ML15261A528

Attachment 1 to NL-15-089, Response to Request for Additional Information Netco Report NET-300067-01, Revision 1, (Non-Proprietary)

ML15261A528
Person / Time
Site:	Indian Point
Issue date:	08/14/2015
From:	Coyle L Entergy Nuclear Northeast
To:	Document Control Desk, Office of Nuclear Reactor Regulation
Shared Package
ML15261A536	List: ML15261A528 ML15261A529 NL-15-089, Response to Request for Additional Information Regarding the Indian Point Nuclear Generating Unit No. 2 - Spent Fuel Pool Criticality Analysis
References
NL-15-089
Download: ML15261A528 (40)

Text

ATTACHMENT 1 TO NL-15-089 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION NETCO REPORT NET-300067-01, REVISION 1, (NON-PROPRIETARY)

ENTERGY NUCLEAR OPERATIONS, INC.

INDIAN POINT NUCLEAR GENERATING UNIT NO. 2 DOCKET NO. 50-247

N L-1 5-089 Docket No. 50-247 Page 1 of 36 RESPONSE TO REQUEST FOR ADDITIONAL IN FORMATION ENTERGY NUCLEAR OPERATIONS. INC.

THE INDIAN POINT NUCLEAR GENERATING UNIT NO. 2 SPENT FUEL POOL CRITICALITY ANALYSIS DOCKET NO. 50-247 In order for the NRC staff to complete their review of the NETCO Report NET-300067-01, "Criticality Safety Analysis of the Indian Point 2 Spent Fuel Pool with Credit for Inserted Neutron Absorber Panels", the NRC Staff have requested additional information. These requests and Entergy's responses are as follows:

Absorber Panel Design

1. Section 3.4, "Absorber Panel Design," credits "a double panel at the [Region 1/Region 2] interface."

Upon installation of the absorber panel inserts, how will it be ensured that the installed insert orientation at the Region 1/Region 2 interface, and elsewhere, is consistent with the modeled orientation credited in the criticality safety analyses? Furthermore, how will proper panel installation be confirmed given that there are three possible panel types with region-specific dependency?

Response

The installation will be procedurally controlled. One insert type will be used.

2.

A footnote to Table 3.5, "Absorber Panel Dimensions," regarding alternate absorber panel adjustments states: "Minor adjustments to these specific dimensions and areal densities are acceptable provided that the panel is shown to be as effective in absorbing neutrons as the primary design." This statement implies that a calculation or calculations will be performed to demonstrate the effectiveness of any minor adjustments to the alternate absorber. Provide clarification regarding how the effectiveness of the final alternate absorber panel will be demonstrated if adjustments to the values in Table 3.5 are made. Also, the notes to Table 3.5 indicate that some values will be restricted to a minimum or maximum value. Will the minor adjustments be consistent with these notes?

Response

The footnote will be removed. If the final panel design fails to meet the assumptions of the criticality analysis, then additional calculations would be required and will be part of the License Amendment Request associated with the final configuration of the spent fuel pool.

2. Section 3.4 states that "all of the loading curve calculations were performed with the

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 2 of 36 primary design," and for the alternate design, the minimum areal density had to be increased relative to the primary design "so that the loading requirements would remain the same." Was the Region 2 alternate panel areal density confirmed to be valid over the entire burnup loading curve range? Also, there are missing footnotes in the second to last paragraph of Section 3.4 for the alternate absorber panel areal densities. Provide the referenced footnotes or remove the references if they were unintentional.

Response

The alternate panel calculations were made at the high burnup of 42.67 GWd/T for 5.0 wt% fuel with no cooling time and also for the low burnup case of 12.31 GWd/T for 2.5 wt% fuel with 25 year cooling. In both cases, the k for the alternate panel case was less than the k for the primary design as shown in the table below:

Burnup Primary Design Alternate Design High (42.67 GWD/MTU) 0.9639 0.9630 Low (12.31 GWD/MTU) 0.9696 0.9683 As a further check, a case was run at 32.04 GWd/T for 4.0 wt% fuel with 5 year cooling.

The k for this case was 0.9645 vs 0.9654 for the primary design. As noted in the response to RA! 2, the footnote will be removed.

Code Validation

4.

The last sentence of Section 5.1, "Limiting Depletion Parameters-Burnable Absorbers," states: "If gadolinium or erbium is used in the future, then this criticality analysis is valid."

How does the current critical experiment benchmarking analysis cover future erbium or gadolinium burnable absorber credit in spent fuel pool (SFP) criticality safety calculations?

Response

No gadolinium or erbium bearing fuel has ever been used at Indian Point and there are currently no plans to include such fuel in the future. The criticality analysis no longer seeks approval for gadolinium or erbium bearing fuel. All references to gadolinium and erbium have been removed.

5.

In Section A.2.5, "Statistical Analysis of the Fresh U02 Critical Benchmark Results,"

the equations for the soluble boron and boron areal density (B-la) trend lines are shown to be identical. Please correct the apparent error and confirm that the uncertainty treatment for the B-10 areal density trend is correct since the limiting criticality code validation uncertainty is based on the areal density trend analysis.

Response

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 3 of 36 Rev. 1 of the criticality analysis had duplicated the trend on soluble boron rather than take the trend on areal density from the spreadsheet. Since Rev. 1 was issued, a couple of small errors in the interpretation of some EXCEL statistical functions have changed the uncertainty in the bias but not enough to change the UO2 limiting bias and uncertainty.

Also, it was decided that it is more accurate to use all the experiments to establish the intercept for the soluble boron and areal density trends. This caused an insignificant difference to the trends but the plots are changed as well as the round off digit of the intercept for the soluble boron trend. After these corrections, the boron content portion of the validation becomes:

Boron Content A fit of the calculated k's as a function of the B-i10 areal density in the absorber plates or the soluble boron ppm was performed using NUREG/CR-6698 [2] equations 10 through 13 and the data from Table A.2. Both fits failed the statistically significance test compared to a zero slope. However, to be conservative, both the zero slope and the calculated fit are used for determining the limiting k as a function of boron content.

The following equations are the best fit of the data for k versus soluble boron and areal density. Figures A.6 and A.7 show the results of the analyses. Th,, uncert.inty

.around;.., ".....,.,,.'

k(ppm soluble boron) = 0.9977-6 + ( 5.24E-08)*ppm k(B-10 Areal Density) - 0.99777 + (&624&-080.0t35)*(B-10 Areal Density) where the B-10 areal density is in gm B-I10/cm 2.

The uncertainty about the mean for the soluble boron trend is 0.0047 at 0 ppm and 0.0050 at 2000 ppm. The uncertainty about the mean for the areal density trend is 0.0050 at 0.015 gm B1°/cm2 and 0.0053 at 0.022 gm B1°/cm2 however since the bias at 0.022 gm B10/cm2 is less than the bias from the EALF the uncertainty in the areal density bias is not limiting.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 4 of 36 1,0 0 2 0 10.9960 0.9 9 4 0 0.9 9 2 0................

0.000 0.010 0.020 0.030 0.040 0.0.50 0.060 0.070 Areal B-10 Density (gin B-.l0/cm'cm)

Figure A.6: keff as a Function of the B-10 Areal Density in the Separator Plates 0.9995 0.9990 0.9985

• 0.9980

~0.9975 0.9970 0.9965 0.9960 4

t.

w 4-4

• *~.

• ~

1000 2000 3000 saeau Bo (mm) 4000 5000 6000 Figure A.7: kef as a Function of the Soluble Boron Content The corrections to the uncertainties found in the other subsections of the U0 2 validation are found on the table below.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 5 of 36 Parameter Previous Uncertainty Corrected Uncertainty Enrichment 0.00435 0.0049 Pin Diameter 0.004343 0.00496 Pitch 0.00427 0.0046 EALF 0.00431 0.0047 The EALF trended uncertainty for the HTC critical benchmarks changed from 0.0057 to 0.0076 but with the new treatment discussed in RAI 7 this has no impact on the final results.

6.

Referring to lattice characteristics in Table A.7, "Area of Applicability (Benchmark Applicability)," the comment mentions that "the expected range of all fuel types, including both pressurized water reactors (PWRs) and boiling water reactors (BWRs) fuel is covered." Why is BWR fuel included in the table if BWR fuel is not stored at Indian Point 2?

Response

Although the lattice characteristics of the critical experiments do cover both PWRs and BWRs, it is agreed that mentioning BWR fuel here was not appropriate and it is removed from Table A.7.

7.

The assessment of the mixed uranium/plutonium oxide (MOX) critical experiments in Section A.3.2, "MOX Critical Experiments," concludes that the average uncertainty weighted k-effective of the fresh U02 critical experiments is less than that of the MOX experiments implying that the fresh fuel critical experiments should be used as the basis for the criticality code bias and bias uncertainty for all the criticality safety analyses. However, the uncertainty for the MOX critical experiments was not reported.

Provide the limiting uncertainty from the MOX critical experiment trending analysis, and the uncertainty for the set as a whole, as was done for the fresh fuel critical experiments and confirm that the criticality code bias and bias uncertainty determined from the fresh fuel critical experiments remains bounding for calculations containing spent fuel.

Response

The NRC is correct that the uncertainty for the MOX set is large. Depending on the size of the other uncertainties that would be statistically combined with the validation uncertainty, the size of the uncertainty could overwhelm the difference in the bias. In order to address this better, a new grouping of the critical experiments will be used. The MOX experiments with less than 2 wt% Pu will be combined with the HTC experiments.

Experiments with greater than 2 wt% PU will be excluded. Note that the weight percent of Pu in spent fuel is always less than 2 wt% Pu. (Highest Pu wt% is about 1.4 wt% Pu).

Since the spent fuel may have a burnup that produces less plutonium than the HTC/MOX set, the more limiting bias and uncertainty from the HTC/MOX or U0 2 sets will be used for

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-15-089 Docket No. 50-247 Page 6 of 36 burned fuel. As a confirmation that this approach is conservative, all the experiments are also combined into one set for statistical analysis. The MOX set including the higher weight percent Pu is maintained in order to clearly demonstrate that the bias decreases with Pu content and the bias decreases with Pu-241 decay.

The statistical analysis still shows that the largest bias comes from the fresh fuel with an EALF of 0.4 eV. That bias is 0.0029 in k for EALF's below 0.4 eV and 0.0039 for EALF's between 0.4 eV and 0.65 eV. The uncertainties for the UO 2 set is 0.005 for EALF's less than 0.4 eV and 0.0054 for EALF's between 0.4 and 0.65 eV.

The HTC/MOX set has a lower bias but a higher uncertainty. Since it is unclear when the higher uncertainty provides more limiting results both the bias/uncertainty from the HTC/MOX analysis and the UO2 analysis are used in the wrap up of biases and uncertainties and the most limiting results are taken. The bias for EALF's less than 0.4 eV from the HTC/MOX set is 0.0021 and for EALF's between 0.4 and 0.65 eV the bias is 0.0027. The uncertainties in these biases from the HTC/MOX set are 0.0087 and 0.0112 for the low and high energy EALF's respectively.

As expected, when the analysis is performed on the set where all the UO2/HTC/MOX are together the biases are less than those from the UO2 set. The uncertainty in the bias was approximately the same as the UO2 set but at the higher EALF the uncertainty was slightly higher 0.0054 versus 0.0051 so the uncertainty above for the UO2 set is derived from the combined set.

