ML063250418

Master Curve Assessment of Kewaunee Power Station Reactor Pressure Vessel Weld Metal, WCAP-16609-NP, Revision 0

ML063250418
Person / Time
Site:	Kewaunee
Issue date:	10/31/2006
From:	Lott R, Server W ATI Consulting, Westinghouse
To:	Office of Nuclear Reactor Regulation
References
WCAP-16609-NP, Rev 0
Download: ML063250418 (62)

Text

ATTACHMENT 2 SUBMITTAL OF IRRADIATED REACTOR VESSEL SURVEILLANCE CAPSULE TEST RESULTS FOR CAPSULE T PER 10 CFR 50 APPENDIX H WCAP-16609-NP, REVISION 0, "MASTER CURVE ASSESSMENT OF KNP RPV WELD METAL," OCTOBER 2006 KEWAUNEE POWER STATION DOMINION ENERGY KEWAUNEE, INC.

Westinghouse Non-Proprietary Class 3 WCAP-16609-NP Revision 0 October 2006 Master Curve Assessment of Kewaunee Power Station Reactor Pressure Vessel Weld Metal ISWestinghouse

WESTINGHOUSE NON-PROPRIETARY CLASS 3 WCAP-16609-NP, Revision 0 Master Curve Assessment of Kewaunee Power Station Reactor Pressure Vessel Weld Metal R. G. Lott

• Westinghouse Electric Company W. L. Server ATI Consulting October 2006 Approved:

Electronically Approved*

J. S. Carlson, Manager Primary Component and Asset Management

• Electronically Approved Records Are Authenticated in ihe Electronic Document Management System.

Westinghouse Electric Company LLC Energy Systems P.O. Box 355 Pittsburgh, PA 15230-0355

©2006 Westinghouse Electric Company LLC All Rights Reserved

This page intentionally blank WCAP-16609-NP October 2006

iii PREFACE This report has been technically reviewed and verified by:

B. N. Burgos:

Electronically Approved*

• Electronically approved records are authenticated in the Electronic Document Management System.

RECORD OF REVISION Revision 0:

Original Issue WCAP-16609-NP October 2006

iv This page intentionally blank WCAP-16609-NP October 2006

V FORWARD The first application of the Master Curve approach for an irradiated reactor vessel weld metal was approved by the NRC for the Kewaunee Power Station (KPS) in 2001 (Safety Evaluation by the Office of Nuclear Reactor Include the Use of a Master Curve-based Methodology for Reactor Pressure Vessel Integrity Assessment, Docket No. 50-305, May 2001). Testing of the next surveillance capsule for KPS included the requirement to perform additional fracture toughness tests to help validate the previous Master Curve evaluation accepted by the NRC. Two reports have been prepared to describe the results and evaluation of the additional fracture toughness testing performed as part of the Capsule T evaluation.

Capsule Requirements In accordance with the NRC Safety Evaluation (SE), removal and testing of one additional capsule at a fluence equivalent to End-of-License-Renewal (EOLR) for the vessel weld of concern would be acceptable for monitoring radiation damage. The currently evaluated fluence for EOLR is documented in WCAP-16641-NP, the Capsule T analysis report, where the value was determined to be 5.37 x 10 " n/cm2 (E>1.0 MeV).

Additionally, the removal and testing of the capsule with fluence equivalent to 60 years will complete the current KPS surveillance program requirements. In accordance with the SE requirement, Capsule T was removed at a calculated fluence of 5.62 x 1019 n/cm2 (E>1.0 MeV), which closely approximates the EOLR vessel fluence.

Master Curve Fracture Toughness T_ Determination The methodology detailed in Appendix A of the NRC SE is the methodology accepted by the NRC. The licensee agreed to use this methodology for future Master Curve fracture toughness testing and to incorporate the results into the KPS licensing basis. All margin terms are defined in Appendix A.

Specific to the testing requirements, the NRC stated the following:

1. Use of ASTM E 1921-97 is acceptable,

2.

The use of multi-temperature maximum likelihood methodology is currently not endorsed (since it was not included in the ASTM Standard).

It was acknowledged that the state of knowledge regarding specific technical topics associated with the Master Curve approach may be improved in the future. Additional conservatisms may be reduced or removed provided technical justification is made. The NRC recognized that it may reconsider it's position based on action within ASME Standards organizations and revisions to ASTM E 1921.

In establishing a valid measurement of To for weld wire heat 1P3571, several sources for the test specimens were deemed acceptable:

1. Charpy V-notch (CVN) weld specimens,

2.

Reconstituted specimens from the weld portion of untested CVN heat-affected-zone (HAZ) specimens, and/or

3. Reconstituted weld specimens from broken halves of the original, broken weld CVN specimens.

WCAP-16609-NP October 2006

vi All specimens for fracture toughness testing were to be single-edge bend, SE(B), geometry as defined in ASTM E 1921; these specimens when fatigue precracked and conforming to CVN size are generally referred to as precracked Charpy V-notch (PCVN) specimens. All of the information in paragraphs 11.1 through 11.2.3 of ASTM E 1921-97 for Capsule T, Capsule S, the Maine Yankee Capsule A35, and any unirradiated specimens used for the current licensee submittal were required to be included in the final reports for Capsule T and the new Master Curve evaluation. Use of Code Case N-629 to define a suitable expression for calculating the RTTo parameter was considered acceptable.

The actual PCVN specimens utilized in deternining the measurement of T0 for Capsule T were fabricated from a combination of the original irradiated CVN weld specimens (eight total) along with the reconstitution of four unbroken CVN HAZ specimen portions to provide a total of twelve specimens.

Details concerning the testing of the PCVN specimens are documented in WCAP-16609-NP (Master Curve Report) and WCAP-1664 1-NP (Capsule T Analysis). In accordance with NRC guidance, the methodology in Appendix A of the SE was used and presented in WCAP-16609-NP. In addition to this methodology, a new methodology has been developed under International Atomic Energy Agency (IAEA) sponsorship which has been applied and documented in WCAP-16609-NP.

The actual test results are presented in WCAP-16641-NP and the analysis is described in WCAP-16609-NP.

Charpy V-Notch Testing and Analysis In accordance with the NRC SE, a full CVN curve was not required to be developed for the surveillance weld, heat 1P3571. However, information regarding material properties was still required to be estimated to include the transition temperatures at 30 ft-lb, 50 ft-lb, and 35 mils along with the drop in upper shelf energy (USE). Accordingly, the methodology used in determining these values was documented in WCAP-16641-NP. Reconstitution of specimens needed to determine material properties was to be performed in accordance with ASTM E 1253, as described in WCAP-1664 1-NP. For the forging and correlation monitor materials, full CVN curves were required and testing/analysis performed in accordance with ASTM E 185-82. CVN impact testing of the HAZ material was not required.

A full CVN curve was not developed for the surveillance weld;, however, the transition temperature values representing 30 ft-lbs, 50 ft-lbs and 35 mils were determined using the methodology presented in WCAP-16641-NP. The drop in the surveillance weld USE was also documented as a part of this analysis. The test results for the forging and correlation monitor materials also were documented in WCAP-16641-NP.

As indicated earlier, four of the HAZ CVN specimens were reconstituted and used as weld metal PCVN specimens to help determine the Master Curve To value for the surveillance weld.

WCAP-16609-NP October 2006

vii TABLE OF CONTENTS L IS T O F T A B L E S.......................................................................................................................................

i x L IS T O F F IG U R E S.....................................................................................................................................

x i E X E C U T IV E SU M M A R Y........................................................................................................................

xiii I

IN T R O D U C T IO N...............................................................................................................................

1-I 2

BACKGROUND ON FRACTURE TOUGHNESS AND TRANSITION TEMPERATURES.......... 2-I 3

CHANGES IN ASTM E 1921 TESTING AND EVALUATION METHODS...............................

3-1 3.1 MULTI-TEMPERATURE To VERSUS SINGLE TEMPERATURE To..................

3-I 3.2 E L A ST IC M O D U L U S....................................................................................................................

3-1 3.3 ELIMINATION OF SUBTRACTION OF 0.3068 TERM IN DETERMINING To........................

3-2 3.4 FACTORS IN THE SE(B) EQUATION for J.................................................................................

3-2 3.5 BIAS EFFECT BETWEEN SE(B) AND C(T) SPECIMENS.........................................................

3-2 3.6 LO A D IN G RA T E EFFEC T S..........................................................................................................

3-3 4

KPS CIRCUMFERENTIAL WELD METAL CHEMISTRY AND AT 30 UPDATE......................

4-1 5

MARGINS APPLIED IN PAST ANALYSES.................................................................................

5-1 6

FRACTURE TOUGHNESS MEASUREMENTS..........................................................................

6-1 6.1

SUMMARY

OF PREVIOUS FRACTURE TOUGHNESS TESTS..........................................

6-1 6.2 CAPSULE T FRACTURE TOUGHNESS TEST RESULTS.........................................................

6-2 7

DETERMINATION OF To FOR CAPSULE T FRACTURE TOUGHNESS RESULTS...............

7-1 8

EVA LU ATIO N FO R K PS V ESSEL..............................................................................................

8-1

8. 1 NRC SE METHODOLOGY AND PROJECTIONS.......................................................................

8-1 8.2 ALTERNATIVE METHODOLOGIES AND PROJECTIONS.......................................................

8-7 8.3 IA E A M E thO D O L O G Y................................................................................................................

8-11 8.4 RECOMMENDED METHODOLOGY AND PROJECTIONS BASED ON CURRENT STA T E O F K N O W LE D G E...........................................................................................................

8-14 9

ART VALUES FOR PTS AND HEAT-UP AND COOL-DOWN CURVES.................................

9-1 10 R E F E R E N C E S..................................................................................................................................

10 -1 APPENDIX A: PREVIOUS FRACTURE TOUGHNESS TESTS......................................................

A-i WCAP-16609-NP October 2006

viii This page intentionally blank WCAP-16609-NP October 2006

ix Table 6-1 Table 8-1 Table 8-2 Table 9-1 Table 9-2 Table 9-3 Table 9-4 Table A-la Table A-lb Table A-Ic Table A-2a Table A-2b Table A-3 Table A-4 LIST OF TABLES Fracture Toughness Results from Capsule T....................................................................

6-2 Alternative Methodologies for Estimating ART for the RPV 1 P3571 Weld Using M aster Curve Fracture Toughness D ata...........................................................................

8-9 Prediction Uncertainties Using ASTM E 900-02 Model Parameters for EOLR and Results fro m C ap su le T..............................................................................................................

8 -12 ART Value to be Used for PTS and Heat-Up and Cool-Down Curves at the 95.6%

Capacity Factor EOLR RPV Maximum Fluence........................................................

9-I ART Value to be Used for PTS and Heat-Up and Cool-Down Curves at the 80%

Capacity Factor EOLR RPV Maximum Fluence........................................................

9-1 ART Value to be Used for PTS and Heat-Up and Cool-Down Curves at the 95.6%

Capacity Factor EOL RPV M aximum Fluence................................................................

9-2 ART Value to be Used for PTS and Heat-Up and Cool-Down Curves at the 80%

Capacity Factor EO L RPV M axim um Fluence................................................................

9-2 Unirradiated Precracked Charpy Specimens Tested at -200'F.......................................

A-I Unirradiated Reconstituted Precracked Charpy Specimens Tested at -200'F................ A-2 Unirradiated '/ T Compact Tension Specimens Tested at -1 87°F...................................

A-3 Irradiated Precracked Capsule S Charpy Specimens Tested at 136 0F............................

A-4 Irradiated Precracked Capsule S Charpy Specimens Tested at 59°F.......................... A-5 Irradiated Precracked Main Yankee Specimens Tested at 210°F....................................

A-6 Unirradiated M ain Yankee W eld Specimen Data............................................................

A-7 WCAP-16609-NP October 2006

x This page intentionally blank WCAP-16609-NP October 2006

xi LIST OF FIGURES Figure 7-1 Calculation of Adjusted Reference Temperature.............................................................

7-3 Figure 8-1 Projections for ART Based on the Different Measurements and the Average U sing the N R C SE M ethodology.....................................................................................

8-6 Figure 8-2 Predictions of RTT-o (No Margins Included) for KPS Weld Metal from C ap su les S an d T............................................................................................................

