ML20129G110

Topical Rept Evaluation of WCAP-14416, W Spent Fuel Rack Criticality Analysis Methodology. Rept Acceptable for Ref in Licensing Application

ML20129G110
Person / Time
Issue date:	10/25/1996
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20129G031	List: ML20129G026 ML20129G110
References
NUDOCS 9610300008
Download: ML20129G110 (20)

Text

. _ _ . . . _ _ _ __ ._ _ . _ _ _ _ _. __ __ _ ~ . _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _

i

~

pa aeog?$

~ f- .. UNITED STATES l

$ E NUCLEAR REGULATORY COMMISSION f WASHINGTON, D.C. 20555-0001

\, ..... / ,

ENCLOSURE SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION i RELATING TO TOPICAL REPORT WCAP-14416-P

" WESTINGHOUSE SPENT FUEL RACK CRITICALITY ANALYSIS METHODOLOGY" WESTINGHOUSE ELECTRIC CORPORATION }

I 1.0 IhTR000CTION In a submittal of July 28, 1995 (Ref. 1), the Westinghouse Owners Group (WOG) i requested U.S. Nuclear Regulatory Commission (NRC) review and approval of topical report WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," June 1995 (Ref. 2). The report presents the current Westinghouse methodology for calculating the effective multiplication factor, k , of spent fuel storage racks in which no credit is taken for soluble b,o,r,on except under accident conditions. The report also presents a new proposed procedure for crediting soluble boron in the spent fuel pool water when performing storage rack criticality analysis for Westinghouse fuel storage pools. A revision to the methodology was submitted on October 23, 1996 (Ref. 28), based on recommendations by the NRC Committee to Review Generic Requirements (CRGR).

General Design Criterion (GDC) 62 (Ref. 3) states that " criticality in the fuel storage and handling system shall be prevented by physical systems or processes, preferably by use of geometrically safe configurations." The NRC n o greater than 0.95) i i

has to established comply with GDC a 5-percent 62 (Ref subcriticality margin (k , biases and uncertainties 4). All of the applicablY '

l should be combined with k to provide a one-sided, upper tolerance limit on k suchthatthetruevaYuewillbelessthanthecalculatedvaluewitha 9Y, percent probability at a 95-percent confidence level (Ref. 5). The

proposed new methodology would permit the use of spent fuel pool soluble boron to offset these uncertainties to maintain k less than or equal to 0.95.

However, the spent fuel rack k , calculatio,n,,would remain less than 1.0 (subcritical) when flooded witN unborated water with a 95-percent probability at a 95-percent confidence level.

2.0 SUP91ARY OF THE TOPICAL REPORT l I

Section 1.0 of the report is an introduction, stating the purpose of the  !

! report and summarizing the individual sections. Section 2.0 explains the computer codes used in.the evaluation of the spent fuel rack k,,, calculations and presents benchmark results. In Section 3.0, the assumptions used to model

the spent fuel storage racks and the reactivity effects of biases and

uncertainties are presented. Section 4.0 discusses reactivity equivalencing 9610300008 961025 PDR TOPRP EMVWEST PDR

j 2 methods that credit fuel assembly burnup and integral fuel burnable absorbus i (IFBA). Section 5.0 describes postulated accidents that are considered in the i spent fuel rack criticality analysis. Section 6.0 of the report, in j conjunction with the supplement (Ref. 28), defines how credit for spent fuel

pool soluble boron will be applied in the reactivity calculations.

3.0 TECHNICAL EVALUATION

The Westinghouse spent fuel rack criticality analysis methodology presented in WCAP-14416-P, and modified by Reference 28, provides a detailed description of both the current methodology, which has been used for many years by Westinghouse to calculate the reactivity of spent fuel' storage racks, and a proposed new methodology with which partial credit for soluble boron in the pool water would be taken. The review of the propm ed new methodology, given in Section 3.7 below, focused on the approximations and assumptions used as well as on revised technical specifications and analysis of dilution events required when crediting baron. The following evaluation is based on the j material presented in the topical report, supplementary information (Ref. 28),

discussions with Westinghouse staff, and responses to our requests for adcitional information (Refs.14 and 26).

3.1 Comouter Code Methods and Benchmarkino Reactivity calculations for the spent fuel storage racks are performed with the KENO-Va (Ref. 6) three-dimensional Monte Carlo computer code. A 227 energy group cross section library is created by NITAWL-II (Ref. 7) and XSDRNPM-S (Ref. 8) from ENDF/B-V data (Ref. 9). This method has been used to analyze a set of 32 low-enriched, water-moderated, U0 critical experiments to establish a method bias and uncertainty (Refs. 10, 11,, 12, 13). These experiments cover a #ange f enrichments varying from 2.35 weight percent (w/o) to 4.31 w/o Uns separated by various materials (B C, 4 borated aluminum, stainless steel, water) at fuel rod spacings from 0 to 6.56 cm. These experiments simulate current PWR spent fuel storage racks as realistically as possible with respect to parameters important to reactivity such as enrichment, assembly spacing, and neutron absorber worth. In response to a ,

staff question (Ref. 14), WOG stated that no significant biases or trends were  !

observed as a function of lattice or fuel parameters, including enrichment.  !

