Materials Reliability Program: Resistance to Primary Water Stress Corrosion Cracking of Alloy 690 in Pressurized Water Reactors (MRP-258)

Materials Reliability Program: Resistance to Primary Water Stress Corrosion Cracking of Alloy 690 in Pressurized Water Reactors (MRP-258)
1019086

Final Report, August 2009

EPRI Project Manager K. Ahluwalia

ELECTRIC POWER RESEARCH INSTITUTE 3420 Hillview Avenue, Palo Alto, California 94304-1338 PO Box 10412, Palo Alto, California 94303-0813 USA 800.313.3774 650.855.2121 askepri@epri.com www.epri.com

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES

THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM: (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTYS INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USERS CIRCUMSTANCE; OR (B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT. ORGANIZATION(S) THAT PREPARED THIS DOCUMENT
Dr. John Hickling, Independent Technical Consultant

NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or e-mail askepri@epri.com. Electric Power Research Institute, EPRI, and TOGETHERSHAPING THE FUTURE OF ELECTRICITY are registered service marks of the Electric Power Research Institute, Inc. Copyright 2009 Electric Power Research Institute, Inc. All rights reserved.

CITATIONS
This report was prepared by Dr. John Hickling, Independent Technical Consultant Off Makariou III Street Prastio-Avdimou CY-4601 Cyprus Principal Investigator J. Hickling

This report describes research sponsored by the Electric Power Research Institute (EPRI). The report is a corporate document that should be cited in the literature in the following manner: Materials Reliability Program: Resistance to Primary Water Stress Corrosion Cracking of Alloy 690 in Pressurized Water Reactors (MRP-258). EPRI, Palo Alto, CA: 2009. 1019086.

iii

PRODUCT DESCRIPTION

Wrought Alloy 600 and its weld metals (Alloy 182 and Alloy 82) were originally used in pressurized water reactors (PWRs) due to the materials inherent resistance to general corrosion in a number of aggressive environments and because of a coefficient of thermal expansion that is very close to that of low alloy and carbon steel. Over the last 30 years, stress corrosion cracking in PWR primary water (PWSCC) has been observed in numerous Alloy 600 component items and associated welds, sometimes after relatively long incubation times. The occurrence of PWSCC has been responsible for significant downtime and replacement power costs. Component repairs and replacements have generally used wrought Alloy 690 material and its compatible weld metals (Alloy 152 and Alloy 52 or 52M), which have been shown to be highly resistant to PWSCC in laboratory experiments and have been free from cracking in operating reactors over periods already up to nearly 20 years. The challenge is to attempt to quantify the longevity of these materials with respect to aging degradation by corrosion in order to provide a sound technical basis for the development of future inspection requirements for repaired or replaced component items. This document updates EPRI report 1009801 (MRP-111), published in 2004, and represents an extensive revision that takes into account recently obtained information on the PWSCC behavior of Alloy 690 base material. The performance of Alloy 152 and 52 weld metals is not considered here, but will be reported on separately at a later date. Some consideration is also given to corrosion fatigue and low temperature crack propagation of thick-walled Alloy 690 material, since these topics were identified as knowledge gaps in the 2004 report (MRP-111). Approach Building on the original MRP-111 report, numerous laboratory tests conducted over the last two decades that were performed with wrought Alloy 690 materials under various test conditions pertinent to corrosion resistance in PWR environments were reviewed, with the main focus on PWSCC. Wherever possible, the existing laboratory test data were evaluated to estimate the improvement factor of Alloy 690 relative to Alloy 600. In addition, Alloy 690 service experience in PWRs is reported to augment the laboratory findings. Results and Findings It is concluded that wrought Alloy 690 is an acceptable and highly corrosion-resistant replacement material for Alloy 600 in PWRs, although further testing is still needed to examine some specific knowledge gaps that have emerged regarding PWSCC. These include the detrimental effect on resistance to crack growth of inhomogeneous cold work (particularly unidirectional cold rolling/tensile straining) and the possibility of enhanced cracking susceptibility in the heat-affected zone following welding. Relative improvement factors of 40 100 times versus Alloy 600 can now be derived for the initiation of cracking, v

but these numbers are clearly conservative, due to an absence of PWSCC in almost all Alloy 690 specimens within the test duration. Prototypical thick-walled Alloy 690 material (for example, as extruded piping for reactor pressure vessel top head penetrations) is extremely resistant to SCC growth from a pre-existing fatigue crack in simulated primary water, even under laboratory test conditions designed to maximize susceptibility. Measured crack growth rates to date are so low (< 5E-9 mm/s or 0.15 mm/year) as to be of no engineering significance and the relative factor of improvement for Alloy 690 CRDM material with respect to Alloy 600MA is thought to be well over 100 times. However, significant rates of intergranular cracking have been observed under constant load in some tests following the introduction of non-uniform cold work into Alloy 690 plate and bar materials that are not thought to be generally representative of plant components. The boundaries of this susceptibility are still being probed in order to confirm that such findings are unlikely to be relevant to long-term PWR operation. No stress corrosion degradation of Alloy 690 has been observed in any replacement application to date. Service experience for inspected material exposed to PWR primary water ranges from approximately 10 to 20 years, depending upon the type of component considered. Applications, Value, and Use This report should be of interest to utility engineers and scientists concerned with all aspects of Alloy 600 cracking in PWR primary water and especially to those involved in developing inspection regimes, making decisions on component replacement, and dealing with plant aging issues. It will be of direct value in obtaining regulatory acceptance for the solutions adopted by the industry to deal with increasing incidences of PWSCC that affect thick-walled Alloy 600 components in existing plants. As part of an ongoing, comprehensive program involving utilities, reactor vendors, and engineering/research organizations, this report will also help to ensure that corrosion degradation of nickel-based alloys does not limit service life and that full benefit can be obtained from improved designs for both replacement components and new reactors. EPRI Perspective The report describes the current knowledgebase on Alloy 690s PWSCC resistance. The data show that Alloy 690 possesses significantly greater resistance to PWSCC than its predecessor material, Alloy 600. Although these results are encouraging and are expected to hold as additional research continues, there are factors such as cold work that can impair its superiority somewhat. On-going and planned research is focused on these factors to more clearly define Alloy 690 resistance to PWSCC and identify any vulnerabilities. Until this additional work is completed, the results presented in this report are for information only and should not be used for prediction of component life for components constructed from Alloy 690. Keywords Alloy 600 Alloy 690 PWSCC Material degradation RPV penetrations

vi

ABSTRACT

Over the last 30 years, stress corrosion cracking (SCC) in pressurized water reactor (PWR) primary water (PWSCC) has been observed in numerous Alloy 600 component items and associated welds, sometimes after relatively long incubation times. Repairs and replacements have generally used wrought Alloy 690 material, which has been shown to be extremely resistant to PWSCC in laboratory experiments and has been free from detected SCC in operating reactors over periods already up to nearly 20 years. It is nevertheless prudent for the PWR industry to attempt to quantify the longevity of this material (and its welding alloys) with respect to aging degradation by corrosion in order to provide a sound technical basis for the development of future inspection requirements for repaired or replaced component items. This document updates EPRI report 1009801 (MRP-111), published in 2004, and represents an extensive revision that takes into account recently obtained information on the PWSCC behavior of Alloy 690 base material. The performance of Alloy 152 and 52a weld metals is not considered here, but will be reported on separately at a later date. Some consideration is also given to corrosion fatigue and low temperature crack propagation of thick-walled Alloy 690 material, since these topics were identified as knowledge gaps in the 2004 report. Building on the original MRP-111 report, numerous laboratory tests conducted over the last 25 years that were performed with wrought Alloy 690 materials under various test conditions pertinent to corrosion resistance in PWR environments are reviewed, with the main focus on PWSCC. Wherever possible, the existing laboratory test data have been evaluated to estimate the improvement factor of Alloy 690 relative to Alloy 600. In addition, Alloy 690 service experience in PWRs is reported to augment the laboratory findings. It is concluded that wrought Alloy 690 is an acceptable and highly corrosion-resistant replacement material for Alloy 600 in PWRs, although further testing is still needed to examine some specific knowledge gaps that have emerged regarding PWSCC. These include the detrimental effect on resistance to crack growth of inhomogeneous cold work (particularly unidirectional cold rolling/tensile straining) and the possibility of enhanced cracking susceptibility in the heat-affected zone following welding. Relative improvement factors of 40 to 100 times can now be derived for initiation of cracking, but these numbers are clearly conservative, due to an absence of PWSCC in almost all Alloy 690 specimens within the test duration. Prototypical thick-walled Alloy 690
a

Note that Alloy 52 exists with both original and modified compositions, whereby the latter are often referred to as Alloy 52M or Alloy 52(M). These terms are used interchangeably in the current report, which is focused on Alloy 690.

vii

material (for example, as extruded piping for reactor pressure vessel top head penetrations) is extremely resistant to SCC growth from a pre-existing fatigue crack in simulated primary water, even under laboratory test conditions designed to maximize susceptibility. Measured crack growth rates to date are so low (< 5E-9 mm/s or 0.15 mm/year) as to be of no engineering significance and the relative factor of improvement for Alloy 690 CRDM material with respect to Alloy 600MA is thought to be well over 100 times. However, significant rates of intergranular cracking have been observed under constant load in some tests following the introduction of non-uniform cold work into Alloy 690 plate and bar materials that are not thought to be generally representative of plant components. The boundaries of this susceptibility are still being probed in order to confirm that such findings are unlikely to be relevant to long-term PWR operation. No stress corrosion degradation of Alloy 690 plate and bar materials has been observed in any replacement application to date. Service experience for inspected material exposed to PWR primary water ranges from approximately 10 to 20 years, depending upon the type of component considered.

viii

ACKNOWLEDGMENTS
The present report was prepared by Dr. John Hickling, acting as an Independent Technical Consultant to EPRI. It represents an extensive revision and update of EPRI report 1009801 (MRP-111), originally published in March 2004. That earlier report was prepared by Framatome ANP (now AREVA) and had the following principal investigators: H. Xu and S. Fyfitch, Framatome ANP, Inc. 3315 Old Forest Road P.O. Box 10935 Lynchburg, VA 24506-0935 P. Scott Framatome ANP, SAS Tour AREVA 92084 Paris La Dfense, France M. Foucault Framatome ANP, SAS Porte Magenta BP 181 71205 Le Creusot Cedex, France R. Kilian and M. Winters, Framatome ANP, GmbH Freyeslebenstr. 1 91058 TGM P.O. Box 3220 91050 Erlangen, Germany

ix

CONTENTS

1 INTRODUCTION ....................................................................................................................1-1 1.1 1.2 1.3 1.4 Background..................................................................................................................1-1 Purpose and Scope of Revising MRP-111 ..................................................................1-2 Causes of Alloy 600 PWSCC.......................................................................................1-4 References ..................................................................................................................1-5

2 ALLOY 690 PROPERTIES AND METALLURGY ..................................................................2-1 2.1 Material Specifications .................................................................................................2-1 Typical PWR Specifications for Thin-Walled Alloy 690 SG Tubing.....................2-3 Specification and Manufacture of Thick-Walled Alloy 690 Components.............2-4 2.1.1 2.1.2 2.2 2.3 2.4 2.5 2.6

Phase Diagram of Alloy 690 ........................................................................................2-4 Carbon Solubility and Dynamic Strain Aging ...............................................................2-5 Intergranular Carbide Precipitation and Sensitization..................................................2-8 Effect of Elevated Temperature Exposure .................................................................2-12 References ................................................................................................................2-13

3 CORROSION BEHAVIOR OF ALLOY 690 APART FROM PWSCC ....................................3-1 3.1 General Corrosion Tests in Primary Water ..................................................................3-1 SG Tubing by Sedricks et al. 1979......................................................................3-1 SG Tubing by K. Smith et al. 1985......................................................................3-2 SG Tubing by Yonezawa et al. 1985...................................................................3-2 Esposito et al. 1991.............................................................................................3-2 Alloy Oxidation Studies related to the Mechanism of PWSCC ...........................3-3 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 3.2 3.3 3.4

Corrosion Fatigue Tests in Primary Water...................................................................3-4 Corrosion Behavior in Secondary Water....................................................................3-15 Low Temperature Crack Propagation (LTCP) ...........................................................3-17 Origins of the Phenomenon ..............................................................................3-17 Recent Studies to Assess the Possible Relevance of LTCP to PWRs .............3-18

3.4.1 3.4.2

xi

3.5

References ................................................................................................................3-23

4 PWSCC OF THIN-WALLED SG TUBING..............................................................................4-1 4.1 Laboratory Testing .......................................................................................................4-1 Early Studies .......................................................................................................4-2 Single U-Bend Test in Saturated Hydrogen Water with B/Li...............................4-2 CERT Tests in Hydrogenated Water with or without B/Li ...................................4-5 RUB Test in Hydrogenated Steam......................................................................4-6 Weibull and Weibayes Analyses of the Test Results ..........................................4-8 Improvement Factor by Weibull Analysis ..........................................................4-15 Improvement Factor with Minimum Alloy 600 Crack Time................................4-18 4.1.1 4.1.2 4.1.3 4.1.4 4.1.5 4.1.6 4.1.7 4.2 4.3

Field Experience ........................................................................................................4-21 References ................................................................................................................4-25

5 PWSCC OF THICK-WALLED ALLOY 690 MATERIAL ........................................................5-1 5.1 Laboratory Testing .......................................................................................................5-2 Crack Initiation Studies .......................................................................................5-3 Testing in Simulated Primary Water by Mitsubishi Heavy Industries ......................................................................................................................5-3 Testing in Pure Supercritical Water at the University of Michigan ................5-8 Testing of Alloy 690 Material not directly related to PWR Components .....5-10 5.1.1.1 (MHI) 5.1.1.2 5.1.2 5.1.2.1 5.1.1

Crack Growth Rate Studies...............................................................................5-10 5.1.2.1.1 Feasibility Studies in Simulated Primary Water by General Electric Global Research (GE-GRC).....................................................................5-11 5.1.2.1.2 Investigations in Simulated Primary Water at Argonne National Laboratory (ANL)...................................................................................................5-15 5.1.2.1.3 Additional Studies in Simulated Primary Water by General Electric Global Research (GE-GRC) as part of the MRP Test Program ...............5-20 5.1.2.1.4 Investigations in Simulated Primary Water at the University of Tohoku in Japan....................................................................................................5-28 5.1.2.1.5 Investigations in Simulated Primary Water at the Institute of Nuclear Safety Systems (INSS) in Japan .............................................................5-29 5.1.2.1.6 5.1.2.1.7 Water Further Investigations in Simulated Primary Water in Japan ..............5-30 Investigations by Bechtel-Bettis in Deaerated High-Temperature ............................................................................................................5-31

5.1.2.1.8 Investigations by Westinghouse in Supercritical Water with Additions of Boron, Lithium and Hydrogen............................................................5-39 5.1.2.2 Testing of Alloy 690 CRDM Material without Deliberate Cold Working ......5-44

xii

5.1.2.2.1 Investigations in Simulated Primary Water at Argonne National Laboratory (ANL)...................................................................................................5-45 5.1.2.2.2 Investigations in Simulated Primary Water at Pacific Northwest National Laboratory (PNNL)..................................................................................5-48 5.1.2.2.3 5.1.2.2.4 Investigations in Simulated Primary Water at Studsvik in Sweden .....5-52 Investigations in Simulated Primary water at MHI in Japan ................5-52

5.1.2.2.5 Investigations by Westinghouse in Supercritical Water with Additions of Boron, Lithium and Hydrogen............................................................5-52 5.1.2.3 Testing of Alloy 690 CRDM Material after Deliberate Cold Working...........5-53 5.1.2.3.1 Investigations in Simulated Primary Water by General Electric Global Research (GE-GRC)..................................................................................5-53 5.1.2.3.2 Investigations in Simulated Primary Water at Pacific Northwest National Laboratory (PNNL)..................................................................................5-58 5.1.2.3.3 Investigations by Westinghouse in Supercritical Water with Additions of Boron, Lithium and Hydrogen............................................................5-59 5.1.2.4 Testing of Heat Affected Zones (HAZ) from Welding of Alloy 690 Material ....................................................................................................................5-60 5.1.2.4.1 Investigations in High-Temperature Water at KAPL............................5-60 5.1.2.4.2 Investigations in Simulated Primary Water at Argonne National Laboratory (ANL)...................................................................................................5-60 5.1.2.4.3 5.1.2.4.4 5.1.2.4.5 Japan 5.2 5.3 Investigations in Simulated Primary Water at Studsvik in Sweden .....5-61 Investigations in Simulated Primary Water at CIEMAT in Spain .........5-62 Investigations in Simulated Primary Water at Tohoku University in ............................................................................................................5-62

Field Experience ........................................................................................................5-63 References ................................................................................................................5-63

6 DISCUSSION..........................................................................................................................6-1 6.1 6.2 6.3 Resistance of Alloy 690 to PWSCC .............................................................................6-1 Other Aspects of Alloy 690 Corrosion Behavior.........................................................6-10 References ................................................................................................................6-10

7 CONCLUSIONS .....................................................................................................................7-1 A TRANSLATED TABLE OF CONTENTS .............................................................................. A-1 (Japanese)................................................................................................................ A-2 / (Korean)...................................................................................................... A-19

xiii

LIST OF FIGURES
Figure 2-1 Carbon solubility diagram for Alloy 690 and Alloy 600 from [18] ..............................2-7 Figure 2-2 Time-Temperature-Sensitization diagram by modified Huey Test, Alloy 690 Heat NX4459HG (0.06%C) from [18] .................................................................................2-9 Figure 2-3 Time-Temperature-Sensitization diagram by modified Huey Test, Alloy 690 Heat NX9217H (0.01%C) from [18]..................................................................................2-10 Figure 2-4 Time-Temperature-Sensitization diagram by modified Huey Test, Alloy 690 Heat NX9780H (0.01%C) from [18]..................................................................................2-10 Figure 2-5 High-temperature tensile properties of annealed Alloy 690. Data shown are a composite of cold-and hot-worked products in the annealed condition and taken from [2] .............................................................................................................................2-12 Figure 3-1 Japanese fatigue data for Ni-base alloys in air at room temperature from [11] ........3-5 Figure 3-2 Japanese fatigue data for Ni-base alloys in simulated PWR water at 325C from [11] .............................................................................................................................3-5 Figure 3-3 Japanese data from [11] on relationship between calculated factor of environmental fatigue enhancement (Fen) and strain rate for Ni alloys in simulated PWR water at 325C ..........................................................................................................3-6 Figure 3-4 Japanese data from [11] on relationship between calculated factor of environmental fatigue enhancement (Fen) and temperature for various materials in simulated LWR water at 325C ..........................................................................................3-7 Figure 3-5 Japanese data from [11] showing results of model predictions for the corrosion fatigue behavior of Ni-base alloys in simulated PWR water at 325C compared with experiments ...............................................................................................3-8 Figure 3-6 Corrosion fatigue initiation data from [12] in high-temperature, deaerated water for high chromium weld alloys and Alloy 690 compared with stainless steels..........3-9 Figure 3-7 Approach to the analysis of environmental effects on cyclic crack growth developed at ANL and now applied to PWSCC testing of Alloy 690 and its weld alloys (from [13]) ..............................................................................................................3-10 Figure 3-8 No environmental enhancement of cyclic crack growth seen for thermally treated Alloy 690 material in either simulated PWR primary water or de-aerated pure water at 320C [14], but under loading conditions not expected to favor EAC ........3-10 Figure 3-9 Appearance of significant environmental enhancement at lower rates of cyclic crack growth for unidirectionally cold-rolled Alloy 690 material (blue points) in simulated PWR primary water at 320C (from [14]) .........................................................3-11 Figure 3-10 Japanese data from [16] indicating that a simulated PWR primary environment can enhance the fatigue crack growth rate of Alloys 600 and 690 by 5 to 10 times over a range of test conditions ......................................................................3-12