8-14 Figure 8-3 Evaluation of ART from the NRC SE Methodology Compared to the IAEA M eth o d o lo g y..................................................................................................................

8 -15 WCAP-16609-NP October 2006

xii This page intentionally blank WCAP-16609-NP October 2006

xiii EXECUTIVE

SUMMARY

Master Curve fracture toughness data on irradiated weld metal heat IP357 1 from the Kewaunee Power Station (KPS) Irradiation Surveillance Program, Capsule T, have been evaluated to derive a direct measurement of irradiated RTTO for use in place of adjusted RTNDT for the reactor pressure vessel limiting circumferential weld. The RTTo for a fluence corresponding to the end of license extension was determined by making direct measurement of irradiated fracture toughness using fatigue precracked Charpy surveillance specimens. Fracture toughness data generated on the same weld wire heat from a previous capsule (Capsule S) and from another nuclear plant surveillance program (Maine Yankee Capsule A-35) were also included in the new evaluation.

The adjusted reference temperature (ART = RTTo + Margin) was determined using several different methods, including the approach used by the NRC in a Safety Evaluation (SE) for the KPS weld in 2001.

Following the NRC approach, the ART value at the inside wall of the reactor pressure vessel is just below the pressurized thermal shock screening criterion currently specified in the Code of Federal Regulations, IOCFR50.61 at an extended license life corresponding to a capacity factor of 0.956. This optimistic capacity factor sets the upper limit for how high the neutron fluence will reach at sixty years of operation.

Lower values of ART will be achieved when less optimistic operation is realized. The Margin term assumed by the NRC Staff in their 2001 SE contains excess conservatism that has now been quantified by a more accurate uncertainty analysis and with three sets of irradiated fracture toughness results. Use of a more realistic Margin term could allow the KPS vessel to operate beyond the sixty year extended license life.

WCAP-16609-NP October 2006

xiv This page intentionally blank WCAP-16609-NP October 2006

1-I 1

INTRODUCTION Operating nuclear power plants maintain surveillance programs to monitor the irradiation induced increase in the fracture toughness transition temperature for the reactor pressure vessel (RPV). Federal regulations (10 CFR Part 50 [1]) require use of the reference transition temperature (RTNDT) for indexing the ASME Code reference fracture toughness curve (Kic curve) [2] throughout the operating life of the RPV as delineated in Appendix G. Federal regulations (10 CFR Part 50, Appendix H [I] ) also require Charpy V-notch impact testing to monitor irradiation induced increases in the ductile-to-brittle transition temperature, which is then used to define the irradiated RTNDT.

The Charpy V-notch impact test provides a simple measure of the ductile-to-brittle transition of ferritic steels. Characterization of the ductile-to-brittle transition requires a series of tests over a range of temperatures. The Charpy impact test provides a quick determination of the ductile-to-brittle transition temperature with a minimal amount of material. It is often used in qualification testing to demonstrate sufficient resistance to brittle fractures at the design operating temperature. Qualification requirements such as RTNDT assume a correlation between the Charpy V-notch transition temperature and the fracture toughness transition temperature. To assure appropriate margins of safety, these correlations are necessarily conservative. Recent advances in fracture testing technology have made it feasible to eliminate these conservative correlations by measuring the fracture toughness transition temperature directly using pre-cracked Charpy sized specimens.

Analysis of the Charpy V-notch data from the Kewaunee Power Station (KPS) surveillance program indicates that the estimated RTNDT value for the circumferential weld (weld wire heat IP3571) will approach the limits set to avoid brittle failure due to pressurized thermal shock (10 CFR Part 50.61 [1])

during the license renewal period. The relatively high RTNDT value estimated for this weld also puts restrictions on the plant heat-up and cool-down curves under the requirements of 10 CFR Part 50, Appendix G [l]. Due to the potential adverse effect on plant operation during the license renewal period, the utility augmented the normal surveillance data from the fourth surveillance capsule (Capsule S [3])

with a supplemental fracture toughness testing program that carried forward into the later testing of Capsule T.

The Charpy and supplemental fracture toughness data from KPS Capsule S were reported in September 1998 (WCAP-14279 RI [3]). The fracture toughness data were derived primarily from testing of single-edged notch bend specimens, SE(B), also commonly referred to as precracked Charpy V-notch (PCVN) specimens reconstituted from the original surveillance Charpy V-notch test specimens. These fracture toughness tests were used to determine the Master Curve fracture toughness transition temperature, To, as defined in ASTM E 1921 [4]. Additional PCVN specimens were reconstituted from weld specimens from the Maine Yankee power plant surveillance program (Capsule A-35). The Maine Yankee and KPS surveillance welds were fabricated from the same weld wire heat (I P3571) using equivalent welding procedures. Subsequent analysis has consistently shown that the Maine Yankee surveillance weld has higher Cu levels and more radiation sensitivity than the KPS surveillance weld.

Both RTNDT and To provide measures of the temperature at which the ductile-to-brittle transition in fracture toughness occurs. However they are not numerically identical because RTNDT references a WCAP-16609-NP October 2006

1-2 bounding toughness curve and To references a median toughness curve. In addition the two parameters are referenced to the corresponding curves at slightly different toughness levels. ASME Code Cases N-629 [5] and N-631 [6] allow an alternative To-based reference temperature for indexing the ASME KIc curve (RTTo). The Code Cases define RTTi as To +35'F for both non-irradiated and irradiated ferritic steels.

A request for an exemption to the existing rules for evaluating reactor vessel integrity to allow use of the To based methodology was originally filed by the operating utility, Wisconsin Public Service Corporation, in 1999 for evaluating the weld metal in the KPS RPV. After continued discussion and detailed review, the Nuclear Regulatory Commission (NRC), granted three specific exemptions to the utility operating KPS, Nuclear Management Company (NMC), in 2001 [7]. The three exemptions were:

1. An exemption to establish a new methodology to meet the requirements of Appendix G to 10 CFR Part 50;

2.

An exemption to establish the use of a new methodology to meet the requirements of 10 CFR 50.61; and

3.

An exemption to modify the basis for the KPS reactor pressure vessel (RPV) surveillance program (required by Appendix H to 10 CFR Part 50) to incorporate the acquisition of fracture toughness data.

The fracture toughness based methodology approved for the exemption was described in Appendix A of the Safety Evaluation (SE) report issued by the NRC [7]. The NRC methodology accepted the use of To to estimate RTNDT, but included margins over and above the levels originally suggested by the utility.

The exemption approval required the testing of one additional surveillance capsule at a fluence equivalent to sixty years exposure in the RPV circumferential weld. This report provides an analysis of the fracture toughness data obtained from the sixty year capsule, KPS Capsule T. Details of the testing program along with data from Charpy V-notch tests and tensile tests are provided in a separate report [8].

The projected fluence for sixty years for the KPS vessel is 5.37 x 1019 n/cm 2 and 3.44 x 10'9 n/cm 2 for forty years using an optimistically-projected capacity factor of 0.956. Note that all neutron fluence values used in this report are for E > I MeV. Traditionally, some utilities have assumed a capacity factor of 0.8 which would correspond to a fluence of 4.99 x 1019 n/cm 2 for sixty years and 3.37 x 1019 n/cm 2 for forty years. The actual fluence at either 40 or 60 years will likely be bounded by the projected values assuming a capacity factor of 0.956.

WCAP-16609-NP October 2006

2-1 2

BACKGROUND ON FRACTURE TOUGHNESS AND TRANSITION TEMPERATURES The primary role of fracture mechanics testing of RPV steels is to provide estimates of the fracture toughness to be used in the analysis of reactor pressure vessel integrity. Ferritic steels undergo a characteristic transition from low toughness brittle failures to high toughness ductile failures with increasing temperature. Therefore, analysis of the fracture behavior of the steel requires characterization of the toughness over a range of temperatures encompassing the ductile-to-brittle transition. The development of the Master Curve testing procedure has provided a powerful new tool for performing this characterization.

ASTM E 616, Standard Terminology Relating to Fracture Testing [9], defines fracture toughness as "...a generic term for measures of resistance to extension of a crack." However, the standard also notes that "The term is sometimes restricted to results of fracture mechanics tests, which are directly applicable to fracture control." There are multiple measures of fracture toughness that come under both definitions.

The broad definition would include tests such as the Sharp Notch Tension Test or the Charpy V-notch test (see ASTM E 23 [10]), which are useful for characterizing material response but do not provide data that can be used directly in a fracture mechanics analysis of the RPV. The narrower definition would be limited to the type of linear-elastic (K stress intensity) and elastic-plastic (J-integral) fracture mechanics tests defined in ASTM E 1820 [11]. All RPV integrity analysis is based on linear-elastic assumptions and the primary measure of fracture toughness is Kic, the stress intensity required for initiation of crack growth. The objective of the fracture toughness testing program is to determine K1c as a function of temperature.

The size requirements for linear-elastic testing are so restrictive that it is impractical to perform an adequate number of tests on unirradiated materials to fully characterize the fracture toughness transition curve. Furthermore, these large specimen size requirements are totally incompatible with the space limitations imposed on RPV surveillance programs. Therefore, RPV surveillance programs have adopted indirect means (i.e. correlations between Charpy V-notch specimens and toughness) to assess the fracture toughness in irradiated reactor pressure vessels. The development of elastic-plastic fracture mechanics techniques introduced a new fracture toughness measure, the J-integral, which can be measured in much smaller specimens. The J-integral has been used both as a measure of ductile fracture toughness and to estimate equivalent linear-elastic toughness values in the lower transition region. The Master Curve provides a framework for using the linear-elastic toughness estimates derived from the J-integral to characterize the fracture toughness in the ductile-to-brittle transition regime.

Ferritic steels used in RPV construction exhibit a characteristic transition from brittle behavior at low temperatures to ductile behavior at higher temperatures. Under normal operating conditions, a nuclear RPV should always be in the high toughness ductile region. The stresses in the RPV must be carefully controlled during heatup and cooldown to avoid brittle fracture. It is therefore important to characterize the temperature at which the ductile-to-brittle transition occurs in a pressure vessel steel. A complete characterization of the transition requires testing at multiple temperatures. The temperature range over which this transition occurs depends on two factors: the properties of the material and the loading conditions. There are numerous tests designed to characterize the ductile-to-brittle transition in ferritic steels. Each test presents a unique combination of specimen geometry and loading and therefore a ductile-to-brittle behavior that is specific to the test method. This test-specific behavior is generally described in WCAP-16609-NP October 2006

2-2 terms of a characteristic transition temperature. While it is common to speak of the ductile-to-brittle transition temperature of a material, there is not a unique definition of this value. In practice, any definition of transition temperature must refer to the test procedure (e.g., Charpy V-notch 30 ft-lb transition, drop weight nil-ductility temperature, and Master Curve To). In each test, the transition temperature describes the ductile-to-brittle transition in the material.

For RPV steels, the most commonly used tests are the nil-ductility drop weight test and the Charpy impact test. The characteristic temperature in the drop weight test is defined as the nil-ductility point at the low temperature end of the transition (NDT). For nuclear RPV steels, the Charpy V-notch transition is usually characterized by the temperature at a specific absorbed energy level (30 ft-lb or 50 ft-lb). However, Charpy V-notch tests may also be characterized in terms of the fracture appearance transition temperature (FATT) or the temperature at a specific level deformation (e.g., 35 mils lateral expansion).

Although it has long been recognized that fracture toughness, as defined in ASTM E 1820, undergoes a ductile-to-brittle transition (which is what is really needed for accurate integrity assessment), a characteristic transition temperature for the fracture toughness has only recently been defined. The development of J-integral based techniques for measuring fracture toughness (Jc) in the transition region has allowed a much clearer definition of fracture toughness behavior in ferritic steels. Based on this experience, it has been observed that ferritic steels have a common temperature dependence of fracture toughness in the transition regime. It was this observation that led Wallin to the definition of a Master Curve [ 12] that allows the fracture toughness for any ferritic steel to be characterized solely in terms of a reference temperature, To, corresponding to a fracture toughness of 100 MPa-m"/2 (91 ksi-in1/2). ASTM E 1921, which was originally adopted in 1997, provides a standard test method for the determination of To. This reference temperature can be used to index the Master Curve or some bounding curve. This behavior is in sharp contrast to the Charpy V-notch behavior, where both the transition curve shape and characteristic temperature vary between materials and after irradiation.

