience from 2005 through mid-2024 as it pertains to LOCA Categories 4, 5, and 6 (defined in accordance with NUREG-1829). The operating experience data came from the database developed under the Organisation for Economic Co-operation and Development/Nuclear Energy Agencys Component Operational Experience, Degradation and Ageing Programme (CODAP). The analysis consisted of four major steps:

(1)

Review applicable operating experience.

(2)

Calculate piping failure precursor frequencies.

(3)

Calculate conditional probabilities of LOCA.

(4)

Calculate LOCA frequencies.

In the first step, the staff reviewed the domestic operating experience applicable to the more risk-significant base case scenarios from NUREG-1829, table 3.7: BWR-1 (12-and 28-inch piping in the BWR recirculation system), PWR-1 (30-inch piping in the PWR hot leg), and PWR-2 (10-inch piping in the PWR pressurizer surge line). The staff also reviewed domestic operating experience for primary system piping branch connections greater than 6 inches in diameter (i.e., piping able to produce at least a Category 4 LOCA). The reviews covered the two distinct time periods (i.e., from before 1970 through approximately 2004, and from 2005 to mid-2024) and assessed differences and underlying factors of influence.

In the second step, the staff calculated the piping failure precursor frequencies using a Bayesian analysis approach. The results are expressed in units of failures per reactor operating year and the number of components exposed to a certain material degradation mechanism, such as SCC. A constrained, noninformative distribution method was used to develop a prior piping failure precursor frequency distribution representative of the state of knowledge in 2004. The staff then updated the prior distribution using a Bayesian approach, with the number of failures

24 as the numerator and the number of reactor operating years and number of components as the denominator.

In the third step, the staff developed the BWR and PWR base case conditional probability uncertainty distributions from NUREG-1829 by first obtaining a geometric mean of the expert elicitation participant results for each base case. This value represented the target LOCA frequency. Next, the geometric means of the precursor failure frequency results of the respective base case analyses were fit to a lognormal distribution. Finally, the median conditional probability of LOCA was obtained by dividing the median target LOCA frequency by the median precursor failure rate. The resulting conditional probability of LOCA represents uncertainty in the 2004 state of knowledge when the NUREG-1829 expert elicitation was completed. These conditional probability of LOCA uncertainty distributions were subsequently updated with evidence of precursor events after 2004 from the operating experience data.

Lastly, in the fourth step, the uncertainty distributions developed in steps 2 and 3 were convolved in a converged Monte Carlo uncertainty propagation to obtain the respective LOCA frequency uncertainty distributions.

The updated LOCA frequency results reflect the state of knowledge in 2024 and, relative to the NUREG-1829 results, indicate the current ones are at least an order of magnitude less than the NUREG-1829 LOCA Category 4, 5, and 6 estimates. The significant reduction reflects the effectiveness of various material degradation mitigation processes that have been implemented for BWRs and PWRs, such as stress improvement processes, water chemistry changes, increased inspections, and more degradation-resistant materials. The analysis methods included consideration of uncertainties in the state of knowledge regarding factors of improvement in reliability and integrity management and operating experience. A complete description of these analyses and results can be found in An Assessment of BWR & PWR Loss-of-Coolant Accident Frequencies Based on Operating Experience, July 2, 2024 (ML24193A122).

4.3 Probabilistic Fracture Mechanics Analyses As part of this effort, using updated PFM tools, data, and expertise, the staff examined the more risk-significant base case scenarios from NUREG-1829, table 3.7: BWR-1 (12-and 28-inch piping in the BWR recirculation system), PWR-1 (30-inch piping in the PWR hot leg), and PWR-2 (10-inch piping in the PWR pressurizer surge line). The BWR base cases are subject to IGSCC, and the PWR base cases are subject to primary water SCC.

To reanalyze the base cases, the NRC used the PFM code xLPR, Version 2.3, described in NUREG-2247, Extremely Low Probability of Rupture Version 2 Probabilistic Fracture Mechanics Code, issued August 2021 (ML21225A736). xLPR was developed specifically to model the effects of primary water SCC and is, therefore, well-suited to reanalyze the PWR base cases. For the BWR-1 base cases, IGSCC was modeled using the assumption of an initial crack at the beginning of plant operation, and a generic SCC model available in the xLPR code was used with parameters adapted to match the IGSCC growth rate published in the 2023 Edition of the ASME Code, Division 1,Section XI, Subsubarticle Y-2310.