xv

Figure 3-11 Comparison of lines from a proposed (very conservative) Japanese model [16] for cyclic crack growth of Ni-base alloys in a PWR environment and experimental data that breach existing ASME curves......................................................3-13 Figure 3-12 Cyclic CGR behavior of Alloys 600 and 690 in Swedish studies [15] ...................3-14 Figure 3-13 Cyclic CGR threshold determined in simulated primary water for Alloy 690 in Swedish studies [15] ........................................................................................................3-14 Figure 3-14 Classification scheme for categories of fracture resistance to LTCP after Brown and Mills [36].........................................................................................................3-19 Figure 3-15 J-R curves determined by Brown and Mills for Alloy 690 in RT air and water at various temperatures [36] ............................................................................................3-20 Figure 3-16 JIC and T values determined by Brown and Mills [36] for Alloy 690 in air and water at various temperatures (values within bars indicate the dissolved hydrogen concentration) ..................................................................................................................3-21 Figure 3-17 Comparison of fractography from J-R testing under various conditions by Brown and Mills [36] of Alloy 690 (top) and Alloy 600 (bottom)........................................3-22 Figure 3-18 Results by Paraventi and Moshier [42] from J-R testing of Alloy 690 plate with additional, non-uniform cold work in 50C water with varying contents of dissolved hydrogen ..........................................................................................................3-23 Figure 4-1 Vickers hardness number as a function of cold reduction % from [2]. Alloy 690 has a higher work-hardening rate than Alloy 600...............................................................4-5 Figure 4-2 Weibull plot from [2] of early Alloy 600 RUB results in primary water at 360C generated in France. The Alloy 690 (three heats) Weibayes line assumes = 5.0 ...........4-9 Figure 4-3 Weibull plot from [2] for RUB test results in deaerated water at 365C reported by Norring et al. [17]. The Alloy 690 (many heats) Weibayes line assumes = 5.0 ..............................................................................................................................4-10 Figure 4-4 Weibull plot from [2] for RUB test results by Norring et al. [17] on special production heats of Alloy 600 in deaerated water at 365C. The Alloy 690 Weibayes line assumes = 5.0 ........................................................................................................4-11 Figure 4-5 Weibull comparison from [2] for different tube diameters from RUB testing by Norring et al. [17] in deaerated water at 365C. The Alloy 690 Weibayes line assumes = 5.0...............................................................................................................4-11 Figure 4-6 Weibull plots from [2] of Japanese Alloy 600MA (one heat) CLT results at 340C (644F) in primary water. No failure was observed in any Alloy 600MA or Alloy 600TT CLT specimens tested at 320C (608F) and in the Alloy 690TT (one heat) CLT specimens tested at 360C (680F). The Weibayes lines for the unfailed specimens assume = 5.0 ..............................................................................................4-12 Figure 4-7 Comparison from [2] of Japanese data for 20% pre-strained RUB specimens of Alloy 600MA or Alloy 600TT, tested in primary water at 320C (608F), with those for Alloy 690TT, tested under the same conditions, but at 360C (680F). The Alloy 690TT Weibayes line assumes = 5.0 ............................................................................4-13 Figure 4-8 Weibull plot from [2] of RUB results at 360C in primary water reported by Vaillant et al. [18]. The Alloy 600 RUB specimens were from four different heats in the MA and TT conditions. The Alloy 690 RUB specimens, also from four different heats in the MA and TT conditions, experienced no failure after up to 54,000 hours of exposure. The Alloy 690 Weibayes line assumes = 5.0............................................4-14

xvi

Figure 4-9 Weibull plot from [2] of SG mockups tested in deaerated water at 680F by Framatome ANP, France. Alloy 690TT SG mockups experienced no failure after 100,000 hours of exposure. The Alloy 690 Weibayes line assumes = 5.0....................4-14 Figure 4-10 Weibull and factors for Alloy 600 tests listed in Table 4-3 ..............................4-17 Figure 4-11 Improvement factors listed in Table 4-2 and Table 4-4 per Equation 4-8 versus test duration..........................................................................................................4-19 Figure 4-12 Swedish RUB testing for crack initiation in Alloys 600 and 690 from [24] ............4-21 Figure 4-13 Worldwide causes of Alloy 600TT SG tube repair by degradation mechanism from [31]........................................................................................................4-23 Figure 4-14 Worldwide causes of Alloy 690TT SG tube repair by degradation mechanism from [31]........................................................................................................4-23 Figure 5-1 MHI test loop for uni-axial constant load studies of PWSCC initiation [4, 5].............5-5 Figure 5-2 Active loading mechanism for uni-axial constant load studies of PWSCC initiation [4, 5] .....................................................................................................................5-6 Figure 5-3 Test specimens for uni-axial constant load studies of PWSCC initiation [4, 5].........5-6 Figure 5-4 Dependence of PWSCC initiation in Alloy 600MA on applied stress and lack of cracking in 690TT BMI material after 58,000 hours of testing [4, 5]...............................5-7 Figure 5-5 Dependence of PWSCC initiation in Alloy 600MA on applied stress and lack of cracking in 690TT CRDM nozzle material after 73,000 hours of testing [4, 5] ...............5-7 Figure 5-6 CGRs derived from a CERT study of various austenitic alloys in pure, deaerated SCW [6].................................................................................................................5-8 Figure 5-7 Cross-section of EPRI Alloy 600 after testing in 400C/25.4MPa deaerated pure SCW [7]......................................................................................................................5-9 Figure 5-8 Cross-section of EPRI Alloy 690 after testing in 400C/25.4 MPa deaerated pure SCW [7]....................................................................................................................5-10 Figure 5-9 Convention with regard to specimen orientation and the principle axis of coldworking in Alloy 690 base material...................................................................................5-12 Figure 5-10 CGR response of cold-worked Alloy 690 plate (with low-temperature mill anneal) tested at GE-GRC for >3000 h at constant stress intensity [9] ...........................5-13 Figure 5-11 CGR response of cold-worked Alloy 690 plate (with high-temperature mill anneal) tested at GE-GRC for >3000 h at constant stress intensity [9] ...........................5-14 Figure 5-12 Predominantly transgranular morphology (but with some intergranular facets) within band of PWSCC crack growth in cold-worked Alloy 690 plate initially tested at GE-GRC [9] .......................................................................................................5-15 Figure 5-13 Details of sample removal from cold-rolled Alloy 690TT plate tested at ANL [11 to 15] ..........................................................................................................................5-16 Figure 5-14 On-line data from ANL testing [11 to 15] of cold-rolled Alloy 690TT plate (S-L specimen orientation).......................................................................................................5-17 Figure 5-15 On-line data from ANL testing [11 to 15] of cold-rolled Alloy 690TT plate (ST specimen orientation)....................................................................................................5-17 Figure 5-16 Macro- and microfractography for the ANL 690TT plate specimen with S-L orientation [11 to 15] ........................................................................................................5-18

xvii

Figure 5-17 Macrofractography for the ANL 690TT plate specimen with S-T orientation [11 to 15] ..........................................................................................................................5-19 Figure 5-18 ANL test showing that the CGR behavior for cold-worked Alloy 690TT plate was unaffected by changing temperature within the range 320 to 300C [13] .................5-20 Figure 5-19 Rapid crack growth in 1D cold-rolled (~26%) Alloy 690 plate supplied by ANL and tested at GE-GRC [17, 18] ................................................................................5-21 Figure 5-20 Second period of rapid PWSCC in 1D cold-rolled Alloy 690 plate from ANL and apparent lack of a CGR response to reducing temperature from 360 to 325, then to 290C [17, 18] ......................................................................................................5-22 Figure 5-21 Change in macroscopic appearance of PWSCC region upon reducing temperature (c372) [17, 18]..............................................................................................5-22 Figure 5-22 Repeat test at GE-GRC on 1-D rolled ANL plate material showing a slight increase in CGR upon raising the dissolved hydrogen concentration [17, 18].................5-23 Figure 5-23 Moderately rapid PWSCC in 20% 1D cold-rolled Alloy 690 GE-GRC forged bar [17, 18] .......................................................................................................................5-24 Figure 5-24 Continued testing of 20% 1D cold-rolled Alloy 690 GE-GRC forged bar with reduction in both test temperature and applied stress intensity [17, 18] ..........................5-24 Figure 5-25 Lack of response to a major change in dissolved hydrogen during testing of a second specimen from a 20% 1-D cold-rolled Alloy 690 forged bar [17, 18] ................5-25 Figure 5-26 Results of testing the same material as in Figure 5-23, but this time in the ST orientation [17, 18] ........................................................................................................5-25 Figure 5-27 Results from a further heat of Alloy 690 tested after 1-D cold-rolling at GEGRC, this time showing expected response to a drop in temperature even for an SL oriented specimen [17, 18]............................................................................................5-26 Figure 5-28 High-resolution fractography (c372) of the 26% 1D-cold rolled ANL 690 plate tested at GE-GRC [17, 18] ...............................................................................................5-27 Figure 5-29 High-resolution fractography of a 20% 1-D cold rolled Alloy 690 specimen from forged bar [17, 18]....................................................................................................5-27 Figure 5-30 Example of crack growth observed at INSS in 20% cold worked 690TT at 360C (from [8])................................................................................................................5-30 Figure 5-31 Apparent effect of temperature and degree of cold work (CW) on measured CGR for 690TT (from [8]) .................................................................................................5-31 Figure 5-32 Increase in CGRs with increasing level of uni-directional cold-rolling in Bettis studies [22, 23].................................................................................................................5-35 Figure 5-33 Cold-rolling is more detrimental than tensile pre-straining (despite lower yield strength) [22, 23]......................................................................................................5-35 Figure 5-34 Effect of test temperature and degree of cold work on CGRs for the VIM/ESR plate [22, 23].....................................................................................................5-36 Figure 5-35 Apparent increase in CGRs at 50 cc/kg hydrogen (blue symbols) vs. 23 cc/kg (pink symbols) for Alloy 690 (left) and comparison (right) with opposite behavior for Alloy 600 [22, 23] .........................................................................................5-37 Figure 5-36 Predominantly intergranular crack advance in an S-L-oriented specimen of VIM/ESR TT plate subjected to only 12% cold-rolling [22, 23].........................................5-37

xviii

Figure 5-37 Detail from fractography of an S-T-oriented specimen of VIM/ESR plate subjected to 24% cold-rolling [22, 23] ..............................................................................5-38 Figure 5-38 Comparison of Alloy 600 and 690 CGRs measured by Bettis [22, 23] .................5-38 Figure 5-39 Attempted back-extrapolation by Bettis of cold-worked Alloy 690 CGRs on the basis of assumed yield strength dependencies [22, 23].............................................5-39 Figure 5-40 Example of transition from transgranular cracking during cyclic loading in SCW to intergranular SCC at ~ constant load in a 10% cold-worked specimen of Alloy 690TT [25] ...............................................................................................................5-43 Figure 5-41 Average SCC growth rate vs. stress intensity factor for Alloy 600 and 690 materials [25]....................................................................................................................5-43 Figure 5-42 Comparison of CGR measured for Alloy 600 MA and Alloy 690 TT in supercritical and primary water [25] .................................................................................5-44 Figure 5-43 Specimen orientation in ANL testing of Alloy 690 CRDM material [11 to 15] .......5-45 Figure 5-44 Reported CGRs during brief test periods at constant load in ANL work on CRDM material [11 to 15].................................................................................................5-46 Figure 5-45 Absence of clear intergranular cracking on fracture surface of the first CRDM specimen tested at ANL [11 to 15] .......................................................................5-47 Figure 5-46 Reported CGRs during constant-load test periods at two test temperatures in further ANL work on CRDM material without deliberate cold working [11 to 15]..........5-48 Figure 5-47 Orientations of CT specimens used by PNNL for CGR testing on two Alloy 690 forgings: left is heat RE243; right is heat WP140 [26]...............................................5-49 Figure 5-48 PNNL data showing the patience used in transitioning to constant K loading [26] ...................................................................................................................................5-50 Figure 5-49 PNNL data showing Alloy 690 CGR response for as-received TT versus a carbide-modified SA condition [26] ..................................................................................5-50 Figure 5-50 PNNL data showing Alloy 690 CGR response for two further heats of material in the as-received TT condition [26] ...................................................................5-51 Figure 5-51 PNNL fractography showing IG cracking limited to isolated grains in specimens without cold-work [26] ....................................................................................5-51 Figure 5-52 Specimen location in GE testing of Alloy 690 CRDM material [17, 18] ................5-54 Figure 5-53 Data from one of two specimens of CRDM material with 20% homogeneous cold work tested at GE-GRC [29] .....................................................................................5-55 Figure 5-54 Intergranular crack morphology in 20% cold-worked Alloy 690 CRDM specimen [29]...................................................................................................................5-55 Figure 5-55 Data from a further specimen of CRDM material with 41% homogeneous cold work showing tendency to crack arrest before reaching constant K conditions (top) and very low CGRs even under periodic partial unloading with a 24h hold time (bottom) [29].....................................................................................................................5-56 Figure 5-56 Macro and micrfractography from the specimen of CRDM material with 41% homogeneous cold work showing extensive out-of-plane secondary cracking [29] ...................................................................................................................................5-57 Figure 5-57 Data from PNNL testing of CRDM material with deliberate cold working (here 17% in the S-L orientation) [26] ..............................................................................5-58

xix

Figure 5-58 Data from PNNL testing of CRDM material with deliberate cold working (here 30% in the T-L orientation) [26] ..............................................................................5-59 Figure 5-59 Details of an Alloy 690 HAZ specimen (CF690) under test at ANL [13] ...............5-61 Figure 5-60 Details of CT specimen being used at Studsvik to examine the CGR behavior in Alloy 690 HAZ material [27] ...........................................................................5-62 Figure 6-1 Summary of laboratory SCC CGR data (as of November 2008) prepared by PNNL [26] and showing the possibility of measuring moderate to high rates in Alloy 690 plate material subjected to non-uniform cold work ......................................................6-7 Figure 6-2 Evidence of microstructural banding in some areas of the 1-D cold-rolled ANL Alloy 690 plate material examined at GE-GRC [18] ...........................................................6-7 Figure 6-3 Microstructural banding perpendicular to the crack plane in Alloy 690 plate material from EPRI orginally tested for the MRP Program by GE-GRC [9]........................6-8 Figure 6-4 Very uniform, homogeneous microstructure in extruded Alloy 690 CRDM material [18] .......................................................................................................................6-8

xx

LIST OF TABLES
Table 1-1 First reported occurrence of Alloy 600 PWSCC for various PWR component items...................................................................................................................................1-2 Table 2-1 ASME specifications of Alloys 690 and 600 ..............................................................2-1 Table 2-2 ASME chemical composition requirement (wt%) .......................................................2-2 Table 2-3 ASME room temperature properties ..........................................................................2-2 Table 2-4 ASME elevated temperature properties for Alloy 690 and Alloy 600 .........................2-3 Table 2-5 Chemical composition (wt%) of Alloy 690 heats used by Sarver et al. [18] ...............2-6 Table 2-6 Effect of heat treatment on Alloy 690 carbide precipitation [18] ..............................2-11 Table 3-1 Estimated typical improvement factors vs. pHT considering all environments [33, 34] .............................................................................................................................3-17 Table 3-2 Chemical composition of Alloys 600, 690, 82, & 52 tested by Brown and Mills [36] ...................................................................................................................................3-18 Table 4-1 Examples of surface IG cracking in Alloy 690 and Alloy 52 specimens [2]................4-3 Table 4-2 Summary of Alloy 690 primary water stress corrosion test data to 2004...................4-7 Table 4-3 Weibull analysis for Alloy 600 tested with Alloy 690 [2] ...........................................4-15 Table 4-4 Summary of Alloy 690 hydrogenated and doped hydrogenated steam stress corrosion test data to 2004...............................................................................................4-19 Table 4-5 Swedish RUB testing for crack initiation in Alloys 600 and 690 [23]........................4-20 Table 4-6 List of operating steam generators manufactured with Alloy 690 tubing (as of December 2008) ..............................................................................................................4-24 Table 5-1 Examples of some replaced PWR RPV heads with CRDM penetrations in Alloy 690 ............................................................................................................................5-1 Table 5-2 Examples of some relatively thick-walled Alloy 690 reactor coolant system original equipment or replacement component items other than CRDM penetrations.......5-2 Table 5-3 Origins and heat treatments of Alloy 690 CRDM nozzles tested by EdF [3]..............5-3 Table 5-4 Chemical composition of materials used in MHI testing for PWSCC initiation [4, 5] ...................................................................................................................................5-4 Table 5-5 Heat treatment and mechanical properties of materials used in MHI testing for PWSCC initiation [4, 5].......................................................................................................5-4 Table 5-6 Environmental test conditions used in MHI testing for PWSCC initiation [4, 5] .........5-5 Table 5-7 Summary of maximum crack depths measured on cross-sectioned samples at the University of Michigan [7] .............................................................................................5-9 Table 5-8 Chemical composition and mechanical properties of all the Alloy 690 materials tested at GE-GRC in the MRP program ...........................................................................5-11

xxi

Table 5-9 Chemical composition of Alloy 690TT plate material tested at ANL [11 to 15] ........5-15 Table 5-10 Material test matrix for Bettis Alloy 690 test program [22, 23] ...............................5-33 Table 5-11 Bettis summary of relative SCC susceptibility for Alloy 690 [22, 23]......................5-34 Table 5-12 Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25] .........................................................................................................................5-40 Table 5-13 Chemical composition of Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25] .................................................................................5-40 Table 5-14 Tensile properties (as reported by material vendors) of the Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25] ..................5-41 Table 5-15 Tensile properties after trial forging of the Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25] ..............................................5-41 Table 5-16 Detailed Results for the Alloy 600 control samples and Alloy 690 plate material used in Westinghouse testing for CGRs in supercritical water [25]....................5-42 Table 5-17 Detailed results for the Alloy 600 control samples and as-received Alloy 690 CRDM materials used in Westinghouse testing for CGRs in supercritical water [25] ......5-52 Table 5-18 Detailed results for the Alloy 600 control samples and Alloy 690 cold-worked CRDM materials used in Westinghouse testing for CGRs in supercritical water [25] ......5-60 Table 6-1 Summary of results of CGR testing in simulated PWR primary water on Alloy 690 CRDM material without deliberate cold working (status: December 2008) .................6-3 Table 6-2 Summary of results of CGR testing in simulated PWR primary water on Alloy 690 CRDM material with deliberate cold working (status: December 2008) ......................6-4

xxii

LIST OF ACRONYMS

ALARA APFIM ATEM ASME ASTM AWS AVT CEA CGR CL CLT CEDM CERT CM CRDM CT DCB DSA

As low as reasonably achievable Atom probe field ion microscope (microscopy) Analytical transmission electron microscope (microscopy) American Society of Mechanical Engineers American Society for Testing and Materials American Welding Society All volatile treatment Commisariat l`nergie Atomique (the French Atomic Energy Commission) Crack growth rate Constant load Constant load test Control element drive mechanism Constant extension rate test, also known as slow strain rate test (SSRT) Carbide modified (heat treatment used in laboratory testing) Control rod drive mechanism Compact tension (specimen) Double cantilever beam specimen Dynamic strain aging xxiii

DSC DTA EdF EFPH EFPY EPRI FOI GTA HAZ ID IG IGA IGSCC LTCP MA MRP NRC OD OTSG ppm ppb PPU PWR xxiv

Differential scanning calorimetry Differentential thermal analysis Electricit de France (the French national electric utility) Effective full power hours Effective full power years Electric Power Research Institute Factor of (relative) improvement (also known as IFR) Gas tungsten arc (welding) Heat-affected zone Inside diameter Intergranular Intergranular attack Intergranular stress corrosion cracking Low temperature crack propagation Mill annealed Materials reliability program Nuclear Regulatory Commission Outside diameter Once-through steam generator Parts per million Parts per billion Periodic partial unloading (trapezoidal waveform used in CGR testing) Pressurized water reactor

PWHT PWSCC RUB

Post weld heat treatment Primary water stress corrosion cracking Reverse U-bend stress corrosion cracking specimens made from split half steam generator tubing, hence also known as split tube U-bend specimens Stress corrosion cracking Supercritical water Scanning electron microscope (or microscopy) Steam generator Slow strain rate test, also known as constant extension rate test (CERT) Scanning transmission electron microscope (or microscopy) Transmission electron microscope (or microscopy) Transgranular stress corrosion cracking Thermal treatment Spanish association of nuclear utilities Unified numbering system Wedge opening loading (specimen) X-ray photoelectron spectroscopy Unidirectional

SCC SCW SEM SG SSRT STEM TEM TGSCC TT UNESA UNS WOL XPS 1-D

xxv

1
INTRODUCTION
1.1 Background

Wrought Alloy 600 and its weld metals (Alloy 182 and Alloy 82) were originally used in pressurized water reactors (PWRs) due to the materials inherent resistance to general corrosion in a number of aggressive environments and because of a coefficient of thermal expansion that is very close to that of low alloy and carbon steel. Over the last thirty-plus years, primary water stress corrosion cracking (PWSCC) has been observed in Alloy 600 component items and Alloy 82/182 welds such as steam generator tubes and plugs, pressurizer heater sleeves and welds, pressurizer instrument nozzles, reactor vessel closure head nozzles and welds, reactor vessel outlet nozzle welds, and more recently in a lower reactor vessel head instrumentation nozzle and weld. Table 1-1 provides a synopsis of the Alloy 600 PWSCC experience in commercial PWRs up to 2003. This table identifies the first commercially observed occurrence of PWSCC for each particular component item in a PWR and lists the approximate service life (in calendar years) at the time PWSCC was identified at that particular location. PWSCC was first observed at very highly stressed tube locations in the hot leg of steam generators in the 1970s. Pressurizer nozzles, which operate at the highest temperature in PWRs, were the next locations to have leakage and failures identified. Currently, PWSCC has been observed in nozzles and welds at nearly all locations where Alloy 600 is utilized throughout the reactor coolant system. The particularly complex service experience with cracking of Alloy 182 and 82 weld metals was recently reviewed [1]. The occurrence of PWSCC has been responsible for significant downtime and replacement power costs at PWRs. Notable examples of equipment failures include extended outages and repairs or replacements at Calvert Cliffs, V.C. Summer, Oconee Nuclear Station, Davis-Besse, and North Anna. Repairs and replacements since the late 1980s have generally utilized wrought Alloy 690 material and its weld metals (Alloy 152 and Alloy 52(M)), which have been shown to be considerably less susceptible to PWSCC in laboratory experiments and are thought to be resistant for all practical purposes under normal operating conditions. Nevertheless, U.S. NRC Bulletin 2002-02, Reactor Vessel Head and Vessel Head Penetration Nozzle Inspection Programs, dated August 9, 2002, originally indicated essentially that no inspection credit would be given for upgraded replacement reactor vessel heads, using Alloy 690 wrought and weld materials, over existing head penetrations using Alloy 600 wrought and weld materials. This situation has changed recently with the issuance of modifications to U.S. NRC Regulation 10 CFR Part 50 [2] that now reference ASME Code Case N729-1.