The ductile-to-brittle transition behavior of a material may be characterized using any combination of the above mentioned tests. The availability of test specimens and the expense of performing the tests generally determine the particular test employed to characterize the material transition temperature.

While there are multiple measures of the ductile-to-brittle transition, the underlying mechanisms of deformation and fracture are clearly inter-related. For this reason, the various measures of transition temperature tend to be correlated. These correlations allow a determination from one test technique to be used to estimate characteristic transition temperatures for the remaining tests. While it is difficult to demonstrate a correlation between NDT and the Charpy V-notch related transition temperature, recent test results have indicated a reasonable correlation between Charpy V-notch transition temperatures and To.

This correlation is fundamental to all Charpy based procedures for estimating toughness values in current RPV integrity analysis.

WCAP-16609-NP October 2006

3-1 3

CHANGES IN ASTM E 1921 TESTING AND EVALUATION METHODS When the KPS Capsule S fracture toughness tests were generated, the first version of ASTM E 1921 was just being finalized. As mentioned earlier, the first published version was ASTM E 1921-97. Since the publication of the 1997 version, there have been several revisions that have involved some minor and more significant changes. The following discussion describes the more significant changes as related to the testing and evaluation performed earlier for Capsule S and more recently for Capsule T. For the I P3571 weld tests, the net effect of the changes in the ASTM E 1921 test method from 1997 to 2005 was a decrease of 1-3"F in the measured To values. Although the old 1997 test method produces values that are slightly conservative with respect to the current procedure, the changes were deemed to be insignificant. Therefore, the previously accepted ASTM E 1921-97 To values have been maintained in this analysis. The Kewaunee Capsule T specimens were analyzed using ASTM E 1921-05, which is the test method in place at the time of the testing of Capsule T.

3.1 MULTI-TEMPERATURE To VERSUS SINGLE TEMPERATURE To One of the most significant changes was the move to make ASTM E 1921 based on the multi-temperature method rather than testing at a single test temperature. The 1997 version was strictly a single temperature method, but later versions allowed both approaches since the single temperature method is a mathematical simplification of the multi-temperature approach. Also, some test results from other test temperatures will not be lost by utilizing the multi-temperature method; often the optimum test temperature(s) can be difficult to define when testing a new or irradiated material. The original data used for the KPS surveillance weld Master Curve submittal were primarily based on the single temperature approach, although the multi-temperature approach was used in some calculations by combining different specimen types, SE(B) - PCVN and compact tension [C(T)] for the unirradiated condition and SE(B) - PCVN and modified IT-WOL* from Capsule S. It is now acknowledged that a specimen bias does exist between SE(B) and C(T) geometries, and this combination of specimen types would not be done today without making adjustments to the Kjc data to account for the differences. The analysis approach used by the NRC Staff in the KPS Master Curve SE [7] only used the single temperature data for the PCVN tests.

The testing for Capsule T was performed at a single temperature to be consistent with the previous testing and the choice of a test temperature was relatively easy due to the extensive testing performed earlier. A temperature dependent value of the elastic modulus also has been used for the determination of Kj, for the KPS 1P3571 weld metal.

3.2 ELASTIC MODULUS The 1997 version of ASTM E 1921 used a plane stress elastic modulus (E), which was a technically questionable value to be used for low toughness material being tested in the early transition region. The 2002 change to ASTM E 1921 (which has been carried forward into the 2005 version) converted the elastic modulus to a plane strain value: E / (I - v2) where v is Poisson's Ratio. This plane strain modulus is applied both in converting the elastic stress intensity term into an equivalent Jelastic value and in t

Modified IT-WOL is a one-inch thick wedge-open loading (WOL) test specimen that has been modified to be tested in a manner similar to a C(T) specimen.

WCAP-16609-NP October 2006

3-2 converting the combined Jelastic and Jplastic values into the equivalent Kj, value. The combination of conversion of K to J and then J back to K effectively cancels out the elastic contribution. However, there is an increase in the Jplastic contribution. The maximum potential increase in the calculated K0 is approximately 5%. The increase in Ko is more typically 2%. This change can lead to a decrease in the To value by 1-20F.

3.3 ELIMINATION OF SUBTRACTION OF 0.3068 TERM IN DETERMINING To A subtraction term of 0.3068 from the number of valid test measurements was included in the 1997 version of ASTM E 1921. This conservative value was eliminated in the 2002 version (and continued in the 2005 version) and the effect results in a decrease in To of less than 1 °F.

3.4 FACTORS IN THE SE(B) EQUATION FOR J Some minor changes have been made that involve the ri factor used in calculating the load line displacement from clip gage mouth opening measurements. These changes have been minor and do not affect the results reported for the testing of the KPS 1P3571 weld metal.

3.5 BIAS EFFECT BETWEEN SE(B) AND C(T) SPECIMENS The latest data relative to specimen bias between SE(B) and C(T) specimens suggests a bias value of about 18'F for non-irradiated specimens. ASTM E 1921 [4] now acknowledges this fact by stating that:

"Median Kj, values tend to vary with the specimen type at a given test temperature, presumably due to constraint differences among the allowable test specimens... Kjc variability among specimen types is analytically predicted to be a function of the material flow properties and decreases with increasing strain hardening capacity for a given yield strength material. This Kic dependency ultimately leads to discrepancies in calculated To values as a function of specimen type for the same material. To values obtained from C(T) specimens are expected to be higher than To values obtained from SE(B) specimens. Best estimate comparisons of several materials indicate that the average difference between C(T) and SE(B)-derived To values is approximately 10'C [18 0F]. C(T) and SE(B) To differences up to 15'C [27°F] have also been recorded.

However, comparisons of individual, small datasets may not necessarily reveal this average trend.

Datasets which contain both C(T) and SE(B) specimens may generate To results which fall between the To values calculated using solely C(T) or SE(B) specimens."

As stated above, the material flow properties and the strain hardening capacity can affect the degree of constraint and the value of bias that can exist between C(T) and SE(B) specimen types. Limited data have suggested that irradiated materials may show a smaller bias than non-irradiated materials [13]. The NRC Staff used a value of bias of 8.5°F in the SE for KPS, which is a reasonable value to use for the analyses to be presented in this report since there is no technical consensus on the proper value to be used.

Additionally, the degree of constraint for a part-through surface flaw in a reactor pressure vessel is quite different than that of a C(T) specimen. It has been inherently assumed that the proper normalization for applying the Master Curve is the C(T) specimen since it has a high level of constraint which results in conservatively high values of T0. The SE(B) PCVN specimen is actually closer to the level of T-stress WCAP-16609-NP October 2006

3-3 constraint of a flaw in an RPV. This aspect also supports using a reasonably small value for a bias correction.

3.6 LOADING RATE EFFECTS Loading rate effects were originally assumed to be inconsequential in the range of static loading covered under ASTM E 1921-97. Later studies have shown that a loading rate dependence exists, even in the range specified by the ASTM Standard. The higher loading rates lead to higher values of To. Later changes in ASTM E 1921 have restricted the loading rate range at the higher end by about a factor of 10.

This loading rate range reduction reduces significantly the potential differences in To that can exist for quasi-static testing. The 2005 version of ASTM E 1921 requires testing at a specific range of stress intensity loading rates of 0.1 to 2 MPa-mni 2/s. All of the testing for both Capsules S and T was in the ASTM E 1921-05 specified loading rate range.

WCAP-16609-NP October 2006

3-4 This page intentionally blank WCAP-16609-NP October 2006

4-1 4

KPS CIRCUMFERENTIAL WELD METAL CHEMISTRY AND AT 30 UPDATE As described in WCAP 15074, Rev. 1 [14], the weld metal Cu and Ni chemistries for weld wire heat 1 P3571 are unchanged. Additional chemical analyses were conducted at Argonne National Laboratory on the LaSalle-I surveillance weld (weld wire heat 1P3571), but these results have been found to be of a confirmatory nature to testing performed by General Electric and have not been as rigorously documented as other measurements generated by the nuclear steam supply system vendors. Therefore, the ANL data have not been included in the determination of the best estimate chemistry of the overall 1 P3571 weld metal. The best estimate chemistry of 0.287 wt% Cu and 0.756 wt% Ni will continue to be used for weld wire heat 1P3571.

Charpy V-notch AT 30 estimates from Capsule T [8] are in agreement with expectations based on previous capsule testing. Therefore, there is no concern that the material and results from Capsule T could be anomalous relative to previous testing and results. The extensive testing that has been conducted on the 1 P3571 weld metal, especially related to the KPS and Maine Yankee surveillance welds, has helped to quantify the specifics on the variability in Cu and Ni content for 1P3571 welds. This variability in Cu and Ni between the KPS and Maine Yankee welds also has been quantified showing differences in the Charpy V-notch and measured fracture toughness properties. This understanding for these two specific welds allows uncertainties to be accurately determined which should allow a reasonable margin to be applied, as discussed later in this report.

WCAP-16609-NP October 2006

4-2 This page intentionally blank WCAP-16609-NP October 2006

5-1 5

MARGINS APPLIED IN PAST ANALYSES The issue of the proper Margin to be applied for the Master Curve methodology as used for the KPS RPV circumferential weld was an area of discussion between the utility and the NRC Staff. The final Margin as applied by the NRC Staff in the SE [7] included several terms that were not proposed by the utility and its contractors. In fact, the final Margin applied to the more appropriate and accurate Master Curve fracture toughness results was greater than the Margin that would have been applied to Charpy V-notch data. This inconsistency was presumably due to the fact that this application was the first direct licensing use of the Master Curve approach. Since the time of the SE, more knowledge and experience have been gained in applying Master Curve data. For the direct use of irradiated Master Curve results, better quantification of the areas of uncertainty has been developed which should be applied in future assessments. These uncertainties and factors included in a proper Margin term are discussed in more detail in Section 8.

WCAP-16609-NP October 2006

5-2 This page intentionally blank WCAP-16609-NP October 2006

6-1 6

FRACTURE TOUGHNESS MEASUREMENTS The fracture toughness of the surveillance specimens is measured in terms of the J-integral. ASTM E 1820 provides a general procedure for conducting J-integral tests. The Master Curve used in ASTM E 1921 describes the temperature dependence of the cleavage initiation toughness. A cleavage event is readily identifiable as a rapid, unstable crack advance. Cleavage initiation implies that this fracture instability must occur prior to the onset of significant stable tearing crack extension. For a known starting crack length, the elastic and plastic contributions to the J-integral can be determined by measuring the specimen load and the amount of plastic work applied to the specimen. The testing procedure requires unloading compliance measurements to demonstrate that stable tearing has not initiated prior to the cleavage failure. If cleavage initiation occurs and the measurement meets the various validity requirements of the testing standard, the value of the J-integral at the point of instability is defined as Jc.

The equivalent linear elastic plane-strain fracture toughness at cleavage instability is Kic. The Master Curve describes the temperature dependence of Kjc.

The determination of the fracture toughness transition temperature, To, requires multiple measurements of Kj,. In the original testing program, fracture toughness measurements were conducted on irradiated weld specimens from KPS Capsule S and Maine Yankee Capsule A-35. Unirradiated archival material from the KPS surveillance weld was also tested. Fracture toughness values for the unirradiated Maine Yankee were provided by the utility. Additional fracture toughness tests were conducted as part of the surveillance testing program for KPS Capsule T.

6.1

SUMMARY

OF PREVIOUS FRACTURE TOUGHNESS TESTS The fracture toughness tests on the unirradiated KPS surveillance welds are summarized in Appendix A, Table A-I. These toughness values were originally reported in WCAP-14279 RI [3]. Two different specimen geometries were tested to determine the distribution of unirradiated Kjc values. PCVN specimens were machined from the archival material and tested (Table A-la). The broken halves of the Charpy specimens were then reconstituted to provide additional unirradiated PCVN specimens for testing (Table A-1 b). All of the PCVN specimens were tested at -2000 F. A series of six /2-T C(T) unirradiated specimens were also tested at -I87"F (Table A-I c).

A total of twelve PCVN specimens were reconstituted from the broken halves of the weld and HAZ specimens from KPS Capsule S. Nine of the PCVN specimens were tested at 136 0F (see Table A-2a) and three at 590F (see Table A-2b). Note that the fluence for Capsule S has been revised in the new fluence evaluation performed for Capsule T [8]. The revised Capsule S fluence is 3.67 x l019 n/cm 2.