The staff concludes that the KENO-Va benchmarking data is'sufficiently diverse I to establish that the method bias and uncertainty will apply to spent fuel storagerackconditionssingartothosecurrentlyinusecontainingfuelrod enrichments up to 5.0 w/o U To minimize the statistical uncertainty of the KENO-Va calculations, at least  !

100,000 neutron his'

• ies are accumulated in each calculation. Experience has  !

shown that this nur~ r of histories is sufficient to assure convergence of KENO-Va reactivity c.lculations. In addition, edits from the KENO-Va calculations provide a visual inspection of the overall convergence of the ,

results.

A method bias of 0.0077 results from the comparison of KENO-Va calculations with the average measured experimental k ,,. The standard deviation of the bias value is 0.00136 ok. The95-percenI. probability /95-percentconfidence

3 level (95/95) one-sided tolerance limit factor for 32 values is 2.20 (Re f. 15) . Thus, there is a 95-percent probability with a 95-percent confidence level that the uncertainty in reactivity due to the method is not greater than 0.0030 Ak (2.20 x 0.00136).

The PHOENIX-P (Ref. 16) transport theory computer code is used to determine reactivity changes due to possible variations (tolerances) in material characteristics and mechanical dimensions in the fuel assembly and spent fuel racks, changes in pool conditions such as temperature and soluble baron, and fuel burnup. PHOENIX-P is a depletable, two-dimensional, multigroup, discrete-ordinates transport theory code that uses a 42 energy group nuclear data library.

PH0ENIX-P has been compared with critical experiments (Refs. 17, 18, 19, 20).

The PH0ENIX-P reactivity predictions acree very well with the critical experiments, showing no significant bias or trends as a function of lattice or fuel parameters. The range of lattice parameters and configurations in the critical experiments encompassed present fuel storage configurations as realistically as possible.

PHOENIX-P has also been compared with isotopic measurements of fuel discharged from Yankee Core 5 (Ref. 21). The PH0ENIX-P predictions agree very well with measurements for all measured isotopes throughout the burnup range.

Based on the above, we conclude that the analysis methods described are acceptableandcapableofpredictingthereactivityofPWRspentfuelstorge racks containing assemblies with maximum fuel rod enrichments of 5.0 w/o U with a high degree of confidence.

3.2 KENO-Va Reactivity Calculations KENO-Va is used to establish a nominal reference reactivity, using fresh (unirradiated) fuel assemblies and nominal rack dimensions, that satisfies the The following assumptions are used in the 0.95 k ' ion: acceptance criterion.

calcu1It (1) The nominal spent fuel rack storage cell dimensions are used.

(2) Fuel assembly parameters for all :ssembly types considered for storage in the spent fuel pool are evaluated. These parameters include number of fuel rods per assembly, fuel rod clad material, fuel rod clad outer diameter, fuel rod clad thickness, fuel pellet outer diameter, fuel pellet density, fuel pellet dishing factor, fuel rod pitch, control rod guide tube material, number of guide tubes, guide tube outer diameter, guide tube thickness, instrument tube material, number of instrument tubes, instrument tube outer diameter, and instrument tube thickness.

(3) The nominal fresh fuel enrichment loaded into each fuel pin is modeled.

The pin locations within a fuel assembly with multiple enrichments are considered, if applicable. The maxim g fuel rod enrichment loaded into the fuel rods is limited to 5.0 w/o U

l*

l l

4 (4) The nominal values for theoretical density and dishing fraction of the fuel pellets are modeled.

(5) If axial blankets are modeled, the length and enrichment of the blanket fuel pellets are considered.

(6) No amount of23U ' or U 23

' is modeled in the fuel pellet.

(7) No amount of material from spacer grids or spacer sleeves is modeled in the fuel assembly.

l (8) No amount of burnable absorber poison material is modeled in the fuel assembly.

(9) No amount of fission product poison material is modeled in the fuel assembly.

(10) The moderator is pure water (no boron) at a zemperature of 68'F and a density of 1.0 gm/cc.

(11) If credit is taken for any fixed neutron-absorbing poison material panels present (except Boraflex), they are modeled using the as-built or manufacturer-specified poison material loadings and dimensions.

Because of the significant Boraflex deterioration observed in some spent fuel racks, additional conservative assumptions are required for racks containing Boraflex as neutron absorber. These assumptions are not part of this technical review but will be reviewed on a case-by-case basis.