The inputs for the PWR base cases mostly came from xLPR-GR-IG Version 1.0, xLPR Group ReportInputs Group, dated December 19, 2017 (ML19337B876), and xLPR-MSGR-WRS Version 1.0, xLPR Models Subgroup ReportWelding Residual Stresses, dated October 5, 2016 (ML16341B049). The material properties and some other inputs for the BWR

25 base cases also largely came from the xLPR Inputs Group report, except for the crack growth rate model parameters mentioned in the previous paragraph. The normal operating loads were selected to match the corresponding values from the NUREG-1829 PFM analyses, and the WRS profiles were created using finite element analysis with the geometries and welding parameters published in EPRI Technical Report NP-1743, Effect of Weld Parameters on Residual Stresses in BWR Piping Systems, dated March 1, 1981. For consistency with the NUREG-1829 results, probabilities for LOCA Categories 1 through 6 were generated considering the effects of leak detection and then translated into annual frequencies.

The PWR analyses were within the range of the NUREG-1829 base case results. The PWR-1 analysis was performed with both a generically representative WRS profile and a conservative WRS profile based on the hot-leg pipe-to-reactor-pressure-vessel nozzle weld at Virgil C. Summer Nuclear Station, Unit 1, which developed a leak due to primary water SCC in 2000. Neither WRS profile resulted in any category of LOCA. As a result, a 95 percent upper confidence bound was estimated for the annual frequencies. For both WRS profiles, the xLPR results were within the range of the NUREG-1829 base case results at 25 and 60 years of operation. For the PWR-2 analysis, a few realizations resulted in a LOCA with leak detection.

The first LOCA realization was a small-break LOCA occurring around 35 years, and medium-and large-break LOCA realizations happened around 75 years of operation. Nevertheless, the resulting annual frequencies were within the range of NUREG-1829 base case results at 25 years of operation and below the NUREG-1829 range at 60 years of operation. Previous sensitivity studies for both PWR analyses are documented in NRC TLR-RES/DE/REB-2021-09, Probabilistic Leak-Before-Break Evaluation of Westinghouse Four-Loop Pressurized-Water Reactor Primary Coolant Loop Piping using the Extremely Low Probability of Rupture Code, issued August 2021 (ML21217A088), and TLR-RES/DE/REB-2021-14-R1, Probabilistic Leak-Before-Break Evaluations of Pressurized-Water Reactor Piping Systems using the Extremely Low Probability of Rupture Code, issued April 2022 (ML22088A006).

The BWR analyses were also within the range of the NUREG-1829 base case results.

NUREG-1829 does not differentiate between the LOCA frequencies for the BWR-1 base case for 12-inch and 28-inch piping. However, the xLPR code generated different LOCA frequency estimates for the two piping sizes because of different loads and WRS profiles, while the same loads and WRS profile were used in the NUREG-1829 PFM analyses for the 12-and 28-inch cases. In the 28-inch BWR-1 analysis, a few realizations resulted in small-break LOCAs with leak detection. Similarly, the 12-inch BWR-1 analysis had a few realizations with all LOCA categories. Nevertheless, the annual frequency estimates remained within or below the range of estimates from NUREG-1829. A few sensitivity studies were performed for the BWR analyses to evaluate the effects of different WRS profiles and water chemistries. The resulting LOCA frequencies were equivalent to or lower than the initial analysis results.

The updated LOCA frequency results based on the xLPR simulations reflect the state of knowledge and PFM modeling capabilities in 2024 and are consistent with the NUREG-1829 base case results. A complete description of these analyses and results can be found in An Assessment of BWR & PWR Loss-of-Coolant Accident Frequencies Based on Operating Experience, July 2, 2024 (ML24193A122).

5.

Conclusion The staff reviewed the conditions, analyses, and results from the expert elicitation process conducted in NUREG-1829 and performed internal and external analyses and elicitation to determine whether the TBS developed in the early 2000s is still applicable, considering the

26 operating history for the last 20 years. The staff focused on conditions and degradation that may impact breaks bigger than the TBS, but it also confirmed the results of the base case analyses conducted as part of the NUREG-1829 effort. The analyses conducted demonstrate that the PFM tools used predicted LOCA frequencies that were bounded by the LOCA frequencies generated from the NUREG-1829 elicitation. In addition, a review and statistical analysis of the operating experience since the early 2000s demonstrate that no substantial degradation has occurred that undermines the NUREG-1829 results or the proposed TBS. Finally, for those material degradation events that have occurred since the completion of NUREG-1829, the staff found that the probability of direct and indirect piping failures associated with these events would also not challenge the proposed TBS.