The following sections of the criticality analysis shall be rewritten as follows:

Last paragraph of Section 4.1:

The most limiting bias and uncertainty from the validation was due to a trend in the spectrum, EALF. From this trend, a bias of 0.0029 for an EALF up to 0.4 eV and 0.0037,9 for EALFs from 0.4 to 0.65 eV has been determined. Cases without soluble boron are in the first range of EALF and will use 0.0029 for the bias. Heavily borated cases can have an EALF greater than 0.4 eV and then would use the 0.003:79 for the bias. The 95/95 uncertainty is 0.0050 for all cases with an EALF less than 0.4 eV and is 0.0054 for cases with an EALF between 0.4 and 0.65 eV. tlfe-aioyree Next to last paragraph of Section 4.2:

The results of the analysis of the three sets of experiments were combined into two sets for the determination of the bias and uncertainty of the burned fuel. The first set was the UO2 set described in Section 4.1. The second set combined the HTC critical experiment results with the MOX experiment results that contain less than or equal to 2 wt% Pu results. The maximum Plutonium content in spent fuel is less than 1.5 wt% Pu so including higher weight percent Plutonium experiments is not appropriate. The most limiting bias and uncertainty from each of the thr-eetwo sets of experiments*.a-rs-O~

b).j H.v TC", --nd c),..M."OX" is used for the bias and uncertainty for the major actinides (U, Pu, Am-241). The most limiting bias a

~i~it comes from the fresh UO2 experiments, since ENDF/B-Vll predicts higher k's for MOX critical experiments relative to U-235 based systems. However, the most limiting uncertainty comes from the HTC/MOX set of experiments. Since it is unclear until the combination of the uncertainties whether the

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 7 of 36 bias or the uncertainty is more important to the final k, the most limiting bias and uncertainty pair from each set is used in the final determination of the k95195. Refer to Section A.3.23 of Appendix A to see the support for this position. This means that the first worth component of the bias and uncertainty for burned fuel i.e., the component for the worth of the major actinides, has a bias of:

a. 0.0029 or 0.003:79 depending on EALF and likewise an uncertainty of 0.0050 or 0.0054 from the UO2 benchmarks, or

b. 0.0021 or 0.0027 depending on EALF with uncertainties of 0.0087 or 0.0112 depending on the EALF derived from the HTC/MOX set.

Table 10. 1 entry corresponding to Section 4*a~ii of DSS-ISG-2010-01:

The bias and uncertainty was calculated from both the UO2 and HTC/MOX critical benchmarks and the final determination of the k95195 used the most limiting set of bias and uncertainty. Thc* MV.and, HTC c.itic.-s.. r.e, not incude in**,

th, anal..i. of,.,

the freh fuel, Section A.3.3 is added:

A.3.3 Determination of Burned Fuel Bias and Uncertainty The bias and uncertainty of burned fuel depends on the amount of plutonium in the burned fuel. As shown on Figure A.9 the bias decreases with plutonium content.

However, the uncertainty increases with plutonium content. In order to determine appropriate biases and uncertainties the HTC and MOX critical benchmarks are combined. The MOX experiments with plutonium content above 2 wt% were useful for confirmation that the bias decreases with plutonium content, but the maximum plutonium content in spent nuclear fuel about 1.*5 wt% plutonium, so using experiments above 2 wt% plutonium needlessly increases the uncertainty. The bias and uncertainty in the bias for the HTC/MOX (2 wt% PU or less) set is controlled by the EALF trend. For EALF's less than 0.4 eV the maximum bias and uncertainty are 0.0021 and 0.0087 respectively.

For EALF's from 0.4 to 0.65 eV the maximum bias and uncertainty are 0.0027 and 0.0112 respectively.

The final criticality analysis for burned fuel should use both,

a. 0.0029 or 0.0039 depending on EALF and likewise an uncertainty of 0.0050 or 0.0054 from the UO2 benchmarks, an..dd

b. 0.0021 or 0.0027 depending on EALF with uncertainties of 0.0087 or 0.0112 depending on the EALF derived from the HTC/MOX set.

to determine the maximum k95195.

To confirm this approach is appropriate the statistical analysis was repeated with all the critical experiments combined into one set. For this combined set, the EALF trend still is most limiting. For the combined set for EALF's below 0.4 the bias and uncertainty is 0.0028 and 0.0050 respectively. Finally, for EALF's from 0.4 to 0.65 the maximum bias and uncertainty from the set with all the experiments is 0.0039 and 0.0054.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 8 of 36 The total bias and uncertainty table 8.5 has been revised and is shown below. The MOX bias and uncertainty resulted in a larger total bias and uncertainty for all cases.

Table 8.5: Total Bias and Uncertainty at Each Burnup Point Enrichment

____Cooling Time (years)

(wt% U-235) 0 1

2 5

10 15 25' 2.0 0.0134 0.0134 0.0134 0.0134 0.0134 0.0134 0.0134 2.5 0.0149 0.0149 0.0149 0.0148 0.0147 0.0147 0.0147 3.0 0.0200 0.0200 0.0199 0.0199 0.0198 0.0198 0.0198 3.5 0.0217 0.0217 0.0216 0.0215 0.0213 0.0212 0.0211 4.0 0.0236 0.0236 0.02.35 0.0234 0.0232 0.0231 0.0229 4.5 0.0252 0.0252 0.0250 0.0248 0.0246 0.0245 0.0244 5.0 0.0264 0.0263 0.0263 0.0262 0.0261 0.0260 0.0258 Depletion Calculations

8.

Section 2.1 discusses a FORTRAN code that is used in addition to SCALE "to interpolate between burnups from the OPUS output and also to decay the isotopic content to the desired cooling time." It is not clear how isotopic content as a function of cooling time is calculated from the discussion in Section 2.1. In Section 6.4, "Interpolation of Isotopics and Cooling Time Verification," more discussion is provided; and it is implied that isotopics are decayed internally within the FORTRAN code without reliance on SCALE/ORIGEN-S.

a.

Is the FORTRAN code managed under a quality assurance program that meets the requirements of Title 10 of the Code of Federal Regulations (10 CFR) 50, Appendix B?

Response

Yes. The Fortran program, INTRPND3.exe, is controlled under NETCO's quality assurance program that meets the requirements of 10CFR50, Appendix B, 10CFR21, and ASME NQA-1. The program has been audited by NUPIC. NETCO maintains documented procedures and assigned responsibilities to control the engineering activities relative to the acquisition, classification, development, testing, evaluation, modification, use, maintenance, retirement, and user notification of computer software utilized by NETCO for applications that are safety related or important to safety. Software is controlled under NETCO Quality Assurance Directives(QADs) in the Quality Assurance Implementing Procedures Handbook. Specifically, QAD 2.4, Software Control, provides procedures for software acquisition, software design, Error Notification, Configuration Control, User Documentation, Verification/Validation, Software TestinglBenchmarking and Run Log maintenance. These features are subject to further procedural control as provided for in NETCO Software Control Procedures, SCP-001, Procedure for Classification of NETCO Software Used for Engineering Calculations, and SCP-002, Procedure for NETCO Computer Identification and Installed Software Inventory.

Additionally, QAD 2.10, Control of Manual and Computerized Calculations, provides

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 9 of 36 procedures for documentation of the accuracy, traceability and verifiability of computerized calculations.

Proprietary (contains source code) and non-proprietary (no source code) versions of the validation reports for INTRPND3 are attached.

b. Provide a description of the cooling time model used in the interpolation program and explain why SCALE isn't used directly for isotopic decay calculations.

[The Response to RAI 8.b being Proprietary is provided in Attachment 4]

c.

Section 6.4 provides verification of the cooling time model in the FORTRAN code for only four cases. Explain why this sample of cases provides assurance that the FORTRAN code will conservatively calculate k-effective relative to SCALE for the burnup and enrichment domain covered by the burnup loading curve.

Response

The four cases were 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, 1 year, 5 years, and 25 years at 40 GWd/T burnup. The model for decaying all nuclides is simple enough that once it is demonstrated to work at both short and long decay times, it is confirmed that the model is working correctly.

There is no burnup dependence of the cooling time model and so there is no reason to suspect that lower burnups would behave any differently (at low burnup, the concentration of nuclides that are decayed is smaller).

9.

Section 5.4, "Limiting Depletion Parameters-Specific Power," states that nominal specific power is used. However, in Section 5.6, "Summary of Depletion Assumptions for Fuel = 3.5 wt percent U-235," Item c. of the first paragraph states that a higher than nominal specific power is used during depletion. Please correct the apparent discrepancy.

Response

The NRC is correct about the discrepancy and it will be corrected in Section 5.4 and Table 10.1 (second column for guidance 2.b.ii). Section 5.6 is correct when it states that the actual specific power used is 1.08 times the nominal specific power. The fuel and moderator temperature effect on reactivity is much greater than the specific power effect.

Therefore Section 2.b.ii of DSS-ISG-2010-01 would allow the use of the specific power that matches the peaking factor used for the fuel and moderator temperature, 1.35 or 1.4.

Depletion with a lower specific power is conservative when using fission products so using 1.08 as a peaking factor rather than 1.35 or 1.4 is conservative.

10.

Section 5.8, "Depletion Analysis Details (Time Steps, etc.)," briefly discusses a few important depletion analysis details.

a. What type of modeling guidance is used when performing depletion calculations using SCALE/TRITON?

For example, NUREG/CR-7041, "SCALE/TRITON Primer: A Primer for Light Water Reactor Lattice Physics Calculations," describes

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 10 of 36 in detail some best practices regarding use of SCALE/TRITON modules that would be applicable to the use of SCALE/TRITON for criticality safety applications.

Response

The depletion analysis uses the t5-depl sequence of TRITON. Since this sequence uses KENO-V.a for the geometric modeling there are not as many modeling issues as if NEWT were used (e.g., no issues regarding mesh, PN,scattering order, number of rings in an absorber, or angular quadrature). However, there are a few modeling details that will be added to Section 5.8. They are:

1. Use 4000 neutrons per generation and 1000 generations. (Shown to be adequate by a convergence study given in Chapter 4 of Reference 10, below)

2. Use a single fuel material for all fuel pins.

3. Use the "flux" option for all burnable absorber materials.

4. Model IFBA as a ring of 0.001 cm thick of ZrB2 meeting the 8-10 areal density. This ring is placed next to the fuel pellet.

5. Small time steps are needed to accurately account for the spectrum change due to Xe and Sm and the initial build in of Pu and other fission products. The initial time steps are 150, 350, 500, 500, 500, followed by steps of 2000 MWd/T until the maximum burnup is reached.

Reference 10: D. B. Lancaster, Utilization of the EPRI Depletion Benchmarks for Burnup Credit Validation, EPRI, Palo Alto, CA, 1025203 (2012).

b. Provide SEP k-effective comparisons by using depleted fuel isotopics from the SCALE/TRITON module used. to support Indian Point 2 criticality safety calculations and a depletion code that has been approved by the NRC for use at Indian Point 2 to provide assurance that SCALE/TRITON depletion modeling is being performed appropriately.

