ML13339A708
| ML13339A708 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 11/26/2013 |
| From: | Robinson T Citizen Power |
| To: | Peter Bamford Division of Policy and Rulemaking |
| References | |
| Download: ML13339A708 (1) | |
Text
CITIZEN POWER PublicPolicy' Research Education and Advocacy November 26, 2013 Peter Bamford, Project Manager Office of Nuclear Reactor Regulation Mail Stop: 08B3 Washington, DC 20555-0001
Dear Mr. Bamford:
As you know, during the latest Beaver Valley Unit 1 refueling outage, a breach of the steel containment liner of approximately .4 inch by .28 inch was discovered on October 4, 2013. This breach followed a 2009 hole in the containment liner of 1 inch by
.375 inch and the 2006 discovery of three significant patches of corrosion of the containment liner.' Citizen Power believes that this pattern of corrosion indicates that there is the potential of a serious ongoing containment liner issue and requests that the Nuclear Regulatory Commission ("NRC") require the FirstEnergy Nuclear Operating Company ("FENOC") modify their Unit 1 containment liner inspection program through License Renewal Commitment 35 in order to meet the requirements of 10 CFR 100 and 10 CFR 50.34(a)(1)(D).
FENOC lists as an intended function of the Unit I and 2 structures the ability to function as an essentially leak tight barrier to protect public health and safety in the event of a postulated design basis accident under 10 CFR 54.4(a)(1).2 As a component of the Unit 1 and 2 structures, FENOC has indicated that an intended function of Unit 1 and 2's steel containment liners is to provide a structural pressure barrier.3 The concrete containment reactor building is conspicuously not listed as providing a pressure barrier in the license renewal application ("LRA"). Under 10 CFR 50.100, if the conditions revealed by any report would warrant the Commission to refuse to grant a license, permit, or approval on an original application, the Commission may be revoked, suspended, or modified. We believe that the current situation at Unit 1, as evidenced by the October 5th 2013 report indicating an unexpected containment liner penetration, in conjunction with the earlier reports, would have warranted the Commission to refuse to grant the LRA unless there were significant modifications to the proposed containment liner inspection program.
In addition, there have been instances of peeled coating and liner corrosion at Unit I dating back to 1992 as indicated in NRC Information Notice 97- 10: Liner Plate Corrosion in Concrete Containments.
2 Beaver Valley'Power Station License Renewal Application, Page 2.4-67.
3 Beaver Valley Power Station License Renewal Application, Table 2.4-22.
Under 10 CFR 50.34(a)(1)(D), an evaluation of a postulated fission product release from the core into the containment during operation at the ultimate power level contemplated must show that using the expected demonstrable containment leak rate "(A)
An individual located at any point on the boundary of the exclusion area for any 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> period following the onset of the postulated fission product release, would not receive a radiation dose in excess of 25 rem total effective dose equivalent (TEDE) [and] (B) An individual located at any point on the outer boundary of the low population zone, who is exposed to the radioactive cloud resulting from the postulated fission product release (during the entire period of its passage) would not receive a radiation dose in excess of 25 rem TEDE." To our knowledge, FENOC has not demonstrated that the containment of Unit 1, in the event that a hole similar to those discovered in 2009 or 2013 existed, would be able to meet the criteria under 10 CFR 50.34(a)(1)(D). Instead, it is assumed by both FENOC and the NRC that the current inspection regime is sufficient to guarantee safety.
However, as described below, the current containment liner inspections do not adequately assure the functionality of the containment liner.
The containment liners at Beaver Valley are examined under a number of programs including the 10 CFR Part 50, Appendix J Program, Option B; the ASME Boiler and Pressure Vessel Code,Section XI, Subsection IWE Program; supplemental random volumetric examinations; 4 and supplemental non-random volumetric examinations.5 In addition, FENOC agreed to evaluate if an appropriate/applicable6 statistical method exists to gain additional insight into potential liner degradation.
Citizen Power believes that the current inspection methodology for Unit 1 should be modified in order to eliminate or at least minimize the amount of time that there is a breach of the containment liner.
For their 10 CFR Part 50, Appendix J Program containment integrated leak rate testing ("ILRT"), FENOC has opted to use Option B, the performance-based approach.