However, the staff identified a concern with the ongoing industry initiative to optimize inspection requirements for components within the reactor coolant loop. The staff concluded that the elimination of inspections in the piping with diameters greater than the TBS would impact the NUREG-1829 conclusions. Therefore, the staff recommends inspecting a risk-informed sample of 10 percent of these welds before implementation of this rule and in every subsequent inspection interval using qualified personnel and procedures. (Repeatedly inspecting the sample welds each interval is acceptable.) With this recommendation, the staff concludes that the results from NUREG-1829 remain adequate in determining the proposed TBS.

27 Appendix AMinority Opinion on Inspection Requirements Section 3.1.3 of the white paper covers the continual decrease in the main loop piping inspection frequency, while section 3.1.4 examines the potential impacts of this trend on the transition break size (TBS). This discussion indicates that the elicitation described in NUREG-1829, Estimating Loss-of-Coolant Accident (LOCA) Frequencies Through the Elicitation Process, issued April 2008 (Agencywide Documents Access and Management System Accession No. ML080630015), presumed that such inspections would continue, and that this presumption was one factor supporting the LOCA frequency estimations that were used as the starting point to select the TBS. Section 3.1.4 also points out that inspections are an important component of performance monitoring to ensure that unexpected degradation can be identified in the future before it progresses to an unacceptable magnitude. Section 3.1.4 further recommends that licensees that wish to implement the voluntary increased enrichment (IE) rulemaking be required to conduct an initial inspection of a risk-informed sample of 10 percent of the Category A welds (in accordance with Boiling Water Reactor Vessel and Internals Program (BWRVIP)-75-A, BWR Vessel and Internals Project, Technical Basis for Revisions to Generic Letter 88-01 Inspection Schedules, EPRI Technical Report 1012621, October 2005) in the largest recirculation loop piping (for boiling-water reactors (BWRs)) or those welds in risk-informed category R1.20 (in accordance with American Society of Mechanical Engineers (ASME) Code Case N-716-2) within the main loop piping (for pressurized-water reactors (PWRs)). The recommendation further stipulates that subsequent inspections of this population shall occur during each inservice inspection (ISI) interval. The recommendation also allows licensees to credit welds within this population that are already part of their inspection sample.

I completely support the recommendation to require a risk-informed inspection of 10 percent of the stated population before implementing changes allowed under the proposed rule, and I also support the notion of allowing credit for such welds that are part of the existing ISI program.

However, I believe that the next inspection (i.e., during the following ISI interval) should choose another risk-informed sample of 10 percent of those remaining welds that have not previously been inspected and are therefore not part of the existing ISI program. Each subsequent ISI interval should then choose a new population of uninspected welds until, if the plant operates long enough, all of these welds will be inspected one time over the course of the plants operation period after implementation of the IE rulemaking.

The basis for this opinion follows. Decades of experimental and computational research as well as operating experience have demonstrated the superior integrity of nuclear primary coolant piping systems. The integrity also generally increases as the piping size increases, such that those systems with diameters greater than the TBS are expected to be the most reliable within the nuclear power plant. These systems are extremely leak-before-break tolerant such that failure due to most localized degradation mechanisms is almost impossible. This is because even if the degradation progresses through the piping wall and causes leaking, there is ample opportunity (i.e., typically months or more) to identify and repair this location before the structural integrity of the piping system is challenged. This attribute has been demonstrated repeatedly in large-scale experimental testing programs, deterministic structural integrity assessments, and, more recently, through probabilistic fracture mechanics (PFM) analyses using, for example, the Extremely Low Probability of Rupture (xLPR) tool.

The only ruptures of concern in these systems are associated with degradation scenarios that produce a break before a leak. A handful of mechanisms have caused such failures in larger (i.e., >6 inches in diameter) nuclear piping systems, including flow-accelerated corrosion, hydrogen deflagration, and stress corrosion cracking (SCC). Of these mechanisms, SCC is

28 believed to be the most credible degradation mechanism that could be active in the main loop (PWRs) or recirculation (BWR) piping.