Eight PCVN specimens were reconstituted from the broken ends of Charpy specimens from Maine Yankee Capsule A-35. All of these specimens were tested at 21 00F (see Table A-3).

Refer to Appendix A for a sulmnary of detailed test information for these previous tests conducted by Westinghouse. The test data provided by the utility for the Maine Yankee unirradiated weld metal are listed in Table A-4, but the details of the testing is not listed.

WCAP-16609-N P October 2006

6-2 6.2 CAPSULE T FRACTURE TOUGHNESS TEST RESULTS Fracture toughness testing of PCVN weld specimens was also conducted as part of the surveillance testing program for KPS Capsule T. These results are included in a separate report [8]. The initial testing was carried out using the eight weld Charpy specimens from Capsule T. The Charpy V-notches in the surveillance specimens were modified to produce sharp crack starters for the pre-cracks. The specimens were side-grooved after pre-cracking to provide uniform crack fronts for testing, which should lead to more valid test results and improved accuracy in determining To. Note that specimens from the previous capsules and for the unirradiated condition were not side-grooved; use of side-grooved specimens is now recommended in ASTM E 1921, but side-grooving is not a specific requirement.

The minimum number of specimens required to determine To is set by ASTM El921. However the standard also indicates that additional tests may be required to minimize the effect of material variability.

All eight of the weld Charpy V-notch specimens were converted to PCVN specimens and tested at 136'F (see Table 6-1). The initial results verified previous observations of significant variability in data from weld 1P3571. For materials with significant variability, the testing of a minimum of twelve specimens is recommended by ASTM E 1921 to minimize the effects of the scatter. In accordance with the recommendations of the standard, an additional four PCVN specimens were reconstituted using the weld portion of unbroken HAZ specimens. These additional weld specimens were also tested at 136°F as indicated in Table 6-1. The ASTM Standard requires that all measurements be included in the To determination. The values for the measured Kjc are listed as well as the IT size adjusted values, K.MICT).

Table 6-1 Fracture Toughness Results from Capsule T Test Temperature Kic KJc(,I)

Specimen Code Specimen Type

("F)

(ksi-in"2)

(ksi-in /2)

W25 PCVN 136 54.6 47.0 W26 PCVN 136 65.0 55.3 W27 PCVN 136 90.1 75.1 W28 PCVN 136 149.7 122.4 W29 PCVN 136 79.6 66.8 W30 PCVN 136 78.9 66.3 W31 PCVN 136 46.0 40.2 W32 PCVN 136 59.6 51.0 H25 Recon. PCVN 136 119.0 98.0 H29 Recon. PCVN 136 75.6 63.6 H30 Recon. PCVN 136 53.1 45.0 H31 Recon. PCVN 136 53.2 46.0 WCAP-16609-NP October 2006

7-1 7

DETERMINATION OF To FOR CAPSULE T FRACTURE TOUGHNESS RESULTS The version of ASTM E 1921 that was in place when the original testing program was conducted in 1998 (ASTM E 1921-97) required multiple tests at a single temperature to determine To. However, multi-temperature methods for estimating were well-established at that time and WCAP-15075 [1 5] reported the results of several alternative methods for determining T o from the measured Ki, values. As expected the multiple analysis methods gave reasonably consistent results. The NRC methodology employed only data collected at a single temperature on a single specimen geometry (PCVN) that met the existing ASTM E 1921 requirements to determine To. The established values for the four measured material conditions were:

Maine Yankee, unirradiated: -158°F Maine Yankee, Capsule A-35: 2230F

• KPS, unirradiated: -l151TF KPS, Capsule S: 136`F l Although the current version of ASTM E 1921 allows multi-temperature determination of To, past practice was used as a guide for the Capsule T testing program, and all of the PCVN specimens were tested at 136 0F. At this temperature, eight tests should have been sufficient for determining To. However, as illustrated in Figure 7-1, one of the eight specimens tested was clearly outside the normal distribution of values. ASTM E 1921 provides the following guidance on data that falls outside the normally expected tolerance bound:

"If the suspected datum is outside the tolerance bounds dictated by Eqs. (14) (for example, Kjc < Kjc (0.02) or Kj, >

Ki, (0.98)) it may be possible to reduce the influence of the outlier datum on Kjk(mcd) by testing additional specimens."

The data set illustrated in Figure 7-1 demonstrates the level of variability in the data. The variability in chemistry and embrittlement response of welds from wire heat 1 P3571 is well established. The ASTM E 1921 Procedure goes on to provide the following additional advice concerning data sets with apparent material variability:

"Typically, a total of 12 replicate specimens is sufficient.

However, outliers shall not be discarded from the data utilized

• Note that a translation error for the value of T0 was made and listed in Table 5-19 in WCAP-14279 RI [3] for the Maine Yankee Capsule A-35 value for the reconstituted PCVN specimens; Table 5-19 in WCAP-14279 RI lists a value of To = 232)F, not To = 223'F, which is the value that the NRC SE uses.

Note that if the multi-temperature approach was used with all of the PCVN data (combining Tables A-2a and A-2b), the value ofT0 would be 142'F, which is very close to the single temperature value of 136°F.

WCAP-16609-NP October 2006

7-2 to calculate KJc(med). The emergence of additional outliers may indicate that the test material is not homogeneous."

Consistent with this advice, an additional four PCVN specimens were constructed by reconstituting the weld ends of HAZ specimens from Capsule T. The results from these additional specimens are also illustrated in Figure 7-1. It appears that these additional toughness values have a distribution similar to the original eight measurements.

The variability in these fracture toughness results appears to be somewhat larger than normally expected in a Master curve distribution for a completely homogeneous material. Examination of the fracture surfaces reveals a non-homogeneous structure, apparently associated with the structure of the weld passes. This structure, which is on the size scale of the test specimen, may help explain some of the excess variability in the test results. Additionally, the Cu and Ni variability in the IP3571 weld metal subjected to radiation exposure can lead to enhanced variability in irradiated fracture toughness properties, as has been observed in other weld metals in which the weld wire has been Cu coated. It should be noted that the 1P3571 weld has exhibited a large degree of irradiated variability in previous testing of the KPS and Maine Yankee surveillance welds, but this large variability has been consistent with the known Cu and Ni contents of the two welds.

The To value for the Capsule T specimens was determined by analyzing all twelve specimens:

KPS, Capsule T: 161T.

The Master Curve and confidence bounds associated with this value are illustrated in Figure 7-1.

WCAP-16609-NP October 2006

7-3 200 180 2%

160

, 140 120

"* 100

)

80 60 40 20 0

-25 0

25 50 75 100 125 Temperature (0C)

• - Weld Charpy Specimens M-Reconstituted HAZ Specimens Figure 7-1 Calculation of Adjusted Reference Temperature WCAP-16609-NP October 2006

7-4 This page intentionally blank WCAP-16609-NP October 2006

8-1 8

EVALUATION FOR KPS VESSEL The original request to use measurements of the Master Curve transition temperature, To, to index the irradiated fracture toughness curves for the KPS circumferential weld was submitted to the NRC by Wisconsin Public Service Corporation in June 1999. Existing analysis indicated that the limiting material in the KPS reactor pressure vessel was the circumferential weld, which was fabricated using weld wire heat 1 P375 1. This same weld wire was used to fabricate the surveillance welds in the KPS and Maine Yankee plants. The original request proposed using data generated using weld specimens from these two surveillance programs to determine the value of To as a function of the irradiation dose. To values were determined using specimens from KPS Capsule S (revised fluence = 3.67 x 10'9 n/cm2 based on the Capsule T evaluation) and Maine Yankee A-35 (fluence = 6.11 x 1019 n/cm 2). Strategies for applying this data to the assessment of the KPS reactor pressure vessel were outlined in WCAP-15075 [15]. In response to the original request, the NRC staff issued a Safety Evaluation (SE) proposing extensive modifications to the WCAP-15075 application strategy. The utility (Nuclear Management Company, LLC) agreed to use those modified procedures. In May 2001, the NRC granted exceptions to allow the use of Master Curve to assess the KPS pressure vessel as indicated in the SE [7].

8.1 NRC SE METHODOLOGY AND PROJECTIONS The SE [7] developed by the NRC staff provided both an evaluation of the available data and a procedure for determining the ART value for the circumferential weld in the KPS vessel. The NRC directed that the same analysis technique be used for the remaining surveillance capsule, which was to be withdrawn at the end of license renewal (EOLR) peak fluence.

The appendix to the SE provided sample calculations to illustrate the approved methodology for determining the adjusted reference temperature (ART), which can then be used for input for pressure temperature curves and pressurized thermal shock (RTPTs) evaluations. To assure consistency with the SE methodology, the fundamental data assumptions used in the SE have been used for the current analysis.

The parameter definitions and data values used in these calculations are outlined next in Section 8.1. 1.

For information and clarity, the data for Maine Yankee Surveillance Capsule A-35 and KPS Capsule S are duplicated from the SE (except for the revised fluence for Capsule S). The data set for KPS Capsule T has been adjusted for the measured To value and the measured capsule fluence. Projections of the circumferential weld properties also require estimates of the fluences in the RPV weld and the vessel weld chemistry. The values used in this analysis duplicate the methodology in the SE. Note that the peak EOLR fluence in the NRC analysis is projected to be 4.7 x 10'9 n/cm2. This fluence is lower than the estimate in the utility submittal because the NRC assumed a much lower capacity factor.

8.1.1 Fundamental Data for Assessing the NRC Method Data for the NRC calculations are presented next. Note that the calculations require use of chemistry factor (CF) values from tables and the fluence function (FF) values as determined in Regulatory Guide 1.99, Revision 2 [16].

(#1) Maine Yankee Surveillance (MY) Capsule A -35, PCVN Surveillance Weld Samples Unirradiated To = -158'F WCAP-16609-NP October 2006

8-2 Fluence = 6.11 x 1019 n/cm 2 (which corresponds to a FFMY-A35 = 1.44)

Irradiated To = TOMY-A35 = 2230 F ATOMY-A35 = [2230F - (-158°F)] = 381 0F CFMY-A3 5 = 237.20F (0.351 wt% Cu, 0.771 wt% Ni, based on Reg. Guide 1.99, Rev. 2)

TirrMY-A35 = The irradiation temperature of MY Capsule A-35 = 532'F BPCVN-MY-U = Bias associated with unirradiated MY PCVN data = 8.50F BPCVNMY.A35 = Bias associated with irradiated MY PCVN data = 8.5°F

(#2) KPS Surveillance Capsule S, PCVN Surveillance Weld Samples Unirradiated To = -151°F (average of whole and reconstituted PCVNs)

Fluence = 3.67 x 10'9 n/cm2 (which corresponds to a FFK-S = 1.34)

Irradiated To = TOK-S = 136°F ATOK-S = [ 136°F - (- 151 F)] = 287°F CFKS = 187.2°F (0.219 wt% Cu, 0.724 wt% Ni, based in Reg. Guide 1.99, Rev. 2)

TirrKS = Irradiation temperature of KPS Capsule S = 532 0F BPCVN-K-U = Bias associated with unirradiated KPS PCVN data = 8.5°F BPCVN-K-S = Bias associated with irradiated KPS PCVN data = 8.5°F

(#2a) KPS Surveillance Capsule T, PCVN Surveillance Weld Samples Unirradiated To = -15 °F (average of whole and reconstitued PCVNs)

Fluence = 5.62 x 10'9 n/cm2 (which corresponds to a FFKT = 1.42)

Irradiated To = TOK-T = 161 'F AToKT= [161°F - (-151°F)] = 312'F CFKT = 187.2-F (0.219 wt% Cu, 0.724 wt% Ni, based in Reg. Guide 1.99, Rev. 2)

TirrK.T = Irradiation temperature of KPS Capsule T = 532°F WCAP-16609-NP October 2006

8-3 BPCVNK.U = Bias associated with unirradiated KPS PCVN data = 8.50F BPCVN-K-T = Bias associated with irradiated KPS PCVN data = 8.5°F The SE supplied by the NRC provided a step by step process for using the measured To values to estimate the ART value at any position in the KPS circumferential weld. The SE methodology is based on the measured shift in To, ATOneas.