Include a range of cases that are representative of the range of fuel depletion conditions at Indian Point. For example, include cases that model control rod, Pyrex, WABA, and IFBA depletion. Since it is stated that gadolinium and erbium may be used in the future, also include comparisons with representative gadolinium and erbium use.

c. Response:

The change in k with burnup derived from using CASMO-5 and SCALE/TRITON depletion is provided for 198 cases. The CASMO delta k's have been published in Reference 11, listed below, for specific benchmark conditions. These include WABA and IFBA. Reference 11 determines a bias for CASMO-5 then applies those biases to determine benchmarks. For this response, the biases are removed from the Reference 11 tables thus reducing the data to pure CASMO values. This response does not depend on the NRC review of Reference 11.

Tables 10.1, 10.2, and 10.3 provide the difference in the delta k of depletion between CASMO-5 and SCALE/TRITON. Notice that a negative value indicates that SCALE/TRITON conservatively under predicts the reactivity of depletion predicted by CASMO-5. Further notice that the maximum deviation is less than 0.0030. Tables 10.4,

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Attachment I Page 11 of 36 10.5, and 10.6, show the percent difference in the delta k of depletion. The maximum percent difference is 1.37%. At these small differences it is unclear which delta k of depletion is correct so the utilization of the 5% uncertainty allowed by DSS-ISG-2010-01 is appropriate.

Reference 11 does not include Pyrex burnable absorbers or control rods in the depletion and so CASMO-5 depletion with these inserts is not available. Pyrex burnable absorbers are only used for analysis of fuel with < 3.5 wt% U-235 enrichment and Pyrex burnable absorbers will not be ordered in the future. Figure 8.1 shows that very few of the old fuel assemblies are close to the loading curve. In fact there are only 5 fuel assemblies that are above the loading curve by less than 5 GWd/T. The analysis assumed 20 finger Pyrex burnable absorbers but the maximum Pyrex loading of these 5 assemblies was 12 fingers. Due to the conservatism in the depletion assumptions and the limited number of assemblies, no further analysis is needed for validation of the depletion with Pyrex burnable absorbers. Control rod depletion is primarily used for the < 3.5 wt% U-235 fuel.

None of the 5 assemblies with discharge burnups within 5 GWd/T burnup of the loading curve were located under a control bank. Although the new fuel is analyzed with a 2 GWd/T burnup under a control rod, this short burnup has a small impact on reactivity.

Based on the above, no further validation of the control rod depletion is done.

Gadolinium and erbium will be removed from the revised criticality analysis.

In addition, Reference 10 has shown excellent agreement between TRITON/NEWT which was used for the analysis of chemical assays and the TRITON/KENO approach used in this analysis. Finally, the TRITON/KENO was also used to analyze the chemical assays and showed agreement better than 5% in the delta k of depletion. (NET-300067-01, Rev.

0, Adams accession number ML14329A195).

Reference 11: K. S. Smith, et al., Benchmarks for Quantifying Fuel Reactivity Depletion Uncertainty, EPRI, Palo Alto, CA, Technical Report Number 1022909 (2011).

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-15-089 Docket No. 50-247 Page 12 of 36 Table 10.1: SCALEITRITON minus CASMO-5 Delta k of Depletion at 100 Hours Cooling Case description Case 10 20 30 40 50 60 3.25% enrichment depletion 1

-0.0007

-0.0012

-0.0015

-0.0025

-0.0022

-0.0020 5.00% enrichment depletion 2

0.0001 0.0002 0.0003 0.0000 0.0004 0.0006 4.25% enrichment depletion 3

0.0004 0.0000

-0.0002

-0.0005

-0.0008

-0.0005 oft-nominal pin depletion 4

-0.0006

-0.0012

-0.0015

-0.0016

-0.0019

-0.0022 20 WABA depletion ___5 0.0002 0.0007 0.0003

-0.0002

-0.0001

-0.0002 104 IFBA depletion 6

0.0012 0.0011 0.0004

-0.0007

-0.0009

-0.0018 104 IFBA, 20 WABA depletion 7

0.0009 0.0016 0.0008 0.0000

-0.0001

-0.0008 high boron depletion = 1500 ppm 8

0.0004

-0.0001

-0.0003

-0.0004

-0.0001

-0.0001 branch to hot rack = 338.7K 9

-0.0002

-0.0003 0.0000

-0.0005

-0.0001

-0.0001 branch to rack boron = 1500 ppm 10

-0.0008

-0.0015

-0.0019

-0.0023

-0.0026

-0.0025 high power density depletion 11 0.0000

-0.0007

-0.0007

-0.0009

-0.0008

-0.0008 Table 10.2: SCALEITRITON minus CASMO-5 Delta k of Depletion at 5 Years Cooling Case description Case 10 20 30 40 50 60 3.25% enrichment depletion 1

0.0000

-0.0005

-0.0009

-0.0014

-0.0008

-0.0004 5.00% enrichment depletion 2

0.0008 0.0011 0.0008 0.0008 0.0010 0.0013 4.25% enrichment depletion 3

0.0010 0.0005 0.0003 0.0001 0.0002 0.0003 oft-nominal pin depletion 4

-0.0001

-0.0003

-0.0010

-0.0011

-0.0010

-0.0009 20 WABA depletion 5

0.0008 0.0011 0.0009 0.0008 0.0008 0.0010 104 IFBA depletion 6

0.0020 0.0014 0.0009 0.0001 0.0001

-0.0005 104 IFBA, 20 WABA depletion 7

0.0020 0.0024 0.0015 0.0012 0.0006 0.0004 high boron depletion = 1500 ppm 8

0.0007 0.0008 0.0007 0.0003 0.0007 0.0010 branch to hot rack = 338.7K 9

0.0004 0.0006 0.0003 0.0004 0.0009 0.0010 branch to rack boron = 1500 ppm 10 0.0002

-0.0007

-0.0010

-0.0017

-0.0015

-0.0014 high power density depletion 11 0.0006 0.0004 0.0003 0.0004 0.0002 0.0008 Table 10.3: SCALEITRITON minus CASMO-5 Delta k of Depletion at 15 Years Cooling Case description Case 10 20 30 40 50 60 3.25% enrichment depletion 1

0.0007

-0.0004

-0.0014

-0.0016

-0.0015

-0.0009 5.00% enrichment depletion 2

0.0011 0.0014 0.0007 0.0008 0.0008 0.0012 4.25% enrichment depletion 3

0.0014 0.0010 0.0002 0.0000 0.0000 0.0003 off-nominal pin depletion 4

0.0006

-0.0004

-0.0006

-0.0014

-0.0013

-0.0010 20 WABA depletion 5

0.0014 0.0018 0.0010 0.0005 0.0006 0.0005 104 IFBA depletion 6

0.0025 0.0019 0.0010 0.0004

-0.0002

-0.0004 104 IFBA, 20 WABA depletion 7

0.0027 0.0028 0.0017 0.0010 0.0008 0.0004 high boron depletion = 1500 ppm 8

0.0011 0.0009 0.0006 0.0001 0.0006 0.0006 branch to hot rack = 338.7K 9

0.0005 0.0004 0.0001 0.0003 0.0006 0.0008 branch to rack boron = 1500 ppm 10 0.0004

-0.0007

-0.0013

-0.0016

-0.0019

-0.0014 high power density depletion 11 0.0011 0.0008 0.0001 0.0000 0.0004 0.0005 Table 10.4: Percent Difference in the Delta k of Depletion at 100 Hours Cooling

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 13 of 36 (SCALE/TRITON minus CASMO-5 Delta k of Depletion over the Delta k of Depletion)

Case description Case 10 20 30 40 50 60 3.25% enrichment depletion 1

-0.50

-0.52

-0.48

-0.62

-0.48

-0.41 5.00% enrichment depletion 2

0.06 0.08 0.10 0.00 0.10 0.12 4.25% enrichment depletion 3

0.31 0.02

-0.06

-0.14

-0.18

-0.09 off-nominal pin depletion 4

-0.49

-0.54

-0.48

-0.42

-0.40

-0.41 20 WABA depletion 5

0.09 0.31 0.09

-0.04

-0.02

-0.03 104 IFBA depletion 6

0.71 0.51 0.12

-0.19

-0.20

-0.36 104 IFBA, 20 WABA depletion 7

0.37 0.65 0.29

-0.01

-0.01

-0.17 high boron depletion = 1500 ppm 8

0.36

-0.06

-0.09

-0.12

-0.02

-0.01 branch to hot rack = 338.7K 9

-0.16

-0.13 0.00

-0.13

-0.02

-0.02 branch to rack boron = 1500 ppm 10

-0.79

-0.83

-0.76

-0.72

-0.68

-0.58 high power density depletion 11

-0.02

-0.34

-0.25

-0.25

-0.19

-0.16 Table 10.5: Percent Difference in the Delta k of Depletion at 5 Years Cooling (SCALE/TRITON minus CASMO-5 Delta k of Depletion over the Delta k of Depletion)

Case description Case 10 20 30 40 50 60 3.25% enrichment depletion 1

-0.01

-0.22

-0.26

-0.32

-0.17

-0.08 5.00% enrichment depletion 2

0.68 0.51 0.28 0.20 0.21 0.25 4.25% enrichment depletion 3

0.79 0.20 0.10 0.01 0.04 0.05 off-nominal pin depletion 4

-0.09

-0.15

-0.30

-0.25

-0.20

-0.16 20 WABA depletion 5

0.38 0.44 0.29 0.19 0.16 0.18 104 lFBA depletion 6

1.16 0.61 0.30 0.03 0.01

-0.10 104 IEBA, 20 WABA depletion 7

0.78 0.95 0.48 0.30 0.12 0.08 high boron depletion = 1500 ppm 8

0.56 0.34 0.23 0.06 0.14 0.19 branch to hot rack =338.7K 9

0.31 0.25 0.10 0.09 0.18 0.18 branch to rack boron = 1500 ppm 10 0.19

-0.40

-0.37

-0.48

-0.37

-0.31 high power density depletion 11 0.47 0.16 0.10 0.09 0.04 0.15 Table 10.6: Percent Difference in the Delta k of Depletion at 15 Years Cooling (SCALE/TRITON minus CASMO-5 Delta k of Depletion over the Delta k of Depletion)

Case description Case 10 20 30 40 50 60 3.25% enrichment depletion 1

0.49

-0.17

-0.37

-0.33

-0.28

-0.15 5.00% enrichment depletion 2

0.92 0.62 0.23 0.18 0.16 0.21 4.25% enrichment depletion 3

1.09 0.40 0.06

-0.01

-0.01 0.05 off-nominal pin depletion 4

0.47

-0.18

-0.17

-0.30

-0.24

-0.16 20 WABA depletion 5

0.66 0.69 0.30 0.10 0.11 0.08 104 IFBA depletion 6

1.37 0.77 0.28 0.09

-0.03

-0.07 104 IFBA, 20 WABA depletion 7

1.03 1.07 0.52 0.24 0.17 0.06 high boron depletion = 1500 ppm 8

0.86 0.37 0.18 0.01 0.11 0.10 branch to hot rack = 338.7K 9

0.38 0.15 0.03 0.06 0.11 0.13 branch to rack boron =1500 ppm 10 0.38

-0.38

-0.45

-0.42

-0.43

-0.28 high power density depletion 11 0.84 0.32 0.03

-0.01 0.07 0.08

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-l15-089 Docket No. 50-247 Page 14 of 36

11.