SCC requires high tensile stresses at the piping inner diameter, an appropriate environment, and a susceptible material. High inner-diameter tensile stresses are possible in these systems depending on the welding process and the specific welding sequence used during fabrication.

Large repair welds as well as aggressive surface treatment (e.g., grinding) and piping system misalignment during fabrication can also induce such stresses. Operating experience has demonstrated that welds with these characteristics exist. For example, the hot-leg, safe-end, dissimilar metal weld at Virgil C. Summer Nuclear Station was subjected to large repairs that significantly contributed to this through-wall cracking event. More recent cracking in safety injection piping near the hot leg at Penly 1 in France was also associated with weld repairs.

Stainless-steel materials are susceptible to SCC, particularly higher carbon alloys that become sensitized during welding or furnace heat treatment. These materials significantly contributed to the intergranular stress corrosion cracking (IGSCC) first identified in BWRs in the late 1970s and early 1980s. However, more recent experience in France has demonstrated that cracking can occur in newer, lower carbon grade stainless steels as well. Operating experience (i.e., IGSCC in BWRs) and research has demonstrated that light-water-reactor environments are conducive to SCC formation. Higher operating temperatures lead to higher crack growth rates (CGRs), so hot-leg CGRs are expected to be higher than cold-leg CGRs. High oxygen environments associated with BWR normal water chemistry also lead to higher CGRs, while much slower CGRs occur in PWR and hydrogenated water chemistry BWR environments.

However, the recent French experience has demonstrated that extensive cracking is possible in what are expected to be low-potential PWR environments.

Prior experimental testing has demonstrated that ruptures in large piping systems typically require extensive cracking around the inner diameter of the pipe (i.e., >180 degrees) and with significant depth penetration (i.e., 50 percent or more) before the structural integrity of the pipe is challenged. Any loss of fracture toughness that may occur during operation (i.e., thermal aging), however, will result in ruptures at smaller crack depths. Typically, the only xLPR realizations that can lead to ruptures are scenarios that start with a large inner-diameter crack, coupled with sufficient applied and residual stresses to continue to drive the crack through the piping thickness due to SCC until the crack reaches a size sufficient to cause rupture before a leak occurs. This is exactly the scenario of concern.

It is recognized that such a scenario requires a confluence of events: high inner diameter tensile stresses due to the fabrication processes, sufficient applied and residual stresses both around the circumference and through the pipe wall thickness, a locally conducive environment to allow the crack to grow as a result of SCC, and sufficient time (i.e., decades) for these very slow processes to occur. Most importantly, such a scenario needs to occur near or within a weld that has not been inspected to uncover such extensive degradation. While such a scenario is certainly unlikely, NUREG-1829 estimates that failures larger than the TBS occur at frequencies on the order of 1x10-5 per year (or less). Given the current understanding of such degradation and the operating experience examples of the presence of associated attributes, the likelihood of such a scenario could be consistent with the TBS frequency, especially in light of uncertainties associated with frequency estimation.

While this is certainly a low-likelihood event, such a rupture could conceivably have high failure consequences. The proposed voluntary IE rulemaking will eliminate such breaks from the design basis. While mitigation of such ruptures is still required, the uncertainty of the success of the mitigation will be greater since best estimate evaluation will be allowed to demonstrate

29 mitigation effectiveness without consideration of other design-basis requirements (e.g., single failure criterion and coincident loss-of-offsite power). The rulemaking also allows broad, risk-informed plant changes to be made such that the plant-specific configuration at the time of such a break is more difficult to predict and model a priori to better understand the uncertainties of mitigating such a break and ultimately the likelihood of success.

For these reasons, I believe all such welds should be inspected approximately one time over the course of the remaining plant life. If such extensive degradation were to occur in the large piping systems, it will require decades before it becomes significant. This attribute decreases the urgency for such inspections. Further, if complete knowledge of weld fabrication and repair and operating conditions of each weld within the population were available, it would be possible to choose a definitive risk-informed sample to rule out the possibility of such degradation. While such complete knowledge is not available, risk-informed sampling provides the best available means for identifying and prioritizing the inspection of the most potentially susceptible welds.

Subsequent new risk-informed sampling of the remaining uninspected welds would then provide continued assurance that no previously unknown attributes are causing extensive degradation that may challenge the structural integrity of the piping system within the uninspected weld population.