The SE methodology includes an adjustment to the measured shift to correct for differences between the surveillance capsule temperature and the vessel temperature. The time weighted average irradiation temperatures for all three capsules are essentially at 532TF and the time weighted average irradiation temperature for the KPS RPV circumferential weld is slightly greater than 532'F since the cycles of operation for KPS since April 2003 have led to an increase in the cold leg temperature of about 8'F due to a power uprate. In the calculations to be presented Sections 8.1.2 to 8.1.5, the correction is conservatively assumed to be 0F since higher irradiation temperatures lead to less embrittlement than lower irradiation temperatures. Additionally, for the KPS vessel, service at the lower temperature for 26 cycles relates to when most of the radiation damage has occurred, and future damage at the higher irradiation temperature (as high as 537"F for EOLR fluences) should have little effect.

After correcting for temperature, the measured shift is normalized to the RPV conditions using the chemistry factors (CF) and fluence function values (FF) from NRC Reg. Guide 1.99, Rev. 2. The irradiated value of To for the RPV can be determined from the estimate of the adjusted shift. The margin terms used to calculate the ART values in the NRC methodology are taken directly from Reg. Guide 1.99, Rev. 2. The primary purpose for the shift based methodology used by the NRC appears to be to provide an apparent link between the margin terms defined in the Reg. Guide and the To estimation used here.

The ART estimate for the vessel also includes a small bias term (8.5°F) designed to account for the observed differences between To values determined in SE(B) specimens as compared to C(T) specimens; thus, the constraint level in the C(T) specimen has been set as the correct value for the NRC calculation method.

This procedure was originally applied to the Maine Yankee Capsule A-35 data (Calculation #1) and the KPS Capsule S data (Calculation #2). The best estimate ART value was determined by averaging the results of the two calculations (Calculation #3). The SE further required that future surveillance capsules be analyzed in a similar fashion and the best estimate be taken as the average of all three determinations.

The KPS Capsule T data has been analyzed (Calculation 2a) and the best estimate value of ART adjusted accordingly to Calculation #3. Each of the three estimates of T0 (Calculation #1, #2, and #2a) provides an independent means of estimating the ART value.

8.1.2 Calculation #1: Determination of Index Parameter ARTTo-X-Y Based on PCVN Data from MY Surveillance Capsule A-35 WCAP-16609-NP October 2006

8-4 The following calculation for the Maine Yankee surveillance weld metal follows the NRC methodology as described in the SE.

ATO-x-Y-MY-A35 = [ATo-MY-A35 - (TirrRpV - TirrMYA 35)]*[(FFx-y / FFMYA35)*(CFRPV / CFMY-A35)]

for X-Y corresponding to EOLR at the Clad-to-Base Metal Interface, FFx-y= 1.39. The calculated ATo is then:

ATo-x-Y-MY-A35 = [381 - (532 - 532)]*[(1.39 / 1.44)*(214 / 237.2)] = 331.8°F The To value estimated at position X-Y as estimated using the data from Maine Yankee Capsule A-35 is then:

To-X-Y-MY-A35 = To-MY-A35 -

(AT0-MY-A35 - ATo-x-Y-MY-A35) = Tounirr + ATo-x-Y-MY-A35 For the clad-to-base metal interface at EOLR, the estimated To value is then:

To-X-Y-MY-A35 = 223 - (381 - 331.8) = -158 + 331.8 = 173.8-F The corresponding ART value is then taken as:

ARTxYMYA 35 = T0-X-Y-MY-A35 + 33 + M + BPCVNMYA35, where M = Margin = 62.5°F For the clad-to-base metal interface the ART value estimated using the Maine Yankee surveillance data is therefore:

ARTTo-X-Y-MY-A35 = 173.8 + 33 + 62.5 + 8.5 = 277.8°F.

8.1.3 Calculation #2: Determination of Index Parameter ARTTo-X-Y Based on PCVN Data from KPS Surveillance Capsule S ATo-x-Y-K-S = [AToK.S - (TirrRpv - TirrK-s)]*[(FFx-y / FFK-S)*(CFRPV / CFKS)]

for X-Y corresponding to EOLR at the Clad-to-Base Metal Interface, FFx-y = 1.39. The calculated To shift is then:

ATo-x-Y-K-S = [287 - (532 - 532)]*[(1.39 / 1.34)*(214 / 187.2)] = 340.3'F The To value estimated at position X-Y as estimated using the data from KPS Capsule S is then:

T0-X-Y-K-S = T0-K-S -

(ATo-K-S - ATo-X-Y-K-S)

For the clad-to-base metal interface at EOLR, the estimated To value is then:

WCAP-16609-NP October 2006

8-5 To-x-Y-K-S = 136 - (287 - 340.3) = 189.3°F The corresponding ART value is then taken as:

ARTX-Y-KS = T0-x-Y-K-S + 33 + M + BPCVN.KS, where M = Margin = 62.5°F For the clad-to-base metal interface the ART value estimated using the KPS Capsule S surveillance data is therefore:

ARTXYKS = 189.3 + 33 + 62.5 + 8.5 = 293.3°F.

8.1.4 Calculation #2a: Determination of Index Parameter ARTTo-X-Y Based on PCVN Data from KPS Surveillance Capsule T (New Data Analysis)

ATo-x-Y-K-T = [ATo-K-T - (TirrRpv - TirrK-s)]*[(FFx-y / FFKT)*(CFRPV / CFK-T)1 for X-Y corresponding to EOLR at the Clad-to-Base Metal Interface, FFx-y = 1.39. The calculated To shift is then:

ATo-X-Y-K-T = [312 - (532 - 532)]*[(1.39 / 1.42)*(214 / 187.2)] = 349. I°F The To value estimated at position X-Y as estimated using the data from UPS Capsule T is then:

T0-X-Y-K-T = T0-K-T - (ATO-KT - ATo-x-Y-K-T)

For the clad-to-base metal interface at EOLR, the estimated To value is then:

ToXYK.T= 161 -(312-349. l)= 198.1°F The corresponding ART value is then taken as:

ARTX-Y-K-T = T0-X-Y-K-T + 33 + M + BPCVNKT, where M = Margin = 62.5'F For the clad-to-base metal interface the ART value estimated using the KPS Capsule T surveillance data is therefore:

ARTXYKT = 198 + 33 + 62.5 + 8.5 = 302.1°F.

8.1.5 Determination of Best Estimate Value for ART1'0-x.

WCAP-16609-NP October 2006

8-6 At the time the initial SE was issued, only data from Maine Yankee Capsule A-35 and KPS Capsule S were available. The SE identified the best estimate of ART for the vessel as the average of these two determinations, which equated to a value of 288TF. Using the revised Capsule S fluence values the best estimate ART is:

ARTTQ-X-y = (ARTT,,-x-Y-MY-A35 + ARTTX-_V-K-S)/ 2 = 285.60 F.

The SE indicates that additional determinations should be similarly averaged. Therefore, the best estimate of the ART value at the peak EOLR fluence defined in the SE (for fluence - 4.7 x 1019 n/cm 2) is:

ARTT,,_x-y = (ARTTo-X-Y-MYA35 + ARTT,,_X-Y-K-S + ARTx-yKT)/ 3 = 29 1. I -F.

The three estimates and their projections are illustrated in Figure 8-1 for Calculations #1, #2 and #2a, as well as the average of the three values (dashed curve for Calculation #3). Note that the Kewaunee Capsule S (Calculation #2) and Kewaunee Capsule T (Calculation #2a) projections are nearly identical.

350 300 0

3OO C.,.

CL 250 200 I--

150 100 0.E+00 1.E+19 2.E+19 3.E+19 4.E+19 Neutron Fluence (n/cm 2) 5.E+19 6.E+19 7.E+19 Figure 8-1 Projections for ART Based on the Different Measurements and the Average Using the NRC SE Methodology (Note that the Curves for Calc 2 for KPS Capsule S and 2a for KPS Capsule T Essentially Lie on Top of Each Other)

WCAP-16609-NP October 2006

8-7 8.2 ALTERNATIVE METHODOLOGIES AND PROJECTIONS Several alternative methods for analyzing the data were outlined in WCAP-15075 [15]. The WCAP provided a table that compared the various terms contributing to the various calculated ART values. A similar comparison is outlined here in Table 8-1, which illustrates the direct measurement approach (no shift required - fourth row in the table), the use of unirradiated RTTo and Charpy shift (the fifth row in the table), and the NRC SE method, which is a shift-based ARTTO approach (rows I through 3 in the table).

Each of the terms represents a different step in the process of using the surveillance capsule data to estimate the ART for the vessel. The process can be broken down into the following steps:

I. Best estimate of the initial (unirradiated) fracture toughness reference temperature value (IRT).

2.

Standard deviation of estimated initial fracture toughness reference temperature value (cad.

3. Estimate of irradiation induced shift in surveillance material fracture toughness reference temperature at fluence corresponding to vessel target vessel (ART).

4. Standard deviation of estimated surveillance material reference temperature shift (CA).

5. Best estimate of irradiated fracture toughness reference temperature for surveillance weld at RPV target fluence (IRT + ART).

6.

Total margin for estimate of irradiated reference temperature (combined a* and 3A).

7. Adjusted reference temperature (ART) for surveillance material at target vessel fluence.

8. Adjustment for difference between surveillance material composition and the "best estimate" vessel composition.

9.

Adjusted reference temperature for the vessel material at the target fluence.

Although the proposed methodologies do not all follow this exact protocol, the calculations can be broken down in a similar manner. The SE methodology requested by the NRC combines the adjustments for vessel fluence and vessel composition in a single step. However, the effective value of the individual steps can be determined by applying the calculations in a sequential manner. In Table 8-1separate entries are shown for each step used in calculating the ART values.

In WCAP-15075, the preferred method for determining ART was direct measurement of the irradiated value (Method 6 in WCAP-15075 and the fourth row in Table 8-1). This method eliminates the need to estimate the first four steps in the process listed earlier. In this direct measurement method, the irradiated value can be determined at the surveillance capsule fluence without reference to the unirradiated value.

The methodology requires that the surveillance capsule measurements correspond to (or bound) the fluence at which the ART is required. In this case, the best estimate of the irradiated reference temperature is based on ASME Code Case N-629, which defines RTTO as T, + 350F. As To is measured directly, there is no error associated with extrapolating this measurement and the margin is determined by the measurement error. Since the direct measurement must be adjusted to match the best estimate WCAP-16609-NP October 2006

8-8 chemistry for the vessel, a heat adjustment term must be determined. ART values corresponding to the three surveillance capsule fluences are also shown in the Table 8-1. Note that the calculations for Maine Yankee Capsule A-35 and KPS Capsule S vary slightly from the WCAP-15075 values because the fundamental data has been adjusted to match the values in the SE. The heat adjustments are smaller in the WCAP-15075 analysis because the calculation is based on the Reg. Guide 1.99, Rev. 2 chemistry factors rather than the To measured chemistry factors.

The WCAP-15075 analyses also used the To shift as a basis for estimating an ART value. The unirradiated RTTo for the Kewaunee surveillance capsule was measured directly at a fluence that exceeds the expected life extension fluence. The standard deviation of the unirradiated value was estimated from the measurement uncertainty. The WCAP analyses fit the To shifts from tile two KPS capsules to get a single best estimate chemistry factor. Since the chemistry factor from two surveillance capsules was based on credible surveillance data, the standard deviation for the shift value was taken as 14WF. The adjustment to the vessel chemistry was based on the difference in the predicted Charpy shifts between the surveillance welds and the vessel.

The WCAP-15075 approaches for calculating ART using the direct Master Curve results produce lower values of ART than the NRC SE method. The difference is in the final Margin applied to the best estimate mean value of RTTo to obtain ART. A newer direct measurement calculation approach was recently developed under an International Atomic Energy Agency (IAEA) Coordinated Research Project (CRP) as discussed next. The IAEA methodology uses the direct measurement of RTTo and applies a detailed and realistic uncertainty analysis to determine an appropriate margin. The IAEA method should be the approach used in assessing the KPS limiting weld metal since it is more rigorous with regard to assessing uncertainties and, therefore, reflects the most accurate condition of the KPS limiting weld metal.

If the NRC wishes to impose the NRC SE method, it should be noted that the SE method contains considerably more margin leading to a much more conservative evaluation of the KPS limiting weld.