Is the use of SCALE, for depletion and criticality calculations, managed under a quality assurance program that meets the requirements of 10 CER 50, Appendix B?

Response

Yes.' Similar to RAI #8, SCALE 6.1.2 is controlled under NETCO's quality assurance program that meets the requirements of 10CFR50, Appendix B, I0CFR21, and ASME NQA-I. The program has been audited by NUPIC.

NETCO maintains documented procedures and assigned responsibilities to control the engineering activities relative to the acquisition, classification, development, testing, evaluation, modification, use, maintenance, retirement and user notification of computer software utilized by NETCO for applications that are safety related or important to safety. Software is controlled under NETCO quality assurance directives in the implementing procedures handbook.

Specifically, QAD 2.4, Software Control, outlines procedures for software acquisition, software design, Error Notification, Configuration Control, User Documentation, VerificationNalidation, Software Testing/Benchmarking and Run Log maintenance.

These features are subject to further procedural control as provided for in NETCO Software Control Procedures, SCP-001, Procedure for Classification of NETCO Software Used for Engineering Calculations, and SCP-002, Procedure for NETCO Computer Identification and Installed Software Inventory. Additionally, QAD 2.10, Control of Manual and Computerized Calculations, provides procedures for documentation of the accuracy, traceability and verifiability of computerized calculations.

Utilization of SCALE 6.1.2 in all cases, includes execution of a test suite to assure that, calculated results on NETCO's dedicated application server, echo pre-programmed test results.

In addition, given that SCALE 6.1.2 has been acquired from ORNL's Radiation Safety Information Computational Center (RSICC) as commercial grade software, then under the provisions of NQA-la-2009 commercial grade dedication of SCALE 6.1.2 is required.

Procedural control for software dedication is provided by NETCO Procedure, SCP-003, Procedure for Commercial Grade Dedication of Safety Related Software.

Using SCP-003, SCALE 6.1.2 has been subjected to Commercial Grade Dedication, as it is classified as a system, structure or component that is used to analyze a safety related component.

As such, for acceptance SCALE 6.1.2 has been demonstrated to comply with 21 critical characteristics per EPRI Technical Report, TR-1 025243, Plant Engineering: Guideline for the Acceptance of Commercial-Grade Design and Analysis Computer Programs Used in Nuclear Safety-Related Applications. The critical characteristics, acceptance criteria and dedication methods are contained in the SCALE 6.1.2 Commercial Grade Dedication Plan, which has been subjected to NUPIC review.

As part of the dedication process, SCALE 6.1.2 has been subject to benchmarking.

SCALE 6.1.2 calculations have been statistically compared with 236 fresh fuel low enriched uranium critical experiments, 63 fresh fuel mixed-oxide critical experiments, 117 HTC critical experiments, as well as isotopic depletion comparisons to 92 chemical assay experiments.

12.

Section 8.4, "Depletion Effect of Hafnium Flux Suppression Inserts,' describes specific analyses that define the basis for the penalty to be applied to corresponding

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 15 of 36 minimum burnup requirements for spent fuel storage of assemblies that were depleted with hafnium inserts. There is no discussion regarding why it was assumed that the hafnium inserts are only depleted for the 8 GWd/MTU and not more, why only enrichments greater than 4 percent were considered, and why the two cooling times chosen were considered. Furthermore, hafnium insert usage is not part of assumption verification during the reload design process as indicated by Table 10.6, "Fuel Assembly Operating Requirements for Fuel Enriched > 3.5 wt%." How do the hafnium insert modeling assumptions bound past, p*resent, and future hafnium insert use at Indian Point 2? Discuss any other inserts that have been used at Indian Point

2. Also recognize that the hafnium insert penalty credited in Note (a) and (b) to Table 10.2, "Region 2 Minimum Burnup (GWd/T [gigawatt-days/ton]) Requirements" is based on these very narrow set of depletion conditions and, therefore, it wouldn't be appropriate to apply these penalties to all possible hafnium insert depletion scenarios without further justification; likewise for footnotes to Table 10.3, "Summary of Loading Restrictions," regarding use of hafnium inserts.

Response

The RAI requests information on why 8 GWd/T burnup, enrichments higher than 4 wt%,

and cooling times of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and 25 years were selected. For the burnup, from operating experience, the maximum depletion with hafnium inserts was only 6 GWd/T so using 8 GWd/T bounds the expected depletion of the fuel while under hafnium inserts.

Only enrichments greater than 4 wt% were used since Hafnium inserts were introduced in later cycles where feed enrichments were always greater than 4 wt%. The 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and 25 years cooling times were chosen to cover the entire range of cooling times.

The hafnium inserts are designed to suppress the flux reaching the vessel welds. The hafnium inserts reduce the power in a set of peripheral assemblies and as such there is little burnup in these assemblies when the hafnium inserts are in the assemblies. As stated above the maximum burnup observed to date in assemblies that contain hafnium inserts is only 6 GWdIT. The analysis conservatively assumed 8 GWd/T in order to obtain a conservative burnup adder for depletion under a hafnium insert. Table 8.8 showed that the maximum increase in the burnup requirement was only 1.31 GWd/T. In order to account for any future use of hafnium inserts this was rounded up to 2 GWd/T.

With the 2 GWd/T burnup adder, a case was selected to see what depletion the 2 GWd/T adder would support. A 5.0 wt% U-235 fuel assembly was depleted with :hafnium/IFBA instead of WABA/FBA for 14 GWd/T. The k at 14 GWd/T was 0.9597 in Region 1. The same fuel was depleted for 12 GWd/T (14 GWd/T minus the 2 GWd/T adder) with WABA/IFBA instead of hafnium/IFBA. This calculated k was 0.9650. Since 0.9650 is greater than 0.9597 it is confirmed that the 2 GWd/T adder is sufficient for hafnium depletion up to even 14 GWd/T. If the burnup under the hafnium insert were as high as 14 GWd/T the insert would not be sufficiently suppressing the flux.

Tables 10.6 and 10.7 will be modified to include a restriction of 8 GWd/T burnup under a hafnium insert.

Section 3.3 describes all the fuel assembly inserts used at Indian Point. All historical fuel was reviewed to assure that the inserts used in the analysis covered how all the inserts were actually used at the plant.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

\\

NL-1 5-089 Docket No. 50-247 Page 16 of 36 Criticality Calculations

13.

Section 2.1, "Computer Codes," and Section 6.5, "Convergence of Calculations,"

gives the number of neutron generations and neutrons per generation used, but not the number of skipped generations.

How many skipped generations were used and what starting source distribution was used? Include justification for these assumptions.

Response

For both the number of generations skipped and the starting source the SCALE default was generally used.

For the number of generations skipped the default is 3, however, SCALE calculates the number of generations to skip that gives the minimum uncertainty in the final result. The k reported for all the calculations is the k with the optimum generations skipped. The number of generations skipped span a wide range but is generally between 100 and 200.

When running a large number of generations (always over 1500) the input number of generations skipped is not significant to the final results.

The default start source is a uniform source over all the fissile materials in the model.

The number of neutrons per generation is always 6000 or greater and for the single and four assembly models this sampling was enough to find the most reactive portion of the model. For the large models the start source was selected to concentrate the neutrons at the feature being investigated. For example, the start source for the dropped assembly case started the neutrons at the dropped assembly. For the interface analysis, the neutrons are started at the interface. For these cases, if the feature being investigated is not the most reactive in the model then convergence may be incomplete but that is not a concern since the intent of the model is to show that the feature is less reactive than the small infinite models used for the final k. However, if the feature being investigated does lead the reactivity of the model then using the start source at the feature assures determination of the maximum k.

14.

Section 2.1 describes a process by which various nuclide concentrations are manually changed for the first 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> of decay after fuel depletion instead of using the SCALE/TRITON module to perform this 72-hour decay for all credited nuclides. Provide verification that manually changing nuclide concentrations is conservative relative to performing the calculation with SCALE/TRITON.

Response

The "manually changed" actually refers to the FORTRAN program that performs the decay (see response to RAI 8). The 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> case was compared to the SCALE derived decay and the results were the same (see Table 6.2 of NET-300067-01, Revision 1).

Decay is only provided at the end of the SCALE analysis. In order to decay at burnups less than the maximum burnup the decay was performed outside of SCALE. Methods have been devised using some of the working files output by SCALE but these are more awkward than using the simple short FORTRAN code. Table 6.2 provides the verification. The verification was done at the highest burnup to assure all the isotopes were included.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-l15-089 Docket No. 50-247 Page 17 of 36

15.

Section 3.3, "Fuel Assembly Insert Designs," describes control rod modeling "for the special case of crediting a control rod in a fresh fuel assembly in the pool" where "the Ag-In-Cd content (density) is reduced by 20%."

a. Are the same assumptions applied to all criticality calculations crediting control rods (e.g. burned fuel that doesn't meet the 12 gigawatt-days/metric ton of uranium (GWd/MTU) burnup requirement in Region 1 or Region 2 control rod credit)?

Response

Yes, the same assumptions are applied to all criticality calculations crediting control rods.

b.

Provide more detail describing how the 20 percent reduction bounds m anufacturing tolerances, any absorber material loss during operation, and any modeling assumptions or simplifications, given that Table 8.1, "Calculated k's in Region 1," shows that the minimum margin case for Region 1 occurs with a case crediting a control rod in the fuel assembly.

Response

The 20% reduction was an engineering estimate that would bound any reasonable allowance for manufacturing tolerances and absorber loss. Measured control rod worths must be verified to be within 10% of the predicted worth. It was determined that a reduction of 65% of the material is needed to obtain a loss of 10% in worth.

The region 1 and region 2 control rod cases were re-run using only 35% of the control rod material remaining and it was confirmed that the k95195 requirements were still met.

The calculated k95195 for a fresh assembly in Region 2 with a control rod k95195 is 0.9772. The calculated kgS/95 for a control rod in a loading curve assembly in Region 2 with a missing absorber panel k95195 is 0.9888. The calculated k95/95 for a control rod in fresh fuel in Region 1 with a missing absorber panel k95195 is 0.9892.

c. Are there different control rod designs or variations in designs available onsite or will there be in the future?

Response

There are currently 2 part length control rods in the IP2 Spent Fuel Pool and 8 part length control rods in the 1P3 Spent fuel pool. Part length control rods are not credited in this analysis. Any future changes to control rod design will be captured in existing processes for changes to the facility.

16.

Section 6.3, "Axial Burnup Distribution," mentions that "the lower 10 nodes were averaged into one node" when accounting for the axial burnup distribution in criticality calculations. Were 'calculations performed to confirm that this modeling simplification does not affect the k-effective calculation? If not, provide confirmation

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 18 of 36 that this simplification is conservative.

Response

The DOE generated axial burnup profiles (and analyzed in NUREG/CR-6801) were used for this analysis. The axial burnup profiles taken from NUREG/CR-6801 are reprinted as Table 6.1 in the criticality analysis. Other than the axial burnup profiles the model is axially symmetric. (Confirmation calculations included the fission gas plenum but this was a small effect.) Table 16.1 shows the ratio of the top half relative power to the bottom half relative power for the twelve burnup bins. For all distributions past 14 GWd/T the top half of the fuel has less burnup and therefore drives the reactivity. On a node by node basis the top half has less or equal burnup in every node for burnups greater than 30 GWdIT.

