ML18065B028

Provides Response to NRC GL 96-04 Re Boraflex Degradation in Spent Fuel Pool Storage Racks

ML18065B028
Person / Time
Site:	Palisades
Issue date:	10/24/1996
From:	Bordine T CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GL-96-04, GL-96-4, NUDOCS 9611010074
Download: ML18065B028 (12)

Text

i**

.~

(

consumers Power l'OWERINli llllClllliAN-S l'lllllillUS Palisades Nuclear Plant: 27760 Blue Star Memorial Highway, Covert, Ml 49043 October 24, 1996 U.S. Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 DOCKET 50-255 - LICENSE DPR PALISADES PLANT Thomas C. Bordlne Manager, Licensing RESPONSE TO NRC GENERIC LETTER 96-04: BORAFLEX DEGRADATION IN SPENT FUEL POOL STORAGE RACKS On June 26, 1996, the NRC issued Generic Letter 96-04, "Boraflex Degradation In Spent Fuel Pool Storage Racks." This Generic Letter requested the licensee with installed spent fuel storage racks containing the neutron absorber Boraflex, to provide an assessment of the physical condition of the Boraflex and to provide a description of proposed actions to confirm that the 5-percent subcriticality margin can be maintained for the lifetime of the storage racks. This submittal provides the requested 120 day response to the NRC Generic Letter 96-04.

The attachment to this letter contains specific responses to the NRC Generic Letter 96-04 requested information.

SUMMARY

OF COMMITMENTS This letteLG9ntairis one revised commitment and the commitmen! is as follows:

Consumers Power Company (CPCo) will perform blackness testing on the Region II spent fuel storage racks during the Spring of 1998. The need and timing of additional blackness testing will be based o"n the results of the 1998 testing and pool information considered to be good indicators of Boraflex

'" ~ 1 I condition (e.g., pool silica concentration).

jJ.. uldo f I 9611010074 961.024 PDR ADOCK **05000255 P

P~DR.

A CMS' ENERGY COMPANY

2 The original commitment was submitted to the NRC by CPCo letter dated May 23, 1995 and stated that periodic neutron attenuation testing would be performed of the Spent Fuel Pool Region II storage racks to monitor the rack Boraflex condition. The next performance of the periodic testing would have been scheduled approximately the first quarter of 1998. The frequency of subsequent testing was to have been determined based on the known condition of the rack Boraflex and industry experience. The new commitment replaces this original commitment to monitor rack Boraflex condition.

-~z-7.

~*-

~t~L Thomas C. Sardine Manager, Licensin*g CC Administrator, Region Ill, USNRC Project Manager, NRR, USNRC.

NRC Resident Inspector - Palisades Attachment

CONSUMERS POWER COMPANY To the best of my knowledge, the contents of this response to NRC Generic Letter 96-04: BORAFLEX DEGRADATION IN SPENT FUEL POOL STORAGE RACKS, is truthful and complete.

~~

Thomas *G. -Bsr-~iAe-Manager, Licensing Sworn and subscribed to before me this L day of g).~

1996.

~'W\\-~~.:*

Alora M. Davis, Notary Public Berrien County, Michigan (Acting in Van Buren County, Michigan)

My commission expires August 26, 1999

[Seal]

ATTACHMENT CONSUMERS POWER COMPANY PALISADES PLANT DOCKET 50-255 CONSUMl;:RS POWER COMPANY'S RESPONSE TO NRC GENERIC LETTER 96-04 BORAFLEX DEGRADATION IN SPENT FUEL POOL STORAGE RACKS 8 Pages

RESPONSES:

NRC GENERIC LETTER 96-04 BORAFLEX DEGRADATION IN SPENT FUEL POOL STORAGE RACKS NRC Generic Letter 96-04, "Boraflex Degradation In Spent Fuel Pool Storage Racks,"

requested information. Below is each request for information and the Consumers Power Company response.

The Palisades spent fuel pool storage racks are divided into Region I and Region II..