WCAP-16609-NP October 2006

8-9 Table 8-1 Alternative Methodologies for Estimating ART for the RPV 1P3571 Weld using Master Curve Fracture Toughness Data Best Estimate of Standard Kewaunee Standard Best Adjusted Adjustment Heat Adjusted Method Initial RT\\ 0 1 Deviation Surveillance Estimate Deviation tor Estimate of Total Margin Reference for Heat Reference fo oRTTF R

),

Irradiated (I),

"F Temperature Uncertainty Temperature V~alue (I RT), "F frI T(),

of Shift (ART), "F ART (OyA), "F Iraitd

()"FO "F

Value, "F (ART), "F (ARTHT),"F (ARTHT), °F NRC SE Unirradiated To +

ART = ATo determined PTS Rule and IRT +ART 2(-i 2+C7,2)1) 2 IRT + ART +

Equivalent For t= 4.7 x 1019:

Cale. #1 33'F + bias =

from CFff.= 38111.44 Reg. Guide M

value for NRC (MY Capsule

= 264.6 1.99, Rev. 2 SE method A-35)

For FF = 1.39

-36.0 277.8

-1 16.5 14 367.8 28 251.3 62.5 313.8 NRC SE Unirradiated To +

ART = AT0 determined PTS Rule and IRT +ART 2(o-i2+o., 2)112 IRT + ART +

Equivalent For 4t = 4. 7x 10'9:

Calc. #2 33"F +- bias =

from CF~ff= 287/1.34 Reg. Guide M

value for NRC (KPS Capsule S)

= 214.2 1.99, Rev. 2 SE method For FF = 1.39 42.6 293.3

-109.5 14 297.7 28 188.2 62.5 250.7 NRC SE Unirradiated To +

ART = AT, determined PTS Rule and IRT +ART 2(ai2+o.2) 1*'

2 IRT + ART +

Equivalent For ýt = 4.7 x 10":

Calc. #2a 33'F + bias from CF*f= 312/142 =

Reg. Guide M

value for NRC (KPS Capsule T) 219.7 1.99, Rev. 2 SE method For FF = 1.39 43.7 302.1

-109.5 14 305.4 28 195.9 62.5 258.4 WCAP 15075 NA NA NA ASTM E 1921 To + 35`F 2 cya,,

lrr RTT,, + M CF Ratio Adj.

For 4t =

(Direct CTT,,=13/'In 12 Measurement for 12 MY:

258 24 282

-33 6.11 x 019 : 249 Each Capsule) 10 KPS-S: 171 24 195 36 3.67x 1019: 231 KPS-T: 196 20 216 38 5.60 x 10": 256 WCAP 15075 Unirradiated ASTM Data Fit, KPS Reg.Guide IRT +ART 2(oi2+aA 2)1,2 IRT + ART +

CF Ratio Adj.

For ýt = 4.7 x 1019:

(Shift To, + 35"F P/In Capsules T and S 1.99, Rev. 2 M

Measurement CFef-= 217 (Credible Data Approach for Assumed) 14 256 KPS Capsules)

-116 8.5 302 186 33 219 37 IAEA NA NA NA ASTM E 1921 T0 + Tp,,j +

2 [o-T.r 2 + CTC,- +

lrr RTT,, + M CF Ratio for For dt = 4.7 x 10'9:

Methodology T0o=13/vn bias + 35"F eNi2 + (yTir,2 + Gy'2 KPS weld (Based on KPS

+ C7,.,j2]1 2

metal to the Capsule T 9.8 RPV 277.4 Results) 198.0 42.1 240.1 37.3 Note that the fluence used for most of the evaluations in this table is that defined in the NRC SE for EOLR (4.7 x 1019 n/cm 2). This value is used to provide a common EOLR fluence for comparisons back to the NRC SE evaluation. Evaluations using current assessments of EOL and EOLR fluences are provided in Section 9. Also note that the results in columns 4 and 5 for the WCAP-15075 measurement methods are based on the direct and shift methods presented in WCAP-15075 [15].

WCAP-16609-NP October 2006

8-10 This page intentionally blank WCAP-16609-NP October 2006

8-11 8.3 IAEA METHODOLOGY The IAEA has recently published guidelines for the application of the Master Curve [17]. The IAEA guidelines suggest extending the direct measurement methodology described in WCAP 15075 by using the measured points as the base for extrapolating shifts to the target fluence. The largest portion of the radiation induced increase in fracture toughness reference temperature occurs in the early stages of irradiation. Beyond a fluence near I x 10'9 n/cm2, the rate of increase in the radiation induced shift is relatively low. The IAEA methodology suggests using an existing model of embrittlement to predict the difference between the shift in transition temperature at the surveillance capsule fluence and the target fluence. For the Kewaunee surveillance weld, the measured RTTo (To + 35°F) value at the Capsule S fluence of 3.67 x 1019 n/cm2 is 171OF. Using the Regulatory Guide 1.99, Rev. 2 fluence function, and a chemistry-based CF of 187.2 for the KPS surveillance weld metal [8] the difference between the shift at the Capsule S fluence and the shift at the NRC SE target EOLR fluence of 4.7 x 10' 9n/cm 2 is lOF.

Because the IAEA methodology does not require a determination of the unirradiated To, there is no uncertainty associated with the initial reference toughness. As long as the fluence extrapolation produces a relatively small change in the shift value, the uncertainty in the RTTO estimation is dominated by the measurement uncertainty in the irradiated T(, value. This situation is analogous to the uncertainty analysis for the direct RTTo measurements. The vessel ART value is estimated by applying the direct measurement uncertainties and heat chemistry adjustments from the direct measurement technique to the IAEA based RTTo estimate. The IAEA methodology is compared to the WCAP 15075 and NRC SE methodologies in Table 8-1 using the Capsule T result since it conservatively predicts the highest values of ART using the NRC SE methodology.

The ART using the direct measurement approach involves a form for extrapolation from the measured irradiated RT (Irr RT), Capsule T in this case, to the fluence of interest:

ART = Irr RT + Tproj + Margin = Irr RT + CF

• AFF(ýt) + Margin where CF is taken as the chemistry-based value of i 87.2°F for the KPS surveillance weld metal; this value for CF was chosen since it is more conservative than using the CF for the vessel. This extrapolation from lrr RT is small (Tprej is a decrease of 6.5'F from the Capsule T fluence) compared to starting at the unirradiated condition (IRTTo); AFF(4t) is the difference in FF(4t) between the irradiated point and the fluence being extrapolated. An uncertainty in projected fluence is then a function of the uncertainty in CF or CFef'.

The development of the Margin is based on the uncertainties in the important parameters affecting irradiation changes in fracture toughness. These parameters and their corresponding uncertainties are:

copper content, nickel content, irradiation temperature (Tirr), and fluence (ýt). Other uncertainties that in general should be considered are the accuracy of the measured irradiated To (@mo) and the projection to other fluences (UProj). The irradiation sensitive parameters can be determined from the ASTM E 900-02 model for embrittlement [18]. The way that an uncertainty in an independent variable propagates through the model to produce an uncertainty in the predicted value can be estimated through a simple error analysis. The effect of the uncertainty in the independent variable on the prediction uncertainty can be determined by taking the partial differential of the prediction equation. The model provides a function WCAP-16609-NP October 2006

8-12 f(x 1,x2,x 3,...,

where x1 represents the ith independent variable. The contribution of uncertainty in x1 to the uncertainty in the prediction, dP, is:

dip ax. dX If the uncertainties in the independent variables are independent, then the composite uncertainty can be determined by taking the square root of the sum of the squares of the individual contributions. The uncertainties in the predictions are not constant and can vary as a function of fluence. Values of dxi must be established from measurements or other methods of assessing the uncertainty of the parameters.

Estimates of uncertainties for the individual ASTM E 900-02 model parameters (Cu, Ni, 4bt, and Tirr) for the KPS RPV material and irradiation conditions have been used to calculated uncertainties in the prediction of RTirr. These prediction uncertainties, the determination of RTTo, and projections from the measured fluence where RTTo have been determined to assess Margin:

Margin = Y [ayTo 2 + aCu2 + GNi2 + C'Tirr2 + a¢ot 2 + a5"j21I/2 Y is a value typically from I to 3 depending upon the judgment of the analyst and/or the regulatory authority; in most cases the value of Y will be between I and 2 since these values are most commonly used in engineering evaluations. (YT, is the uncertainty calculated as 13/N" 2, where N is the number of valid test results used to determine the irradiated To, and 13 is the statistical quantity from ASTM E 1921

[4]; for an individual data set of 12 valid tests, cYT 1 is 9.80F for the results from Capsule T. cyproj is partially included in the cyat estimate and can be determined by the uncertainty in CF (or CF0,f) produced by the uncertainties in Cu, Ni, irradiation temperature, and fluence; OTProj is typically of the order of I°F when on the plateau region of the shift-ýt curve; i.e., at fluences indicative of EOLR. CYCu, ayNi, GTirr, and ayt are based on the statistics and best estimates of uncertainties in the measured Cu and Ni values and the irradiation temperature (Tirr) and fluence (4t) values. The cumulative value of uncertainty using this process leads to a Margin (using a value of 2 for Y) as listed in Table 8-I for a fluence of 4.7 x 1019 n/cm 2.

Table 8-2 lists the uncertainties used at a fluence of 4.7 x 1019 n/cm2. The low value of prediction uncertainty due to Cu, even when a 25% uncertainty in the value of Cu can exist, is due to the saturation of Cu at values above 0.301 wt% as defined in ASTM E 900-02 [18].

For a good estimate of a 2ay Margin, the results using the uncertainty in prediction values in Table 8-2 give 2a = 42.1°F. All of the a values in Table 8-2, except for aYTm, are a function of fluence,* but the net effect on the overall cy is a net change of no more than I °F over a large fluence range around the Capsule T value of 5.62 x 10'9 n/cm2; this essentially constant value of 2cy is illustrated in Figure 8-3. It could be argued that three measurements of Master Curve RTTo for the 1 P3571 weld metal should be adequate to assure credibility and thus reduce the assumed value of Y equal to 2. One key argument is that the

+2ca lines in Figure 8-3 almost encompass the three measured data points, even before they are corrected for known chemistry differences.

• The effect of decreasing fluence decreases the cy associated with Cu, Ni, and irradiation temperature, whereas decreasing fluence increases the cy associated with fluence and fluence projection back from the Capsule T fluence used as the basis for this calculation.

WCAP-16609-NP October 2006

8-13 In the SE methodology, the NRC staff included an additional contribution for the uncertainty in the initial transition temperature, a,. Although the direct measurement and IAEA methodologies do not rely on the measurement of the initial transition temperature, the NRC staff argued that this term accounts for an inherent material variability. The 7°F difference in the initial To values measured for the MY surveillance weld (-158°F) and the KPS surveillance weld (-15 IF) is less than the 0,To uncertainty in the measurement.

This mTo term is an indirect measure of material variability as indicated by the master curve. Note that the effects of irradiation are characterized by the other uncertainties in Cu, Ni, Tir, and 4t. Therefore, the IAEA method does not typically include a separate amat term.

Table 8-2. Prediction Uncertainties using ASTM E 900-02 Model Parameters for EOLR and Results from Capsule T Approximate la Approximate la Uncertainty Parameter Uncertainty in in Prediction at EOLR, ýt =

Parameter 4.7 x 10'9 n/cm2 ("F)

Cu 25% for Cu =.287 wt%

8.0 Ni 0.065 wt%

13.8 Tirr 5)F 6.5

ýt 5%

7.t To Per ASTM E 1921 9.8 Projection back from 4tpro1 Capsule T, ýt = 5.62 x 0.6 10'9 n/Cm 2; 10% in CF The plot of the IAEA method for irradiated RTirr in Figure 8-3 shows that there is excess Margin included in the NRC SE approach. Therefore, all values calculated using the NRC SE methodology are overly conservative by 10 - 15')F. It is reemphasized here that the IAEA methodology is the most technically detailed approach for assessing the KPS limiting weld metal since it directly uses the measurements of RTTo and applies a thorough uncertainty analysis leading to a reasonable 2cy Margin.