(12) If all storage cells are not loaded with the same fuel assembly type and enrichment, the specific storage configuration will be modeled. l Different types of configurations include checkerboard patterns, empty cell locations, specific pool configur ations, and other layouts as defined.

is calculated with KENO-Va Using to showthese assumptions, that k,,, is less thanthe or spent equal fuel rack k,Sith to 0.95 no credit for soluble boron. A tempera ture bias, which accounts for the normal operational temperature range of the spent fuel pool water, and the method bias, determined from the benchmarking calculations, are included. In additio1. if neutron absorber panels are used, a reactivity bias is added to correct far the modeling assumption that individual B D atoms are homogeneously distributed within the absorber material rather than clustered around ear 5 B 4C particle. The staff concludes that these assumptions tend to maximize tFo rack reactivity and are, therefore, appropriately conservative and accep'.a.ble.

! 3.3 PH0ENIX-P Tolerance / Uncertainty Calculations

, PH0ENIX-P is used to calculate the reactivity effects of possible variations l in material characteristics and mechanical / manufacturing dimensions. The

, following tolerances and uncertainties are considered:

l

.. ~ -- . -. ._ . . . . .

5 235 (1) Enrichment tolerance of 10.05 w/o U about the nominal fresh reference enrichments (2) Variation of 12.0% about the nominal reference V0 2 theoretical density (3) Variation in fuel pellet dishing fraction from 0% to twice the nominal dishing (4) Tolerance about the nominal reference storage cell inner diameter, center-to-center pitch, and material thickness i l

(5) Tolerances about the nominal width, length, and thickness of j j neutron absorber panels l

(6) Tolerances about the nominal poison loading of the neutron l absorbing panels, if the nominal poison loading assumed in the KENO-Va model is not the minimum manufacturer-specified loading (7) Asymmetric positioning of fuel assemblies within the storage cells 4 The manufacturing tolerance uncertainties are based on the reactivity difference between nominal and maximum tolerance values and, therefore, meet i the 95/95 probability / confidence level requirement. These uncertainties are combined statistically with the 95/95 calculation uncertainty on the KENO-Va

nominal reference k,,,

i' the benchmarking bias determined for the KENO-Va methodology.and the 95/95 me benchmarking bias of 0.0077 Ak, the water temperature bias, and the BThemegodology sel f-summation before shielding bias, comparison if applicable, against the 0.95 k are included in the final The following formuk,,fa is used to determine the 95/95 k,,, for tNe, limit.

spent fuel storage racks:

k,,, = komu + Bm.,w + B,, + B,,, + Boom 4

where:

l Li- nominal conditions KENO-Va k,,,

B .,w -

method bias determined from benchmark critical comparisons 1

B,,m, - temperature bias B, - B' self-shielding bias, if applicable B,

o

= E(tolerance . . .or. . . uncertainty,)2

i -

i 6

The staff concludes that the final k calculated using ' above methodology will satisfy the NRC guidance that tNe' fuei storage rack reactivity be less

! than or equal to 0.95 when fully flooded with unborated water, including all

' appropriate uncertainties at the 95/95 probability / confidence level (Refs. 4, 5). Therefore, the documented methodology is acceptable, i

{ 3.1 Fuel Assembly Burnuo Credit Reactivity equivalencing is used to allow s t higher initial enrichments (up to 5.0 w/o U}grage ofthose

) than fuel assemblies with found acceptable i using the previously described methodology. This concept is predicated upon the reactivity decrease associated with fuel depletion. For burnup credit, a series of reactivity calculations are performed with PHOENIX-P to generate a i

set of initial enrichment versus fuel assembly discharge burnup ordered pairs

} that all yield an equivalent k

are stored in the spent fuel sI,o,ra(ge racks.no greater than 0.95) when fuel assemblies j The CINDER computer code (Ref. 22) was used to determine the most reactive 1 time after reactor shutdown of an irradiated fuel assembly. CINDER is a i point-depletion code that has been widely used and accepted in the nuclear l industry to determine fission product activities. The fission products were permitted to decay for 30 years after shutdown and the fuel reactivity was i found to reach a maximum at approximately 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />. At this time, the major

fission product poison, Xe*, has nearly completely decayed away. Therefore, i

the most reactive time for an assembly after sgtdown of the reactor can be

. conservatively approximated by removing the Xe

An uncertainty associated with the depletion of the fuel assembly and the j reactivities computed with PHOENIX-P is accounted for in determining the

reactivity equivalence limits. This uncertainty is based on the PHOENIX-P i

comparisons to the measured isotopics from the Yankee Core 5 experiments and

is used to account for any depletion history effects or calculational uncertainties not included in the depletion conditions that are used in

PHOENIX-P. The staff concludes that this uncertainty, which increases

j. linearly with burnup from o at 0 burnup to 0.02 ok at an assembly average j burnup of 60,000 MWD /MTU, is conservative and acceptable.

j The effect of axial burnup distribution on fuel assembly reactivity has been

! evaluated by modeling depleted fuel in both two dimensions and three j dimensions. These evaluations show that axial burnup effects can cause assemblyreactivitytoincregeatburnup-enrichmentcombinationsgreaterthan j 40,000 MWD /MTU and 4.0 w/o U . Westinghouse has stated that this effect j will be accounted for as an additional bias if burnup credit limits reach

j. the e combinations.

e i An additional conservatism is } pat the depletgn calculations do not take credit for effects, such as Pu decay and Am growth, that are known to

} substantially reduce reactivity during long-te'a storage. However, the staff j does not consider this to be a requirement.