1-1

Introduction Table 1-1 First reported occurrence of Alloy 600 PWSCC for various PWR component items Component Item Steam Generator Hot Leg Tubes Pressurizer Instrument Nozzles Steam Generator Cold Leg Tubes Pressurizer Heaters and Sleeves Steam Generator Channel Head Drain Pipes Pressurizer Heater Diaphragm Plate Weld Control Rod Drive Mechanism Nozzles Hot Leg Instrument Nozzles Power Operated Relief Valve Safe End Pressurizer Nozzle Welds (Repair) Steam Generator Tubesheet Plate Cladding Cold Leg Piping Instrument Nozzles(b) Hot Leg Nozzle Welds Reactor Vessel Hot Leg Nozzle Buttering/Piping Welds Pressurizer Instrumentation Nozzle Welds Control Rod Drive Mechanism Nozzle/RV Head Welds Surge Line Nozzle Welds Reactor Vessel Lower Head In-Core Instrumentation Nozzles/Welds Date PWSCC Initially Observed 1971 1986 1986 1987 1988 1989 1991 1991 1993 1994 1995 1997 2000 2000 2000 2000 2002 2003 Service Life(a) (Calendar Years) 2 2 18 5 1 16 12 5 22 1 13 13 13 17 27 27 21 14

(a) This listing identifies the first reported occurrence of identified cracking for each component item. Leakage has occurred in some component items in less than one year of service life and in other component items after nearly 30 years of service. (b) One plant identified suspect visual evidence of boric acid leakage around two nozzles during a visual inspection; nozzles were preventively repaired without investigating whether leakage had in fact occurred.

1.2

Purpose and Scope of Revising MRP-111

Based on excellent test results and field performance, Alloy 690 has become the replacement material of choice for degraded Alloy 600 component items in PWRs. The objective of the original MRP-111 report [3], published in 2004, was to document everything that was then known about the resistance of Alloy 690 and its weld metals to PWSCC, both from laboratory testing and field experience, and to quantify factors of improvement in behavior over Alloys 600, 182 and 82. The summary conclusions reached were as follows:

1-2

Introduction

Wrought Alloy 690 and its weld metals (Alloys 52 and 152) are acceptable and highly corrosion-resistant replacement materials for Alloy 600 and its weld metals in PWRs, although limited, further testing is needed to examine some specific knowledge gaps that have been identified. Wherever possible, the existing laboratory test data have been evaluated to estimate the improvement factor of Alloy 690 relative to Alloy 600. Average improvement factors of at least 26 relative to Alloy 600MA material and 13 relative to Alloy 600TT material can be derived, but these numbers are clearly conservative, due to an absence of PWSCC in most Alloy 690 specimens within the test duration. Of the various knowledge gaps mentioned in 2004, the most important clearly involved the Alloy 52 and 152 weld metals, where only one investigation each had been reported concerning PWSCC testing. The second most important gap concerned the possible effects of Alloy 690 product form (plate, tube, rolled bar, forged bar, and extruded bar), since nearly all testing up to that time had been performed with specimens obtained from thin-walled steam generator tubing. Mention was made in [3] of possible effects from subtle changes in chemical composition and thermo-mechanical processing. It was also noted in 2004 that Given the welcome difficulties in initiating cracking of Alloy 690 and its weld metals, no plausible estimates of crack growth rates for base metal, HAZ and weld metals are available at this time. Furthermore, a lack of data was identified concerning the combined effects of exposure to primary water and fatigue on Alloy 690 and its weld metals, although it was postulated that models developed for Alloy 600 may also be applicable here. Finally, it was mentioned that additional efforts might be needed to confirm the lack of relevance of low temperature crack propagation (LTCP) for Alloys 690/152/52 in PWR primary water, even though the available operating experience did not suggest that a practical problem exists. Extensive studies, particularly on the crack growth behavior of thick-walled Alloy 690 materials, have been carried out since 2004 and were summarized in 2008 [4]. Many of these investigations are still ongoing and, together with the research branch of the U.S. NRC, EPRI is currently leading efforts to establish an international research collaboration on this topic. Nevertheless, it is judged that most of the abovementioned knowledge gaps have already been sufficiently closed for Alloy 690 base material to make revision of MRP-111 worthwhile. Note, however, that this revised report does not cover the weld metals (Alloys 152 and 52(M)), since too much additional data on these is expected shortly. These results will be incorporated in a separate report, currently scheduled for publication in 2010. The present report includes some of the original material from MRP-111, but does not entirely replace it. The report structure has been re-arranged considerably to help focus on the key issue of demonstrating adequate, long-term, PWSCC resistance of thick-walled Alloy 690 components exposed to primary water environments. It is also intended to provide a technical basis for future development of the most appropriate inspection requirements for these materials.

1-3

Introduction

1.3

Causes of Alloy 600 PWSCC

Stress corrosion cracking of metals and alloys is caused by the synergistic effects of environment, material condition, and stress. In a PWR primary water environment, intergranular stress corrosion cracking of wrought Alloy 600 material and its weld metals (Alloy 182 and Alloy 82) is commonly referred to as PWSCC. The occurrence of stress corrosion cracking of Alloy 600 in high-purity water has been extensively studied since the first reported observation of cracking in laboratory tests by Coriou et al. [5] in 1959. The mechanism of this cracking phenomenon is still not completely understood, and prediction of crack initiation time has proven to be very difficult due to the uncertainty of numerous contributory variables. These include metallurgical condition, cold work, and residual stress. Although the crack initiation time can vary tremendously from heat to heat, wrought Alloy 600 material and its weld metals are generally susceptible to PWSCC when the total stress level is close to, or exceeds, the yield strength at operating temperatures. PWSCC is a thermally-activated mechanism that can be correlated with an Arrhenius relationship (exponential) and is very temperature dependent. The vast majority of PWSCC at steam generator roll expansion transitions has occurred first on the hot-leg side of the tubesheet due to the 27-38C (50-70F) higher temperatures. However, failures of Alloy 600 material have also been reported in France [6] to have occurred in reactor vessel upper head nozzle material at a temperature of approximately 290C (554F). On at least one other occasion, PWSCC has been cited on a component item at a significantly lower water temperature of 217C (423F), but the details leading to this conclusion have not been independently verified [7]. The susceptibility of Alloy 600 depends on several factors including the chemical composition, metallurgical condition during manufacture of the material, heat treatment during fabrication of the component item, and its operating parameters [8]. The carbon and chromium contents appear to be the most important chemical composition variables. These, in turn, affect chromium carbide precipitation during thermo-mechanical processing. Microstructural conditions, such as grain size and location relative to carbide precipitation, are also important variables that determine the susceptibility of a particular material to PWSCC. Finally, fabrication parameters and heat treatment determine the overall yield strength and degree of cold work. Alloy 600 that has been low temperature mill-annealed, with grain boundaries poorly decorated with carbides, and has relatively high yield strength (due, e.g., to some degree of remaining cold work) is generally observed to be the most susceptible to PWSCC. Work is ongoing in the USA on an empirical/theoretical hybrid model [9] that shows promise in predicting the crack propagation rate, in particular of thick-walled Alloy 600 material, on the basis of the different engineering tensile properties of each heat (while taking into account changes in applied stress intensity factor, dissolved hydrogen, and water temperature). Efforts also continue in France to refine engineering predictions of the initiation of PWSCC in Alloy 600 on the basis of so-called susceptibility index models [10, 11]. Tensile stresses, resulting from both residual and operating stresses, can be significant for some Alloy 600 component items. A stress close to the high-temperature material yield strength is generally necessary for PWSCC to initiate. Operating stresses arise from mechanical (pressure) and thermal loading, while residual stresses are generated as a result of fabrication, installation, and welding processes. Residual stresses are more difficult to quantify than operating stresses. In many instances, they are of a higher magnitude and usually a major factor leading to premature failure. Note, however, that initiation of cracking may still occur only after very long periods of operation, particularly in heats of Alloy 600 exhibiting low susceptibility. 1-4

Introduction

Once PWSCC initiates in some thin wall components, such as pressurizer heater sleeves, the part concerned has essentially reached the end of its service lifeb. This is not true, however, for thickwalled components such as RPV head penetrations, where an appreciable amount of time may be taken up with crack growth before safety margins related to structural integrity are breached. One of the key factors governing crack growth rate (CGR) is the stress intensity developed at the crack tip and in some cases this may drop as the crack grows into a region of decreasing residual stress, thus leading to much slower growth, or even to crack arrest. Further consideration of Alloy 600 CGRs and their importance for determining the residual service life of components known to contain cracks from PWSCC are given in related MRP reports [12, 13]. In summary, PWSCC requires three key factors to be present simultaneously: an environment that promotes intergranular stress corrosion cracking, susceptible material, and significant, prolonged tensile stress. Eliminating any one of these three factors will mitigate cracking in principle, although in practice it is prudent to attack all of these factors at once, wherever feasible.

1.4

References

1. P.M. Scott and M.-C. Meunier, Materials Reliability Program: Review of Stress Corrosion Cracking of Alloys 182 and 82 in PWR Primary Water Service (MRP-220). EPRI, Palo Alto, CA: 2007. 1015427. 2. NRC, Industry Codes and Standards; Amended Requirements; 10 CFR Part 50, Federal Register: September 10, 2008 (Volume 73, Number 176). 3. H. Xu et al., Materials Reliability Program (MRP), Resistance to Primary Water Stress Corrosion Cracking of Alloys 690, 52, and 152 in Pressurized Water Reactors (MRP-111). EPRI, Palo Alto, CA: 2004. 1009801. 4. J. Hickling, EPRI Materials Reliability Program: Resistance of Alloys 690, 152 and 52 to Primary Water Stress Corrosion Cracking (MRP-237, Rev 1): Summary of findings from completed and ongoing test programs since 2004. EPRI, Palo Alto, CA: 2008. 1018130. 5. H. Coriou, et al., High Temperature Stress Corrosion Cracking of Inconel in Water, Third Metallurgical Symposium on Corrosion (Aqueous and Gaseous), 1959, North Holland Publishing Co., Amsterdam, published in 1960, pp. 161-169. 6. F. Champigny, F. Chapelier, et al., Maintenance Strategy of Inconel Components in PWR Primary System in France, paper presented at NRC/ANL Conference on Vessel Head Penetration Inspection, Cracking, and Repair in Gaithersburg, MD from September 29 October 2, 2003. 7. B. Gronwall, L. Ljungberg, et al., Intercrystalline Stress Corrosion Cracking of Inconel 600 Inspection Tubes in the Agesta Reactor, Atomenergi, (Rapp.) AE, AE-245, 1966.

An exception to this rule is that PWSCC in kiss-rolled transitions in SG tubes appears to grow so slowly, or even to arrest, such that many SG tubes remain in operation ten or more years following detection of PWSCC.

1-5

Introduction

8. C.A. Campbell and S. Fyfitch, PWSCC Ranking Model for Alloy 600 Components, Proceedings of 6th International Conference on Environmental Degradation of Materials in Nuclear Power SystemsWater Reactors, TMS, 1993, p. 863. 9. E. Eason, Program on Technology Innovation: A Preliminary Hybrid Model of Nickel Alloy Stress Corrosion Crack Propagation in PWR Primary Water Environments. EPRI, Palo Alto, CA: 2008. 1016546. 10. F. Vaillant et al., Development of a Predictive Model for SCC Initiation of Alloy 600 in Primary Water, Workshop on Detection, Avoidance, Mechanisms, Modeling, and Prediction of Stress Corrosion Cracking Initiation in Water-Cooled Nuclear Plants. EPRI, Palo Alto, CA; TEPCO R&D Center, Yokohama, Japan; AREVA NP, Technical Center, Le Creusot, France; Institut de Radioprotection et de Sret Nuclaire (IRSN), Fontenay-aux-Roses, France; Institute of Nuclear Safety System, Incorporated (INSS), Fukui, Japan; The Materials Aging Institute (MAI), Moret Sur Loing Cedex, France; and EDF R&D, Moret Sur Loing Cedex, France: 2009. 1018908. 11. I. de Curires and M.-C. Meunier, Prediction of PWSCC initiation in Steam Generator Alloy 600 Tubes by the Index Model: Comparison between Model and Experience, Workshop on Detection, Avoidance, Mechanisms, Modeling, and Prediction of Stress Corrosion Cracking Initiation in Water-Cooled Nuclear Plants. EPRI, Palo Alto, CA; TEPCO R&D Center, Yokohama, Japan; AREVA NP, Technical Center, Le Creusot, France; Institut de Radioprotection et de Sret Nuclaire (IRSN), Fontenay-aux-Roses, France; Institute of Nuclear Safety System, Incorporated (INSS), Fukui, Japan; The Materials Aging Institute (MAI), Moret Sur Loing Cedex, France; and EDF R&D, Moret Sur Loing Cedex, France: 2009. 1018908. 12. G. White et al., Materials Reliability Program Crack Growth Rates for Evaluating Primary Water Stress Corrosion Cracking (PWSCC) of Thick-Wall Alloy 600 Materials (MPR-55) Revision 1. EPRI, Palo Alto, CA: 2002. 1006695. 13. G. White et al., Materials Reliability Program: Reactor Vessel Closure Head Penetration Safety Assessment for U.S. Pressurized Water Reactor (PWR) Plants (MRP-110): Evaluations supporting the MRP Inspection Plan. EPRI, Palo Alto, CA: 2004. 1009807.

1-6

2
ALLOY 690 PROPERTIES AND METALLURGY
2.1 Material Specifications

Alloy 690 and its predecessor Alloy 600 were first developed by Inco Alloys International under the trade names Inconel Alloy 690 and Inconel Alloy 600 [1, 2]. The present owner of the trade name of Inconel is Special Metals Corporation. Currently, both alloys have been adopted by ASTM (American Society for Testing and Materials), ASME (American Society of Mechanical Engineers), and other international materials societies. Even though these alloys are listed under the UNS (Unified Numbering System) numbers in ASTM or ASME standards (see Table 2-1), these alloys are generically referred to as Alloy 690 or Alloy 600 in the nuclear power industry. Table 2-1 provides a summary of the most commonly used product forms in the ASME Boiler and Pressure Vessel Code Section II material specifications for the wrought Alloy 690 materials. Because Alloy 690 was developed to replace its predecessor Alloy 600 for light water nuclear power reactors, both are listed in the same ASME material specifications.
Table 2-1 ASME specifications of Alloys 690 and 600
Reprinted from ASME 2001 BPVC, Section II-B and 2001 BPVC, Section II-D, by permission of The American Society of Mechanical Engineers. All rights reserved.

Specification ASME SB-163c ASME SB-166d ASME SB-167e ASME SB-168f

c

Alloy UNS N06690 (Alloy 690) UNS N06600 (Alloy 600) UNS N06690 (Alloy 690) UNS N06600 (Alloy 600) UNS N06690 (Alloy 690) UNS N06600 (Alloy 600) UNS N06690 (Alloy 690) UNS N06600 (Alloy 600)

Product Form Seamless Tubing Rod, Bar, Wire Seamless Pipe and Tube Plate, Sheet, Strip

ASME SB-163, Specification for Seamless Nickel and Nickel Alloy Condenser and Heat-Exchanger Tubes, ASME Boiler and Pressure Vessel Code, Section II Part B, Nonferrous Material Specifications, 2001. ASME SB-166, Specification for Nickel-Chromium-Iron Alloys (UNS N06600, N06601, N06603, N00690, N06025, and N06045) and Nickel-Chromium-Cobalt-Molybdenum Alloy (UNS N06617) Rod, Bar, and Wire, ASME Boiler & Pressure Vessel Code, Section II Part B, Nonferrous Material Specifications, 2001. ASME SB-167, Specification for Nickel-Chromium-Iron Alloys (UNS N06600, N06601, N00690, N06025, and N06045) Seamless Pipe and Tube, ASME Boiler & Pressure Vessel Code, Section II Material Specifications, Part B, 2001. ASME SB-168, Specification for Nickel-Chromium-Iron Alloys (UNS N06600, N06601, N06603, N00690, N06025, and N06045) and Nickel-Chromium-Cobalt-Molybdenum Alloy (UNS N06617) Plate, Sheet, and Strip, ASME Boiler & Pressure Vessel Code, Section II Material Specifications, Part B, 2001.

2-1

Alloy 690 Properties and Metallurgy

Table 2-2 lists the chemical composition requirements for Alloys 690 and 600. The ASME chemical composition requirements for these alloys remain identical to the original specification developed by Inco Alloys International. However, stricter requirements on chemical composition, mechanical properties and heat treatment process are imposed on Alloy 600 or 690 by utilities and vendors for applications in PWRs (see below).
Table 2-2 ASME chemical composition requirement (wt%)
Reprinted from ASME 2001 BPVC, Section II-B and 2001 BPVC, Section II-D, by permission of The American Society of Mechanical Engineers. All rights reserved.

Alloy 690 600

Ni 58.0 min 72.0 min

Cr 27.031.0 14.017.0

Fe 7.011.0 6.010.0

C 0.05 max 0.15 max

Mn 0.50 max 1.0 max

Si 0.50 max 0.50 max

S 0.015 max 0.015 max

Ti

Nb + Ta

Cu 0.50 max 0.50 max

Al

Mo

Other

Table 2-3 lists the room temperature properties of Alloys 690 and 600 and Table 2-4 lists the ASME properties of these materials at elevated temperatures. Such properties are used in the design of Alloy 600 and 690 component items for repair or replacement and for calculating the operating stress. International specifications for Alloy 690 may differ from the above in various respects. It should also be pointed out that some components in the USA may have been fabricated to older specifications that differ slightly from those given here.
Table 2-3 ASME room temperature properties
Reprinted from ASME 2001 BPVC, Section II-B and 2001 BPVC, Section II-D, by permission of The American Society of Mechanical Engineers. All rights reserved.

Alloy 690 Density Poisons Ratio Min. Tensile Strength Min. Yield Strength Min. Elongation
g h i

Alloy 600 0.300g 8.30g 0.29h 80i 550i 30i 205i 35i

lb/in ksi

0.293 8.11g 0.29h 85i 586i 30i 205i 35i

kg/cm3

MPa ksi MPa %

ASME Boiler & Pressure Vessel Code, Section II Part D, Materials Properties, 2001, Table NF-2. ASME Boiler & Pressure Vessel Code, Section II Part D, Materials Properties, 2001, Table NF-1. ASME SB-167, Specification for Nickel-Chromium-Iron Alloys (UNS N06600, N06601, N00690, N06025, and N06045) Seamless Pipe and Tube, ASME Boiler & Pressure Vessel Code, Section II Material Specifications, Part B, 2001.

2-2

Alloy 690 Properties and Metallurgy Table 2-4 ASME elevated temperature properties for Alloy 690 and Alloy 600j
Reprinted from ASME 2001 BPVC, Section II-B and 2001 BPVC, Section II-D, by permission of The American Society of Mechanical Engineers. All rights reserved.

Youngs Modulus Temp E(a), x106 ksi F 70 100 200 300 400 500 600 650 700
(a) (b) (c) (d)

Coefficient of Thermal Expansion (b), x10-6 in/in/F 690 7.7 7.8 7.9 7.9 8.0 8.1 8.2 8.2 8.3 600 6.8 6.9 7.1 7.3 7.5 7.6 7.8 7.8 7.9

Design Stress Intensity(c) Sm, ksi 690 23.3 23.3 23.3 23.3 23.3 23.3 23.3 23.3 23.3 600 23.3 23.3 23.3 23.3 23.3 23.3 23.3 23.3 23.3

Yield Strength(d) Sy, ksi 690 35.0 35.0 31.7 29.8 28.6 27.9 27.6 27.5 27.5 600 35.0 35.0 32.0 31.2 30.7 30.3 29.9 29.7 29.4
j

Tensile Strength(d) Su, ksi 690 85.0 85.0 85.0 84.0 82.0 80.8 80.2 80.0 79.8 600 80.0 80.0 80.0 80.0 80.0 80.0 80.0 80.0 80.0

690 30.3 30.1(a) 29.5 29.1 28.8 28.3 28.1 27.9(a) 27.6
j

600 31.0 30.8 30.2 29.8 29.5 29.0 28.7 28.5 28.2

From Table TM-4 of . Values of Youngs modulus for 100F and 650F are obtained by linear interpolation. Mean coefficient of thermal expansion going from 70F to the indicated temperature, from Table TE-4 of . From Table 2B of . For annealed Alloy 690 and Alloy 600 of SB-163, SB-166, SB-167, and SB-168, some exceptions j exist: see Table 2B of for details. From Table Y-1 of . For annealed Alloy 690 and Alloy 600 of SB-163, SB-166, SB-167, and SB-168, some exceptions j exist: see Table Y-1 of for details.
j j

2.1.1 Typical PWR Specifications for Thin-Walled Alloy 690 SG Tubing For Alloy 690 SG tubing, the EPRI Guidelines require the carbon content to be between 0.015% and 0.025% [3]. The lower limit ensures continuous or semi-continuous intergranular carbide precipitation while the upper limit keeps intragranular carbide precipitation to a minimum upon thermal heat treatment. Alloy 690 has higher yield and tensile strengths than Alloy 600 due to the increased Cr content. Therefore, the lowered carbon level does not affect the allowable stresses for Alloy 690. The high cycle fatigue properties of Alloy 690 were recently investigated and found to be comparable to those of Alloy 600 [4].