WCAP-1 6609-NP October ?006

8-14 8.4 RECOMMENDED METHODOLOGY AND PROJECTIONS BASED ON CURRENT STATE OF KNOWLEDGE The determination of the RTTo for the Kewaunee surveillance weld requires an extrapolation from the values measured in Capsules S and T. The methodologies outlined in WCAP-15075, the SE written by the NRC Staff, and the IAEA guidelines all use slightly different extrapolation schemes. However, as illustrated in Figure 8-2, within the relevant range of fluences, these three extrapolation schemes are keyed to the data and give very similar results.

Although the proposed methodologies give consistent predictions for RTTo, there is a much wider variation in the corresponding ART values. Figure 8-3 illustrates the magnitude of the margin included in the NRC SE method of ART determination for the KPS RPV weld. The RTTo values plotted for the two Kewaunee surveillance capsules and the Maine Yankee capsule are direct measurements. The measured To values have been adjusted to include an 8.50F Charpy bias. The measured RTTo values have been fit to create a projection for the "best estimate" vessel weld chemistry. The difference between the "best estimate" projection and the ART curve is the margin implied by the NRC SE methodology. The IAEA methodology for Margin determination is more structured and adds up to estimates of ART that are less than the NRC SE method. The NRC SE recommended method conservatively uses all of the full Charpy uncertainties from Regulatory Guide 1.99, Rev. 2 assuming no credible data as applied to the measured RTTo values. Although the predictions were based on surveillance capsule results that are very consistent, no credit was given for credible data or the fact that the data from measured fracture toughness tests are more relevant to the integrity of the KPS vessel weld.

The plot of the IAEA method (IAEA + 2 s.d., meaning IAEA + 2 cy, for the Capsule T results) for irradiated RTir. in Figure 8-3 shows that there is excess Margin included in the NRC SE approach, (which is the average of three capsules). Therefore, values calculated using the NRC SE methodology are overly conservative by 10 - 150F with the difference decreasing as the fluence is reduced to lower values. If a comparison is made just for the Capsule T method using the NRC SE methodology, the excess Margin using the NRC SE method can be as high as 25°F. It is reemphasized here that the IAEA methodology is the most technically detailed approach for assessing the KPS limiting weld metal since it directly uses the measurements of RTTo and applies a thorough uncertainty analysis leading to a reasonable 2cy Margin.

WCAP-16609-NP October 2006

8-15 250 200 U-0; 150

._ 100

-Direct Measurement Projection - IAEA U

Shift Based Projection - NRC S0" -Shift Based - WCAP 15075 50 Surveillance Capsule 0

0 1E+19 2E+19 3E+19 4E+19 5E+19 6E+19 7E+19 Neutron Fluence (nlcm 2)

Figure 8-2 Predictions of RT(,, (No Margins Included) for KPS Weld Metal from Capsules S and T (Note that RT1,1 Definition Includes an 8."'F Bias AdjustmenO WCAP-16609-NP October 2006

8-16 350 o

300 ART - NRC Staff Evaluation o..

3 0 0 IA E +"2*

• '*="

Dirett Mea sure=*

2 U)

- 0 M.Y.

Best Estimate RTT.

0-250 Direct MeasureDe a-S200 Kew "T"

-IEA-2sd 4-Direct Measure IAA-

-d 150 E

100 50 I-0 0.E+00 2.E+19 4.E+19 6.E+19 8.E+19 1.E+20 Neutron Fluence (n/cm2)

Figure 8-3 Evaluation of ART from the NRC SE Methodology Compared to the IAEA Methodology WCAP-16609-NP October 2006

9-1 9

ART VALUES FOR PTS AND HEAT-UP AND COOL-DOWN CURVES The value of ART that should be used for PTS (RTpTS) for the limiting circumferential weld should be under 300°F at the fluence expected for EOL and EOLR. EOL fluence estimates for the KPS circumferential weld range from 3.37 x 10'9 n/cm 2 (80% capacity factor) to 3.44 x l019 n/cm 2 (95.6%

capacity factor). Similarly, EOLR fluence estimates for the KPS circumferential weld range from 4.99 x 1019 n/cm 2 (80% capacity factor) to 5.37 x 1019 n/cm 2 (95.6% capacity factor). Using the SE methodology, the average value of RTPTS is 297.5°F as indicated in Table 9-I at EOLR using the most optimistic capacity factor. The value of RTPTS associated with an 80% capacity factor is 294.0°F as indicated in Table 9-2. Both values are less than the PTS screening criterion.

For heat-up and cool-down curves, the values of ART to be used corresponding to EOLR can be determined utilizing the SE methodology for the Master Curve results, as well as the IAEA method. The range of results using both approaches for the 1/4-T and 3/-T locations in the KPS RPV are listed in Tables 9-1 and 9-2. Similarly the results for EOL are listed in Tables 9-3 and 9-4. Note that the IAEA column in each table includes the 2o Margin reflecting the sensitivity analysis for the key contributing parameters.

Notice that the extrapolation of the NRC SE and the IAEA approaches gives very similar values at the lowest fluences as seen in the last row of Tables 9-3 and 9-4.

Table 9-1. ART Values to be Used for PTS and Heat-up and Cool-down Curves at the 95.6%

Capacity Factor EOLR RPV Maximum Fluence

ýt (at Calc #1 Calc #2 Calc #2a NRC SE IAEA Location location)

FF Average with 2c-(x 1019 n/cmZ)(F

°(F (TF)

(TF)

Inside Surface 5.37 1.4161 284.0 299.7 308.7 297.5 283.3 1/4-T 3.64 1.3352 264.7 279.9 288.4 277.7 265.2 3/4-T 1.67 1.1408 218.3 232.3 239.5 230.0 223.1 Table 9-2. ART Values to be Used for PTS and Heat-up and Cool-down Curves at the 80%

Capacity Factor EOLR RPV Maximum Fluence

ýt (at Calc #1 Calc #2 Calc #2a NRC SE IAEA Location location)

FF (T)

(T)

(T)

Average with 2a (x10' 9 n/cm')

(_F)

(_F)

(_F)

(TF)

("F)

Inside Surface 4.99 1.4020 280.7 296.3 305.1 294.0 280.1 I/4-T 3.38 1.3185 260.7 275.8 284.2 273.6 261.5 3/4-T 1.55 1.1209 213.6 227.5 234.6 225.2 218.9 WCAP-16609-NP October 2006

9-2 Table 9-3. ART Values to be Used for PTS and Heat-up and Cool-down Curves at the 95.6%

Capacity Factor EOL RPV Maximum Fluence Ot (at Cale #1 Calc #2 Cale #2a NRC SE IAEA Location location)

FF Average with 2a (x,019 n/cm 2)

("F)

(OF)

(OF)

(OF)

(OF)

Inside Surface 3.44 1.3227 261.7 276.8 285.2 274.6 262.5 1/4-T 2.33 1.2284 239.2 253.8 261.5 251.5 241.9 3/4-T 1.07 1.0183 189.1 202.3 208.8 200.1 197.0 Table 9-4. ART Values to be Used for PTS and Heat-up and Cool-down Curves at the 80%

Capacity Factor EOL RPV Maximum Fluence Ot (at Calc #1 Calc #2 Calc #2a NRC SE IAEA Location location)

FF (OF)

(OF)

(F)

Average with 2a (xlO'9 n/cm 2)

("F)

(°F)

(_F)

(OF)

(OF)

Inside Surface 3.37 1.3179 260.6 275.7 284.0 273.4 261.4 1/4-T 2.28 1.2231 238.0 252.5 260.2 250.2 240.8 3/4-T 1.05 1.0126 187.7 200.9 207.3 198.7 195.8 WCAP-16609-N P October 2006

10-1 10 REFERENCES

1. Title10 Code of Federal Regulations, Part 50, "Domestic Licensing of Production and Utilization Facilities."

2.Section XI of the ASME Boiler and Pressure Vessel Code, Appendix G, "Fracture Toughness Criteria for Protection Against Failure."

3. C.C. Kim, E. Terek, and S.L. Anderson, Evaluation of Capsule S from the Kewaunee and Capsule A-35 from the Maine Yankee Nuclear Plant Reactor Vessel Radiation Surveillance Programs, WCAP-14279, Revision 1, September 1998.

4. ASTM E 1921, "Standard Test Method for Determination of Reference Temperature, To, for Ferritic Steels in the Transition Range," American Society for Testing and Materials, West Conshohocken, PA, Editions from 1997 to 2005.

5.

ASME Boiler and Pressure Vessel Code Case N-629, Use of Fracture Toughness Test Data to Establish Reference Temperature for Pressure Retaining Materials,Section XI, Division 1, ASME, 1999.

6. ASME Boiler and Pressure Vessel Code Case N-631, Use of Fracture Toughness Test Data to Establish Reference Temperature for Pressure Retaining Materials Other Than Bolting for Class I VesselsSection III, Division 1, ASME, 1999.

7. Safety Evaluation by the Office of Nuclear Reactor Include the Use of a Master Curve-based Methodology for Reactor Pressure Vessel Integrity Assessment, Docket No. 50-305, May 2001.

8.

WCAP-16641-NP, Analysis of Capsule T from Dominion Energy Kewaunee Power Station Reactor Vessel Radiation Surveillance Program, B.N. Burgos et al, September 2006.

9.

ASTM E 616, "Standard Terminology Relating to Fracture Testing," American Society for Testing and Materials, West Conshohocken, PA, 1989.

10. ASTM E 23, "Standard Test Methods for Notched Bar Impact Testing of Metallic Materials,"

American Society for Testing and Materials, West Conshohocken, PA, 2005.

11. ASTM Standard E 1820, "Standard Test Method for Measurement of Fracture Toughness,"

American Society for Testing and Materials, West Conshohocken, PA, 2006.

12. K. Wallin, "The Scatter in KIc Results," Engineering Fracture Mechanics, Vol. 19, pp. 1085-1093, 1984.

WCAP-16609-NP October 2006

10-2

13. W. Server, T. Griesbach, and S. Rosinski, "Application of Master Curve Data for Reactor Vessel Steels," 2003 PVP Conference, Cleveland, OH, July 2004.

14. WCAP 15074 Revision 1, 2006 - New Reference with latest chemistry update.

15. WCAP-15075, Master Curve Strategies for RPV Assessment, R.G. Lott, M.T. Kirk, and C.C.

Kim, September 1998.

16. Regulatory Guide 1.99, Revision 2, "Radiation Embrittlement of Reactor Vessel Materials,"

U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research, May 1988.

17. IAEA Technical Report Series No. 429, Guidelines for Application of the Master Curve Approach to Reactor Pressure Vessel Integrity in Nuclear Power Plants, Vienna, 2005.

18. ASTM E 900, "Standard Guide for Predicting Radiation-Induced Transition Temperature Shift in Reactor Vessel Materials, E706 (IIF)," American Society for Testing and Materials, West Conshohocken, PA, 2002.

19. M.T. Erickson Kirk et al., Technical Basis for Revision of the Pressurized Thermal Shock (PTS) Screening Limit in the PTS Rule (10CFR50.61), Proceedings of the 18th International Conference on Structural Mechanics in Reactor Technology (SMiRT-18), Beijing, China, August 2005, SMiRT18-D06-4.

WCAP-16609-NP September 2006

A-1 APPENDIX A: PREVIOUS FRACTURE TOUGHNESS TESTS Table A-la Unirradiated Precracked Charpy Specimens Tested at -200'F SPECIMEN ID wps201 1wgs202 wps20 wps204_

wps205 gs206 1wps207 wps208 Iwps29 wps210 Pmax (Ibs) - determined from test 1341 1253 1550 1261 1250 1274 1253 1364 1374 1468 LL Compliance (mils/lb) 0.0062 0.0081 0.0071 0.0115 0.0081 0.0068 0.0087 0.0059 0.0079 0.0059 Total Area (in-Ibs) 15.40 6.92 9.49 16.60 7.52 6.12 7.76 8.56 10.65 9.66 Plastic Area (in-ibs), calculated 9.87 0.58 0.97 7.43 1.21 0.57 0.95 3.12 3.17 3.33 Ao, Initial Crack length. (in.)

0.2080 0.2060 0.1988 0.1988 0.2100 0.2030 0.2100 0.2050 0.2030 0.1988 bo, Remaining Ligament (in.)