1 i The staff concludes that adequate conservatism has been incorporated in the l methodology used to determine burnup credit.

l 3

l 4

7

3.4 Intearal Fuel Burnable Absorber (IFBA) Credit <

l i Another reactivity equivalencing technique for storage of fuel enrichments j greater than those allowed by the previous methodology is based on the

! reactivity decrease associated with the addition of integral fuel burnable

absorbers (IFBA) to Westinghouse fuel. IFBAs consist of neutron-absorbing material applied as a nonremovable thin zirconium diboride (ZrB2 ) coating on the outside of the U02 pellet. PH0ENIX-P is used to generate a set of initial assembly enrichment versus number of IFBA rods per assembly ordered pairs that

all yield the equivalent k ,, (no greater than 0.95) when fuel assemblies are i

storedinthespentfuelsIorageracks. The following assumptions are used

for the IFBA rod assemblies in the PHOENIX-P calculations

(1) The fuel assembly is modeled at its most reactive point in life. This includes any time in life when the IFBA has depleted and the fuel i assembly becomes more reactive.

l (2) The B" leading for each IFBA rod, determined from Westinghouse IFBA i design specifications for the given fuel assembly type, is the i minimum standard loading offered by Westinghouse for that fuel assembly j type.

(3) The IFBA B" loading is reduced by 5 percent to account for 1

manufacturing tolerances and by an amount which corresponds to the minimum absorber length offered for the given fuel assembly type (e.g.,

a 144-inch fuel length with a minimp absorber length of 108 inches j would result in a 25 percent IFBA B loading).  !

! A calculational uncertainty of approximately 10 percent is included in the  !

l development of the IFBA requirements by adding an additional number of IFBA rods to each data point. To demonstrate that reactivity margin exists in the i l

! IFBA credit limit to accommodate future changes in IFBA patterns, calculations  !'

are also performed with nonstandard IFBA patterns. If a future change is made to the standard IFBA pattern designs, the reactivity difference between the

! new patterns and the old patterns will be calculated in order to assess the j impact on both core reactivity and spent fuel rack IFBA credit limits.

) The staff concludes that adequate conservatism has been incorporated in the methodology for determining IFBA requirements and that assemblies that comply with the enrichment-IFBA requirement curve developed by this methodology will have a k,,, no greater than 0.95 when placed in the spent fuel pool storage racks.

3.5 Infinite Multiolication Factor An alternative method for determining the acceptability of fuel storage in a specific spent fuel rack is based on a PHOENIX-P calculation of the infinite multiplication factor (k,,) for a fuel assembly in the reactor core geometry as  :

a reference point. The fuel assembly model is based on a unit assembly configuration (infinite in the lateral and axial dimensions) in reactor geometry and is modeled at its most reactive point in life t.nd moderated by

I

• l 8

i l pure water (no boron) at a temperature of 68*F with a density of 1.0 g/cc. A l

0.01 Ak reactivity bias is added to this reference k, to account for calculational uncertainties. The spent fuel storage rack is then modeled with r

these assemblies to ensure that the storage rack reactivity will be no greater than 0.95.

The staff concludes that fuel assemblies that have a reference k, less than or  !

equal to the value calculated with the above assumptions and methodology will have a k,,, no greater than 0.95 when placed in the spent fuel pool storage l

racks.

3.6 Postulated Accidents  :

The criterion that k,,, be no greater than 0.95 exists even for postulated l accidents. Two types o f accidents that can occur in a spent fuel storage rack ,

may cause a reactivity increase: (1) a fuel assembly misplacement and (2) a pool water temperature change. However, for any of these accidents, the double contingency principle (Ref. 23) can be applied. According to this principle, it is unnecessary to assume two unlikely, independent, concurrent events to ensure protection against a criticality accident. Thus, for these postulated accidents, the presence of soluble boron in the pool water can be assumed as a realistic initial condition since assuming its absence would be a second unlikely event. PH0ENIX-P boron worth calculations are used to ,

l determine the amount of soluble baron required to offset the highest reactivity increase caused by any postulated accident and to maintain k ,,

less than or equal to 0.95, which is also the staff's acceptance criterlon for accident conditions.