For burnups of 14 through 30 GWd/T the top ends are clearly more limiting than the bottom ends but some of the more central nodes have less burnup at the bottom half.

However, for all these central nodes the relative burnup is greater than 1 so the uniform burnup assumption would be more limiting if the central nodes were important.

The averaging of the bottom 10 nodes into an average node effectively moves the low burnup of the bottom toward the top. Table 16.1 also shows the relative burnup of the node derived by averaging the bottom ten and the ratio of this average node to the NUREG/CR-6801 1 0 th node. As can be seen from Table 16.1 the burnup at the 1 0 th node was decreased by using the average and this would provide conservative results (but the conservatism is negligible).

As stated in the second paragraph of Section 6.3, for burnups less than 18 GWd/T the 14 to 18 GWd/T shape was used since it was the most limiting shape for all bins less than 18 GWd/T. This position is not obvious from the relative burnups alone so the Region 2 case for 2.5 wt% U-235 at 13.97 GWdIT and the Region 1 case with 5 wt% U-235 at 12 GWd/T were rerun with the full 18 nodes and the 10 to 14 GWd/T burnup profile. As can be seen on Table 16.2 the 14 to 18 GWdIT burnup shape was more limiting.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 19 of 36 Table 16.1: Ratio of Relative Burnups, Top Nodes Divided by Bottom Nodes (Node 1 is bottom of fuel) 1 2

3 4

5 6

7 8

9 10 11 12 Burnup 46 42-46 38-42 34-38 30-34 26-30 22-26 18-22 14-18 10-14 6-10

<6 (G~d/T Sum of Nodes 10-18 0.96 0.96 0.96 0.93 0.93 0.95 0.93 0.93 0.78 1.25 1.04 0.95

÷~ Nodes 1-9 NoeRto 0.88 0.77 0.80 0.69 0.71 0.74 0.71 0.56 0.44 0.81 0.86 0.83 18/1 NoeRto 0.91 0.86 0.88 0.78 0.76 0.76 0.74 0.65 0.46 0.87 0.89 0.84 17/2 NoeRto 0.93 0.96 0.95 0.91 0.90 0.84 0.82 0.81 0.51 1.05 0.96 0.88 16/3 NoeRto 0.95 0.99 0.98 0.96 0.96 0.95 0.94 0.96 0.62 1.39 1.03 0.93 15/4 NoeRto 0.96 0.99 0.99 0.97 0.98 1.01 1.00 1.05 0.84 1.61 1.10 0.98 14/5 NoeRto 0.97 1.00 0.99 0.98 0.99 1.03 1.02 1.08 0.96 1.69 1.14 1.01 13/6 NoeRto 0.98 1.00 0.99 0.99 0.99 1.04 1.02 1.07 0.99 1.63 1.15 1.02 12/7 NoeRto 0.99 1.00 1.00 0.99 1.00 1.03 1.01 1.04 1.01 1.26 1.12 1.02 11/8 NoeRto 1.00 1.00 1.00 1.00 1.00 1.01 1.01 1.01 1.00 1.06 1.05 1.01 10/9 Average of bottom 10 1.029 1.026 1.026 1.042 1.040 1.038 1.044 1.046 1.130 0.926 0.992 1.028 nodes Average of Bottom 10 ÷-

0.94 0.95 0.94 0.95 0.95 0.91 0.92 0.92 0.95 0.74 0.92 0.97 Node_10 Table 16.2: Calculated keff Using Bin 10 (10-14 GWd/T) and Bin 9 (14-18 GWd/T) Shapes Bin 10 (10-14 GWd/T) Shape Bin 9 (14-18 GWd/T) Shape (Conservatively used in criticality analysis) 2.5 wt% U-235 at 13.97 0.9573 0.9698 GWd/T In Region 2 5.0 wt% U-235 at 12.0 0.9605 0.9650 GWd/T in Region 1

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 20 of 36

17.

Table 7.1, "Tolerance Reactivity Effects," indicates that only positive fuel pin pitch changes were considered when determining the fuel pin pitch tolerance reactivity effect.

Please verify that negative fuel pin pitch changes do not result in larger reactivity effects.

Response

More water between pins in the assembly increases k because the assembly is under-moderated. Consequently, less water between pins in the assembly will decrease k. In order to confirm this, the confirmation case at 5.0 wt%, 43.07 GWdIT was run with reduced pin pitch. The calculated k was 0.9593 with a reduced pin pitch, as compared to the confirmation k of 0.9636 with an increased pin pitch.

18.

Section 8.1 discusses criticality calculations involving fresh fuel crediting integral fuel burnable absorber (IFBA). Due to the large flexibility allowed by SCALE regarding cross-section processing options and geometry modeling, which could significantly impact k-effective estimation, describe how the IFBA is modeled and provide a sample SCALE input deck for a case crediting IFBA.

Response

The IFBA is modeled as a ring 0.00 1 cm thick on the outside of the fuel rod pellet. The ring material is ZrB2. The B-I10 atom density is set to match the linear B-I10 specification (1.77 mg B1°/inch) and the Zr atom density is set to match the stoichiometry. No smearing of the IFBA within the pellet or gap is done. The resonance treatment is done ignoring the IFBA using standard lattice cell input. To confirm that the IFBA does not require any special resonance treatment, a case was analyzed where the lattice cell input was replaced by "multiregion" input. The multiregion input consisted of cylinders of pellet, followed by IFBA, followed by gap, followed by clad, followed by moderator where the moderator area was preserved. For this test case, as with the standard analysis, the resonance treatment was the same for the fuel pins whether they had an IFBA or not.

This would maximize the effect of the treatment change. As can be seen on Table 18.1 the impact of the change in treatment from lattice cell to multiregion was not significant and the lattice cell approach that was used was very slightly conservative.

The sample input deck follows.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 21 of 36 Table 18.1: keff for Region I with 5 wt% U-235 Fuel with 48 IFBA Rods Using Two Resonance Treatments Resonance Treatment Calculated k Sigma Lattice Cell 0.9717 0.0001 Multiregion 0.9714 0.0001

=csas5

'Input File Begins Region 1 -

5 wt% with 48 IFBA -

0 ppm 4 panels v7-238 read camp uo2 10 0.975 277 92235 5

92238 95 end zirc4 2 1 277 end h2o 3

denr~l 1 277 end h2o 31 den=l 1 277 end h2o 32 den~l 1 277 end ss304 4

1.0 end

'Snap In Borated Al (x gmBl0/cm2) b4c 6

Den=2.65 0.17960713 end AI-27 6

Den=2.65 0.82039287 end

'ifba coating at iX (1.77 mg/inch) zr 7

0

.0322187 277 end b-l0 7

0 0.0143942 277 end end camp READ CELLDATA latticecell squarepitch pitch=l.43002 3 fueld=0.93091 10 gapd=0.96647 0

cladd=1.06807 2 end

'MULTIREGION CYLINDRICAL RIGHTBDY=WHITE END 1i0 0.465455 7

.466455 0 0.483235 2 0.534035 3 0.806802 END ZONE end celidata read parm tme=900.0 gen=4000 npg=16000 run=yes plt=no htm=no far=no NB8=500 end parm read geom unit 1 com=

• Fuel Pin Cell

• cylinder 10 1 0.465455 365.76 0.0 cylinder 0 1 0.483235 365.76 0.0 cylinder 2 1 0.534035 365.76 0.0 cuboid 3 1 4p0.71501 365.76 0.0 unit 43 com= 'ifba rod cut to 120 inches' cylinder 10 1 0.465455 304.8 0.0 cylinder 7 1 0.466455 304.8 0.0 cylinder 0 1 0.483235 304.8 0.0 cylinder 2 1 0.534035 304.8 0.0 cuboid 3 1 4p0.7l501 304.8 0.0 unit 42 com 'short 12 inch fuel pin' cylinder 10 1 0.465455 30.48 0.0 cylinder 7 1 0.466455 30.48 0.0 cylinder 0 1 0.483235 30.48 0.0

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-15-089 Docket No. 50-247 Page 22 of 36 cylinder 2 1 0.534035 30.48 0.0 cuboid 3 1 4p0.71501 30.48 0.0 unit 41 array 4 0.0 0.0 0.0 unit 2 com='guide-tube cylinder 3 1 0.65278 365.76 0.0 cylinder 2 1 0.69088 365.76 0.0 cuboid 3 1 4p0.71501 365.76 0.0 unit 9 com=~'instrument tube cylinder 3 1 0.65278 365.76 0.0 cylinder 2 1 0.69088 365.76 0.0 cuboid 3 1 4p0.71501 365.76 0.0 unit 3 com=' fuel assembly, array 1 3*0.0 SRegion 1 units unit 25 com='steel box Region 1 cuboid 3 1 22.4155 0.1905 22.4155 0.1905 365.76 0.

hole 3 0.57785 0.57785 0.0 hole 35 0.1905 0.2667 0.

hole 36 0.40896 22.19706 0.0 cuboid 4 1 22.606 0.0 22.606 0.0 365.76 0.

unit 21 com='boraflex box top' cuboid 3 1

19.4691 0.0889 0.28448 0.0 365.76 0.0 cuboid 4 1

19.558 0.0 0.37338 0.0 365.76 0.0 unit 22 com='boraflex box bottom' cuboid 3 1

19.4691 0.0889 0.37338 0.0889 365.76 0.0 cuboid 4 1

19.558 0.0 0.37338 0.0 365.76 0.0 unit 23 com='boraflex box right' cuboid 3 1

0.28448 0.0 19.4691 0.0889 365.76 0.0 cuboid 4 1

0.37338 0.0 19.558 0.0 365.76 0.0 unit 24 com='boraflex box left' cuboid 3 1

0.37338 0.0889 19.4691 0.0889 365.76 0.0 cuboid 4 1

0.37338 0.0 19.558 0.0 365.76 0.0 "unit 26 com='Region 1 vertical separation plates' cuboid 4 1

0.1905 0.0 2.089149 0.0 365.76 0.0 unit 27 com=r'Region 1 horizontal separation plates' cuboid 4 1

2.368548 0.0 0.1905 0.0 365.76 0.0 unit 35 com='vertical snap in Reg 1' cuboid 6 1 0.21844 0.0 22.1488 0.0 365.76 0.0 unit 36 com='horizontal snap in with cut out Reg 1' cuboid 6 1 21.930359 0.0 0.21844 0.0 365.76 0.0 unit 700 com='Region 1 single assembly with Snap in' cuboid 3 1 27.3431 0.0 26.7843 0.0 415.76

-50.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Attachment I Page 23 of 36 hole 25 2.36855 2.08915 0.0 hole 21 3.89255 24.69515 0.0 hole 22 3.89255 1.71577 0.0 hole 23 24.97455 3.61315 0.0 hole 24 1.99517 3.61315 0.0 hole 26 2.36855 0.0 0.0 hole 26 24.78405 0.0 0.0 hole 26 2.36855 24.69515 0.0 hole 26 24.78405 24.69515 0.0 hole 27 0.0 2.08915 0.0 hole 27 0.0 24.50465 0.0 hole 27 24.97455 2.08915 0.0 hole 27 24.97455 24.50465 0.0 global unit 99

'Region 1

array 7 0.0 0.0

-50.0 end geom read array ara=7 com='2 by 2 of region i1 nux=2 nuy=2 nuz=1 till 700 700 700 700 end fill ara=4 com='stacking for IFBA pin!