ASME Boiler & Pressure Vessel Code, Section II Part D, Materials Properties, 2001.

2-3

Alloy 690 Properties and Metallurgy

2.1.2 Specification and Manufacture of Thick-Walled Alloy 690 Components Many replacement RPV head penetrations made of thick-walled Alloy 690 were manufactured prior to the 2008 publication of EPRI Guidelines for the manufacture of such components [5], which suggest limiting the allowable carbon content to between 0.015% and 0.035%. For Alloy 690 bars, plates, and heavy section tubing, however, the carbon content has routinely been specified to be in a range between 0.01% and 0.04% or narrower, instead of the max. 0.05% carbon specified in ASME Section II. Because of the potential importance of both the exact chemical composition of Alloy 690 and the manufacturing process in ensuring good PWSCC resistance (as discussed in later sections), efforts were recently started to assess the actual characteristics of replacement CRDM nozzles that have been installed in U.S. PWRs within the last few years. To date, the picture remains somewhat incomplete [6], although further progress is expected shortly. The cited report provides generic information on the various practices that a melt shop could have used in the production of Alloy 690 pipe or rod material. Information from the three main suppliers (MHI, AREVA, and BWC), while not complete, identified the sources (melt shop) and some processing parameters of the Alloy 690 material, including the annealing and thermal heat treatments. Perhaps the most significant variable identified in this exercise was that straightening of the pipe or rod after final thermal treatment appears to be a common practice at the Alloy 690 suppliers. To avoid introducing cold work, such a procedure would not be allowed under the new EPRI guidelines [5] for procurement of such Alloy 690 material. Note, also, that the currently practised final thermal treatments (typically between 5 and 15 hours at around 700C) to produce Alloy 690 TT, with extensive carbide decoration of the grain boundaries, were developed considering Alloy 600 PWSCC experience and were then adjusted a small amount to improve resistance to IGSCC for SG tubing in caustic environments. They have not yet been shown to be either necessary, or even beneficial, for obtaining PWSCC resistance in Alloy 690.

2.2

Phase Diagram of Alloy 690

The first comprehensive study of Alloy 690 was published by Sedriks et al. [7] of Inco Alloys International, Inc. in 1979. This was about the time that Alloy 690 began to be introduced for fabricating steam generator tubing in PWRs. The physical metallurgy and properties of Alloy 690 from this study and other sources are summarized below. Figure 2-1 in [8] showed the Ni-Cr-Fe phase diagram with the location of the /+ solvus line indicated from 816 to 1260C (1500 to 2300F). The lowest temperature at which the /+ solvus has been determined is 816C (1500F). The locations of Alloy 690 and 600 compositions are both well within the austenite field (). The range of melting temperatures is 1343-1377C (2450-2510F) [2] for Alloy 690 and 1354-1413C (2470-2575F) [1] for Alloy 600. The phase is a chromium-rich phase, which is very similar to the iron-chromium phase both in morphology and hardness. The phase can cause embrittlement in alloys whose compositions lie in the two-phase + field when precipitated during prolonged high temperature exposure. Precipitation of phase is a mechanism of aging embrittlement of austenitic stainless steel welds or castings containing small amounts of ferrite and of martensitic, precipitation-hardenable stainless steel 2-4

Alloy 690 Properties and Metallurgy

when exposed to elevated operating temperatures in PWRs [9]. Since Alloys 600 and 690 are both stable austenitic solid-solution alloys from room temperature to the melting temperature, precipitation of phase is not expected and has never been found in Alloys 600 and 690. K. Smith et al. evaluated the possible occurrence of an ordered Ni2Cr phase in Alloy 690 [10]. This concern was raised due to the fact that the Ni/Cr (atomic) ratio for an alloy with 70% Ni-30%Cr by weight is about 2.1 to 1, which is very close to that of Ni2Cr [11]. The brittle and hard intermetallic Ni2Cr phase is stable below 580C (1076F) and could have a similar embrittling effect as the iron-chromium intermetallic phase in austenitic alloys. However, K. Smiths research found that iron has an inhibiting effect on the formation of Ni2Cr phase and that the minimum specified iron content in Alloy 690 was high enough to inhibit any Ni2Cr formation. Later, Larsson et al. performed extensive testing for the same phenomenon including hardness tests, tensile tests, transmission electron microscopy (TEM), differential scanning calorimetry (DSC), and differential thermal analysis (DTA) on commercially produced Alloy 690TT that had been thermally aged for 3000 hours at 400 and 500C (752 and 932F), as well as on an Alloy 690 corrosion specimen exposed at 365C (689F) for 32,961 hours [12]. The study confirmed that no evidence of long range ordering of type Ni2Cr was found in the thermally aged Alloy 690. Nevertheless, reactor vendors usually specify higher minimum iron contents than the ASME Code allowed minimum in order to provide a significant margin against the formation of an ordered Ni2Cr phase in Alloy 690 and both the EPRI Guidelines for steam generator tubing [3] and the recent guidelines for pressure vessel nozzles [5] require a minimum content of 9 wt%. Some work is still in progress on defining the minimum iron content required more closely in the context of very long plant operating times [13, 14]. The microstructure of Alloy 690 is similar to Alloy 600, i.e., an austenitic matrix with the secondary phases being predominantly chromium carbides precipitated both intergranularly and intragranularly. The other minor secondary phases in Alloys 600 and 690 are titanium nitrides, titanium carbides, and carbonitrides. Because the austenitic matrix phase is stable up to the melting temperature, neither alloy is heat-treatable via phase changes (such as the austenite to martensite transformation in carbon and low-alloy steels) and cannot be hardened through secondary phase precipitation (such as through precipitation in Alloy X-750 or Alloy 718).

2.3

Carbon Solubility and Dynamic Strain Aging

The extent of intergranular and intragranular carbide precipitation depends on the thermal mechanical history and carbon content. Most research efforts on Alloys 690 and 600 have been focused on the grain boundary microstructure, especially due to observations of generally higher PWSCC resistance of Alloy 600 with a microstructure containing continuous intergranular carbides and few intragranular carbides [15]. The intergranular carbide precipitates are found to be both M7C3 and M23C6 types in Alloy 600 and mostly globular M23C6 type in Alloy 690 [16, 17]. It is generally recognized that the solubility of carbon in Alloy 690 is lower than in Alloy 600. This is due to the higher Cr content which lowers the solubility while increasing the propensity for carbide precipitation. The carbon solubility curve of Alloy 690 was investigated by Sarver et al. [18]. The Alloy 690 heats used are listed in Table 2-5 and the results are shown in Figure 2-1 with the Alloy 690 solubility line drawn to separate the specimens having no visible carbides from the specimens having visible carbides. Figure 2-1 also shows the comparison of carbon solubility curves for Alloy 690 and Alloy 600. The equations fitted to these curves are: 2-5

Alloy 690 Properties and Metallurgy

Alloy 690 F C = = 2647.5 + 120 ln(%C) 1453.1 + 66.9 ln(%C)

Equation 2-1

Alloy 600 F C = = 2640 + 234.5 ln(%C) 1449 + 130.3 ln(%C)

Equation 2-2

Table 2-5 Chemical composition (wt%) of Alloy 690 heats used by Sarver et al. [18]
Heat Ni Cr Fe C Mn Si S Ti Nb + Ta Cu P Al Mo

Intergranular Carbide Precipitation Study NX9217H NX9780H NX4459HG NX4401H Bal Bal Bal Bal 27.27 29.20 28.25 29.25 10.22 8.85 8.86 10.22 0.01 0.01 0.06 0.01 0.15 0.33 0.20 0.21 0.10 0.43 0.10 0.25 0.003 0.001 0.004 0.001 0.25 0.46 0.32 0.28 0.15 0.12 0.04 0.26 0.17 0.26 0.17 0.15

Carbon Solubility Study EXP 1 EXP 2 NX4458H EXP 3 EXP 4 NX05E1H NX10C1H EXP 5 EXP 6 EXP 7 EXP 8 Bal Bal Bal Bal Bal Bal Bal Bal Bal Bal Bal 28.7 28.8 27.9 29.9 28.7 29.9 29.8 29.0 29.1 29.4 29.5 9.2 9.8 9.8 9.6 9.3 9.5 9.4 9.1 9.1 10.3 9.8 <0.001 0.01 0.016 0.02 0.02 0.021 0.039 0.04 0.058 0.06 0.06 0.02 0.06 0.19 0.03 0.02 0.21 0.15 0.02 0.02 0.03 0.01 0.001 0.06 0.10 0.05 0.001 0.39 0.15 0.001 0.001 0.06 0.05 0.001 0.003 0.002 0.003 0.001 0.001 0.008 0.002 0.001 0.003 0.003 0.03 0.02 0.26 0.01 0.03 0.28 0.30 0.02 0.02 0.03 0.02

Heat Treatment Study NX4401H NX4588H NX2184H NX4308 (Alloy 600) Bal Bal Bal Bal 29.25 29.92 28.82 15.11 10.22 9.49 8.98 7.60 0.01 0.03 0.01 0.03 0.21 0.18 0.18 0.35 0.25 0.21 0.24 0.21 0.001 0.001 0.001 0.007 0.28 0.27 0.26 0.26 0.26 0.24 0.24 0.29 0.15 0.21 0.30 0.50

2-6

Alloy 690 Properties and Metallurgy

It should be noted that thermo-mechanical processing of both of these materials may cause grain boundaries to move by recrystallization after all the available carbon has precipitated as carbides on the old grain boundaries, thus creating a ghost grain boundary carbide network. Coring, or non-equilibrium solidification of the original ingot, can also manifest itself in the form of carbide banding. Solution heat treatments at temperatures above the carbon solubility curves for prolonged periods followed by rapid quenches can minimize the observed effects, but cannot eliminate them. Prohibitively long solution heat treatment would be required to dissolve the carbides and allow the carbon to diffuse away from the original sites in order to prevent the carbon from re-precipitating at the old sites during carbide precipitation heat treatments.
2400 2200 2000 1800 1600 1400 1200 0.001
Alloy 690 Alloy 600 No Carbides, Alloy 690 Carbides, Alloy 690

Temperature

0.010
Carbon, wt%

0.100

Figure 2-1 Carbon solubility diagram for Alloy 690 and Alloy 600 from [18]

The phenomenon of dynamic strain aging (DSA) has been extensively studied in recent years (often using measurements of so-called internal friction) because of a suspected link between the resulting localization of deformation within a material and its SCC susceptibility. Because of its lower carbon solubility than Alloy 600, Alloy 690 might have been expected to show more susceptibility to DSA, but investigations by Hnninen et al. show that this is not the case [19]. The authors consider that the higher Cr content of Alloy 690 substantially increases its creep resistance, probably by reducing the stacking fault energy of the alloy, and that these positive effects of Cr are more pronounced than the possible negative effects of reduced carbon solubility. Additionally, higher Cr content will affect the oxidation reactions controlling vacancy injection and hydrogen uptake in the material (see Section 3.1.5), which also play key roles in strain localization and subsequent crack initiation.

2-7

Alloy 690 Properties and Metallurgy

2.4

Intergranular Carbide Precipitation and Sensitization

The time-temperature chromium carbide precipitation curves for Alloy 600 and Alloy 690k were determined by Yonezawa et al. [20]. The highest precipitation rate (the nose of the C-curves) is at approximately 850-950C (1550-1750F). The Alloy 690 and Alloy 600 heats were annealed at 1100C (2012F) and contained 0.03% carbon, but the exact chemical composition was not listed. From this work, chromium carbides precipitate first at the grain boundaries followed by precipitation inside the grains. In addition, carbide precipitation is faster in Alloy 690 than in Alloy 600 due to the higher chromium content of Alloy 690. For Alloy 600, the precipitation of the grain boundary carbides is complete after a short period of time at 704C (1300F). However, an extended 10-15 hours period is required to fully replenish the Cr-depleted grain boundary with Cr from the bulk of the grains to avoid a sensitized microstructure [21, 22]. It was reasoned that since Alloy 690 has a much higher Cr content, full replenishment of the grain boundaries is not necessary and a 4 to 5 hour thermal treatment should be adequate [23]. Recent thinking on this by one major reactor component vendor, as reported in 2005, also makes a distinction between the most suitable mill-annealing temperatures (as a function of carbon content) for either cold or hot finished Alloy 690 TT [24]. Sedriks et al. [7] found that if the carbon content is kept to 0.02% or below, Alloy 690 cannot be sensitized as defined by the Huey test, i.e., the boiling nitric acid test per ASTM A 262 Practice C [25]. This means the Cr level near the grain boundary was sufficiently high in Alloy 690 containing 0.02% carbon to prevent intergranular attack by boiling nitric acid even after intergranular carbide precipitation. In the Huey test, specimens are placed in boiling 65% (by weight) nitric acid, typically for five 48-hour periods, with a fresh nitric acid solution being used for each period. It should be noted that the Huey test for sensitization is normally used for detecting susceptibility to intergranular attack in austenitic stainless steels, not for high-nickel alloys such as Alloy 690 or Alloy 600. Sensitization, as usually used, refers to grain boundary chromium depletion in austenitic stainless steels such as Type 304 and Type 316 exposed to a temperature range of 427-816C (800-1500F) for a prolonged period of time that varies with carbon content. Precipitation of chromium carbides along the grain boundaries causes Cr depletion near the grain boundaries, which are preferentially attacked under certain oxidizing environmental conditions (for example in nitric acid, or in in conjunction with stress in the oxygenated water chemistry sometimes used in Boiling Water Reactors). For Alloy 600 steam generator tubing, a modified Huey Test (25% nitric acid with two 24-hour boiling periods) is often used, instead of the 65% nitric acid with five 48-hour boiling periods specified in ASTM A 262, Practice C. Due to the higher resistance of Alloy 690 to sensitization, Sarver et al. used 65% nitric acid for sensitization testing of several heats of Alloy 690 with varying carbon contents and heat treatments [18]. The time-temperature-sensitization diagrams plotted by Sarver et al. are shown in Figure 2-2 to Figure 2-4. Figure 2-2 shows very low corrosion rates for heat NX4459HG (0.06%C) with a 1038C (1900F) anneal before the sensitization treatment. The corrosion rate was highest (1.7 mil/month) after a sensitization treatment of 538C (1000F) for 100 hours. Figure 2-3 shows the corrosion rates for heat NX9217H (0.01%C), which was mill annealed at 1038C (1900F) for one hour. The corrosion rate was also low except when sensitized at 538C (1000F) for 20-100 hours. Figure 2-4 shows the results for heat NX9780H, also containing 0.01%C, but having a slightly higher Cr content than NX9217H (29.20% vs. 27.27%). Heat
k

See Figure 2-3 in [8].

2-8

Alloy 690 Properties and Metallurgy

NX9780H also showed a very low corrosion rate, even when sensitized at 538C (1000F) for 10-20 hours. The highest degree of sensitization for Alloy 690, as defined by the Huey test, was produced by heat treatment at 538C (1000F) for 20 to 100 hours. The complete chemical compositions of heats NX4459HG, NX9217H, and NX9780H studied by Sarver et al. are summarized in Table 2-5.

Corrosion rate in 0.001inch/month

2000 1800
Sensitization Temp, o F

1600 1400 1200 1000 800 0.01

0.7

0.5

0.8

0.7

0.6

0.9

0.6

0.5

0.6

0.5

0.7

0.6

0.3

0.5

0.7

0.6

0.6

0.9

1.7

0.4

0.1

10

100

1000

Sensitiz ation Time, hour

Figure 2-2 Time-Temperature-Sensitization diagram by modified Huey Test, Alloy 690 Heat NX4459HG (0.06%C) from [18]

2-9

Alloy 690 Properties and Metallurgy

Corrosion rate in 0.001inch/month

2000
1.2

1800
Sensitization Temp, o F

0.5

0.4

0.7

1.9 0.7

1600 1400 1200 1000 800 0.01

0.4

0.5

0.4

0.4

0.5

0.4 0.4

0.5 0.3 0.6 0.4 0.7

0.6

0.3

1.1 0.3

0.7 0.5 9.1 28.5 78.3

0.4

0.3

0.1

10

100

1000

Sensitization Time, hour

Figure 2-3 Time-Temperature-Sensitization diagram by modified Huey Test, Alloy 690 Heat NX9217H (0.01%C) from [18]
Corrosion rate in 0.001inch/month

2000 1800
Sensitiz ation Temp, o F

1600 1400 1200

0.5

1000 800 0.01

0.4

0.6 0.5

0.1

10

100

1000

Sensitization Time, hour

Figure 2-4 Time-Temperature-Sensitization diagram by modified Huey Test, Alloy 690 Heat NX9780H (0.01%C) from [18]

2-10

Alloy 690 Properties and Metallurgy

It should be noted that, for Alloy 600, a sensitized microstructure with continuous intergranular carbides (and associated grain boundary Cr depletion) is more resistant to PWSCC than the same heat without intergranular carbides, despite the grain boundary Cr depletionl. Sarver et al. also investigated the effect of carbon content, annealing temperature, and one-hour precipitation heat treatments on the carbide morphology. The results are shown in Table 2-6. Both the annealing temperature and carbon level affected the temperature at which heavy intergranular carbide precipitation occurred. Depending upon the desired annealing temperature and carbon content, a one-hour heat treatment could be chosen to produce heavy intergranular carbide precipitation. The high carbon (0.03%C) material was much more likely to show intragranular carbides than the lower carbon (0.01%C) material, although intergranular precipitation occurred equally in both materials. Although heavy intergranular carbide precipitation was produced by many heat treatments, no significant intergranular sensitization occurred and corrosion rates in the Huey tests were low (less than 0.4 mm/year) except in one case: the 0.03% carbon material annealed at 1149C (2100F) and heat treated at 649 and 704C (1200 and 1300F) showed rates of 1.0 and 1.5 mm/year, respectively, indicating that intergranular sensitization had occurred.
Table 2-6 Effect of heat treatment on Alloy 690 carbide precipitation [18] 0.01%C NX4401H Initial Anneal Temp, 20 Minutes 1800F Carbide Precipitation Temp, 1 hour 1200F 1300F 1400F 1500F VL VL H VL 1900F VL L H H 2000F L H Intra H H 2100F VL L H H 0.03%C NX4588H Initial Anneal Temp, 20 Minutes 1800F VL L L Intra L Intra 1900F H H L L Intra 2000F H Intra H Intra H Intra H Intra 2100F L L H Intra H

(a)

Heat treatment code: VL very light intergranular carbides; L light intergranular carbides. H heavy intergranular carbides; Intra intragranular carbides.

For Alloy 600, the precipitation of intergranular carbides usually implies Cr-depletion and a sensitized microstructure, because of the limited Cr content of the alloy. This makes the alloy susceptible to intergranular attack in standard tests (such as the Huey test) and should also be avoided as the material condition in an oxidising HT-environment, such as a BWR on NWC. It is not relevant under low-potential conditions in PWR primary water, however, where the benefit of the carbides (e.g., in reducing grain boundary creep) dominates. This indicates that the mechanism of PWSCC is somewhat different from the slip dissolution/oxidation mechanism generally accepted to apply to SCC in oxidising BWR water.

2-11

Alloy 690 Properties and Metallurgy

2.5

Effect of Elevated Temperature Exposure

Figure 2-5 shows the results of short-term tensile tests performed on annealed Alloy 690 at temperatures ranging from room temperature up to 982C (1800F) [2]. The curves represent average values for both cold- and hot-worked products in the annealed condition and show that annealed Alloy 690 retains more than 90% of its room-temperature tensile properties (yield and tensile strengths, as well as elongation) up to 427C (800F). Only at temperatures over 540C (1000F) does the tensile strength start to decline substantially. This retention of room temperature tensile properties, together with the long-term, high-temperature stability discussed below, are reflected in the ASME design stress intensity values for Alloy 690 and Alloy 600 (see Table 2-4).