3.7 Soluble Boron Credit Methodoloov f

In the proposed methodology for performing spent fuel rack reactivity calculations with credit for soluble boron in the pool water, a 95/95 rack k is first calculated b,,r,on o credit. This k which remains below 1.0 (subcritical) with no soluble calculation uses the same assumptions described in Section3.2above,incYudingtheassumptionofnosolubleboroninthegool water. As previously described, a temperature bias, a method bias, a B self-shielding bias, and the 95/95 uncertainties associated with the calculation uncertainty, the methodology uncertainty in the benchmarking bias, I

and the manufacturing tolerances are included in the k,:, calculation.

The final equation for determining the k,,, requirement is k,,, = k o,ma + B,,,, +Bw+ B,,,, +B < 1.0 where:

t ,

k o,m u - nominal condition KENO-Va k,,,

B,,,, - temperature bias for normal operating ran9e

5

?

9 Bm,inog -

method bias from benchmark critical comparisons

8,,, - B" self shielding bias l

B .,,

=

E(tolerance . . .or. . . uncertainty,)2 i

i

To determine the amount of soluble boron required to maintain k j

KENO-Va is used to establish a nominal reference k ,, and PH0ENIY# s; 0.95, P is used to i evaluate the reactivity effects of possible variatlons in material

! characteristics and mechanical manufacturing dimensions. These calculations i contain the same assumptions, biases, tolerances, and uncertainties previously l described except for the assumption regarding the moderator soluble boron concentration. Borated water is assumed instead of pure water. The tolerance I

calculations are, therefore, performed assuming the presence of soluble boron. l The abovefinal 95/95 and must be k,Tess than or equal to 0.95 with allowances for biasescalcu tolerancer, and uncertainties including the presence of the determined concentration of soluble boron.

! For enrichments higher than those assumed in the k calculation, reactivity

! equivalencing methodologies are used to determine b,u,rnup or IFBAgredit.  ;

i However, the maximum fuel rod enrichment is limited to 5.0 w/o U . Soluble j i

boron credit is used to offset the uncertainties associated with each of these i j equivalencing methodologies, as appropriate. l Postulated accidents are considered in the same manner as discussed in Section 3.6 except that the previously determined amount of soluble boron for i the 95/95 k calculation, plus the amount determined for the reactivity equivalenci,n,g, calculation, if required, is assumed present. The results of f

1 PHOENIX-P calculations of the reactivity change due to the presence of soluble boron are used to determine the amount of soluble boron required to offset the

maximum reactivity increase caused by postulated accident conditions.

4 The final soluble boron credit requirement is determined from the following

summation:

\

a SBCrogg - SBCef, + SBC,, + E,,

l where:

a

)

SBCrorn - total soluble boron credit requirement (ppm) j SBCg ,n - soluble boron credit required for 95/95 k,,, less than

or equal to 0.95 (ppm) 4 i

i 10 i

SBC u - soluble boron credit required for reactivity equivalencing methodologies (ppm)

SBC n - soluble boron credit required for k,,, less than or  ;

equal to 0.95 under accident conditions (ppm) '

Thus the total soluble boron credit requirement will maintain the spent fuel  !

rack k perceni,, confidence level.less than or equal to 0.95 with a 95-percent probability at a 95-The total soluble boron required to maintain k less than or equal to 0.95 is normally well below the large amount of soliaSle baron which is typically in spent fuel pool water. Therefore, a significant margin to criticality would

^

generally still exist. However, a boron dilution analysis will be performed for each plant requesting soluble baron credit to ensure that sufficient time" is available to detect and mitigate the dilution before the 0.95 k design basis is exceeded and submitted to the NRC for review (Ref. 29). INeanalysis should include an evaluation of the following plant-specific features:

1. Spent Fuel Pool and Related System Features a) dilution sources

' b) dilution flow rates c) boration sources d) instrumentation -

e) administrative procedures f) piping g) loss of offsite power impact

2. Boron Dilution Initiating Events (including operator error)

3. Boron Dilution Times and Volumes '

a 4.0 SUMARY AND CONCLUSIONS The topical report WCAP-14416-P and supporting documentation provided in References 14, 26 and 28 have been reviewed in detail. A major portion of this review focused on a proposed new methodology whereby partial credit could

be taken for soluble boron in the spent fuel pool to meet the NRC-recommended criterion or equal tothat the at 0.95, spent fuel rack probability, a 95-percent multiplication factor (k con 95-percent ,fidence level.g ) be less than

. The staff concludes that the proposed new methodology for soluble boron credit is acceptable for the following reasons:

(1) Uncertainties in mechanical tolerances and storage rack dimensions are determined at the 95-percent probability, 95-percent confidence level and are incorporated in a conservative direction.

(2) Conservative uncertainties are incorporated for depletion calculations.