The Region I storage racks were manufactured by NUS and placed into the pool in 1977. The Region i-racks utiiize cinteredbornn-caibide plates iri its wa!!s as a neutron poison and can accommodate fuel assemblies enriched to 4.4 weight percent. The Region II storage racks were manufactured by Westinghouse and placed into the pool in 1987. The Westinghouse racks utilize Boraflex in their walls as a neutron poison and can accommodate fuel assemblies with an equivalent new fuel enrichment of 1.5 weight percent. The as-manufactured Boraflex contained a minimum boron-1 O areal.*

density of 0.006 gm/cm2.

NRC Reguested Action:

1.

Provide assessment of the physical condition of the Boraflex, including any deterioration, on the basis of current accumulated gamma exposure and possible water ingress to the Boraflex and state whether a subcritical margin of 5 percent can be maintained for the racks in unborated water. Monitoring programs or calculational models in effect or being developed, or an estimation of anticipated*

concerns based on the specific rack design, are considered an appropriate basis for this response.

Consumers Power Company Response:

The physical condition of the Boraflex at Palisades is assessed to be in good overall condition and fully capable of performing its criticality control function. This assessment is based upon the 1995 blackness testing results and the 1993 coupon inspection data. The blackness test data indicates that minimal gapping has occurred in the panels and that the gapping is related to the accumulated exposure. The surveillance coupon data indicates that the amount of thinning due to interaction with pool water has been minimal.

The blackness test data extrapolated over all the storage cells in Region II predicts:

1.

More than half of the Boraflex panels do not have any gaps greater than 0.5

inches,

2.

The maximum size of any gap is one inch or less,. and

3.

The maximum accumulated amount of gap in any panel is less than 2.3 inches including measurement uncertainty.

2 A detailed discussion of the results of the 1995 blackness test may be found in the section describing the results of blackness testing performed on the racks. Blackness testing is capable of detecting the presence of gaps in Boraflex due to shrinkage or

'vvas!:lGut, but does not provide a quantitative measure of the amount of boron-10 loss through thinning as a result of silica dissolution. However, neutron attenuation. tests on Boraflex from the 1993 surveillance coupon inspection revealed that the Boron-10 areal density had not changed in the Boraflex that remained, even though most of the Boraflex in the coupons was missing.

The 1993 Boraflex surveillance coupon inspection consisted of removing several Boraflex surveillance coupons to conduct visual inspections and neutron attenuation testing. The coupons had been placed in the spent fuel pool in 1988. Inspection of the coupons revealed that most of the Boraflex material was missing. 1 Neutron attenuation testing was performed on the Boraflex material remaining in the coupons. Test data showed that the Boraflex boron-10 areal density had not changed significantly from the as-manufactured boron-10 areal density. This result indicates that minimal thinning had occurre~ to the Boraflex sheets in the coupons. This result can be extrapolated to the Region II Boraflex panels since a _rack has a much "tighter" design than the surveillance.coupons. A "tighter" design reduces the interaction of Boraflex with pool water, thus minimizing a necessary factor in the phenomenon of Boraflex thinning.

Presently, a subcritical margin of 5 percent can be maintained by the Palisades Region II racks. A criticality calculation (EA-SFP-94-02, Determination of Reactivity Changes Due to Gaps in Boraflex Panels of Regiod 2 Spent Fuel Pool Racks) has been performed that demonstrates a 5 percent subcriticality margin can be maintained in the rack with up to 5 inches of gap in each Boraflex panel. The maximum amount of gap detected by the blackness test in any one Boraflex panel was 2.3 inches, including 20 percent measurement uncertainty. If every Region II Boraflex panel were cons.ervatively assul]led to contain 2.3 inches of gap, the rack would still be bounded by the criticality analysis and the subcritical margin would be greater than 5 percent.

This circumstance resulted in_ Palisades License Event Report LER 93-007, Degradation of Boraflex Neutron Absorber in Surveillance Coupons. The 1995 blackness testing demonstrated that the washout that had occurred in the coupons was not indicative of the state of the Boraflex panels in the Region II racks.