Figure 2-5 High-temperature tensile properties of annealed Alloy 690. Data shown are a composite of cold-and hot-worked products in the annealed condition and taken from [2]

Reference [2] also lists room-temperature tensile results for annealed Alloy 690 after exposure to elevated temperatures for various periods of time. The long-term, high-temperature stability of Alloy 690 was demonstrated by Charpy impact testing after long periods at 566 to 760C (1050 to 1400F). This range of temperatures is similar to that of post-weld, stress-relieving heat treatment (PWHT) temperatures for carbon and low alloy steel vessels or piping in PWRs. This work showed that the room-temperature Charpy impact energy (189 J un-aged) was virtually unchanged after 12,000 hours or 500 days at elevated temperatures. This property is known to be a very sensitive indicator for the precipitation of or phases in alloys with 2-12

Alloy 690 Properties and Metallurgy

an austenitic matrix. For example, aging at 400C (752F) for 10,000 hours (417 days) reduced the Charpy impact energy from ~220 J to 50 J for a CF-8 cast austenitic stainless steel (the equivalent of wrought Type 304 stainless steel) containing 24% ferrite [26]. Possible long-term aging of Alloy 690 is being investigated at EDF in France and a new paper [27] shows that it is possible after 60000 h at 420C for a material with 7.2% Fe, but that industrial products are expected to be free from aging. Hence, it can be concluded that thermal aging embrittlement is not a concern for Alloy 690 and Alloy 600 components from exposure to high temperatures either during fabrication (including repair welding or PWHT heat treatment), or from long-term exposure at PWR operating temperatures of up to 343C (650F).

2.6

References

1. INCONEL Alloy 600, Publication Number SMC-027, Special Metals Corporation, 2002 (Sept. 02). 2. INCONEL Alloy 690, Publication Number SMC-079, Special Metals Corporation, 2002 (Sept. 02). 3. Guidelines for PWR Steam Generator Tubing Specification and Repair, Vol. 2, Rev. 1: Guidelines for Procurement of Alloy 690 Steam Generator Tubing. Final Report, April 1999. EPRI Report TR-016743-V2R1. 4. G. Chai, J. Frodigh and H. Trnblom, Fatigue Behavior of Alloy 690 and Alloy 800 SG Tubing at Temperatures up to 330oC in Very High Cycle Fatigue Regime, Proceedings of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, August 1923, 2007, Published by the Canadian Nuclear Society. 5. A. Mcllree and G. Ilevbare, Materials Reliability Program: Guidelines for Thermally Treated Alloy 690 Pressure Vessel Nozzles (MRP-241). EPRI, Palo Alto, CA: 2008. 1015007. 6. G. Theus et al., Materials Reliability Program: Material Production and Component Fabrication and Installation Practices for Alloy 690 Replacement Components in Pressurized Water Reactor Plants (MRP-245). EPRI, Palo Alto, CA: 2008. 1016608. 7. A.J. Sedricks, J.W. Schultz, and M.A. Cordovi, Inconel Alloy 690 A New Corrosion Resistant Material, Corrosion Engineering (Boshoku Gijutsu), vol. 28, pp. 82-95, 1979, Japan Society of Corrosion Engineering. 8. H. Xu et al., Materials Reliability Program (MRP), Resistance to Primary Water Stress Corrosion Cracking of Alloys 690, 52, and 152 in Pressurized Water Reactors (MRP-111). EPRI, Palo Alto, CA: 2004. 1009801. 9. H. Xu and S. Fyfitch, Aging Embrittlement Modeling of Type 17-4PH at LWR Temperatures, the 10th International Conference on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, NACE International, Houston, Texas (2001).

2-13

Alloy 690 Properties and Metallurgy

10. K. Smith, A. Klein, P. Saint-Paul, J. Blanchet, Inconel 690, A Material with Improved Corrosion Resistance for PWR Steam Generator Tubes, Proceedings of 2nd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1985 pp. 319-328. 11. A. Marucco, Atomic ordering in the Ni-Cr-Fe system, Materials Science & Engineering, Vol A189 (1994) pp 267-276. 12. T. Larsson, J. O. Nilsson, and J. Frodigh, On the Possibility of Forming Ordered Ni2Cr in Alloy 690, Proceedings of 9th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Newport Beach, CA, 1999, pp. 143 to 147. 13. W. Lunceford et al., EPRI Materials Degradation Matrix, Revision 1. EPRI, Palo Alto, CA: 2008. 1016486. 14. W. Lunceford et al., Materials Reliability Program: Pressurized Water Reactor Issue Management Tables Revision 1 (MRP-205). EPRI, Palo Alto, CA: 2008. 1018400. 15. S.M. Bruemmer and C.H. Henager, Microstructure, Microchemistry, and Microdeformation of Alloy 600 Tubing, Proceedings of 2nd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1985 pp. 293-300. 16. K. Norring, K. Stiller, and J. Nilsson, Grain Boundary Microstructure, Chemistry, and IGSCC in Alloy 600 and Alloy 690, Proceedings of 5th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1991. 17. K. Stiller, J. Nilsson, and K. Norring, Structure, Chemistry and Stress Corrosion Cracking of Grain Boundaries in Alloys 600 and 690, Metallurgical and Materials Transactions, Vol. 27A-No.2, February, 1996. 18. J.M. Sarver, J.R. Crum, and W.L. Mankins, Carbide Precipitation and the Effect of Thermal Treatments on the SCC Behavior of Inconel Alloy 690, Proceedings of 3rd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Traverse City, MI, 1987, pp. 581-586. 19. H. Hnninen et al, Dynamic Strain Aging Of Ni-Base Alloys Inconel 600 and 690, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp. 1423 to 1430. 20. T. Yonezawa et al., Effect of Heat Treatment on Corrosion Resistance of Alloy 690, Proceedings of 2nd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1985 pp. 593-600. 21. G.P. Airey, Effect of Processing Variables on the Caustic Stress Corrosion Resistance of Inconel Alloy 600, Corrosion, vol. 36(1), 1980, pp. 9-17. 22. J.R. Crum, Effect of Composition and Heat Treatment on Stress Corrosion Cracking of Alloy 600 Steam Generator Tubes in Sodium Hydroxide, Corrosion, vol. 36(1), 1982, pp. 40-45. 23. Proceedings: Workshop on Thermally Treated Alloy 690 Tubes for Nuclear Steam Generators. July 1986. EPRI report NP-4665M-SR. 2-14

Alloy 690 Properties and Metallurgy

24. T. Yonezawa et al., Materials Reliability Program: Proceedings of the 2005 International PWSCC of Alloy 600 Conference and Exhibit Show (MRP-154). EPRI, Palo Alto, CA, AREVA, Lynchburg, VA, Westinghouse, Madison, PA, Structural Integrity Associates, Inc., Centennial, CO, and Welding Services, Inc., Norcross, GA: 2005. 1012089. 25. ASTM Standard A 262, Standard Practices for Detecting Susceptibility to Intergranular Attack in Austenitic Stainless Steels. 26. O.K. Chopra and H.M. Chung, Aging Degradation of Cast Stainless Steels: Effects on Mechanical Properties, Proceedings of the 3rd Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, pp. 737-748, September 1987, Traverse City, Michigan. 27. F. Delabrouille et al., Long range ordering in Ni Alloys containing 30% Cr, 14th Int Symp on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Virginia Beach (VA), August 2009 (to be published by ANS).

2-15

3
CORROSION BEHAVIOR OF ALLOY 690 APART FROM PWSCC

3.1

General Corrosion Tests in Primary Water

This section gives some examples of the available laboratory test data pertinent to general corrosion of Alloy 690 in PWR primary water. General corrosion can be defined as uniform deterioration of a metal surface by chemical or electrochemical reaction with the environment. Nickel-base Alloy 600 and Alloy 690 are essentially immune to damage through general corrosion in PWR environments due to the formation of an adherent Cr-rich oxide on the surface. The very low general corrosion rate of Alloy 690 in flowing primary water is nevertheless of particular interest for PWR steam generator tubing materials. Material lost to the water can become radioactive by coming in contact with the reactor core and then redeposit on the surfaces of the primary loop. Hence, the metal release rate has a strong implication for a plants ALARA practice to keep plant shutdown dose rates to a level as low as reasonably achievable. 3.1.1 SG Tubing by Sedricks et al. 1979 Sedriks et al. [1] evaluated Alloy 690 general corrosion in two simulated PWR water environments using a high velocity test loopm. Borated water was added to simulate the primary side of the steam generator tubing and ammoniated water was used to adjust the pH value to simulate a secondary water environment. The Alloy 690 test specimens and Alloy 600 and Alloy 800 control specimens were heat treated for hour at 980C (1800F), followed by air cooling, and the surfaces were prepared by grinding on a wet 120 grit silicon carbide belt to a 0.75 m finish. The specimens were weighed and then exposed to the borated water for 2,250 hours and to the ammoniated water for 1,000 hours at a flow velocity of 5.5 m/sec (18 ft/sec). After exposure, the specimens from the ammoniated water tests were descaled by cathodic charging in 5% H2SO4 inhibited with quinoline ethiodide. The specimens from the borated water tests were descaled by the alkaline permanganate acid method. The terminology and reporting method followed NACE standard TM-02-74, covering the following terms: 1. Descaled metal loss (metal consumed): the difference between initial weight and weight after removal of adherent corrosion film. 2. Corrosion film weight (adherent corrosion film): the difference between the weight before descaling and the weight after descaling.

See MRP-111 for further details.

3-1

Corrosion Behavior of Alloy 690 Apart from PWSCC

The difference between the descaled metal loss and the corrosion film weight represents the base metal lost to the flowing water. Alloys showing the least loss of material to the water would be expected to produce the lowest activity levels in the PWR primary coolant system. For the three alloys tested in the borated water, Alloy 600, Alloy 690, and Alloy 800, the general corrosion rate decreased with increasing Cr content. Alloy 690, having the highest Cr content of the three, lost the least amount of material to the high velocity simulated PWR water. The observed corrosion films formed on Alloy 690 after the borated and ammoniated exposures were of the thin tarnish type and appeared to be extremely adherent to the base metal. The study reported a standard experimental error of about 25% and suggested that the data should be used as a guide rather than a precise corrosion rate measurement in a high temperature PWR water environment. 3.1.2 SG Tubing by K. Smith et al. 1985 K. Smith et al. [2] investigated the corrosion rate and metal release rate of Alloy 690 and Alloy 600 SG tubing in flowing simulated PWR primary watern. The Alloy 690 specimens were fabricated from full length SG tubes (also called pre-series tubes) that were produced by an established industrial route and were intentionally chosen to span the limits of the specification requirements for C and Cr content. The Alloy 690 specimens for the general corrosion tests were in the mill annealed condition (MA) as no difference between the MA and the thermally treated (TT) condition was expected. Weight changes before and after chemical de-filming were used to calculate the amount of total corrosion from the SG tube I.D. and the amount released to the water. The test results showed that Alloy 690s general corrosion rate and metal release rate were reduced by a factor of 2 to 4 compared with Alloy 600 material. 3.1.3 SG Tubing by Yonezawa et al. 1985 Yonezawa et al. [3] investigated the weight loss of Alloy 690, Alloy 600, and Alloy 800L SG tubing n in simulated PWR primary water chemistry . The test coupons were made from split SG tubes and the test was conducted at 360C (680F) for up to 4,000 hours. Even though the data were scattered and the test duration was limited, the results showed that the general corrosion behavior of Alloy 690 and Alloy 600 in primary water is about the same. 3.1.4 Esposito et al. 1991 Esposito et al. [4] investigated the corrosion rate and metal release rate of Alloy 690 and Alloy 600 in a simulated PWR primary water environment with and without Zn addition (Zn was used in other n tests to assess its effectiveness in mitigating Alloy 600 PWSCC initiation) at 330C (626F) . Test coupons were fabricated from Type 304, Type 316, Alloy 600MA, Alloy 600TT, Alloy 690TT, Alloy 750, and Stellite, suspended in the autoclave and exposed for times up to 2,500 hours with or without adding zinc borate. Unlike in the general corrosion rate studies for SG tubing, the test coupons in this study were not subjected to any significant rate of coolant flow. At the conclusion of the test, some samples were descaled to a constant weight by cathodically stripping the oxide film.
n

See MRP-111 for further details.

3-2

Corrosion Behavior of Alloy 690 Apart from PWSCC

The metal corrosion and corrosion release were calculated from the coupon weight change measurements. Total metal corroded = W0 Wd Oxide film weight = Wa Wd Metal release = (W0 Wd) 70%*(Wa Wd) Where: W0 = original sample weight Wa = weight after coupon exposure Wd = weight after descaling 70% is the assumption that metals in the oxide corrosion film represent 70% of the total film weight. The above values were then divided by the specimen area to obtain the corrosion release per unit area. The test results clearly show that the Alloy 690TT has the lowest general corrosion rate among all the materials tested in PWR primary water, both with and without zinc addition. The results also confirm that the general corrosion rate of Fe-Cr-Ni alloys in primary water environment decreases with increasing Cr content. In addition, the test results also indicated that zinc addition could significantly reduce the general corrosion rate of most commonly used materials in the PWR primary coolant system. 3.1.5 Alloy Oxidation Studies related to the Mechanism of PWSCC F. Scenini et al. [5] have tested the intergranular oxidation model of PWSCC by exposing several Ni-based alloys to water and steam environments containing hydrogen, at temperatures up to 480C, and analyzing the resultant oxidized surfaces. If the surfaces were mechanically prepared without subsequent electropolishing, short-circuit outward diffusion of Cr occurred, and the differences between alloys or alloy conditions (600, 690, 600TT) were minimal. Only on deformation-free surfaces were the differences in oxidation response of the materials displayed clearly. For hydrogen partial pressures greater than that corresponding to the Ni/NiO equilibrium, Alloy 600 showed intergranular and internal oxidation with the growth of nodules of pure Ni within the grains. The oxidation morphology was strongly dependent on grain orientation. Nodule-free zones appeared near grain boundaries, which were also enriched in metallic Ni. Thermal treatment to precipitate grain boundary carbides eliminated or strongly modified the intergranular oxidation and the presence of nodules. No signs of internal oxidation were observed on Alloy 690, which underwent some thermal etching but formed a nearly continuous external oxide. Extensive studies of the oxidation behavior of Alloys 600 and 690 in PWR primary water carried out in France have also identified important differences that support the expectation of better PWSCC resistance of Alloy 690 [6]. This includes ingenious work on the way in which applied stress acts to increase oxide thickness and decrease the chromium content of the oxide scale [7], as well as attempts to relate the oxides developed within SCC cracks to that found at the free surface of Ni-base alloys exposed to primary water [8].

3-3

Corrosion Behavior of Alloy 690 Apart from PWSCC

3.2

Corrosion Fatigue Tests in Primary Water

Psaila-Dombrowski et al. [9] performed S-N fatigue tests for Alloy 690 plate and Alloys 82, 182, and 152 weld metals in air and in simulated primary water at 315C (600F)o. A number of weld build-ups and a 1 m long composite plate were manufactured in a manner consistent with SG fabrication practices in which Alloy 52 is used to attach the Alloy 690 plate to the Alloy 82 buildup. The mock-up composite plate began as a 60 mm (2.36) thick A508 carbon steel plate. One edge of the A508 plate was overlaid with a 5 mm thickness of Alloy 82 and then stress relieved. The test block was completed by making a modified double V-groove weld between the 5.5 mm thick Alloy 690 plate and the Alloy 82 overlay using Alloy 152. The plate was not stress relieved following the welding. Modified tensile specimens was used to conduct the low cycle fatigue tests oriented such that the specimen axis was always perpendicular to the direction of welding. In all cases, the gauge sections of the specimens contained only the test material of interest. The tests were conducted in a 76 liter autoclave using a load frame with high lateral stiffness to maintain alignment of the specimen during testing. Most fatigue tests were completed using a strain range of 0.008 and a zero mean strain. However, several tests were performed at different strain levels to expand the data base. With one exception, all tests in the simulated primary water were performed at a strain rate of 0.001 -1 sec . Higher strain rates were used in the air tests because no effect of frequency was expected in an air environment. The fatigue test results showed that the Alloy 690 fatigue life in the primary water (one data point only) decreased by about 40% compared to the fatigue life in air. However, it was still well above the ASME design curve. The fatigue properties of Alloy 690 in the primary water and in air were in good agreement with predictions made using the following model for Alloy 600 at 150 to 350C (302 to 662F) [10]. In air, In water, ln (N25) = 6.94 1.776 ln (a 0.12) + 0.498 ln (N25) = 6.94 1.776 ln (a 0.12) + 0.498 0.401
Equation 3-1 Equation 3-2

Where, N25 = the fatigue life defined as the number of cycles for the peak tensile stress to drop 25% from its initial value a = the applied strain amplitude in % This work also showed that the Alloy 82 weld metal exhibited the highest fatigue life among the four materials tested in air. In primary water, all four materials exhibited roughly similar fatigue lives, which were all well above the ASME design curve. Thus, it appeared from this early study that primary water has a measurable influence on Alloy 690 fatigue life at the strain range and rates used, but more data under other test conditions were clearly required to provide a better characterization. Unfortunately, additional S-N studies of Alloy 690 corrosion fatigue using standard tensile specimens appear not to have been carried out, with the exception of work in Japan that has not yet been fully published. In a 2006 paper [11], Higuchi et al. present a number of diagrams that include data for both Alloy 600 and 690 (as well as their weld alloys) in air (Figure 3-1) and simulated PWR primary water (Figure 3-2).
o

See MRP-111 for further details.

3-4

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-1 Japanese fatigue data for Ni-base alloys in air at room temperature from [11]

Figure 3-2 Japanese fatigue data for Ni-base alloys in simulated PWR water at 325C from [11]

3-5

Corrosion Behavior of Alloy 690 Apart from PWSCC

The effects of PWR primary water on fatigue life were slightly less for Alloy 690 (and Alloy 152 weld metal) than for Alloys 600/132 and essentially the same for base and weld metals. Considering that Ni-base alloys basically have a low sensitivity to environmental fatigue, it was decided in Japan to obtain the trend line for dependence upon strain rate (Figure 3-3) without any distinction between the 600 and 690 alloys. Modeling to calculate the factor of environmental fatigue enhancement (Fen) was based upon the approach previously used for austenitic stainless steels (which show much larger effects of high-temperature PWR or BWR environments than Alloys 690 and 600 see Figure 3-4). The comparison between predicted and experimentally determined environmental effects shown in Figure 3-5 for Alloy 690 exposed to PWR primary water appears to be very 3 5 satisfactory over cyclic fatigue lives between ~ 10 and 10 cycles.

Figure 3-3 Japanese data from [11] on relationship between calculated factor of environmental fatigue enhancement (Fen) and strain rate for Ni alloys in simulated PWR water at 325C

3-6

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-4 Japanese data from [11] on relationship between calculated factor of environmental fatigue enhancement (Fen) and temperature for various materials in simulated LWR water at 325C

3-7

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-5 Japanese data from [11] showing results of model predictions for the corrosion fatigue behavior of Ni-base alloys in simulated PWR water at 325C compared with experiments

Etien et al. [12] have also shown corrosion fatigue data in high-temperature, deaerated water for high-chromium weld alloys and Alloy 690 relating to the number of cycles at 0.2 Hz and K = 31 MPam (28 ksiin) required to initiate an approximately 0.5 mm (20 mil) deep crack in a notched 1TCT specimen (root radius = ~0.13 mm) exposed to hydrogenated water at 338C. Figure 3-6 compares these results with baseline data for 308L stainless steel weld metal, showing improvement factors of 2 to 4 times. It is also interesting, however, to compare the extent to which the presence of HT-water reduces the number of cycles to fatigue crack initiation under these loading conditions when compared with the values measured for the same alloy in HT-air. The three high-chromium weld metals appear to exhibit reduction factors of between 2.2 and 3.3, which are somewhat smaller than those measured for Alloy 690 (~ 3.9) or for 308L stainless steel (~ 4.5).

3-8

Corrosion Behavior of Alloy 690 Apart from PWSCC

100000

Cycles to 20-mil Crack Extension

Air: filled symbols Water: open symbols

10000

1000

2x

3x

2.5 x

4x

100

MLTS24Cr

MLTS27Cr

MLTS30Cr

A690 Bar

308L Stainless

Figure 3-6 Corrosion fatigue initiation data from [12] in high-temperature, deaerated water for high chromium weld alloys and Alloy 690 compared with stainless steels

A different aspect of corrosion fatigue relates to the degree of environmental enhancement measured during cyclic crack growth. This has been studied extensively over many years at Argonne National Laboratory [13] (see Figure 3-7) and results for Alloy 690 have recently been reported by Alexandreanu [14] in the context of ongoing studies of PWSCC behavior (described later). For two heats of Alloy 690 plate material in the as-received condition (with a final thermal treatment at ~ 720C) there was no sign of any environmental enhancement during cyclic loading, -7 even as the crack growth decreased (Figure 3-8)p. Instead, at growth rates < ~10 mm/s, signs of a K threshold effect became apparent. In contrast, a different plate material, tested under loading conditions favorable to EACq, showed slight environmental enhancement even in the as-received condition and entirely different behavior after being subjected to ~26% cold work (introduced by unidirectional cold rolling in three passes). Figure 3-9 shows that, in this case, significant -7 environmental acceleration of cyclic crack growth rates was observed starting at values > ~10 -8 mm/s and continuing down to well below 10 mm/s. The nature of the changes that can be induced in thick-walled Alloy 690 by extensive, non-uniform cold working, and the extent to which they could be relevant to actual reactor component behavior, are discussed in more detail in later sections of this report dealing with PWSCC testing of such materials.