.m ..- _4 _ . _ _ - . _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _

j.-

i

!~

l 11 (3) A substantial margin to criticality would be available since the spent

fuel rack k,,, will be less than or equal to 0.95, at a 95-percent

probability, 95-percent confidence level, with an amount of soluble boron j significantly less than that amount normally available in the pool.

3 i (4) The fuel rack k , will remain less than 1.0 (subcritical), at a 95-percentprobabilTty,95-percentconfidencelevel,evenwithnosoluble boron'in the spent fuel pool, thereby conforming to Criterion 62,

" Prevention of criticality in fuel storage and handling".of Appendix A to 10 CFR Part 50.

'~

The staff concludes that the methodology documented in WCAP-14416-P and .

. Reference 28 can be used in licensing actions with the following provisions j

which are stated in WCAP-14416-P and Reference-28:

(1) If axial and planar variations of fuel assembly characteristics are i present, they should be explicitly addressed, including the locations of j burnable absorber rods.

1 l 235 (2) The maximum fuel rod enrichment shall be limited to 5.0 w/o U .

I (3) The spent fuel storage racks should be assumed to be infinite in lateral i extent or surrounded by a water reflector and concrete or structural material as appropriate to the design. The fuel may be assumed to be

infinite in the axial dimension, or the effect of reflector on the top j and bottom of the fuel may be evaluated.

a j (4) If credit for the reactivity depletion due to fuel burnup is taken, j operating procedures should include provision for independent

! confirmation of the fuel burnup, either administrative 1y or

experimentally, before the fuel is placed in burnup-dependent storage

, cells.

I' (5) A reactivity uncertainty due to uncertainty in the fuel depletion l

history effects and depletion calculations should be included.

l (6) A correction for the effect of the axial distribution in burnup should be 3

determined and added to the reactivity calculated for uniform axial

burnup distribution if it results in a positive reactivity effect.

4

! In addition, as stated in the letter of October 18, 1996, from Westinghouse to

the NRC (Ref. 28), the following items will be submitted by all licensees proposing to use the methodology described above

l (1) All licensees proposing to use the new method described above for soluble

i. boron credit should submit a 10 CFR Part 50.36 technical specification j change containing the following:

i d

a. k shall be less than or equal to 0.95 if fully flooded with water i b,o,r,ated to [1050] ppm which includes an allowance for uncertainties l as described in WCAP-14416-P.

l i

---

,y,-m

12

b. k shall be less than 1.0 if fully flooded with unborated water wNchincludesanallowanceforuncertaintiesasdescribedin WCAP-14416-P.

c. The spent fuel pool boron concentration shall be greater than [2300]

ppm and shall be verified at a frequency of [7 days).

Licensees using the Westinghouse Improved Standard Technical Specifications (ISTS) described in NUREG-1431 (Ref. 27), should adopt specification 3.7.16, " Fuel Storage Boron Concentration," and 4.3.1, Fuel Storage-Criticality," as shown in section 5.0 below.

(2) All licensees proposing to use the new method described above for soluble boron credit should identify potential events which could dilute the spent fuel pool soluble boron to the concentration required to maintain the 0.95 k limit (as defined in (1)a above) and should quantify the time span N these dilution events to show that sufficient time is available to enable adequate detection and suppression of any dilution event. The effects of incomplete boron mixing, such as boron stratification, should be considered. This analysis should be submitted for NRC review and should also be used to justify the surveillance interval used for verification of the technical specification minimum pool boron concentration.

(3) Although Boraflex deterioration is not addressed in this topical report, appropriate analyses are required to account for Boraflex degradation in storage racks that credit the negative reactivity effect of Boraflex.

These analyses should be submitted for NRC review.

(4) Plant procedures should be upgraded, as necessary, to control pool boron concentration and water inventory during both normal and accident conditions.

i

l' e

I i

1 13 I 5.0 TECHNICAL SPECIFICATIONS 1

i l

i i

I 1

l l

l I

14 3.7 PLANT SYSTEMS 3.7.16 Fuel Storage Pool Boron Concentration LC0 3.7.16 The fuel storage pool boron concentration shall be 2 [2300] ppm.

APPLICABILITY: When fuel assemblies are stored in the fuel storage pool and  ;

a fuel storage pool verification has not been performed  !

since the last movement of fuel assemblies in the fuel storage pool.

ACTIONS '

CONDITION REQUIRED ACTION COMPLETION TIME A. Fuel storage pool ------------NOTE-------------

boron concentration LC0 3.0.3 is not applicable.

not within limit. -----------------------------

A.1 Suspend movement of Immediately fuel assemblies in the fuel storage pool.

)

l MQ A.2.1 Initiate action to Immediately restore fuel storage pool boron concentration to within limit.

QB A.2.2 Verify by Immediately administrative means

[ Region 2] fuel storage pool verification has been performed since the last movement of fuel assemblies in the fuel storage pool .