nux=1 nuy=l nuz=3 fill 42 43 42 end fill ara=l com='layout of pins in the assembly nux=15 nuy=15 nuz=l fill 1

1 11 11 11 11 11 111 11 11 11 1

1 1111 11 1

1 1

2 41 1

2 41 1 41 2

1 41 2

1 1

1 1 41 1 41 41 41 2 41 41 41 1 41 1

1 1

1 1 41 2 41 1

1 1 41 2 41 1

1 1

1 1

2 41 41 1

1 1

1 1 41 41 2

1 1

1 1 41 41 1

1 1

1 1

1 1 41 41 1

1 1

1 12 11 19 11 12 111 1

1 41 41 1

1 1

1 1

1 1 41 41 1

1 1

1 2 41 41 1

1 1

1 1 41 41 2

1 1

1 1

1 41 2 41 1

1 1 41 2 41 1

1 1

1 1 41 1 41 41 41 2 41 41 41 1 41 1

1 1

1 2 41 1

2 41 1 41 2

1 41 2

1 1

1 1

1111 11 1

111 11 1

1 1

111 1

111 1

11 111 end fill end array read bnds xfc=periodic yfc=periodic zfc=void end bnds read plot scr=yes pic=mat ipi=l0 clr=

0 0

0 0

1 255 0

0 2

255 255 255 3

0 0 255 4

255 255 0

6 73 205 50

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 24 of 36 10 255 0

0 end color ttl-'2-d radial cross section' xul=-.3 yui=53.8686 zul=.5 xlr=54. 9862 ylr=-.3 zlr=. 5 ndn= 5000 uax~l vdn=-l end ttl='2-d axial cross section' yul=6, xul= -.3 zui=415.76 ylr=6.

xlr=53. 8686 zlr= -50 nax=5 000 uax~l wdn=-l end end plot end data end

19.

In Table 8.1, explain why the k-effective of 0.9717 doesn't match the corresponding k-effective of 0.97182 in Table 8.10 given that these are supposed to be the same case.

Response

The value on Table 8.10 was from an early model and should have been updated to 0.9717. This did not have any effect on the full pool models. It will be corrected.

20.

The following statements in Section 8.2.1, "Curve Fit," regarding the curve fitting process of the data in Table 8.2, "Minimum Burnup Requirements (GWd/T) in Region 2" are misleading:

The coefficients contain an adjustment to ensure that all burnups calculated by the equation are greater than the burnups from the table. Using the curve fit results in a maximum penalty of 0.2 GWd/TI for low enrichments and 0.4 GWd/T for high enrichments when compared to the tabulated values shown in Table 8.2.

Confirmatory analysis shows that the defined fits can actually be non-conservative between 3 percent and 3.5 percent enrichment for cooling times of zero and one year, and for enrichments between 4.5 percent and 5 percent for cooling times of 15 and 25 years. The confirmatory analysis shows that approximately 10 percent of burnups are non-conservative with respect to the non-fitted loading requirements, defined by linearly interpolating between points in Table 8.2, up to a maximum of approximately 0.4 GWd/MTU.

Please revise the discussion and/or update the fits accordingly.

Response

The reviewer is correct. Although all curve fit burnups are equal to or greater than the tabulated burnups, some data points using the curve fit result in a burnup requirement that is less than the burnup obtained by interpolating from the table values. These lower burnup requirements were used in four special runs (3.25 wt% at 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and 1 year cooling and 4.75 wt% at 15 year and 25 year cooling). The k595/ was still less than 0.99 in 3 of the 4 cases while the k95195 at 4.75 wt%, 15 year cooling was 0.9914. After raising

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 25 of 36-the curve fit by 0.4 GWd/T for enrichments greater than 3.5 wt%, and including the effect of fission gas release (see response to RAI 21), the final k95195 for the worst case interpolated point is now 0.9894 (4.75 wt% at 15 year cooling). The quoted text in the criticality analysis has been revised to read as follows The coefficients contain an adjustment to ensure that all burnups calculated by the equation are greater than the burnups from the table. Using the curve fit, however, does result in some curve fit burnups being slightly less than obtained by linear interpolation between table values. For these cases, con firm atory calculations were done at the lower burnup given by the curve fit to confirm that the resulting k95/95 meets the requirements.

However, only the curve fit will be used to determine the required burnup and linear interpolation will not be used. The statement "The table can be interpolated to find the required burnup at any enrichment/cooling time combination." will be removed from NET-300067-01.

21.

Section 8.6, "Volatile Fission Gases," estimates the reactivity effect associated with a 10 percent release of volatile fission gases from the spent fuel citing Regulatory Guide (RG) 1.183, "Alternative. Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors," as the basis for using 10 percent. However, the SEP environment is not the focus of RG 1.183, and fuel in the SEP can be stored there for several decades or more. Consequently, it is not clear that the 10 percent value taken from RG 1.183 would be applicable to long term storage in a SEP environment.

Please re-consider the volatile fission gas release fraction assumed in the criticality safety analyses taking into account long term storage of spent fuel in the SFP.

Response

When the fuel is in the spent fuel pool, the temperature of the fuel is too low to have much increased release of fission gases. Olander in his textbook, Fundamental Aspects of Nuclear Reactor Fuel Elements, states in Section 15.1:

"At low temperatures (less than about 13000 K), the mobility of fission-gas atoms is too low to permit appreciable gas-atom movement, either to release surfaces of even to sites where bubbles can form. The fission gases are frozen into the matrix of the solid, and only the gas formed very close to an external surface can escape. Release occurs both by direct flight from the fuel while the gas atom is still an energetic fission fragment (recoil) or by interaction of a fission fragment, a Collision cascade, or a fission spike with a stationary gas atom near the surface (knockout). These release mechanisms are independent of both temperature and temperature gradient. Since they affect only the outer layer of the fuel (within ~ 10 pm of the surface), the fraction of the total fission gas released by recoil or knockout is quite small" Although the SEP environment should not increase the fission gas release, the use of Regulatory Guide 1.183 was reviewed. Regulatory Guide 1.183 has a revision draft for review where the release fractions have changed. Further PNNL-1 8212 Rev.1 was issued to support the revision to the Regulatory Guide. The PNNL-1 8212 Rev. 1 was completed June 2011. The draft Regulatory Guide was issued October 2009. The values for release fractions will be based on the newer PNNL-1 8212 Rev. 1. Table 2.9 of

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247.

Page 26 of 36 the PNNL-18212 gives recommended release fractions. For Kr-85 the maximum release fraction given is 38%. Table 2.1 of PNNL-1 8212 shows that the uncertainty component of the 38% is 6.61%. All the noble gases will be reduced to 68% of their initial values (which is a reduction of 32%, a roundup of 38% minus 6.61%). Table 2.9 gives 8% for the release of noble gases other than Kr-85 but all the other noble gases used in the PNNL study are short lived and the criticality analysis is crediting stable noble gases so it was conservatively assumed that the difference in release fractions is due to the decay time so all Xe isotopes and Kr isotopes are reduced by 32%.

The iodine and bromine isotopes are reduced to 90% of their initial values. This was to be consistent with item C.1I.d of Regulatory Guide 1.25, Assumptions Used For E~valuating The Potential Radiological Consequences Of A Fuel Handling Accident In The Fuel Handling And Storage Facility For Boiling And Pressurized Water Reactors. PNNL-18212 Table 2.9 would have allowed a maximum release fraction of 8%. The halogens have low reactivity worth so the difference between 8% and 10% is not significant.

The Rb isotopes are reduced to 56% of the initial value. This was determined by taking the maximum alkali metals release fraction of 50% from Table 2.9 of PNNL-1 8212 and reducing it by the 6% uncertainty from Table 2.1.

Cs isotopes are not reduced. The key Cs isotope is Cs-I133 which is one of the isotopes used in transport analysis as listed in ISG-8 Rev. 3. Cesium is volatile and will move toward the outside of the pellet but at the fuel clad interface, the temperatures are low enough and the oxygen potential of the pellet is high enough that the Cs forms an oxide that then stays in place. Cesium is reactive and reacts with other fission products, which also stops the migration. There are a number of measurements that support the assertion that cesium does not migrate up or down from the plane where it was generated.

1.

Cs-137 has been used as a burnup index for chemical assays. Although Nd-148 is the preferred way of establishing burnup of a sample, Cs-I137 has also been used and generally gives the same burnup determination.

• 2.

Cs-I137 is the isotope used for the BNFL burnup measurement device. The measurements performed matched the burnup records of the utilities where the device has been employed. (Andrew S Chesterman, et. al., Burnup Credit Measurements for Cask Loading Compliance - A Review of Techniques and Calibration Philosophies, WM2011I Conference, February 27 - March 3, 2011, Phoenix, AZ.)

3.

The PNL series of measurements on the Approved Testing Material (ATM) used for the chemical assays showed that the Cs-I137 remained in place. (PNL-51 09-103, 104, 105, 106). ATM-106 did observe some movement of the Cs-137 in Rod NBD107 that blurred the expected dips under the grids.

The PNNL-1 8212 is conservative since it assumes a high power to be bounding for all fuel. The power for the Indian Point fuel is generally less than that assumed in PNNL-18212. As a final check, the fission gas release was measured in the ATM samples. The maximum fission gas released in all the ATM samples was 11.2%. (PNL-51 09-106 Section 2)

The confirmatory calculations given in Table 8.4 of the criticality analysis have now been

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 27 of 36 re-run with the noble gases reduced by 32%, the halogens reduced by 10% and the Rb reduced by 44%. At the maximum burnup of the loading curves the impact of reducing the isotopes described here is only 0.0014 in k.

In response to this RAI and also RAI 7, the loading curve has been moved up by 0.4 GWd/T for enrichments greater than 3.5 wt%. For the confirmatory calculations, the plenum is modeled as 6 inches long containing a steel spring that occupies 20% of the plenum volume. This plenum model is more conservative due to reduced neutron absorption by the amount of spring steel than the one described in NET-300067-01, Revision 1, which assumed that steel comprised 50% of the plenum volume. The maximum k95195 in Table 8.6 is now 0.9899 and the maximum k96195 for any interpolated point in the table is now 0.9894 (4.75 wt%, 15 year cooling, 36.56 GWdIT).