Note, however, that these results were obtained using fast rise times and, in some cases, high R values (i.e. under loading conditions where significant environmental enhancement of cyclic CGR would not be expected). Sawtooth waveform with slow rise time and less extreme load ratios.

3-9

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-7 Approach to the analysis of environmental effects on cyclic crack growth developed at ANL and now applied to PWSCC testing of Alloy 690 and its weld alloys (from [13])

Figure 3-8 No environmental enhancement of cyclic crack growth seen for thermally treated Alloy 690 material in either simulated PWR primary water or de-aerated pure water at 320C [14], but under loading conditions not expected to favor EAC

3-10

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-9 Appearance of significant environmental enhancement at lower rates of cyclic crack growth for unidirectionally cold-rolled Alloy 690 material (blue points) in simulated PWR primary water at 320C (from [14])

The ANL results on cyclic crack growth rate were obtained largely as a by-product of their experimental technique for SCC studies, but dedicated testing for such effects in simulated primary water has now also been reported from both Sweden [15] and Japan [16] over a wider range of test conditions. The intent of the latter work was to propose draft fatigue crack growth rate curves for Ni-base alloys to the Japanese standards organization JSME. Figure 3-10 shows that a simulated PWR primary environment can enhance the fatigue crack growth rate of Ni-base alloys by some 5 to 10 times over a fairly wide range of test conditions, with the effect being slightly more pronounced for Alloy 600 than for Alloy 690. These experimental data breach the cyclic CGR curves that would be calculated from existing ASME rules (see Figure 3-11), but are described in a very conservative manner by the proposed Japanese disposition lines from the newly developed model. In contrast, the Swedish study reported in reference [15], which was carried out on archive CRDM material from the replacement head for the Ringhals 2 plant, indicated that there was no significant influence of the environment on da/dN in comparison with the cyclic CGRs measured in air either for Alloy 600 or for 690 (Figure 3-12). This work showed a threshold K value of ~ 4 MPam for Alloy 690 in the primary water environment (Figure 3-13).

3-11

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-10 Japanese data from [16] indicating that a simulated PWR primary environment can enhance the fatigue crack growth rate of Alloys 600 and 690 by 5 to 10 times over a range of test conditions

3-12

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-11 Comparison of lines from a proposed (very conservative) Japanese model [16] for cyclic crack growth of Ni-base alloys in a PWR environment and experimental data that breach existing ASME curves

3-13

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-12 Cyclic CGR behavior of Alloys 600 and 690 in Swedish studies [15]

Figure 3-13 Cyclic CGR threshold determined in simulated primary water for Alloy 690 in Swedish studies [15]

3-14

Corrosion Behavior of Alloy 690 Apart from PWSCC

3.3

Corrosion Behavior in Secondary Water

The original MRP-111 report included, in an appendix, a considerable amount of information on the corrosion behavior (in particular stress corrosion cracking) of Alloy 690 in both nominal and faulted secondary water chemistry. The research on this has been extensive, given the early use of Alloy 690 as a steam generator (SG) tubing material, but is not very helpful to understanding and predicting PWSCC behavior, which is the main goal of this report. The reason for this is the very different water chemistry that exists on the SG secondary side, not only with regard to chemical composition (absence of lithium, boron and hydrogen additions, presence of volatile alkalizing agents), but also with regard to the much higher electrochemical corrosion potential that results at the Alloy 690 secondary-side surface compared with the uniformly low values established on the primary side by deliberate addition of gaseous hydrogen. Nevertheless, the main conclusions of Appendix C from the 2004 MRP-111 report are given below for the sake of completeness: 1. In an AVT water environment, laboratory tests have demonstrated that Alloy 690 (MA or TT) is very highly resistant to IGSCC. 2. In deaerated caustic solutions, various laboratory tests consistently showed that Alloy 690TT material usually has significantly higher IGA/SCC resistance than Alloy 600MA and Alloy 600TT materials. Alloy 690TT may have a slightly lower SCC resistance in highly caustic water (50% NaOH) at 315C (600F). The observed improvement for Alloy 690 is derived from longer crack initiation times and lower crack propagation rates compared to the Alloy 600 materials. However, these tests also showed that Alloy 690TT material can undergo IGA/SCC in deaerated caustic environments. 3. In neutral or acidic water, or AVT water doped with lead, Alloy 690TT material is more resistant to IGA/SCC than Alloy 600TT material, which is more resistant than Alloy 600MA material. However, all of these materials are considered by some investigators to be susceptible to cracking in neutral or acidic environments containing lead. 4. In caustic water doped with PbO , Alloy 690 TT material has sometimes shown a somewhat lower resistance than Alloy 600MA and Alloy 600TT, but all three materials are susceptible to IGA/SCC. 5. In chloride-containing solutions in the absence of oxygen, laboratory test data confirm that both Alloy 600 and Alloy 690 materials possess very high SCC resistance. However, they can be susceptible to SCC in highly acidified chloride solutions, or when both oxygen and chloride are present. Alloy 690 material has a higher SCC resistance than Alloy 600 material under such extreme conditions. 6. In acidic solutions containing sulfate and chloride, laboratory tests show that Alloy 690TT material has a much higher IGSCC resistance than Alloy 600TT&MA materials. 7. In deaerated neutral, or slightly caustic, sulfate solutions, laboratory tests indicate that Alloy 690TT is highly resistant to IGA/IGSCC, even when copper or copper oxides are present (see, e.g., [17]).
r

Such caustic environments are considered very unlikely actually to form in operating steam generators.

3-15

Corrosion Behavior of Alloy 690 Apart from PWSCC

More recent work in this area has focused, in particular, on the role of lead contamination in impairing the SCC resistance of all Ni-base alloys [18, 19, 20, 21], whereby Alloy 690 is generally considered to be more resistant than Alloy 600. EPRI results [22, 23, 24] indicate that Alloy 690TT is only susceptible to cracking in environments with lead that are alkaline; in those environments where cracking was observed, Alloy 690TT was much less susceptible than Alloy 600MA or 600TT. Although there are numerous potential sources for low-level contamination of PWR secondary water by lead [25], the general opinion is that primary water will not be affected by such issues, although some slight uncertainty concerning possible effects on PWSCC remains (see Section C.3.5 of reference [26]). Another area that continues to be studied extensively on the secondary side is the effect of reduced sulfur species on outer-diameter SCC of SG tubing (see Section 2.3.4.1 of [27] and reference [28]). In this case, however, there are also two confirmed plant incidents of sensitized Alloy 600 SCC having occurred as a result of such species being generated on the primary side following resin intrusions (TMI and Zorita). One of these (1994 incident in Spain) in fact led to PWSCC of 20 out of 37 CRDM penetrations. In part as a consequence of this, the current EPRI Primary Water Chemistry Guidelines limit permissible sulfate concentrations in the reactor water (see Sections 2.3.2.10 and 3.6.1 of [29]). Results from a recent destructive examination of a cracked Alloy 600 CRDM penetration from the Davis Besse reactor also implicated reduced sulfur species in the Alloy 600 PWSCC that was observed [30]. Recent Japanese work to determine the CGR of Alloy 600 SG tubing in severely faulted secondary water (both highly acidic and highly alkaline) includes a limited comparison with the behavior of 2% cold-worked Alloy 690TT material [31]. Even under very aggressive conditions that led to rapid crack propagation in Alloy 600, the 690TT tubing material failed to show SCC. A similar result was reported by de Bouvier et al. [32] in so-called complex secondary side environments. In terms of the implications of secondary side studies for the relative factors of improvement in SCC behavior for Alloy 690 over Alloy 600 in various environments, reference is made to the 2003 EPRI prediction methodology described in reference [33] and a more recent updates of this [34] that comes to the conclusions shown in Table 3-1. Overall, the following conclusions reached in the 2004 MRP-111 report remain valid. In summary, even though it cannot be considered immune to IGA/SCC, Alloy 690TT material has nevertheless demonstrated far superior IGA/SCC resistance compared to Alloy 600MA or Alloy 600TT materials under most conditions pertinent to faulted secondary water. Lead-doped caustic water is an exception here. Furthermore, Alloy 690 appears to be just as resistant to IGSCC in AVT water as it is to PWSCC in primary water.

Note that a further update of references [33] and [34] is expected to be published by the end of 2009.

3-16

Corrosion Behavior of Alloy 690 Apart from PWSCC Table 3-1 Estimated typical improvement factors vs. pHT considering all environments [33, 34]

3.4

Low Temperature Crack Propagation (LTCP)

An emerging issue of concern with PWRs is a possible reduction in the fracture resistance of structural materials, particularly nickel alloys, during (over)loading in water, rather than air. There is growing evidence that the fracture resistance of some structural materials can be degraded significantly by stressing in water at low temperatures, especially after prior, long-term exposure to high-temperature water. Some data exist to characterize elements of the problem, but the broad nature and extent of potential concerns as a function of loading, temperature, material/microstructure, and environment are not yet well defined. At present, there is consensus that a hydrogen mechanism is involved, but insufficient information to quantitatively predict the effects [35]. 3.4.1 Origins of the Phenomenon J-R tests measure fracture resistance (both fracture toughness and tearing modulus), and - starting with classic work at the Bettis laboratories by Brown and Mills (see [36]) - various investigators have observed large reductions in these properties for precracked specimens of certain nickel alloys (see Table 3-2) when the tests were performed at specific strain rates in low-temperature (< 150C) water. In these situations hydrogen was present either as an added constituent to the water, as a result of corrosion at the crack tip, or sometimes as a result of prior exposure to a high-temperature hydrogenated environment. These particular results are important in assessments of the integrity of nickel-base alloys operating in PWR primary circuits, since the extent of the decrease in fracture resistance depends on the alloy, and on environmental and dynamic loading conditions that are 3-17

Corrosion Behavior of Alloy 690 Apart from PWSCC

relevant to reactor shut down. The phenomenon is usually referred to as Low Temperature Crack Propagation (LTCP) and it is not yet clear whether it represents a genuine reduction in fracture properties of the material, or rather a form of rapid, subcritical crack growth due to the environment [37]. Both aspects may be involved. It is important to recognize that accelerated cracking at low temperature can also occur in some highly susceptible materials at constant load or K, not just under J-R tearing conditions. Some key general aspects of LTCP behavior are: Fracture resistance increases with increasing water temperature. Above 149C, LTCP is not an issue. Cracking resistance is recovered at stress intensity loading rates above 1000 MPam/h. (Insufficient time to embrittle grain boundaries ahead of crack.) LTCP does not initiate at as-machined notches, but can initiate at sharp weld defectst. Decreasing the hydrogen content in LT water produces a large increase in fracture resistance for many susceptible materials.
Table 3-2 Chemical composition of Alloys 600, 690, 82, & 52 tested by Brown and Mills [36]
Heat
EN82, heat 1 EN82H, heat 2 EN82H, heat 3 EN82H, heat 4 EN82H, heat 5 EN82H, heat 6 EN52 Alloy 600, plate Alloy 690, bar (a)

Ni
73.2 76.3 72.9 73.6 73.7 72.8 60.4 75.4

Cr
18.7 19.93 20.00 19.75 19.54 19.8 28.97 15.54

Fe
1.5 0.68 1.10 0.73 0.77 1.31 8.98 7.76

C
0.007 0.037 0.04 0.04 0.04 0.041 0.03 0.07

Mn
2.4 2.73 2.90 2.93 2.92 2.87 0.23 0.25

Si
<0.1 0.06 0.06 0.09 0.18 0.07 0.17 0.29

S
0.0007 0.001 0.002 <0.001 0.001 0.004 <0.001 <0.001

Ti
0.4 0.32 0.41 0.30 0.30 0.28 0.56 0.35

Nb + Ta

Cu
0.006

P
<0.001 0.014 0.003 0.004 0.002 0.003 0.004 0.007

Al
0.09

Mo

Co
<0.1 0.02

2.44 2.52 2.51 2.49 2.5

0.11 0.02 0.07 0.02 0.07 <0.01 0.11

0.63 0.17

0.01 0.04

59.8

29.54

8.25

0.026

0.29

0.01

0.0005

0.32

0.01

0.001

0.31

<0.001

EN82 heat 1 is the as-welded chemistry. All other EN82 and EN52 were the filler metal chemistry.

3.4.2 Recent Studies to Assess the Possible Relevance of LTCP to PWRs EPRI-sponsored research is ongoing in an attempt to assess the relevance of such environmental influence on J-R tearing resistance to reactor components during certain phases of PWR plant operation [38] using both testing [39] and analysis. While historical plant operation provides confirmation of degradation predictions for SCC and corrosion fatigue, there is little (if any)
t

Or at the tip of pre-existing cracks, e.g. from high-temperature SCC, as studied in the EPRI work [39].

3-18

Corrosion Behavior of Alloy 690 Apart from PWSCC

plant experience to characterize the nature and magnitude of the environmental fracture issue. The consequence of the phenomenon might be a reduction in structural integrity margins and an increase in the risk of loss-of-coolant accidents to an extent that depends on specific materials of construction and component operating conditions [40]. Most of the ongoing work is focused more on Ni-alloy weld metals (both original and replacement materials) rather than Alloy 690 and it would be premature to draw final conclusions at this time. With regard to the relative susceptibility of various materials in low temperature (<150C) water, Brown and Mills [36] have proposed the classification scheme shown in Figure 3-14. Their work suggests the following for Alloys 600/690: Alloy 600 is not susceptible to LTCP. Alloy 690 exhibits mostly moderate (category II) degradation in fracture resistance but this improves to borderline category II/III behavior at lower concentrations of dissolved hydrogen (see Figure 3-15 and Figure 3-16). Degradation in Alloy 690 fracture resistance is due to hydrogen-induced intergranular u cracking, but this is not seen in Alloy 600 with its lower Cr content (see Figure 3-17).

Figure 3-14 Classification scheme for categories of fracture resistance to LTCP after Brown and Mills [36]

This may be partly due to lower solubility of hydrogen in Alloy 600 and thus reduced hydrogen pick-up.

3-19

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-15 J-R curves determined by Brown and Mills for Alloy 690 in RT air and water at various temperatures [36]

3-20

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-16 JIC and T values determined by Brown and Mills [36] for Alloy 690 in air and water at various temperatures (values within bars indicate the dissolved hydrogen concentration)

3-21

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-17 Comparison of fractography from J-R testing under various conditions by Brown and Mills [36] of Alloy 690 (top) and Alloy 600 (bottom)

Bruemmer and Toloczko are performing limited testing for LTCP at the end of some of their HT PWSCC tests (see Section 5.1.2.2.2) and have observed some intergranular crack propagation at 50C in Alloy 690TT CRDM material, but during fatigue loading, not at constant K [41]. This suggests that homogeneous material representative of plant components may show only limited susceptibility. In contrast, Paraventi and Moshier [42] have reported severe degradation in the fracture resistance for Alloy 690 plate material subjected to large amounts of additional, nonuniform cold working (see Figure 3-18)v. This mirrors the high SCC CGRs for such material discussed in Section 5.1.2.1.7.
v

It is noteworthy in Figure 3-18 that even the L-T specimen orientation (where fracture resistance values were less affected) showed extensive, out-of-plane cracking along the 1-D rolling direction.

3-22

Corrosion Behavior of Alloy 690 Apart from PWSCC

Figure 3-18 Results by Paraventi and Moshier [42] from J-R testing of Alloy 690 plate with additional, non-uniform cold work in 50C water with varying contents of dissolved hydrogen

3.5

References

1. A.J. Sedricks, J.W. Schultz, and M.A. Cordovi, Inconel Alloy 690 A New Corrosion Resistant Material, Corrosion Engineering (Boshoku Gijutsu), vol. 28, pp. 82-95, 1979, Japan Society of Corrosion Engineering. 2. K. Smith et al., Inconel 690, A Material with Improved Corrosion Resistance for PWR Steam Generator Tubes, Proceedings of 2nd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1985 pp. 319-328. 3. T. Yonezawa et al., Effect of Heat Treatment on Corrosion Resistance of Alloy 690, Proceedings of 2nd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1985 pp. 593-600. 4. J.N. Esposito et al., The Addition of Zinc to Primary Reactor Coolant for Enhanced PWSCC Resistance, Proceedings of 5th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1991. pp. 495-501. 5. F. Scenini et al., Alloy Oxidation Studies Related to PWSCC, Proceedings of the Twelfth International Conference on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, Salt Lake City, UT, TMS, 2005. 6. P. Combrade et al., Oxidation of Ni Base Alloys in PWR Water: Oxide Layers and Associated Damage to the Base Metal, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp. 883 to 890.

3-23

Corrosion Behavior of Alloy 690 Apart from PWSCC

7. F. Delabrouille et al., Effect of the Chromium Content and Strain on the Corrosion of Nickel Based Alloys in Primary Water of Pressurized Water Reactors, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp. 903 to 911. 8. L. Legras et al., ATEM & SEM Study of the Oxides Developed in SCC Cracks and at the Surface of Nickel Based Alloys Exposed in Primary Water, Proceedings of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, August 19-23, 2007, Published by the Canadian Nuclear Society. 9. M.J. Psaila-Dombrowski et al., Evaluation of Weld Metals 82, 152, 52, and Alloy 690 Stress Corrosion Cracking and Corrosion Fatigue Susceptibility, Proceedings of 8th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Amelia Island, FL, 1997. pp. 412 to 421. 10. J. Keisler, O.K. Chopra, and W.J. Shack, Fatigue Strain-Life Behavior of Carbon and Low-Alloy Steels, Austenitic Stainless Steels, and Alloy 600 in LWR Environments, NUREG/CR-6335 (ANL-95/15), Argonne National Laboratory, 1995. 11. M Higuchi et al., Revised and new proposal of environmental fatigue life correction factor (Fen) for carbon and low-alloy steels and nickel base alloys in LWR water environments, Proc. ASME Pressure Vessels and Piping Division Conference July 23-27, 2006, Vancouver, Canada, PVP2006-ICPVT-11-93194. 12. R. Etien et al., EAC Behavior and Mechanical Properties of Improved Alloy 690 Filler Metals. EPRI MRP PWSCC Expert Panel Meeting, St. Petersburg, Florida (Nov. 2007). 13. W.J. Shack and T.F. Kassner: Review of Environmental Effects on Fatigue Crack Growth of Austenitic Stainless Steels, NUREG/CR-6176 ANL-94/1, (May 1994). 14. B. Alexandreanu, SCC CGRs of Alloys 690 and 52/152 Welds in PWR Water, Alloys 690/52/152 PWSCC Research Test Materials Meeting, Industry/NRC RES, July 17-18, 2008, Rockville, MD http://www.nrc.gov/reading-rm/adams/web-based.html, ADAMS Accession Number: ML082140693. 15. A. Jenssen, K. Norring and P. Efsing, Swedish Activities on Alloy 690 and its Weld Metals. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 16. K. Tsutsumi, Fatigue Crack Growth Rate Curve for Nickel Based Alloys in PWR Environment. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 17. F. Vaillant et al., Comparative behaviour of alloys 600, 690 and 800 in caustic environments, 7th Int Symp on Environmental Degradation of materials in Nuclear Power Systems Water Reactors, Breckenridge, Colorado, Published by NACE. 18. J. Gorman and C. Marks, Proceedings: 2005 EPRI/ANL/NRC Workshop on Effects of Lead and Sulfur on the Performance of Secondary Side Tubing of Steam Generators in PWRs. EPRI, Palo Alto, CA: 2005. 1012780. 19. B.T. Lu et al., Passivity of Nuclear Steam Generator Tube Alloy in Lead Contaminated Crevice Chemistries with different pH, Proceedings of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, August 19-23, 2007, Published by the Canadian Nuclear Society. 3-24