15 SURVEILLANCE REQUIREMENTS SURVEILLANCE FREQUENCY SR 3.7.16.1 Verify the fuel storage pool boron [7 days]

concentration is within limit.

l 16 l

1 4.0 DESIGN FEATURES 4.3 Fuel Storage 4.3.1 Criticality 4.3.1.1 The spent fuel storage racks are designed and shall be maintained with:

a. Fuel assemblies having a maximum U-235 enrichment of [4.5] weight percent;

b. k,,, < l.0 if fully flooded with unborated  !

water which includes an allowance for uncertainties as described in WCAP-14416-P;

c. k,f, s 0.95 if fully flooded with water borated to [1050] ppm which includes an allowance for uncertainties as described in WCAP-14416-P;

[d. A nominal [9.15] inch center to center distance between fuel assemblies placed in [the high 4

density fuel storage racks];)

" [e. A nominal [10.95] inch center to center distance between fuel assemblies placed in [ low density fuel storage racks];)

[f. New or partially spent fuel assemblies with a

' discharge burnup in the " acceptable range" of Figure [3.7.17-1] may be allowed unrestricted storage in [either] fuel storage rack (s); and]

. [g. New or partially spent fuel assemblies with a discharge burnup in the " unacceptable range" of Figure [3.7.17-1] will be stored in compliance with the NRC approved [ specific document containing the analytical methods, title, date, or specific configuration or figure).]

l i

17

6.0 REFERENCES

1) Newton, R.A. (WOG), letter to Document Control Desk (NRC), Westinghouse Owners Group Transmittal of Reports: WCAP-14416-P (Proprietary) and WCAP-14417-NP (Non-proprietary) Entitled " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," OG-95-062, July 28,1995.

Newayer, W.D., " Westinghouse Spent Fuel Rack Criticality Analysis

.2)

Methodology," WCAP-14416-P, June 1995.

3) Code of Federal Regulations, Title 10, Part 50, Appendix A, Criterion 62, " Prevention of Criticality in Fuel Storage and Handling."

4) U.S. Nuclear Regulatory Commission, Standard Review Plan, Section 9.1.2, NUREG-0800, July 1981.

5) U.S. Nuclear Regulatory Commission, Letter to All Power Reactor Licensees from B.K. Grimes, "0T Position for Review and Acceptance of Spent Fuel Storage and Handling Applications," April 14, 1978.

6) Petrie, L.M. and Landers, N.F., " KEN 0 Va--An Improved Monte Carlo Criticality Program With Supergrouping," NUREG/CR-0200, Vol. 2, Section Fil, November 1993.

7) Greene, N.M., "NITAWL-II: SCALE System Module for Performing Resonance Shielding and Working Library Production," NUREG/CR-0200, Vol. 2, Section F2, June 1989.

8) Greene, N.M., "XSDRNPM-S: A One-Dimensional Discrete-Ordinates Code for Transport Analysis," NUREG/CR-0200, Vol. 2, Section F3, November 1993.

9) Ford, L'.E., III, "CSRL-V: Processed ENDF/B-V 227-Neutron-Group and Pointwise Cross-Section Libraries for Criticality Safety, Reactor and Shielding Studies," ORNL/CSD/TM-160, June 1982.

10) Baldwin, M.N., " Critical Experiments Supporting Close Proximity Water Storage of Power Reactor Fuel," BAW-1484-7, July 1979.

11) Biernan, S.R., and Clayton, E.D.,23, Criticality Separation Between Suberitical Clusters of 2.35 WtX U Enriched U0, Rods In Water with Fixed Neutron Poisons," PNL-2438, October 1977

12) Bierman, S.R., and Clayton, E.D.,g", Criticality Separation Between Subcritical Clusters of 4.29 Wt% U Enriched U0, Rods in Water with Fixed Neutron Poisons," PNL-2615, August 1979.

13) Bierman, S.R., and Clayton, E.D., " Criticality Subcritical Clusters of 2.35 WtX and 4.31 Wt% jxperiments U Enrichedwith 00 Rods in Water at a Water-to-Fuel Volume Ratio of 1.6," PNL-3314, July,1980.

l

- - _ - - - . .. - - - -- 1

18

14) Bush,'L. (WOG), letter to' Document Control Desk (NRC), Westinghouse Owners Group Transmittal of Response to Request for Additional Information (RAI) Regarding WCAP-14416-P Entitled " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," Second Round of Questions, January 17, 1996.

15) Owen, D.B., " Factors for One-Sided Tolerance Limits and For Variables Sampling Plans," Sandia Corporation, SCR-607, March 1963.

16) Nguyen, T.Q., et. al., " Qualification of the PH0ENIX-P/ANC Nuclear Design System for Pressurized Water Reactor Cores," WCAP-Il597-A, June 1988.