Under this approach, an ILRT of the overall leakage rate is performed once every ten years.7 Prior to the ILRT, the exposed interior surface is visually examined to detect and repair any degradation. The NRC has determined that FENOC's Appendix J Program is consistent with the Generic Aging Lessons Learned ("GALL") Aging Management Program ("AMP") XI.S4 and further found that the-ILRT approach will provide assurance that the containment liners at BVPS will continue to perform their intended functions for the period of extended operation.8 As stated in the GALL AMP XI.S4, "Implementation of Option B for testing frequency must be consistent with plant-specific operating experience.'9 In our opinion, performing an ILRT on Unit 1 once every ten years is not consistent with current experience since breaches of the containment liner have now occurred twice and a ten year inspection interval can leave a breach undetected 4 Unit I License Renewal Commitment 32, Unit 2 License Renewal Commitment 33.
5 Unit I License Renewal Commitment 33, Unit 2 License Renewal Commitment 34.
6 Unit I License Renewal Commitment 35, Unit 2 License Renewal Commitment 36.
This is in the process of being extended to fifteen years as per ML I3067A219.
8 Beaver Valley Safety Evaluation Report, Pages 3-12 to 3-14.
9 NUREG- 1801, Rev.2,Section XI.S4. Page, XI S4-2.
for up to ten years. Additionally, in the case of both the 2009 and 2013 breaches, the hole was repaired and an integrated leak rate test was passed.. However, the relevant inquiry for Unit 1, given its history of corrosion and the possibility of an existing containment liner breach, is not whether the containment passes a leak rate test as specified by Appendix J to 10 CFR 50 after a hole is patched, but whether it could pass a leak ratetest with a hole in place.
The second inspection program under 10 CFR 50.55a imposes the inservice inspection ("ISI") requirements of the American Society of Mechanical Engineers Boiler and Pressure Vessel Code,Section XI, Subsection IWE on steel containment liners. The main ISI method used under IWE is visual examination though volumetric examination and surface examination may also be necessary.' 0 These examinations typically use ten year inspection intervals under inspection Program B. " FENOC, in response to the corrosion found in 2006, added additional examination requirements to the IWE Program; the examination of paint or coatings by a VT-3 inspector when removed for further inspection and the use of surface or volumetric examinations of areas where2 the visual examination detected flaws that could impact the leak tightness of the liner.'
However, since the IWE Program is mainly based upon visual examinations, it has a difficult time detecting corrosion on the exterior of the containment liner.
According to 10 CFR 50.55a(b)(2)(ix)(A)(i), "[t]he applicant or licensee shall evaluate the acceptability of inaccessible areas when conditions exist in accessible areas, that could indicate the presence of or result in degradation to such inaccessible areas." Similarly, GALL AMP X 1.S 1 states that the ASME Code XI paragraph IWE-1240 requires augmented examinations of containment surface areas that are subject to degradation.
FENOC has indicated that loss of material due to corrosion is not significant for inaccessible areas (embedded Containment steel liner) based upon several conditions being met such as the concrete meeting American Concrete Institute ("ACT") Code 318 and the moisture barrier being monitored for aging effects by the ASME Section, Subsection IWE program.' 3 In addition, FENOC has claimed that Units 1 and 2 do not meet the criteria for ASME Code augmented examinations as defined in ASME Code Section XI paragraph IWE-1240. "4 However, after the second breach of the containment liner, it is apparent that these conclusions are incorrect. The amount of construction debrisremaining is unknown, but the two separate breaches do indicate that it is possible that there could be further degradation of the inaccessible areas. The current IWE Program for Beaver Valley Unit 1 does not adequately detect corrosion on the exterior side of the containment liner and does not properly apply IWE- 1240's requirement of augmented examinations 5 including volumetric examinations on the inaccessible areas of the containment liner.'
'0NUREG- 1801, Rev.2,Section XI.S I. Page, XI SI - I.
' NUREG-1801. Rev. 2.Section XI.S I. Page XI S 1-3.
12 Beaver Valley Safety Evaluation Report, Pages 3-106.
13Beaver Valley Power Station License Renewal Application, Page 3.5-47.
14 Beaver Valley Safety Evaluation Report, Pages 3-107.
"5NUREG- 1801, Rev. 1.Section XI.S1. Page XI SI-3. ASME Code XI paragraph IWE-1241(b) states that areas requiring augmented examinations include "interior and exterior containment surface areas that are In addition to these inspection requirements, both Unit 1 and Unit 2 at Beaver Valley have added UT examinations of the containment liner as relicensing conditions.
The current corrosion issues with the containment liner appear to be the result of two separate mechanisms, each of which prompted a separate UT examination procedure. The first corrosion mechanism is associated with construction debris left in the concrete adjacent to the steel containment liner. There have now been two through-wall breaches of the containment liner in 2009 and 2013 attributed to construction debris left behind in the concrete. After the first through-wall breach in 2009, the NRC required that FENOC perform non-random UT examinations of eight areas in Units 1 and 2 that were determined to be more likely to have leftover construction materials. There was no confidence level associated with these non-random inspections and these examinations failed to discover the location of the 2013 through-wall breach. Citizen Power believes that the current examination regimen does not adequately assure that the containment liner is capable of performing its intended function as a leak tight barrier and recommends that a supplemental inspection methodology is introduced for at least Unit 1.