The revised burnup requirements and confirmatory tables are shown below.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 28 of 36 Table 8.2: Minimum Burnup Requirements (GWd/T) in Region 2 Enrichment Cooling Time (years)

(wt% U-235) 0 1

2 5

10 15 25 2.0 3.20 3.10 3.08 3.00 3.00 3.00 3.00 2.5 15.17 14.85 14.66 13.97 13.20 12.84 12.31 3.0 21.28 21.17 20.98 20.66 20.26 19.98 19.65 3.5 27.53 27.10 26.63 25.56 24.29 23.50 22.45 4.0 34.22 33.83 33.45 32.44 31.12 30.10 28.84 4.5 39.38 39.05 38.39 36.89 35.07 34.09 33.00 5.0 43.07 42.54 42.18 41.18 40.12 39.36 38.08 Table 8.4: Calculated k Values at Each Burnup Point Enrichment Cooling Time (years)

(wt% U-235) 0 1

2 5

10 15 25 2.0 0.9722 0.9713 0.9713 0.9712 0.9703 0.9698 0.9691 2.5 0.9700 0.9696 0.9694 0.9697 0.9702 0.9696 0.9695 3.0 0.9660 0.9652 0.9656 0.9651 0.9646 0.9643 0.9640 3.5 0.9672 0.9671 0.9672 0.9669 0.9667 0.9664 0.9668 4.0 0.9638 0.9632 0.9630 0.9625 0.9626 0.9635 0.9634 4.5 0.9613 0.9610 0.9631 0.9639 0.9634 0.9619 0.9615 5.0 0.9636 0.9630 0.9616 0.9609 0.9597 0.9593 0.9622 Table 8.6: k9 s/9s at Each Burn up Point For Region 2 Enrichment Cooling Time (years)

(wt% U-235) 0 1

2 5

10 15 25 2.0 0.9856 0.9847 0.9847 0.9846 0.9837 0.9832 0.9825 2.5 0.9849 0.9845 0.9842 0.9845 0.9849 0.9843 0.9841 3.0 0.9859 0.9852 0.9856 0.9850 0.9845 0.9841 0.9838 3.5 0.9889 0.9887 0.9888 0.9884 0.9880 0.9877 0.9879 4.0 0.9874 0.9868 0.9866 0.9859 0.9858 0.9866 0.9864 4.5 0.9865 0.9862 0.9881 0.9887 0.9880 0.9864 0.9859 5.0 0.9899 0.9894 0.9879 0.9871 0.9858 0.9853 0.9879

22.

In Section 8.10.2, "Results of Reduced Periphery and Region 1/Region 2 Interface Analysis," regarding the pool refiector sensitivity studies, it is stated that "the concrete used is a conservative mixture created by Oak Ridge-- named as orconcrete."

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 29 of 36 Explain why this concrete is considered to be a conservative mixture.

Response

The previous analysis did not attempt to generate a conservative concrete. For this revised analysis a conservative concrete is determined by taking the 4 concrete compositions given in SCALE and using the maximum weight percent of the positive worth elements and the minimum weight percent of the negative worth elements. The positive worth elements are C, 0, Na, Mg, Al, and Si. The negative worth elements are H, N, 5, Cl, K, Ca, Ti, Mn, and Fe. In order to keep the atom densities of these elements the same in the conservative concrete as in the SCALE concretes, the density has been increased. The conservative concrete composition is given in Table 22.1.

Table 22.1 Conservative Concrete Composition (Density = 2.91 g/cm 3)

Element Weight Percent Fe 0.45 H

0.26 C

13.97 0

42.41 Na 2.31 Mg 7.51 Al 2.71 Si 26.87 Ca 3.51 This concrete composition is the same as given in the EPRI Sensitivity Analyses for Spent Fuel Pool Criticality (Technical Report Number 3002003073, issued December 2014).

The limiting keff of the Region 2 is taken from the 2x2 model with periodic boundary conditions and therefore is not sensitive to the concrete composition. The relaxed peripheral burnup requirements still produced a lower keff in the full pool model than the 2x2 model.

However, it is possible that the new conservative concrete could change that conclusion. To test this, the case with concrete closest to the rack (only a 1 cm water gap) from Table 8.13 was run using the dry conservative concrete (The dry conservative concrete uses the same atom densities as derived from Table 22.1 but the hydrogen atom density is zeroed. This is even more conservative than the Table 22.1 concrete).

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 30 of 36 The calculated k was 0.9648 where the k using the previous concrete was 0.9647. Since the Monte Carlo sigma on both runs was 0.0001 it can be concluded that the concrete composition makes no impact on Region 2. Further the keff from the 2x2 model is 0.9668.

Thus the conclusion that the concrete composition does not affect the limiting keff in Region 2 is confirmed.

In Region 1, however, the limiting keff is due to the peripheral assemblies and therefore the results can be dependent on the concrete composition. The model used in the criticality analysis was rerun with dr concrete. The keff increased from 0.9759 (See Table 8.12) to 0.9849. Previously, the full pool model assumed a 1.25 inch gap on all sides between the racks and pool walls, yet still met the target 1% margin so further model improvements were not necessary. However, with the new very conservative concrete the margin was reduced from 1% to 0.4%. In order to remedy this situation, the full pool model was improved. The gap between the North wall and the Region 1 rack is only 1.25 inches (which was used for all faces in the criticality analysis) but the gap between the East wall and the Region I rack is 2.125 inches. The model was updated to have the correct gap on both sides and the calculated keff dropped to 0.9814. Assuming that the concrete has lost more hydrogen than the driest of the SCALE concrete compositions is too conservative so the composition given in Table 22.1 was then used. The calculated keff is 0.9793 which becomes 0.9904 after adding the bias and uncertainties. This is the highest k195/9 for normal operation of the spent fuel pool. It assumes a conservative concrete composition and all the peripheral assemblies are fresh 5 wt% U-235 assemblies with no burnable absorbers. Although such fuel is allowed in the pool in these locations, high enrichment fuel without any burnable absorbers has never been ordered by Indian Point.

23.

In Section 8.11, "Failed Fuel Canisters," an analysis is described where "36 fuel pins could be loaded into each failed fuel container with no criticality concern." This conclusion is valid given strict control on the configuration that was analyzed. What controls will ensure that any failed fuel canisters will remain in the geometry analyzed? Alternatively, what will be the actual range of variation of fuel pin geometry allowed in the failed fuel canister and why does the analysis presented bound all potential fuel pin geometry variations?

Response

Cases were run in which the pin pitch of the 36 fuel pins was varied. When the pin pitch is increased, k decreases because the initial arrangement is over-moderated. When the pin pitch is decreased, k decreases again because although the reactivity of the 36 pins increases due to the over-moderation, the 36 pins become further isolated from the other assemblies in the pool such that the net effect is a decrease in k. As a final check, the 36 pins were arranged in an optimum pitch and placed close to each other and the other fuel assemblies (see Figure below). For this case, the k was larger than the original k (0.9458 vs 0.9374) but after bias and uncertainty is added, the k51/95 is 0.9722 which still meets the requirements. So any arrangement of the 36 pins is permissible. Section 10.2 will be modified to state that the failed fuel canisters cannot contain more than 36 fuel pins.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 31 of 36

24.

Provide more discussion regarding Section 8.12, "Fuel Rod Storage Basket," to address the following:

a.

How is geometry controlled?

Response

Fuel Rod Storage Basket contains a structure which holds fuel rods in the geometry shown.

b.

Are there spaces where other fuel rods could be placed other than the analyzed configuration?

Response

Fuel rods can only be placed in the configuration shown. The structure limits the locations.

c.

Why isn't a missing fuel rod analysis necessary?

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Attachment I Page 32 of 36

Response

The arrangement is over-moderated so removing fuel decreases k. As confirmation, the storage basket case was run with one fuel rod missing. The k was 0.9099 compared to the original k of 0.9195.

d.

Is it possible that another movable fuel rod storage basket design could be introduced at the plant in the future?

Response

Any other designs to the fuel rod storage basket will be captured in existing processes for changes to the facility.

25.

In Section 8.13, "Assemblies with Missing Fuel Rods," it was not stated if an analysis was done to verify that the burnup worth associated with 4 GWd/MTU burnup adder to cover reactivity increases for an assembly with any number of missing rods is appropriate, nor was it stated what burnup the modeled reconstituted fuel assemblies were analyzed at. Consequently, provide justification for the 4 GWd/MTU burnup adder to cover reactivity increases for an assembly with any number of missing rods at any burnup, given that burnup worth changes as a function of burnup.

Response

The worst case of 36 missing fuel rods was analyzed at a burnup of 42.67 GWdFI" and resulted in a k of 0.9823 or an increase of 0.01 84 from the base case. The reactivity worth due to burnup is smallest at the highest burnup. The reactivity worth of burnup between 42 and 46 GWdIT is 0.01 94 delta-k. Since the reactivity worth due to burnup is larger at smaller burnups, the 4 GWdFI" adder would apply at all lower burnups.

26.

Section 8.13 implies that the 4 GWd/MTU adder will only apply to reconstituted fuel assemblies that do not install stainless steel 'rods --this means that an additional check would be necessary to indicate that the 4 GWd/MTU adder is needed in some cases, but not for others, creating implementation complexity.

Consequently, what controls ensure that the 4 GWd/MTU will be added to the fuel assembly

/

burnup for applicable fuel assemblies before comparison to minimum burnup requirement? That is, how will it be ensured that a reconstituted fuel assembly where stainless steel rods are not installed (i.e.. requiring a burnup adder) can be distinguished from a reconstituted fuel assembly where stainless steel rods are installed (i.e. not requiring a burnup adder), and then be ensured that the fuel assembly burnup is adjusted appropriately for comparison to the corresponding minimum burnup requirement?

Response

This would be procedurally controlled and details will be part of the LAR.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

N L-1 5-089 Docket No. 50-247 Page 33 of 36

27.

The missing fuel rod sensitivity study discussed in Section 8.13 does not seem to have analyzed the sensitivity to missing fuel rod orientation for the various cases analyzing different numbers of missing fuel rods.

Provide justification that the selected orientations are bounding relative to other potential orientations.

Response

Numerous cases were analyzed with missing rods in different locations in the assembly.

It became apparent that removing rods near water holes had the largest effect.

Therefore, the locations used for missing rods in the final analysis maximized the k. The table below summarizes the cases analyzed.

Table 27-1. Missing Pin Analysis Cases Case Description K (uncorrected)

TESTPIN0 2x2 array, 0 pins missing 0.96387 TESTPIN1 2x2 array, I pins missing in lower right assembly 0.96417 TESTPIN2 2x2 array, 2 pins missing in lower right assembly 0.96467 TESTPIN4 2x2 array, 4 pins missing in lower right assembly 0.96528 TESTPIN8 2x2 array, 8 pins missing in lower right assembly 0.96652 TESTP12D 2x2 array, 12 pins missing in lower right assembly 0.96745 TESTP16B 2x2 array, 16 pins missing in lower right assembly 0.96805 TESTP20B 2x2 array, 20 pins missing in lower right assembly 0.96849 TESTP24B 2x2 array, 24 pins missing in lower right assembly 0.96828 TESTP28B 2x2 array, 28 pins missing in lower right assembly 0.96831 TESTP32B 2x2 array, 32 pins missing in lower right assembly 0.96827 TESTP36B 2x2 array, 36 pins missing in lower right assembly 0.96882 TESTP364 2x2 array, 36 pins missing in all assemblies 0.98227 TESTP404 2x2 array, 40 pins missing in all assemblies 0.97830 TESTP444 2x2 array, 44 pins missing in all assemblies 0.97095 TESTP484 2x2 array, 48 pins missing in all assemblies 0.96710 TESTP524 2x2 array, 52 pins missing in all assemblies 0.96299 Accidents

28.

The criticality safety analysis needs to be consistent with existing administrative controls credited in future license amendment requests. Currently, there is no control rod removal accident analyzed. If there are no explicit controls for fuel assemblies crediting the presence of control rods, this accident should be considered for both Region 1 and Region 2. Missing panels are also part of normal operation as shown in Section 8, "Results." Consistent with DSS-IGS-201 0-01, "accidents should be considered with respect to all normal conditions,

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 34 of 36 e.g. fuel inspections and fuel reconstitution." If there are no explicit controls for missing panels, then all accident analyses should be updated to account for the maximum number of missing panels allowed during normal operation in both Region 1 and Region 2.