0.1860 0.1880 0.1953 0.1953 0.1840 0.1910 0,1840 0.1890 0.1910 0.1953 Ao/1*, Crack/vAdth Ratio 0.528 0.523 0.504 0.504 0.533 0.515 0.533 0.520 0.515 0.504 f(Ao/W) 2.915 2.867 2.700 2.700 2.965 2.796 2.965 2.843 2.796 2.700 Ke (ksi-inA0.5) 63.24 58.09 67.70 55.08 59.95 57.61 60.10 62.71 62.14 64.12 Je (in-lbs/in2)/l000 0.1333 0.1125 0.1528 0.1011 0.1198 0.1106 0.1204 0.1311 0.1287 0.1371 Jp (in-lbs/in2)/l 000 0.2558 0.0149 0.0240 0.1834 0.0317 0.0145 0.0250 0.0795 0.0800 0.0822 Jt(in-lbs/in,)/l000 0.3891 0.1274 0.1768 0.2845 0.1515 0.1251 0.1454 0.2106 0.2087 0.2193 KJC (ksi-i-A0.5) 108.0 61.8 72.8 92.4 67.4 61.3 66.1 79.5 79.1 81.1 KJC(iTAdiusted) - (ksi-inA0.5) 89.4 52.8 61.5 77.0 57.2 52.3 56.1 66.8 66.5 68.0

• Invalid precracks WCAP-16609-NP September 2006

A-2 Table A-lb Unirradiated Reconstituted Precracked Charpy Specimens Tested at -200"F SPECIMEN ID rkwl rkw3 rkw6 rkw7 I rkw8 rkwlO rkwll rkw2 Pmax (lbs) - determined from test 1255 1378 1472 1421 1582 1468 1422 1556 LL Compliance (mils/lb) 0.0045 0.0049 0.0065 0.0059 0.0043 0.0043 0.0057 0.0040 Total Area (in-lbs) 10.12 8.02 6.01 8.54 11.56 14.12 6.32 19.90 Plastic Area (in-ibs), calculated 6.58 3.37 0.00 2.59 6.18 9.49 0.56 15.01 Ao. Initial Cracklength, (in.)

0.2040 0.1970 0.1880 0.1940 0.1870 0.1920 0.1910 0.1920 bo, Remaining Ligament (in.)

0.1900 0.1970 0.2060 0.2000 0.2070 0.2020 0.2030 0.2020 Ao/*W Crack/Width Ratio 0.5178 0.5000 0.4772 0.4924 0.4746 0.4873 0.4848 0.4873 f(Ao/W) 2.819 2.663 2.480 2.599 2.460 2.558 2.538 2.558 Ke (ksi-inAO.5) 57.22 59.32 59.02 59.72 62.97 60.73 58.37 64.38 Je (in-lbs/in2)/lO00 0.1091 0.1173 0.1161 0.1189 0.1322 0.1229 0.1136 0.1382 Jp (in-lbs/in2)/l000 0.1669 0.0825 0.0000 0.0624 0.1439 0.2266 0.0133 0.3584 Jt (in-lbs/in2)/l000 0.2761 0.1998 0.1161 0.1813 0.2761 0.3495 0.1269 0.4966 KJC (ksi-jiA0.5) 91.0 77.4 59.0 73.7 91.0 102.4 61.7 122.1 KJC(1TAd/usted) - (ksi-inA0.5) 75.9 65.1 50.5 62.2 75.9 84.9 52.7 100.5 WCAP-16609-NP September 2006

A-3 Table A-ic Unirradiated 2 T Compact Tension Specimens Tested at -187°F SPECIMEN ID wpsl01 wps]02 wps03 wpsI04 wps0I 5 wps106 Pmax (lbs) - deterimined from test 3869 2867 3750 3445 3433 2416 LL Compliance (mils/1b) 0.0026 0.0029 0.0025 0.0029 0.0027 0.0024 Total Area (in-Ibs) 24.02 13.69 22.25 17.49 17.13 7.13 Plastic Area (in-lbs), calculated 4.56 1.77 5.00 0.41 1.14 0.04 Ao, Initial Crack length, (in.)

0.5200 0.5280 0.5230 0.5250 0.5220 0.5170 bo, Remaining Ligament (in.)

0.4800 0.4720 0.4770 0.4750 0.4780 0.4830 Ao/W, Crack/width Ratio 0.5200 0.5280 0.5230 0.5250 0.5220 0.5170 f(Ao/W) 10.134 10.400 10.233 10.299 10.200 10.038 Ke (ksi-inAO.5) 78.42 59.63 76.74 70.96 70.03 48.50 Je (in-lbs/in 2)/1O00 0.2050 0.1185 0.1963 0.1678 0.1635 0.0784 Jp (in-lbs/in2)/i000 0.0427 0.0169 0.0471 0.0039 0.0108 0.0004 Jt (in-lbs/in2)/i000 0.2477 0.1354 0.2434 0.1717 0.1742 0.0788 KJC (ksi-nIA0.5) 86.2 63.7 85.5 71.8 72.3 48.6 KJC(ITAdjusted) - (ksi-inA0.5) 75.4 56.5 74.8 63.3 63.7 43.8 WCAP-16609-NP September 2006

A-4 Table A-2a Irradiated Precracked Capsule S Charpy Specimens Tested at 136°F SPECIMEN ID w24 W19 h17 h18 w23 Ih 20*

h19 w17 h21 Pmax (lbs) - detelmined from test 1514 1349 1433 1285 1490 1319 1273 1323 1076 LL Compliance (mils/lb) 0.0069 0.0070 0.0071 0.0073 0.0071 0.0071 0.0071 0.0071 0.0082 TotalArea (in-Ibs) 15.80 8.49 25.90 20.36 23.03 29.60 9.97 15.57 15.09 PlasticArea (in-lbs), calculated 7.90 2.12 18.61 14.34 15.13 23.42 4.19 9.35 10.33 Ao, Initial Crack length, (in.)

0.1920 0.1920 0.1960 0.2030 0.1990 0.2000 0.1970 0.1970 0.2100 bo, RemainingLigament (in.)

0.2020 0.2020 0.1980 0.1910 0.1950 0.1940 0.1970 0.1970 0.1840 Ao/1W, Crack/vidth Ratio 0.4873 0.4873 0.4975 0.5152 0.5051 0.5076 0.5000 0.5000 0.5330 f(Ao/W) 2.5585 2.5585 2.6412 2.7960 2.7059 2.7280 2.6625 2.6625 2.9652 Ke (ksi-inAO.5) 62.65 55.84 61.20 58.10 65.20 58.20 54.81 56.97 51.60 Je (in-lbs/in2 )/lO00 0.1308 0.1039 0.1248 0.1125 0.1417 0.1129 0.1001 0.1082 0.0888 Jp (in-Ibs/ln2)/lO00 0.1887 0.0505 0.4534 0.3620 0.3742 0.5821 0.1025 0.2289 0.2707 Jt (in-lbs/in2)/l000 0.3195 0.1545 0.5782 0.4745 0.5159 0.6950 0.2026 0.3371 0.3594 KJC (ksi-inA0.5) 97.9 68.1 131.7 119.3 124.4 144.4 78.0 100.6 103.8 KJC(1TAdjusted) - (ksi-inA0.5) 81.3 57.7 108.1 98.3 102.3 118.2 65.6 83.5 86.1 WCAP-16609-NP September 2006

A-5 Table A-2b Irradiated Precracked Capsule S Charpy Specimens Tested at 59"F SPECIMEN ID

]

W2l W20 W22 Prnax (lbs) - dete; mined from test 1100 1220 1330 CLL (mils/lb) - determined from initial unloadings of test 0.0071 0.0071 0.0071 Total Area beneath Load-Disp Curve 3.77 3.77 3.85 Plastic Area (in-Ibs), calculated 0.00 0.00

-2.43 Ao, Initial Crack length, (in.)

0.2 0.2 0.2 bo, Remaining Ligament (in.)

0.194 0.194 0.194 Ao/1W, Crack/width Ratio 0.508 0.508 0.508 f(Ao/W) 2.728 2.728 2.728 Ke (ksi-in-AO.5) 48.54 53.83 58.68 Je (in-lbs/in2)/1000 0.0785 0.0966 0.1148 Jp (in-lbs/in2)/lO00 0.0000 0.0000

-0.0604 Jt (in-lbsin 2)/lO00 0.0785 0.0966 0.0544 KJC (ksi-inAO.5) 48.5 53.8 40.4 KJC(iTAdjusted) - (ksi-inAO.5) 42.2 46.4 35.8 WCAP-16609-NP September 2006

A-6 Table A-3 Irradiated Precracked Maine Yankee Specimens Tested at 210'F SPECIMEN ID 322 36a 313 371a 33u 375 371b 1 37ua Pmax (lbs) -determined from test 1482 1220 1373 1570 1259 1436 1457 1722 LL Compliance (mils/lb) 0.0065 0.0071 0.0071 0.0059 0.0073 0.0071 0.0060 0.0064 TotalArea (in-lbs) 9.96 5.28 14.77 35.00 7.26 10.57 9.52 15.82 Plastic Area (in-lbs), calculated 2.82 0.00 8.07 27.70 1.48 3.26 3.18 6.35 Ao, Initial Crack length, (in.)

0.1840 0.2020 0.1930 0.1860 0.2020 0.1960 0.1900 0.1970 bo, Remaining Ligament (in.)

0.2100 0.1920 0.2010 0.2080 0.1920 0.1980 0.2040 0.1970 Ao/W, Crack/width Ratio 0.4670 0.5127 0.4898 0.4721 0.5127 0.4975 0.4822 0.5000 f(Ao/W) 2.4042 2.7731 2.5788 2.4415 2.7731 2.6412 2.5185 2.6625 Ke (ksi-in'0.5) 57.63 54.72 57.27 62.00 56.47 61.34 59.35 74.15 Je (in-lbs/in2)/1000 0.1107 0.0998 0.1093 0.1281 0.1063 0.1254 0.1174 0.1833 Jp (in-lbs/in2)/1000 0.0648

-0.0001 0.1936 0.6423 0.0372 0.0794 0.0752 0.1553 Jt (in-lbs/in2)/l000 0.1755 0.0997 0.3029 0.7704 0.1435 0.2048 0.1927 0.3386 KJC (ksi-*MA0.5) 72.6 54.7 95.3 152.0 65.6 78.4 76.0 100.8 KJC(ITAdjusted) - (ksi-inA0.5) 61.3 47.1 79.3 124.2 55.8 65.9 64.0 83.6 WCAP-16609-NP September 2006

A-7 Table A-4 Unirradiated Maine Yankee Weld Specimen Data SPECIMEN ID C04-4 C04-5 C04-2 C04-7 C04-8 C04-3 C04-6 Pmax (Ibs) - determined froa test 1070 1110 1230 1150 1170 1180 1180 LL Compliance (mils/lb) 0.0090 0.0090 0.0089 0.0095 0.0096 0.0088 0.0089 TotalArea (in-lbs) 7.01 7.98 12.93 12.29 13.42 13.61 14.09 Plastic Area (in-lbs), calculated 1.86 2.44 6.20 6.00 6.84 7.49 7.89 Ao, Initial Crack length, (in.)

0.2031 0.2040 0.2003 0.2084 0.2032 0.2060 0.2045 bo, Remaining Ligament (in.)

0.1909 0.1900 0.1937 0.1856 0.1908 0.1880 0.1895 Ao/1W Crack/width Ratio 0.5155 0.5178 0.5084 0.5289 0.5157 0.5228 0.5190 f(Ao/W) 2.7984 2.8193 2.7347 2.9252 2.8007 2.8667 2.8310 Ke (ksi-inAO.5) 49.14 51.35 55.20 55.20 53.77 55.51 54.82 Je (in-Ibs/in2)/l000 0.0792 0.0865 0.0999 0.0999 0.0948 0.1011 0.0986 Jp (in-lbs/in2)/l000 0.0469 0.0618 0.1543 0.1560 0.1730 0.1920 0.2008 Jt (in-Ibs/in2)/1000 0.126 0.148 0.254 0.256 0.268 0.293 0.299 KJC (ksi-inA0.5) 62.0 67.2 88.0 88.3 90.4 94.5 95.5 KJC(JTAdiusted) - (ksi-inA0.5) 52.9 57.1 73.5 73.8 75.4 78.7 79.5 WCAP-16609-NP September 2006

A-8 This page intentionally blank WCAP-16609-NP September 2006