17) Strawbridge, L.E., and Barry, R.F., " Criticality Calculations for Uniform Water-Moderated Lattices," Nuclear Science & Engineering, Vol.

23, pp. 58-73 (1965).

18) Baldwin, M.N., and Stern, M.E., " Physics Verification Program Part III, Task 4: Summary Report," BAW-3647-20, March 1971.

19) Baldwin, M.N., " Physics Verification Program Part III, Task 11:

Quarterly Technical Report January-March 1974," BAW-3647-30, July 1974.

20) Baldwin, M.N., " Physics Verification Program Part III, Task 11:

Quarterly Technical Report July-September 1974," BAW-3647-31, February 1975.

21) Melehan, J.B., " Yankee Core Evaluation Program Final Report," WCAP-3017-6094, January 1971.

22) England, T.R., " CINDER - A One-Point Depletion and Fission Product Program," WAPD-TM-334, August 1962.

23) American Nuclear Society, "American National Standard for Nuclear Criticality Safety in Operations with Fissionable Materials Outside Reactors," ANSI /ANS-8.1-1983, October 7, 1983,

24) Kopp, L. (NRC), letter to R. Newton (WOG), Request for Additional Information on Licensing Topical Report WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," August 30, 1995.

25) Kopp, L. (NRC), letter to L. Bush (WOG), Request for Additional Information on Licensing Topical Report WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," Second Round of Questions, October 16, 1995.

26) Bush, L. (WOG), letter to Document Control Desk (NRC), Westinghouse Owners Group Transmittal of Response to Request for Additional Information (RAI) Regarding WCAP-14416-P Entitled " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," OG-95-080, September 28, 1995.

27) U.S. Nuclear Regulatory Commission, " Standard Technical Specifications, Westinghouse Plants," NUREG-1431, Rev.1, April 7,1995.

j

19

28) Stringfellow, J. (WOG), letter to Document Control Desk (NRC),

Transmittal of Page Revisions to WCAP-14416, Rev. O, Entitled

" Westinghouse Spent Fuel Rack Criticality Analysis Methodology" to Address CRGR Review Comments - MUHP-3001, OG-96-092, October 23, 1996.

29) Cassidy, B., et. al., " Westinghouse Owners Group Evaluation of the Potential for Diluting PWR Spent Fuel Pools," WCAP-14181, July 1995.

l 1

i k

i l

, August 12, 1996

. MEMORANDUM TO: Edward L. Jordan, Chairman i Committee to Review Generic Requirements FROM: Frank J. Miraglia, Deputy Director Office of Nuclear Reactor Regulation

SUBJECT:

CREDIT FOR SOLUBLE BORON IN PWR SPENT FUEL POOLS

REFERENCES:

1. Letter from R.C. Jones (NRC) to R.A. Newton, Chairman, l Westinghouse Owners Group, " Acceptance for Referencing of Licensing Topical Report WCAP-14416-P, ' Westinghouse Spent Fuel Rack Criticality Analysis Methodology'"

2. WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology," June 1995 The Office of Nuclear Reactor Regulation (NRR) is proposing (Ref.1) to accept l the Westinghouse Owners Group (WOG) methodology for crediting the soluble boron in Westinghouse-designed spent fuel pools (Ref. 2). The basis for this acceptance is that a significant margin to criticality would still be available and, thus, the probability of an inadvertent criticality in the spent fuel pool would not be increased to any appreciable extent. In fact, the new methodology would still require subcriticality even if no soluble l boron were present. The large reactivity margin available in the soluble boron in the spent fuel pool water could be used by licensees to design spent fuel storage racks with expanded storage capacity.

The staff considers this a relaxation of our existing position, which only allows credit for soluble baron under accident conditions. The CRGR charter requires review of proposed relaxations or decreases in current requirements which affect such a generic group of plants. Therefore, we are requesting CRGR review of the attached package.

Attachment 1 is the staff Safety Evaluation Report (Ref. 1). Attachment 2 contains copies of Topical Report WCAP-14416-P (Ref. 2). Please note that Westinghouse has reclassified the report as nonproprietary. Attachment 3 is 4

the response to the questions contained in Section IV.B of the CRGR Charter.

We will work with your staff to schedule a meeting on this subject after the Committee has reviewed the enclosed package. The staff considers this to be Category 2.

Attachments:

1. WCAP-14416-P SER

• See previous concurrence f g ,c,3 c j OFC SRXB:DSSA SRXB:DSSA D:DSSA TSB:ADPR OGj r* ADT NR gRR NAME LK0PP* RJ0NES* GHOLAHAN* CGRIMESM ATHA,DANI LIA DATE 6/13/96 6/20/96 6/20/96 .,~, _ //>/ f4 #

,\

OFFICIAL RECORD COPY G:\BCREDIT.WOG I

g OQs ,J

\

3 .,

t)

.a - n ,:.a ..

• g g, ,iJ ii