3 This scenario reveals that margin exists for Boraflex thinning and gap growth which the industry has observed to occur with increased exposure to radiation.and pool water.

NRC Requested Action:

2.

A description of any proposed actions to monitor or confirm that this 5 percent subcriticality margin can be maintained for the lifetime of the storage racks; also describe what corrective actions could be taken in the event the 5 percent margin cannot be maintained.

Consumers Power Company Response:

Palisades has adopted a defense-in-depth philosophy to monitor the potential for degradation of Boraflex in the Region II spent fuel storage racks. Since actual physical confirmation of the condition of the Boraflex panels in the racks is not possible, the ability of the Boraflex to perform its design function must be inferred from data that the industry has shown to indicate Boraflex condition. Defense-in-depth provides a logical methodology that utilizes obtainable information to predict the condition of the Boraflex and provides mitigational options should the Boraflex be determined not able to perform its design function. The implementation of the levels of defense will be dependent on conclusions drawn using information from the level of defense elements.

The first level of defense consists of surveillance tools which are used as indicators of the condition of the Boraflex. These surveillance tools include blackness testing, monitoring spent fuel pool silica concentrations, computer simulation of Boraflex performance, and Boraflex industry experience. None of these surveillance tools by themselves provides sufficient-information to determine the condition of Boraflex in the racks. However, taken as a whole, the cumulative surveillance tool information provides data indicative of the severity of Boraflex degradation.

Blackness testing at Palisades was first performed in 1995 and is currently scheduled for 1998. The storage cells tested in 1995 will be tested again in 1998 and, compared with the 1995 baseline data. Blackness testing trends provide gross indication of Boraflex degradation over time.

Spent fuel pool silica level can be indicative of Boraflex degradation due to the dissolution of Boraf!~x silica. Th~ pool reactive silica concentration is currently measured on a weekly basis and trended. The pool silica level is compared to that of the industry. Elevated pool silica levels (relative to the industry) may be an indication of Boraflex degradation. The trend of the pool silica is evaluated since increasing silica trends may indicate Boraflex degradation.

e.

Palisades is presently in the process of initializing the computer code RACKLIFE, developed by the Electric Power Research Institute (EPRI) to simulate the behavior of Boraflex in a spent fuel pool system. This computer simulation tracks the behavior of each panel of Boraflex in a rack installation subject to the irradiation history of each panel and conditions in the pool. RACKLIFE predicts the quantity of Boraflex and boron-1 O lost as a function of service history. RACKLIFE can also be used as a spent fuel/Boraflex rack management tool. Using RACKLIFE information, the storage of fuel assemblies in the Boraflex racks may be managed in a way that minimizes Boraflex degradation.

4 As part of the ongoing EPRI Boraflex program, an industry-wide database is maintained with respect to pooi siiica histories* and associated Boraflex performance. Palisades will continue to compare its data with the data of other, plants that have racks of similar design and vintage to obtain a sense of the relative condition of the BorafleX. EPRI continues to provide a forum for the industry to exchange information and ideas that relate to the Boraflex issue.

Should the. first level of defense elements indicate non-acceptable Boraflex performance, a second level of defense may be implemented. The second level involves quantitative in-situ measurement of the boron-10 areal dens!ty to ascertain the actual blackness of the Boraflex panels. EPRI is currently developing the Boron Areal Density Gage for Evaluating Racks (BADGER) Test which provides panel average and local boron-1 O areal density measurement with a measurement uncertainty comparable to the range of variation in the as-manufactured areal density of Boraflex. The utilization of BADGER or other in-situ boron-10 measurement methods will depend on the degree of estimated Boraflex degradation. If the degradation appears to be significant enough, the third level of defense may be initiated, bypassing the second level of defense:

Should the results of the first and/or second levels of defense indicate suspect Boraflex performance, a third level of defense would be implemented. The third level of defense elements assume that Boraflex degradation is significant and use other mitigation measures to ensure that the storage rack 5 percent subcriticality margin is maintained.