The second corrosion mechanism is the pitting corrosion that appeared in 2006 for which the root cause was never definitively determined. ' 6 In response to the pitting corrosion, the NRC required that FENOC undertake random UT examinations of the containment liner to determine whether there is an underlying pitting corrosion problem.
Any corrosion that was found to be the result of construction practices would not be counted as a failure for the purpose of these examinations. These inspections were intended to ensure that there would be a 95% confidence that 95% of the containment liner did not exhibit pitting corrosion greater than 10% of the nominal thickness. Citizen Power does not agree with the current methodology and disagrees that it is designed to detect corrosion of the containment liner with the stated degree of statistical reliability.
As evidenced by the latest breach, it is clear that the current condition of the steel containment liner for Unit 1 is unknowable under the current inspection programs.
Although it is possible that there are no other locations with construction materials embedded next to the containment liner, it is also possible that there are one or more areas that currently have corrosion that will eventually progress to the point of where the liner is again breached. Since both the 2009 and 2013 breaches were discovered by a visual inspection after they had gone through-wall, it is apparent that the containment liner had operated in a compromised state for a not insignificant amount of time during the last few years. Given the time between visual examinations of the inside of the containment liner and ILRTs combined with the possibility of additional areas of corrosion, it cannot be assumed that the containment liner will be intact in the event there is a design basis accident. As a result of this situation, either additional inspection procedures must be put into place to ensure that there are no serious ongoing corrosion subject to excessive wear from abrasion or erosion that causes a loss of protective coatings, deformation, or material loss."
16 "The probable cause was identified as corrosion of the liner that occurred during construction where the liner was exposed to oxygen and water." Beaver Valley Power Station License Renewal Application. Page 3.5-47.
issues with the containment liner or, in the alternative, FENOC must demonstrate that Unit 1 can operate in compliance with the requirements of Title 10 of the Code of Federal Regulations and Beaver Valley's. existing technical specifications regarding leak rates when an existing containment liner breach is in place. Citizen Power believes that the following measures would ensure that Unit l's containment liner is properly serving its function as a leak-tight barrier:
- 1. Unit 1 should supplement its ILRTs by requiring on-line monitoring.
On-line monitoring ("OLM") has the advantage of being able to provide a continuous indication of leakage rates over a period of time. OLM has the advantage of being able to detect leaks within one day to several weeks, minimizing the amount of time that the containment liner is not able to perform its function as a leak tight barrier. Continuous OLM has been successfully used in many European nuclear power plants.
- 2. Future holes in the containment liner should be subjected to leak rate testing before and after repair.
Both FENOC and the NRC have been unclear regarding what components are being credited as part of the containment system in place at Beaver Valley. Specifically, although only the steel containment liner is credited as providing a leak-tight barrier in the LRA, references to the concrete as performing a containment function have been mentioned by both FENOC and the NRC. For instance, in a 3/8/10 letter from Eric Leeds to Citizen Power, the Mr. Leeds states "...a successful ILRT does provide assurance, as a defense-in-depth indication, that the whole containment system (steel liner and 54-inch thick concrete) is functioning as intended, consistent with regulatory requirements."
Although the concrete would undoubtedly function to some degree as a containment barrier, it is unclear to what extent it would do so. Citizen Power believes it would be helpful to determine what the impact of a containment liner hole is upon actual leak rates at design basis accident pressures in order to accurately assess the safety implications for both Beaver Valley and other plants.
- 3. IWE inspections of the containment liner should be performed under IWE- 1240.
As indicated above, the corrosion history of the containment liner for Unit 1 indicates that there are conditions present (namely stranded construction materials) that could result in degradation to the containment liner. This trend of corrosion and containment liner breaches requires that the entire containment liner be subject to augmented examinations under IWE-1240. Furthermore, since the containment liner is accessible from only one side, volumetric examination should be performed.17 After the 2009 hole in the containment liner, the NRC added a licensing commitment requiring non-random testing of the containment liner at eight locations for both Unit 1 and Unit 2. These non-random inspections were supplemental to the established inspection programs and were not established under IWE-1240 or as an alternative under 7 NUREG- 1801, Rev. 2.Section XI.S I. Page XI SI -3.