Corrosion Behavior of Alloy 690 Apart from PWSCC

20. H.P. Kim et al., Stress Corrosion Cracking of Steam Generator Tubing Materials in Lead Containing Solution, Proceedings of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, August 19-23, 2007, Published by the Canadian Nuclear Society. 21. J. Lumsden and A. McIlree, Factors affecting PbSCC in Alloy 600 and Alloy 690 Steam Generator Tubing, Proceedings of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, August 19-23, 2007, Published by the Canadian Nuclear Society. 22. B.P. Miglin, Investigation of Lead as a Cause of Stress Corrosion Cracking at Support Plate Intersections. EPRI, Palo Alto, CA 1991. NP7367-S. 23. J. Lumsden, Resistance of Alloy 600 and Alloy 690 Tubing to Stress Corrosion Cracking in Environments With and Without Lead. EPRI, Palo Alto, CA: 2004. 1009532. 24. J. Lumsden, Factors Affecting PbSCC in Alloy 600/Alloy 690 Steam Generator Tubing. EPRI, Palo Alto, CA: 2007. 1014990. 25. C. Marks, Pressurized Water Reactor Lead Sourcebook: Identification and Mitigation of Lead in PWR Secondary Systems. EPRI, Palo Alto, CA: 2006. 1013385. 26. G. White et al., Materials Reliability Program: Reactor Vessel Closure Head Penetration Safety Assessment for U.S. Pressurized Water Reactor (PWR) Plants (MRP-110): Evaluations supporting the MRP Inspection Plan. EPRI, Palo Alto, CA: 2004. 1009807. 27. Pressurized Water Reactor Secondary Water Chemistry Guidelines Revision 6. EPRI, Palo Alto, CA: 2004. 1008224. 28. C. Mansour et al., Behavior of Sulfur Species in Steam Generator Conditions of PWRs towards an update of the Secondary Side Corrosion Cracking Model based on Laboratory Tests in Sulfate Environments, Proceedings of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, August 19-23, 2007, Published by the Canadian Nuclear Society. 29. Pressurized Water Reactor Primary Water Chemistry Guidelines: Volume 1, Revision 6. EPRI, Palo Alto, CA: 2007. 1014986. 30. L. Thomas et al., Materials Reliability Program: Characterization of Cracks in a Control Rod Drive Mechanism (CRDM) Nozzle Base Metal and Weldment from Davis-Besse Reactor MRP-193. EPRI, Palo Alto, CA: 2006. 1013419. 31. Y. Yamamoto et al., Evaluation of Crack growth Rate for Alloy 600TT SG Tubing in Primary and Faulted Secobdary Water Environments, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp. 1243 to 1253. 32. O. de Bouvier et al., Stress Corrosion Cracking of Nickel Alloys in Complex (Liquid and Vapor) Environments, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp.1255 to 1266. 33. J. Harris, V. Maroney and J. Gorman, Pressurized Water Reactor Generic Tube Degradation Predictions: U.S. Recirculating Steam Generators with Alloy 600TT and Alloy 690TT Tubing. EPRI, Palo Alto, CA: 2003. 1003589. 3-25

Corrosion Behavior of Alloy 690 Apart from PWSCC

34. C. Marks, Alloy 690 Improvement Factor Update: Application of Improvement Factor Data to the Analysis of a Secondary System Chemistry Upset at Ginna. EPRI, Palo Alto, CA and Constellation Energy Group, Inc., Baltimore, MD: 2006. 1013640. 35. C. Marks, Evaluation of Altering the Hydrogen Concentration for Mitigation of Primary Water Stress Corrosion Cracking. EPRI, Palo Alto, CA: 2007. 1015017. 36. C.M. Brown and W.J. Mills, Low Temperature Crack Propagation Behavior of Alloy 82H Welds, Alloy 52 Welds, Alloy 600 and Alloy 690 in Water. EPRI MRP LTCP Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 37. T. Yonezawa, The Possibility of Metallurgical Effects on LTCP of Alloy 690 and its Weld Metal in Hydrogenated Water. EPRI MRP LTCP Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 38. A. Demma et al., EPRI Low Temperature Crack Propagation Projects. EPRI MRP LTCP Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 39. J. Peng et al., Effects of Dissolved Hydrogen and Hydrogen Peroxide on the Fracture Resistance of Weld Metals 182, 52 and 152 in Simulated PWR Shutdown Environment, 13th International Conference on Environmental Degradation of Materials in Nuclear Systems-Water Reactors, Whistler, B.C., Canada, August 19-23, 2007, Published by CNS. 40. W. Lunceford et al., EPRI Materials Degradation Matrix, Revision. EPRI, Palo Alto, CA: 2008. 1016486. 41. S. Bruemmer and M. Toloczko, Intergranular Crack Growth in Alloy 152 and Alloy 690 during Cyclic Loading at 50C. EPRI MRP LTCP Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 42. D.J. Paraventi and W.C. Moshier, Alloy 690 SCC Growth Rate Testing. EPRI Workshop on Cold Work in Iron and Nickel Base Alloys, Toronto (2007).

3-26

4
PWSCC OF THIN-WALLED SG TUBING
The move to introduce Alloy 690 as a replacement for Alloy 600, starting during the late Eighties, was driven primarily by the need to replace the tube bundles in PWR steam generators (SGs) that had undergone cracking, either from the primary or the secondary side, although the new material was also used extensively from about 1989 for all small diameter nozzle and pressurizer heater sleeve repairs [1]. Accordingly, virtually all of the experience with the new materials resistance to PWSCC up to 2004 was obtained with thin-walled SG tubing.

4.1

Laboratory Testing

The vast majority of laboratory testing for PWSCC behavior was carried out either on pre-production tubing, at a time when the manufacturing process was being fine-tuned, or on commercial tubing material. In isolated cases, laboratory heats of Alloy 690 were also manufactured and tested and the results from these have been included in this section, since although not strictly SG tubing they were also relatively thin-walled. Appendix A to MRP-111 [2] contained the detailed results from a comprehensive review of available laboratory test data pertinent to SCC in environments relevant to primary water for thinwalled Alloy 690 material up to 2004. Certain aspects were updated in [3] and further assessment is currently in progress in the context of the EPRI Steam Generator Program. Numerous investigations have been performed under a variety of environmental conditions including temperatures to 360C (680F) in water with dissolved oxygen levels < 20 ppb, lithium concentrations up to 3.5 ppm, boron concentrations up to 1800 ppm, hydrogen concentrations up to 100 cc/kg H2O, additions of chlorides and zinc, and tests in doped 400C (752F) steam. Even though the wording simulated primary water conditions was often used by the authors of published work, it is clear that some test conditions employed were outside the normal range of primary water chemistry. Accelerated testing was performed on double U-bend (simulating crevice conditions), reverse U-bend (RUB), constant load test (CLT), four-point bend, and steam generator tubing mock-up specimens. The literature review of Alloy 690 PWSCC behavior, whose results are summarized below, comprised essentially studies of crack initiation and was divided into three groups: 1. Deaerated water 2. Deaerated water with additions of boron and lithium 3. Steam with the addition of hydrogen

4-1

PWSCC of Thin-Walled SG Tubing

4.1.1 Early Studies The first published stress corrosion cracking test results on Alloy 690 material in deaerated high temperature water containing 20 ppb or less dissolved O2 were reported by Sedricks et al. of Inco in 1979 [4]. No cracking was produced in any of the Alloy 690MA U-bend specimens fabricated after a maximum exposure of 10,000 hours at 360C (680F). Impressed by this test outcome, the authors were among the first to suggest that, in high temperature deaerated water, Alloy 690 was effectively immune to the SCC that had been observed in this environment in other high nickel alloys, especially Alloy 600. It is noted that a majority of the Alloy 600MA specimens in Sedricks high temperature deaerated water tests did not develop cracking by the end of the tests, which may be an indication that the testing condition was not aggressive enough, or that the duration of the tests was not long enough. Airey later investigated Alloy 600 PWSCC in both pure hydrogenated water and primary water (containing boric acid and lithium) [5]. The results showed that the SCC initiation time was shorter in pure water than in primary water. Hence, a pure hydrogenated water environment can be considered slightly more aggressive than hydrogenated water containing boric acid and lithium when testing for susceptibility to PWSCC. Since that time, laboratories in several countries have employed progressively more severe forms of stress corrosion testing under more and more aggressive conditions and to longer test durations in order to assess Alloy 690s resistance to PWSCC. Most of these studies have confirmed that Alloy 690 material has extremely high PWSCC resistance, as no cracking was produced in the Alloy 690 specimens at the termination of testing in the vast majority of cases. Appendix A in [2] reviewed 15 different studies of Alloy 690 PWSCC in high temperature deaerated water with or without boric acid and lithium. Approximately 300 U-bend, double U-bend, and constant load specimens from about 40 heats of Alloy 690 have been tested. The carbon contents ranged from 0.001% to 0.065% and the heat treatment included MA, TT, and thermally aged conditions. The very few cases where cracking of Alloy 690 material were reported - and the reasons why they are not representative of in-service PWR conditions - are further discussed immediately below. 4.1.2 Single U-Bend Test in Saturated Hydrogen Water with B/Li In 1987, Nakayama et al. reported results of Alloy 690 single U-bend tests in saturated hydrogenated water containing 1000 ppm boron and 2 ppm lithium at 330C (626F) for 3000 hours [6]. Slight intergranular cracking was detected in two out of the four Alloy 690 conditions tested. The maximum intergranular crack depth was 70 m (0.0028) in the Alloy 690TT material with an aging treatment of 24 hours at 500C (932F) and 30 m (0.0012) in the Alloy 690MA material with an aging treatment of 100 hours at 500C (932F). The Alloy 690 grain size was not reported, but the 70 m maximum intergranular crack depth would correspond to two grains deep for a typical grain size of ASTM no. 6.5. Hence, these cracks were shallow and more comparable to an intergranular attack (IGA) depth than to the much deeper PWSCC cracks routinely seen in Alloy 600 material. In addition, SEM examinations revealed that the intergranular cracks in the Alloy 690TT specimen (70 m max. depth) were at the tips of dislocation pile-ups. This indicates that the Alloy 690 intergranular cracking could have been caused by mechanical strain from U-bending, rather than by PWSCC. It is noteworthy that similar mechanically induced surface intergranular cracks have been observed in bent Alloy 690 and Alloy 52 specimens in several other studies. These observations and 4-2

PWSCC of Thin-Walled SG Tubing

attributed causes are summarized in Table 4-1 for a side-by-side comparison. It is evident from these reports that Alloy 690 and Alloy 52 can be prone to develop surface defects (including intergranular microfissuring) induced by plastic straining from bending. Shallow, surface-modified layers from SG tubing fabrication processes are seemingly responsible for this mechanical cracking propensity. The CEA and EdF studies (see Appendix A in [2]) showed that the plastic strain at the apex of a RUB is typically between 30 to 40%. This level of plastic straining approaches the room temperature ductility limit of Alloy 690. In addition, Alloy 690 shows a slight drop in ductility as the temperature is increased from room temperature to 315C (600F; see Figure 2-5). Alloy 690 also has a higher cold work hardening rate as shown in Figure 4-1, which plots the Vickers hardness number as a function of cold reduction percentage. Hence, an Alloy 690 surface is expected to be more sensitive to work hardening during cold work fabrication steps. Any highly cold-worked surface layer is likely to re-crystallize during thermal treatment and is observed to be more prone to developing surface defects compared to Alloy 600. However, the studies listed in Table 4-1 demonstrated that these surface defects or microfissures did not propagate (through PWSCC) by the end of the tests. Hence, it appears that the short IG cracking reported by Nakayama et al. was also due to strain applied during specimen bending.
Table 4-1 Examples of surface IG cracking in Alloy 690 and Alloy 52 specimens [2] Origin (MRP-111) Miglin, 1986 Matl. Alloy 690MA&TT SG Tubing Description of Surface IG Cracks and Defects Longitudinal shallow and blunt defects were found on the I.D. surface of the Alloy 690 RUB apex. These defects were not typical of IGSCC. Metallographic examinations showed that some of these defects appeared to be like mechanical grooves while others appeared to be separation along the grain boundaries. No evidence of further IGSCC cracking emanating from the base of these defects were found. SEM examinations of archive Alloy 690 RUB specimens found similar longitudinal defects, but they were shorter in length than those detected after the AVT autoclave exposure (0.1 to 0.35 mm long in archive Alloy 690 RUB vs. ~1.0 mm in Alloy 690 RUB after the AVT exposure). The less strained leg portions of the RUB specimens were free from these defects. Short IG cracks (depth not reported) on the RUB specimen I.D. surface produced by flattening at room temperature. The RUB specimen had been exposed to hydrogenated steam at 380C (716F) for 13,824 hours. Attributed Cause The growth in length observed was due to an opening process of closed or partially closed defects. During the high temperature exposure, the surface oxide growth (in the defects) would enhance an opening process by a wedging effect to give an appearance of longitudinal growth.

Sui, 1997

Alloy 690TT SG Tubing

Embrittlement of grain boundaries near the surface from the exposure.

4-3

PWSCC of Thin-Walled SG Tubing

Table 4-1 Examples of surface IG cracking in Alloy 690 and Alloy 52 specimens [2] (continued) Origin (MRP-111) Vaillant, 1999 Matl. Alloy 690TT SG Tubing Description of Surface IG Cracks and Defects Shallow cracks (10 m or 0.0004 deep) appeared at the apex of the Alloy 690 RUB specimens after bending, but before immersion inside the autoclave. So the stress measured in the Alloy 690 RUB specimen apex surface was not welldefined. However, the stress level measured after removing the 10 m deep crack was the same as for the Alloy 600 RUB specimens. Hence, the stress at the tip of the short crack caused by bending is at least as high as the apex surface stress of the Alloy 600 RUB specimens. No crack growth observed in the Alloy 690 RUB specimens during the testing. IG cracks on the inner tube surface in the transition zones of the kiss rolling (1.5 mm and their width 2 m, see Appendix A for details). Examination of the flattened half tube revealed the presence of many microcracks, some of the opened microcracks corresponded to those which were also observed before flattening but numerous others were new ones opened or created during the mechanical flattening operation. This type of flaw was also observed on parts of the tube which were not rolled, that is remote from the rolled zone. The crack depth was always limited to the thickness of the perturbed surface layer on the internal surface of the tube (about 10 m thick in this case). These microcracks were intergranular but did not propagate in the tube during exposure to hydrogenated water at 360C during 60,000 hours exposure. Attributed Cause Cracking was due to bending.

Framatome, France 2003

Alloy 690 TT, SG Tubing

Strain induced by rolling and are limited to the hard perturbed surface layer already present on the inside surface of the tube in the as-received condition.

4-4

PWSCC of Thin-Walled SG Tubing

400

350

Vickers Hardness Number

300

Alloy 690 Alloy 600

250

200

150

100 0 10 20 30 40 50 60 70 Cold Reduction, %

Figure 4-1 Vickers hardness number as a function of cold reduction % from [2]. Alloy 690 has a higher work-hardening rate than Alloy 600

In addition to the factors discussed above, the Alloy 690 plate used to fabricate the U-bend specimens tested by Nakayama et al. was made from remelted (in a vacuum induction furnace) Alloy 690 SG tubing. Only the chemical compositions of the tubing were reported and it is unclear whether remelting might have caused changes to the Alloy 690 chemical composition. Interestingly, these Alloy 690 heats contained a rather high aluminum content of 0.39 or 0.40%. (Note, aluminum is not specified in ASME Section II and is often unreported). A. Smith et al. reported that aluminum increases the propensity to caustic IGA in Alloy 690 [7, 8]. Unfortunately, the Alloy 690 microstructure (cracked or not cracked) was not examined to confirm that the thermal treatment had produced intergranular carbide precipitations. 4.1.3 CERT Tests in Hydrogenated Water with or without B/Li Other than the studies by Nakayama et al., and some CEA and EDF tests using high deformation U-bend or RUB specimens in which only mechanical surface intergranular cracking was detected, reports of intergranular cracking in deaerated water originate from CERT studies at strain rates on -7 -1 the order of 10 sec. or less. These are very severe tests for evaluating the susceptibility of Alloy 600 or Alloy 690 to stress corrosion cracking. The specimens are loaded well past the original yield strength to maintain a continuous plastic strain rate, and are often continued until the specimens fail by ductile overload, if no SCC occurs. Such continuous plastic deformation situations are not encountered in nuclear power plants. Angeliu et al. produced intergranular cracking in a 0.002% 4-5

PWSCC of Thin-Walled SG Tubing

carbon laboratory heat of Alloy 690 in deaerated water at 360C (680F) during CERT tests [9, 10]. Since the same small amount of intergranular cracking (2%) was also produced in argon, the intergranular cracking in the deaerated water should not be considered an indication of Alloy 690 susceptibility to PWSCC. Other than the study by Angeliu et al, a study by CEA and EdF also reported Alloy 690 cracking during CERT tests (see [2] for more details). Boursier et al. considered the CERT test as measuring the stress corrosion crack growth rate, rather than propensity to crack initiation [11]. They concluded that Alloy 690 material did not necessarily have a lower CGR than Alloy 600 material and suggested that the CGR obtained in the CERT tests is mainly dependent on the grain boundary carbide precipitation and the creep rate, while the crack initiation time in the RUB tests was mainly a function of Cr contents. It is noted that the most susceptible Alloy 690 heat in this study was a pre-production heat involving a heat treatment that gave rise to a very low density of grain boundary carbides. Compact tension (CT) specimens, wedge opening load (WOL) specimens, double cantilever beam (DCB) specimens, or other forms of precracked specimens per ASTM E 399 or E 813 are commonly used for PWSCC CGR testing of Alloy 600 or Alloy 182 materials [12, 13]. During the CGR test, the specimen is loaded under plane strain condition to a pre-determined stress intensity factor which does not necessarily increase during the test. On the other hand, the tensile load in CERT tests increases with test time to keep the elongation rate constant. Hence, CERT tests are very different from the typical CGR tests used for Alloy 600 and Alloy 182 materials. 4.1.4 RUB Test in Hydrogenated Steam In 1997, Sui et al. reported stress corrosion cracking of Alloy 690 specimens in hydrogenated steam at 380C (716F) [14, 15]. Two RUB specimens from the same heat of Alloy 690 developed cracks between 12,600 and 13,824 hours of exposure. Both heats of Alloy 690 had been thermally treated at ~715C (1319F) for 15 hours. However, the cracked Alloy 690 Heat B had an unusually (deliberately) low final mill anneal temperature compared to the un-cracked Alloy 690 Heat A (1769F vs. 1958F). The average grain size was 25 m (ASTM no. 7.5) for Heat A and 15 m (ASTM no. 9.0) or less for Heat B. The small grain size in Heat B is reflected in its higher room temperature yield strength. All the grain boundaries in Heat A had almost continuous carbide coverage. The carbides were determined to be the M23C6 type by TEM. In contrast, Heat B was essentially free of intergranular carbides. The cracked Alloy 690 heat, Heat B, also had a higher aluminum content (0.14%), which tends to result in a finer grain size and therefore a lower density of intergranular carbides. Interestingly, the Alloy 690 heats used by Nakayama et al. also had a rather high aluminum content. The results reported by Sui et al. show that Alloy 690 could be susceptible to IGSCC in a hydrogenated steam environment and potentially susceptible to PWSCC if the final mill anneal temperature is too low to produce intergranular carbide precipitation upon subsequent thermal heat treatment, or in the event of re-crystallization after thermal treatment. Because such a low temperature solution anneal heat treatment is prohibited by material specifications for Alloy 690 (or Alloy 600) for use in PWRs, and the intergranular carbide precipitation must be verified by optical microscopy and/or SEM, cracking of the type seen by Sui et al. will be prevented. On the other hand, re-crystallization may be a detrimental factor to take into account for the heat affected zones of welds. In any event, the times 4-6

PWSCC of Thin-Walled SG Tubing

to failure at 380C (716F) reported by Sui et al are such that when extrapolated to normal PWR operating temperatures using an Arrhenius relationship and a standard value of the activation energy for PWSCC, the predicted lives (equivalent to 98 EFPYs at 315C based on the 50 kcal/mole PWSCC activation energy commonly used for Alloy 600, see Table 4-2) would be well in excess of current expectations for plant life extension. Nevertheless, this possible limited susceptibility induced by the absence of grain boundary carbides is an aspect of Alloy 690 behavior requiring further consideration.
Table 4-2 Summary of Alloy 690 primary water stress corrosion test data to 2004
Ref. # in MRP111[2] Test Temp (F) Alloy 690 Heat Number Alloy 690 Heat Cond. MA, TT, MA+CW, MA+Weld MA, TT MA, TT 20 MA 680 One Alloy 690 heat Two industrial heats A, E F 35 37 RUB Deaerated water 689 H, G, A, B, D, C, I PP Y, Z RUB 43 CLT Deaerated water + B + Li Deaerated water + B + Li with or without Zn Deaerated water + B + Li Deaerated water 680 One Alloy 690 heat TT TT 644 TT MA MA, TT TT MA TT 20 10000 23.0 1142 3 2 5 2 6 47 6 8 40 16000 6600 12000 7000 1500 33000 25000 23000 20500 10000 36.8 15.2 27.6 16.1 0.9 103.6 78.5 72.2 64.3 23.0 3000
(c)

Test

Test Environ.

Total Speci (a) No.