The third level of defense consists of mitigation measures including, but not limited to additional analysis, administrative controls, credit for soluble boron in the spent fuel pool, and/or the use of absorber inserts to provide additional negative reactivity.

Palisades has previously performed criticality analyses on the Region II racks due to*

the severe degradation of the Boraflex in the surveillance coupons in 1993. Following this discovery, a Region II criticality analysis (Criticality Analysis to Support Current Fuel Storage in the Palisades Region 2 Spent Fuel Racks With No Boraflex Panels, Westinghouse) was performed assuming zero Boraflex and zero soluble boron to ascertain the maximum fuel assembly equivalent fuel enrichment that could be stored in

Region II with no Boraflex. This analysis concluded that the acceptable maximum fuel assembly equivalent new fuel enrichment would be 1.0 weight percent.

A second criticality analysis (Criticality Analysis of the Palisades Region 2 Spent Fuel Storage Racks Without Considering the Boraflex Poison Panels, Westinghouse) was performed again assuming zero Boraflex and zero soluble boron for a three out of four checkerboard storage configuration. The fourth storage cell would be left empty or contain a control blade. The maximum fuel assembly equivalent new fuel enrichment was calculated to be 1.46 weight percent.

5 A third criticality analysis (EA*SFP-94-02) was performed assuming zero soluble boron which calculated that e~:rch -aoraflex panel GSt,J!d have a total of five inches of gap without reducing the 5 percent subcriticality margin for the racks.

NRG Requested Action:

3.

A description of the results from any previous post operational blackness tests and

-state whether blackness testing, or other in-situ tests or measurements, will be periodically performed.

Consumers Power Company Response:

Blackness testing was performed on the Palisades Region II spent fuel storage racks during January 1995 (EA-MLB-95-01, Spent Fuel Pool Region II Boraflex Condition).

The evolution tested 31 storage cells. 17 additional cells were effectively tested because they shared walls with cells that were blackness tested for a total of 48 storage cells tested. This represents 14% of the usable Region II cells.

The blackness testing results revealed that the Boraflex panels in the 48 cells tested were in satisfactory condition. The 48 storage cells contained 98 full-length Boraflex panels. Of the 98 Boraflex panels tested, 63 (64%) panels had no measurable gaps.

There were 45 measurable gaps distributed among the remaining 35 (36%) panels.

The minimal detectable gap size for the test was considered to be approximately 0.5 inches with a measurement uncertainty of 20%. The average gap size for the 45 measurable gaps was 0. 7 inches. The largest single gap measure.d was approximately one inch. The maximum amount of gap measured in any _one pariel was 1.9 inches resulting from three gaps in the panel occurring at different heights. The number of Boraflex panels with more than one gap was 7 (7% ). The gaps detected were of the

type induced by Boraflex shrinkage. 2 This type of gapping has been observed in the industry.

6 The state of the Boraflex panels in the Region II cells tested are a good representation of the Boraflex condition in the remaining untested Region II cells. The cells tested included the "accelerated exposure region" and "regio"n of longest exposure." The accelerated exposure region consists of storage cells that have purposely received the maximum radiation exposure relative to other cells in Region II. The region of longest exposure consists of cells that have been purposely exposed to radiation longer than other cells of Region 11. All of the Boraflex panels with more than one gap were located in close proximity to the accelerated exposure region. The cells tested in the region with the longest exposure didfiot-exhisit-this trend. Ibis result agrees with the EPRI conclusion that the magnitude of the accumulated exposure is a major contributor in Boraflex degradation (EPRI NP-6159, An Assessment of Boraflex Performance in Spent-Nuclear-Fuel Storage Racks).