Provide justification for not including a control rod removal accident, which could include more than one control rod for a cluster of assemblies in either Region 1 or Region 2. Include a discussion of controls that are, or will be, in place at Indian Point 2 to ensure that credited control rods cannot be removed. Alternatively, model the maximum number of allowed missing panels, a normal condition of operation, as a base condition for a control rod removal accident.

Response

The control rod removal accident is covered in the analysis. The RAI specified that the control rod withdrawal should happen when the fuel assembly is in a cell that has a missing panel. The end result of this accident would be a maximum reactivity assembly (5 wt% with no IFBA and no burnup) in a cell with no insert. This scenario was analyzed, but the mechanism to get to the same end condition was to drop the maximum reactivity assembly into a cell with no insert. In fact the dropped assembly condition was even more reactive since it was assumed that the grid failed and the fuel pins evenly distributed in the cell, adding to the reactivity. Using the Tech Spec soluble boron, the k of this accident condition is 0.9181 (see Table 9.2 of NET 300067-01, Rev. 1 )after biases and uncertainties which is much less than 0.95.

The PAl suggests that this accident should be considered with multiple missing absorber panels since this is allowed as part of the normal operation. It is agreed that the initial condition for the accident consider all the allowed normal operation starting conditions. It is not a normal condition to have fuel in a cell without an absorber panel unless it has a control rod inserted. The reactivity of a cell missing an absorber panel with no fuel assembly and the reactivity of a cell with a missing absorber panel and fuel assembly with a control rod inserted is less than the reactivity of a cell with an absorber panel and fuel at the loading curve enrichment and burnup. Therefore, the accident analysis started with the most reactive initial condition which is Region 2 filled with fuel at the loading curve burnup.

The RAl may also be asking about a multi failure event. Similar to the multiple misload of fuel it may be possible to have multiple removals of control rods. The second sentence of Section 1.1 states:

"Due to this fact, Entergy, the operator of the Indian Point Plant, will no longer take credit for the BoraflexTM but rather install new neutron absorber panels into eveny cell in the spent fuel pool."

Although missing absorber panels are allowed, it is not anticipated that they will be in sufficient numbers to warrant a multiple event analysis. It is anticipated that the missing panels will be widely dispersed such that interactions will be rare. Control rods can also be used to store fresh fuel in Region 2 cells with an absorber panel. The multi misload analysis covers this condition with the analysis of fresh 5.0 wt% fuel having 64 IFBA rods at a 1.25X IFBA loading in every cell. Section 9.4 points out that less than 64 IFBA rods

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 35 of 36 is so rare that considering such assemblies in a multi misload analysis would not be appropriate. The response to RAl 29 adds new calculations for missing panels with the multiple misload assemblies.

29.

The claim that no checkerboarding is credited as part of the justification for why a multiple mislead is unlikely in Section 9.4, "Multiple Misleads," isn't accurate given that Footnote (f) to Table 10.1 states: "...a Region 2 cell which does not contain an absorber panel does not affect the loading requirements of any other cell in Region 2, so long as the cell which is missing an absorber panel does not contain a fuel assembly...". This operational flexibility would allow for checkerboard ing of fuel and also one-out-of-four storage with the empty cells containing no absorber panels.

Please revise the statement in Section 9.4 and/or other related statements accordingly.

Response

Crediting checker boarding normally means taking credit for the lower reactivity of adjacent cells. No such credit is taken in this criticality analysis. Checker boarding is not a normal occurrence. If a cell does not have an absorber panel:k it must remain empty but multiple missing absorber panels near each other would be highly unusual. Even though

• an empty cell with no absorber panels has a lower reactivity than a cell with fuel at the loading curve enrichment and burnup, no relaxed requirements are given for the assemblies next to these empty cells.

The single assembly misload analysis was done where the misload location did not contain an absorber panel. However, the multi assembly misload analysis assumed all the cells contained absorber panels. Since it is allowed to have missing absorber panels,

'the multi assembly misload analysis has been redone. For Region 1 it is assumed that all the absorber panels are missing. The calculated k of Region 1 with unburned 5 wt% U-235 enriched fuel with no IFBA and with no absorber panels is only 0.8667 when the Tech Spec 2000 ppm is used in the model. After adding the bias and uncertainty this k is less than 0.90 which leaves a large margin to the 0.95 criterion. Region 2 has less reactivity margin for the multiple misload, so it is assumed that one cell in every 36 cells was missing an absorber panel. It is the design objective to have absorber panels in all

• cells. If during placement of the absorber panels this is not met additional criticality analysis will be needed which will accompany the LAR associated with the actual absorber inserts. Assuming one out of every 36 cells is missing an absorber panel and all of the cells in Region 2 are filled with fresh 5.0 wt% fuel having 64 IFBA rods at a 1.25X IFBA loading, the calculated k is 0.9263. The bias and uncertainty for Region 2 (updated with values from RAI 7) is 0.0113 so the k95195 is 0.9376 which meets the less.

than 0.95 criterion. Assuming one out of every 36 cells is missing an absorber panel and all of the cells in Region 2 are filled with 5.0 wt% fuel with 18 GWd/T burnup, the calculated keff iS 0.9244. The bias and uncertainty for this fuel in Region 2 (updated with values from RAI 7) is 0.0169 and the final k95195 is 0.9403, which meets the keff less than 0.95 criterion.

Reactor Operating Limits and Allowable Fuel Loading Checks

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

NL-1 5-089 Docket No. 50-247 Page 36 of 36

30.

Note (b) to Table 10.2 and the footnotes to Table 10.3 are confusing as written.

An analysis was performed in Section 8.4 accounting for fuel assembly depletion with hafnium inserts and it states that: "For simplicity and to provide margin, a 2 GWdIT increase in the loading curve is required for any assembly having any burnup with a hafnium insert [emphasis added]."

Based on this statement, if a fuel assembly contained a hafnium insert at any time, why does the amount of burnup a fuel assembly has achieved factor into the application of the 2 GWd/MTU burnup requirement penalty as specified in Note (b) to Table 10.2 and the footnotes to Table 10.3?

Response

The amount of burnup prior to hafnium depletion was mentioned to clarify that if an assembly already meets the loading curve prior to hafnium depletion, no adder is required for hafnium depletion after that point. This is because any subsequent burnup with hafnium will always lower k. For simplicity, however, this caveat will be removed since an assembly will always achieve a burnup greater than 2 GWd/T in its final cycle.

In other words, the footnote and Note (b) will be revised to state that the adder must always be applied (whether the assembly already meets the loading curve requirement prior to its final cycle or not).

31.

In Section 10.5, "Reactor Operation Limits," there is a footnote that states: "If fuel less than or equal to 3.5 wt% is ever used in the future, the same requirements shown in Table 10.6 apply. Table 10.6 should be used for all future fuel regardless of enrichment." However, the burnup requirements in Table 10.2 are based on different sets of depletion conditions that depend on enrichment. Therefore, why is it acceptable to treat all lower enrichment fuel, specifically newer fuel, with depletion conditions specified in Table 10.6 given that Table 10.2 doesn't account for the Table 10.6 depletion conditions for lower enrichment fuel?

Response

It is not clear from the values in Tables 10.6 and 10.7 but the conditions given in Table 10.7 for the old fuel produce more reactive fuel after depletion and therefore, are more limiting than the conditions in Table 10.6 for the new fuel. This was confirmed by depleting 3.5 wt% fuel under both sets of conditions. The k (0.9650) using the newer depletion condition was less than the k (0.9665) using the original depletion condition, a difference of -0.0015SAk. That means that the burnup requirements shown on Table 10.2 which were generated for old low enriched fuel are conservative for new low enriched fuel meeting the Table 10.6 depletion requirements.

• Deleted text/tables/figures are crossed-out and insertions/changes are shown in red

ENCLOSURE 1 TO NL-15-089 AFFIDAVIT EXECUTED BY NETCO ENTERGY NUCLEAR OPERATIONS, INC.

INDIAN POINT NUCLEAR GENERATING UNIT NO. 2 DOCKET NO. 50-247

731 GTU' Ave Phone 845.382.6900. Faz: 846.382.6917 AFFIDAVIT I, Brian J. Dassatti, General Manager of Scientech, a business unit Curtiss-Wright Flow Control Service Corporation, do hereby affirm and state:

1. I am the General Manager of Scientech, a business unit Curtiss-Wright Flow Control Service Corporation, and am authorized to execute this affidavit on its behalf. I am further authorized to review information submitted to the Nuclear Regulatory Commission (NRC) and apply to the NRC for the withholding of reformation from disclosure.

2. The information sought to be withheld is contained in the Response to "Request for Additional Information - NET-300067-01, Revision 1, Criticality Safety Analysis of the Indian Point Unit 2 Spent Fuel Pool with Credit for Inserted Neutron Absorber Panels) and (INTRPND3: Verification and Validation Report). The proprietary information is contained in the Response to RAI 8.b and Appendix A of Attachment 2 (Fortran code).

3. In making this application for withholding of proprietary information of which it is the owner, NETCO relies on provisions of NRC regulation 10 CFR 2.390(a)(4). The information for which exemption from disclosure is sought is confidential commercial information.

4. The proprietary information provided by NEICO should be held in confidence by the NRC pursuant to the policy reflected in 10 CFR 2.390(aX4) because:

a) The information sought to be withheld in the Attachments (see paragraph 2 above) is and has been held in confidence by NETCO.

b) This information is of a type that is customarily held in confidence by NETCO, and there is a rational basis for doing so because the information contains methodology, data and supporting information developed by NETCO and its subcontractors, that could be used by a competitor as a competitive advantage.

c) This information is being transmitted to the NRC in confidence.

PagI1of2 d) This information sought to be withheld, to the best of my knowledge and belief, is not available in public sources and no public disclosure has been made.

e) The information sought to be withheld contains developed methodology, data and supporting information that could be used by a competitor as a competitive advantage, and would result in substantial harm to the competitive position of NETCO and its subcontractors. This information would reduce the expenditure of resources and improve his competitive position in the implementation of a similar product. Third party agreements have been established to ensure maintenance of the information in confidence. The development of the methodology, data and supporting information was achieved at a significant cost to NETCO and its subcontractors. Public disclosure of this information sought to be withheld is likely to cause substantial harm to NEr'CO's competitive position and reduce the availability of profit-making opportunities.

5. Initial approval of proprietary treatment of a document is made by the General Manager of Scientech, the person most likely to be familiar with the value and sensitivity of the information and its relation to industry knowledge. Access to such information within NETCO is on a "need to know" basis.

6. Accordingly, NETCO requests that the designated document be withheld from public disclosure pursuant to 10 CFR 2.390(a)(4).

I declare under penalty of perjury that the foregoing affidavit and statements therein are true and correct to the best of my knowledge, information and belief.

Brian J. Daseattl, P.E.

General Manager, Outage and Fuel Management Solutions utl-WlhSCientech, Nuclear Division Date:

lII*"

/

SiERSO DAfLVEI~RANr R

UJ SlffATh OF CI1CU Page 2 of 2 MYt OCII EiP.019 NEITCO 731 Grant Ave

• Lake Kairine, NY 12449

• Phone: 845.382.6900'* Fax: 845.382.8917