IWE-1240. Under IWE-1240 we believe that a much greater amount of the containment liner needs to be subject to volumetric examinations; 100% surface of the containment liner should be examined using UT unless a statistical sampling methodology can be developed under license commitment 35 for Unit 1 that ensures the proper functioning of the containment liner. This hypothetical sampling methodology should not mirror the 95/95 criteria used in the non-random examinations. The 95/95 criteria was designed to identify and flag corrosion well before it would impact the functionality of the containment liner and it was therefore proper that it allowed 5% of the liner to have corrosion greater than 10% of the nominal thickness. Instead, we believe that the proper goal for any IWE- 1240 inspection should be a 95% percent probability that the containment liner is 100% free of corrosion that impacts the ability of the containment to function as a leak-tight barrier.
- 4. The current random inspection methods do not accurately determine the amount of pitting corrosion of the containment liner.
Although the majority of this letter addresses our concerns regarding the latest through-wall breach of the containment liner, we would like to take this opportunity to expand on our objections to the random examination of the containment liner as indicated in our letter of May 1 6 th. In that letter, Citizen Power had asked the Nuclear Regulatory Commission ("NRC") to re-examine the inspection methodology of FENOC and the NRC's determination that FENOC had completed licensing commitments 32 through 34 for Unit 1. Citizen Power's concern was that FENOC's inspection methodology was designed to discover areas where the thickness of the containment liner is less than 90%
of the nominal thickness as opposed to meeting the license renewal commitment to sample areas in order to determine if degradation has taken place. One could argue that as long as the containment liner is 90% of nominal it does not matter if corrosion is occurring. However, degradation of the corrosion liner may be continuous and areas that are greater than 90% of nominal today may not be 90% of nominal tomorrow (or in twenty years). Citizen Power believes strongly that the amount of liner degradation rather than liner thickness should be what is measured in FENOC's random examinations.
Another concern is that by using wall thickness as a proxy for degradation, the applicable sampling methodologies are modified. Specifically, FENOC's reliance on binomial sampling, as described in chapter 21 of NUREG 1475, is misplaced. Any calculation of probabilities related to the assurance-to-quality process (e.g. the 95/95 criteria being used by FENOC) must meet the four requirements of a binomial experiment as explained on page 21-8 of NUREG 1475:
- 1) The experiment is conducted in n trials, or items, all of which are conducted under identical conditions.
- 2) The i"' trial (i = 1, ... , n) results either in a success (i.e., the attribute of interest is identified in the i"' item) and recorded as a success with Yi = 1 or in a failure with Yi = 0.
- 3) The probability of success, denoted by the parameter 7t, is consistent from trial to trial.
- 4) The n trials are independent.
When the attribute being examined was degradation, as indicated in the license commitments, these four requirements were arguably met. However, when the attribute was changed to the existing containment liner thickness, the sampling methodology violated two of the four requirements of a binomial experiment. This is because the determination of whether a location was a success or a failure became dependent on two separate characteristics, the amount of the degradation and the original thickness of the containment liner. Although the amount of degradation may be randomly distributed, the original thickness of the containment liner consists of steel plates which appear to have different thicknesses and is therefore unsuitable for a binomial experiment. First, the probability of success (defined as the wall thickness being greater than 90% of the nominal thickness) varies from trial to trial in contravention of rule 3 because of the variability of the original wall thickness. Second, the trials are not independent in contravention of rule 4 because the original thickness of the wall for a specific location would be associated with the original thickness of adjacent locations since variances in the manufacturing process related to the thickness of the steel plate are likely to be less within a given plate than between plates. It is this variation in the original thickness that would invalidate the use of NUREG 1475.
Given the identified issues with the original sampling procedure and the recent identification of a breach, Citizen Power respectfully requests that the NRC re-examine whether FENOC has completed license commitments 32-34 for Unit 1. In addition, in accordance with license commitment 35 for Unit 1, Citizen Power believes that any supplemental statistical analysis of the state of the containment liner should take into account the recently discovered breach in the sampling methodology. Finally, containment liner degradation has been identified as a potential issue with aging plants.
Citizen Power believes that given this fleet-wide issue, FENOC should perform an integrated leak rate test before patching the hole. This would provide information to the NRC regarding the validity of any existing computer modeling.
Sincerely, Theodore S. Robinson, Esquire Staff Attorney Citizen Power 2121 Murray Avenue Pittsburgh, PA 15217 Cc: Erin Bonney
.-.aon fitizen Power 2121 Murray Avenue Pittsburgh, PA 15217 Peter Bamford, Project Manager Office of Nuclear Reactor Regulation Mail Stop: 08B3 Washington, DC 20555-0001 S.-...-
I111ijil1*11i IIIIIIiIII IJJ IIli)) 11
/