The untested cells in Region II have received less radiation exposure than the. area tested. Likewise, the flow characteristics in the untested cells are not significantly:

different than those in the. area tested. Thus, the untested Boraflex panels are not expected to exhibit any degradation significantly greater than that seen in the tested accelerated exposure region. If the results of the 1995 blackness test at Palisades are extrapolated to the remainder of Region II, the Boraflex panels in the Region II racks are in satisfactory condition, fully capable of performing their criticality control function.*

Palisades has committed to the NRC to perform blackness testing during the Spring of 1998. The need and timing of additional blackness testing campaigns will be based on the results of the 1998 testing and pool information considered to be a good indicatqr of Boraflex condition (e.g., pool silica concentration).

NRC Reguested Action:

4.

Provide chronological trends of pool reactive silica levels, along with the timing of significant events such as refuelings, pool silica cleanups, etc., and describe how these pool silica.levels relate to Boraflex performance.

2 Gaps of one inch or less are known to occur due to radiation-induced shrinkage of the Boraflex organic binder which results in the subsequent tearing of the Boraflex in response to the shrinkage stresses (EPRI TR-101986, Boraflex Test Results and Evaluation).

Consumers Power Company Response:

The following is a summary of the pool silica history from March 1991 to the present as shown below on the graph of Palisades Spent Fuel Pool Silica History.

Palisades Spent Fuel Pool Silica History 3/19/91 10/5/91 4122/92 11/8/92 5/27/93 12/13/93 7/1/94 1/17/95 8/5/95 2/21/96 9/8/96 7

Spent fuel p'ool silica was not analyzed prior to March 1991 and then infrequently until.

October 1992. The pool silica concentration increased from less than 2 ppm to slightly more than 3 ppm during that time period. Beginning in October 1992, the pool boron concentration was increased in preparation for dry fuel storage loading. This resulted in an increase in pool silica because the pool boron was being raised using water from the Safety Injection and Refueling *Water Tank (SIRWT) which had a silica concentration of 3-4 ppm. Prior to the Cycle 10/11 refueling outage, pool silica had reached 6-7 ppm. This declined to 3-4 ppm during the refueling outage. Following the outage, pool boron was once again increased for dry fuel storage activities using the SIRWT. By the time dry fuel storage loading resumed, pool silica had increased to 7 ppm. Pool silica subsequently increased to 20 ppm over the next eight months during dry fuel storage loading and associated activities. Much of this silica increase occurred by step changes of 1-3 ppm. These relatively large silica increases coincided with several of the dry fuel storage loadings. It is suspected that these silica increases resulted from the dry fuel storage activities since no further such step increases have been observed following the cessation of dry fuel storage activities. Pool silica dropped to 9 ppm during the Cycle 11/12 refueling outage. Early in Cycle 12, evaporator bottoms were added to the pool which increased pool silica to 11 ppm.

Since September 1995, pool silica has averaged 11 ppm with little or no increase.

\\.

~

The dates of Palisades operating cycles since the Region Ii racks have been installed are listed below.

PALISADES CYCLE LENGTHS Cycle 7 03/04/86-08/08/88 Cycle 8 11 /29/88~09/15/90 Cycle 9 03/15/91-02/07 /92 Cycle 10 04/18/92-06/04/93 Cycle 11 11 /08/93-05/22/95 Cycle 12 08/21/95-8 The relationship of the current pool silica level with the condition oft.he Boraflex is.

complicated by past dry fuel storage activities. What is known is that the present pool silica level is higher than it would be without silica additions from dry fuel storage activities. With that in mind, the following two generalization can be made: 1) the pool silica from. Boraflex degradation is lower than the levels that have been demonstrated to be of concern (i.e., <10 ppm); and 2) the flat silica trend during the past 12 months reveals that the rate of Boraflex degradation based on pool silica concentration is slow.

The conclusion reached from these generalizations is that currently.the Boraflex in the

• Region II racks is in satisfactory condition based on pool silica concentration data.

Based on all available information, the Palisades Region II spent fuel storage racks containing Boraflex as a neutron poison are fully capable of maintaining the 5 percent subcriticality margin, as required by General Design Criteria (GDC) 62. Therefore, Palisades is in compliance with GDC 62 of Appendix A to Part 50 of Title 10 of the Code of Federal Regulations.