Test Time at Test Temp. (Hour) 8064 8064 13000

Equatio n Test Time at (b) 600F (Year) 0.9 23.2 29.9

Time to First Alloy 600 Failure (Hour) 3024 No 600 control 1500 2000 340 1144 No 600 control

Improve Factor (IFR)

Double U-Bend Double U-Bend

Deaerated water Deaerated water Deaerated water + B + Li Deaerated water + B + Li Deaerated water

600 680 680

Y24A7L NX4458H NX4460H Three preseries, Heats 1-3

52 8 20

2.7 N/A 6.5 6.4 19.4 35.3 6.1 N/A 41.3

11

RUB

17

RUB CLT

32, 33

C-Ring

800

31.3 28.8 25.6 3.3 8.8

44

RUB

662

752246

TT

20

7500

9.1

5500

1.4

46 Framatome, France (new data in [2]) (a) (b) (c)

RUB

680

9.092Exp 9.592Exp 9.799Ind 9G4 WE094, Pre-series

TT, MA

54000

124.1

500

108

SG Mocku p

680

TT

16

100000

229.8

800

125

Total number of specimens of similar heat treatment and test condition and duration. The equivalent test time at 600F for Alloy 690 is calculated based a PWSCC crack initiation time of 50 kcal/mole (Equation 4-7). 1142 hours is the equivalent of the 4179 hours failure time at 644F for the Alloy 600 control CLT specimens based on Equation 4-7. Only two of four Alloy 600 MA series results were reported. Hence, the time to the first Alloy 600 failure could be less than 4179 hours.

4-7

PWSCC of Thin-Walled SG Tubing

4.1.5 Weibull and Weibayes Analyses of the Test Results The results of the laboratory tests reviewed in Appendix A of [2] were examined to obtain a quantitative or semi-quantitative assessment of Alloy 690s resistance to PWSCC relative to Alloy 600. This is desirable for developing a statistically-based estimation of a hypothetical PWSCC event in Alloy 690 component items that can be used in selecting and justifying an inspection method. The two-parameter Weibull distribution of Equation 4-1 is by far the most widely used distribution for stress corrosion cracking analysis [16].

F (t ) = 1 e
where:

Equation 4-1

F(t) = the cumulative fraction of cracking by time (t). = the characteristic time (scale parameter) equal to the time to a 63.2% cumulative fraction of cracking. = slope or shape parameter e = 2.71828, the base for natural logarithms.

The significance of is briefly summarized below: <1 =1 1< < 4 >4 Implies infant mortality, failure rate decrease with time. Implies random failure. Implies early wear out. Implies old age and rapid wear out. A steeper means less material variation or tighter quality control. A steeper means subsequent failures will occur quickly after the first failure.

Weibull analysis has been used to analyze the failures of the Alloy 600 control specimens in the studies reviewed in [2]. If sufficient data were reported or available, a best-fit regression analysis was then performed to determine the Weibull characteristic time () and slope (). In addition to Weibull analysis for the Alloy 600 control specimens, a Weibayes analysis (an estimated Weibull) was used in [2] for the Alloy 690 specimens that did not develop cracking by the end of the test period. In the Weibayes method, the slope is assumed to be the same as that from the Alloy 600 analysis. The Alloy 690 specimen characteristic time is estimated by Equation 4-2 [16].

N ti = i =1 r

1/

Equation 4-2

4-8

PWSCC of Thin-Walled SG Tubing

where:

= the slope or shape parameter from Alloy 600. ti = test time of each Alloy 690 specimen tested. N = total number of Alloy 690 specimens tested. r = number of Alloy 690 specimens cracked.

As no constant deformation (deflection) or constant load Alloy 690 specimen developed SCC cracks by the end of the test period (except in one study in hydrogenated steam at 380C (716F)), it is conservatively assumed that the first (i.e., r = 1) Alloy 690 specimen failure could occur immediately should the test continue. The confidence level that the true Alloy 690 Weibull line lies to the right of the Weibayes line is 63%. The failure number r can be set to produce any level of confidence desired. For example, Weibest uses 0.693 for r to produce a confidence level of 50%, which is less conservative than the 63% confidence level used by Weibayes. Figure 4-2 is the Weibull plot from [2] of some early Alloy 600 RUB test results at 360C (680F) generated in France. The differences in the water chemistry used (beginning-of-fuel-cycle vs. end-of-fuel-cycle) have little impact on the Alloy 600 (one heat) RUB cracking compared to the heat treatment. Hence, the Alloy 600 RUB cracking data in the two primary water environments are combined in Figure 4-2. The Alloy 690 Weibayes characteristic life is calculated based on a total population of 40 RUB specimens from three different heats in either the MA or TT condition since there was no failure of any Alloy 690 RUB specimens. The Alloy 690 Weibayes line in Figure 4-2 assumes a slope of 5.0 ( = 5.0).

90

Cumulative Probability, %

75 63.2 50 25 10 600MA

690MA&TT Weibayes

600TT 1,000 10,000 100,000

Time to Failure, Hour

Figure 4-2 Weibull plot from [2] of early Alloy 600 RUB results in primary water at 360C generated in France. The Alloy 690 (three heats) Weibayes line assumes = 5.0

4-9

PWSCC of Thin-Walled SG Tubing

Figure 4-3 and Figure 4-4 are the Weibull plots of the Alloy 600 RUB results at 689F reported by Norring et al. [17]. Figure 4-3 plots Alloy 600 RUB results of different heats of real SG tubing while Figure 4-4 plots the Alloy 600 RUB results of EPRI special production tubes. Figure 4-4 shows that the Alloy 600MA annealed at 1024C (1875F) has a higher PWSCC resistance than the Alloy 600MA annealed at 927C (1700F). Figure 4-5 compares the influence of the SG tubing diameter on the time to cracking. Even though the RUBs were from the same heats, the ~22mm (7/8) dia. RUB developed cracking much earlier than the ~19mm (3/4) dia. RUB. This difference in time to cracking was attributed to different stress levels after the reverse bending of different diameter tubing. However, no stress calculations or measurements were reported. The Alloy 690 Weibayes characteristic life is calculated based on a total population of 67 RUB specimens from three different heats of Alloy 690. These specimens were heat treated in the MA or TT condition with additional variations in heat treatment temperature and duration. None of the 67 Alloy 690 RUB specimens developed cracking after 20,500 to 33,000 hours of exposure. The Alloy 690 Weibayes line in Figure 4-3, Figure 4-4, and Figure 4-5 assumes a slope () of 5.0.

Cumulative Cracking Probability, %

90
gh al 4 (3/4 Ringh "d al 3 ( i a. 3/4 " d ) i a.)

y 60 0M A

25 10

Ri ng ha ls 2

(7/8 " di a.)

75 63. 2 50

ia. )

(7/ 8" d

Allo

Rin

Alloy

600T T

690MA&TT Weibayes

100

1,000
Tim e to Failure, Hour

10,000

100,000

Figure 4-3 Weibull plot from [2] for RUB test results in deaerated water at 365C reported by Norring et al. [17]. The Alloy 690 (many heats) Weibayes line assumes = 5.0

4-10

PWSCC of Thin-Walled SG Tubing

Cumulative Cracking Probability, %

90 75 63.2 50 25 10

B&W TP 1 7 00F

Huntin

gt on 1 700F

0F

Sand vik 1 70

Sand vik 1 87

5F

B&W TP 1

875F

690MA&TT Weibayes 100 1,000

Time to Failure, Hour

10,000

100,000

Figure 4-4 Weibull plot from [2] for RUB test results by Norring et al. [17] on special production heats of Alloy 600 in deaerated water at 365C. The Alloy 690 Weibayes line assumes = 5.0

Cumulative Cracking Probability, %

90 75 63.2 50
7/8 "d ia. A ng

25

10

3/4 "d ia. All oy 600 tub ing

llo y

60 0t

ub i

7/8" = Ringhals 2, NX1991 and NX2650, 18 specimens. 3/4" = Ringhals 3 and 4, 28 specimens

690MA&TT Weibayes 10,000 100,000

100

1,000
Tim e to Failure, Hour

Figure 4-5 Weibull comparison from [2] for different tube diameters from RUB testing by Norring et al. [17] in deaerated water at 365C. The Alloy 690 Weibayes line assumes = 5.0

4-11

PWSCC of Thin-Walled SG Tubing

Extensive constant load tests (CLT) using specimens of Alloy 600M, Alloy 600TT and Alloy 690TT have been performed in Japan (see [2] for further details) and the overall results from these are summarized in Figure 4-6. None of the Alloy 690TT CLT specimens tested at 360C (680F) cracked and the Alloy 690 Weibayes characteristic life was calculated in MRP-111 based on a total population of 20 CLT specimens tested in four variations of primary water chemistry, again assuming a slope () of 5.0.

90

Cumulative Probability, %

75 63.2 50 25 10
600MA at 644oF 600MA&TT at 608 oF Weibayese 690TT at 680 oF Weibayese

1,000

10,000

100,000

Time to Failure, Ho ur

Figure 4-6 Weibull plots from [2] of Japanese Alloy 600MA (one heat) CLT results at 340C (644F) in primary water. No failure was observed in any Alloy 600MA or Alloy 600TT CLT specimens tested at 320C (608F) and in the Alloy 690TT (one heat) CLT specimens tested at 360C (680F). The Weibayes lines for the unfailed specimens assume = 5.0

In addition to the CLT tests, RUB tests with various degrees of prestrain were performed under the same primary water conditions. Table 27 in Appendix A of [2] showed that variation of the primary water did not have a significant impact on the Alloy 600 RUB specimens cracking. Hence, the RUB test data from the four primary water chemistries were combined before applying the Weibull analyses reported in MRP-111. Figure 4-7 compares the results of 20% prestrained RUB specimens from the Alloy 690TT, the Alloy 600TT, and the Alloy 600MA materials. None of the Alloy 690TT specimens tested at the higher temperature of 360C (680F) developed cracking after 10,000 hours of exposure. The Alloy 690TT Weibayes characteristic life was calculated based on a total population of 40 RUB specimens assuming a slope () of 5.0.

4-12

PWSCC of Thin-Walled SG Tubing

90

Cumu lative Probability, %

75 63.2 50 25 10 20% 600MA at 608oF

20%690TT Weibayes at 680oF

20% 600TT at 608oF 10,000 100,000

1,000

Time to Failure, Hour

Figure 4-7 Comparison from [2] of Japanese data for 20% pre-strained RUB specimens of Alloy 600MA or Alloy 600TT, tested in primary water at 320C (608F), with those for Alloy 690TT, tested under the same conditions, but at 360C (680F). The Alloy 690TT Weibayes line assumes = 5.0

The RUB tests in primary water at 360C (680F) reported by Vaillant et al. [18] were also evaluated with Weibull analysis. RUB specimens from different heats of Alloy 600 were separated into the MA and TT conditions and the results are plotted in Figure 4-8. None of the Alloy 690TT and Alloy 690MA specimens developed cracking after up to 54,000 hours of exposure. The Alloy 690 Weibayes characteristic life is calculated based on a combined TT and MA population of 4 RUB specimens listed in the report. The Alloy 690 Weibayes line assumes a slope () of 5.0. Further information on the extensive studies of PWSCC in Alloy 600 and 690 SG tubing at EDF in France is given in references [19, 20, 21]. The results of SG mockups tested in deaerated, hydrogenated water at 680F by Framatome ANP, France are plotted in Figure 4-9. None of the Alloy 690TT mockups developed any cracking after 100,000 hours of exposure. The Alloy 690 Weibayes line assumes a slope () of 5.0. The results of the Weibull and Weibayes analyses are used in the next section for estimating the improvement factor (IFR) of Alloy 690 relative to the Alloy 600 in PWR primary water conditions.

4-13

PWSCC of Thin-Walled SG Tubing

90

Cumulative Probability, %

75 63.2 50 25 10
600MA 690MA&TT Weibayese

600TT

100

1,000

10,000

100,000

Time to Failure, Hour

Figure 4-8 Weibull plot from [2] of RUB results at 360C in primary water reported by Vaillant et al. [18]. The Alloy 600 RUB specimens were from four different heats in the MA and TT conditions. The Alloy 690 RUB specimens, also from four different heats in the MA and TT conditions, experienced no failure after up to 54,000 hours of exposure. The Alloy 690 Weibayes line assumes = 5.0

90

Cumulative Probability, %

75 63.2 50 25

10

600MA WD281
600TT NX3335

690TT Weibayes
600TT WD281

100

1,000

10,000

100,000

1,000,000

Time to Failure, Hour

Figure 4-9 Weibull plot from [2] of SG mockups tested in deaerated water at 680F by Framatome ANP, France. Alloy 690TT SG mockups experienced no failure after 100,000 hours of exposure. The Alloy 690 Weibayes line assumes = 5.0

4-14

PWSCC of Thin-Walled SG Tubing

4.1.6 Improvement Factor by Weibull Analysis The Weibull parameters of Alloy 600 materials tested with Alloy 690 are listed in Table 4-3. Using the Weibull analysis, the relative improvement factor, IFR, can be defined by Equation 4-3.
IFR = Weibayes (r = 1) , Alloy 690 Weibull , Alloy 600

Equation 4-3

Table 4-3 Weibull analysis for Alloy 600 tested with Alloy 690 [2]
Figure Figure 4-2 Figure 4-2 Figure 4-2 Figure 4-3 Figure 4-4 Figure 4-5 Figure 4-3 Figure 4-3 Figure 4-3 Figure 4-3 Figure 4-3 Figure 4-4 Matl. 690MA&T T 600MA 600TT 690MA&T T 600 600MA 600 600 600TT 600MA Heat Pre-series 3 Heats No info No info Many heats PWR Ringhals 2, 7/8 dia. Alloy 600MA, 7/8 dia. PWR Ringhals 4, dia. PWR Ringhals 3, dia. Alloy 600TT, dia EPRI Provided B&WTP, 1700oF EPRI Provided Huntington, 1700oF EPRI Provided Sandvik, 1700oF EPRI Provided B&WTP, 1875oF EPRI Provided Sandvik,1875o F 7/8 dia., Ringhals 2, NX 1991, NX 2650 3/4 dia. Ringhals 3 and 4 One heat Test RUB RUB RUB RUB RUB RUB RUB RUB RUB RUB Environ. Deaerated water + B + Li Deaerated water + B + Li Deaerated water + B + Li Deaerated water Deaerated water Deaerated water Deaerated water Deaerated water Deaerated water Deaerated water Tem p (oF) 680 680 680 689 689 689 689 689 689 689 Weibull, (Hour) 30,950 8,112 14,539 59,748 910 1,440 4,590 5,620 8,060 740 Weibull Slope, (a) 5.00 1.96 5.08 5.00 1.74 3.91 2.47 6.05 5.52 4.53 IFR (b) N/A 3.8 2.1 N/A 65.7 41.5 13.0 10.6 7.4 80.7

Figure 4-4

600MA

RUB

Deaerated water

689

2,480

10.43

24.1

Figure 4-4

600MA

RUB

Deaerated water

689

3,490

5.77

17.1

Figure 4-4

600MA

RUB

Deaerated water

689

10,260

5.48

5.8

Figure 4-4

600MA

RUB

Deaerated water

689

10,990

5.27

5.4

Figure 4-5

600

RUB

Deaerated water

689

1,118

1.91

53.4

Figure 4-5 Figure 4-6

600 690MA&T T

RUB CLT

Deaerated water Deaerated water + B + Li

689 680

5,397 18,206

3.05 5.00

11.1 N/A

4-15

PWSCC of Thin-Walled SG Tubing

Table 4-3 Weibull analysis for Alloy 600 tested with Alloy 690 [2] (continued)
Figure Figure 4-6 Figure 4-7 Matl. 600MA 690TT Heat One heat One heat Test CLT RUB 20% prestrain RUB 20% prestrain RUB 20% prestrain RUB RUB RUB SG Mock-up SG Mock-up SG Mock-up SG Mock-up Environ. Deaerated water + B + Li Deaerated water + B + Li Deaerated water + B + Li Deaerated water + B + Li Deaerated water + B + Li Deaerated water + B + Li Deaerated water + B + Li Deaerated water Deaerated water Deaerated water Deaerated water Temp (oF) 644 680 680 608 680 608 680 680 680 680 680 680 680 680 Weibull, (Hour) 5,358 1464 (c) 20,913 6,051 414
(c)

Weibull Slope, (a) 5.91 5.91 5.00 2.10 2.10 3.64 3.64 5.00 1.04 1.85 5.00 0.95 1.28 2.29

IFR (b) 12.4 N/A 51.2 32.2 N/A 30.6 13.8 N/A 19.4 15.5 8.6

Figure 4-7

600MA

One heat

Figure 4-7 Figure 4-8 Figure 4-8 Figure 4-8 Figure 4-9 Figure 4-9 Figure 4-9 Figure 4-9
(a) (b) (c)

600TT 690TT 600MA 600TT 690TT 690MA 690TT 690TT

One heat Four heats Three heats Four heats One heat WD281 WD281 NX3335

9,530 652 54,284 1,776 3,945 174,110 8,954 11,199 20,231

is time to Weibull 63.2% cumulative failure for Alloy 600 and the equivalent Weibayes for Alloy 690. IFR per Equation 4-6. The equivalent time at 680 F per Equation 4-7 based a PWSCC crack initiation time of 50 kcal/mole.
o

Hence, IFR is the ratio of the time to a 63.2% cumulative fraction of failure between the Alloy 690 and the Alloy 600 specimens being tested. Transformation of Equation 4-1 yields Equation 4-4, which in turn produces Equation 4-5.

1 t = 1 F (t )

Equation 4-4

t690 t600

=
to a given F ( t )

690 600

Equation 4-5

Equation 4-5 shows that, if the Alloy 690 Weibayes line assumes the same Weibull distribution slope () as for the Alloy 600 (either TT or MA condition), the IFR to any cumulative fraction of cracking would be a constant. However, the slope of Alloy 690, although unknown due to a lack of cracking in the tests, may not vary as much as seen in Alloy 600 materials. Assuming a small 4-16

PWSCC of Thin-Walled SG Tubing

Weibull slope value for Alloy 690 can result in a higher Weibayes (hence a higher IFR), making the IFR less conservative. A review of the Alloy 600 Weibull plots indicates that the slope increases with increasing cracking resistance, i.e., Alloy 600TT tend to have a higher than Alloy 600MA in the same investigation. However, significant variation exists in the PWSCC susceptibility of different heats of Alloy 600. The variation in the Alloy 600 material can be seen in Figure 4-10, which plots the Weibull vs. listed in Table 4-3, except the data in Figure 4-5 which were derived from the data in Figure 4-3 and Figure 4-4. The average for Alloy 600MA is 4.12 while the average for Alloy 600TT is 3.28. Hence, the slope for the Alloy 690 Weibayes line when arbitrarily set to 5.0 is higher than the average value for Alloy 600. This introduces additional conservatism in the calculation of the Alloy 690 Weibayes . No differentiation is made in this case between Alloy 690 in the MA condition and in the TT condition because neither has cracked in the hydrogenated, deaerated water. Therefore, the improvement factor, IFR, is redefined as
IFR = Weibayes (r = 1, = 5.0) , Alloy 690 Weibull , Alloy 600
Equation 4-6

12

Alloy 600MA
10
Alloy 600 Weibull Slope,

Removed from PWR

Alloy 600TT

6 4

2 0 0 5,000 10,000 15,00 0 20,000 25,000 Alloy 600 Weibull () , hour

Figure 4-10 Weibull and factors for Alloy 600 tests listed in Table 4-3

The resulting IFR values are listed in Table 4-3. The Alloy 600 test data shown in Figure 4-6 and Figure 4-7 are normalized per the Arrhenius relationship of Equation 4-7 to 360C, the Alloy 690 test temperature in the same study. Overall, the average IFR of Alloy 690 listed in Table 4-3 is 26.5 relative to Alloy 600MA and 13.3 relative to Alloy 600TT.

4-17

PWSCC of Thin-Walled SG Tubing

t t ref

Q exp i R

1 1 T Tref

Equation 4-7

where Qi = 50 kcal/mole. 4.1.7 Improvement Factor with Minimum Alloy 600 Crack Time Unfortunately, not all the studies reviewed in 2004 (see Appendix A of [2]) had either obtained or reported sufficient data to allow a Weibull or Weibayes analysis. In order to include these data, an alternative way to calculate the improvement factor using Equation 4-8 can be used:
IFR = Test Time of Alloy 690 Time to first cracking in Alloy 600
Equation 4-8

Here, the time to first cracking in Alloy 600 refers to the time when the first failure in the Alloy 600 control specimens was observed. The following factors are considered for the IFR values determined from Equation 4-8. 1. The PWSCC resistance of Alloy 600 has been observed to vary greatly and depend on several factors. Like the Alloy 690 specimens, some Alloy 600 TT specimens did not develop any cracking by the end of the test. Hence, the calculated IFR is relative to Alloy 600 materials that are highly susceptible to PWSCC, such as Alloy 600MA with low mill anneal temperature and a high degree of intragranular carbides. The IFR so calculated would be a more realistic improvement factor, somewhat less conservative than that by Equation 4-6. 2. The IFR calculated per Equation 4-8 is still considered conservative because the actual cracking time of the Alloy 600 specimens was less than the accumulated test time at inspection intervals. Often, the first Alloy 600 specimen cracking was observed at the first scheduled inspection interval. In addition, the actual cracking time of Alloy 690 specimens could be much longer than the test duration, if-in fact-they were ever to develop cracking at all. The resulting IFR values calculated by Equation 4-8 are listed in Table 4-2 and Table 4-4. In addition, the equivalent EFPYs at 315C (600F) for each test is also determined per Equation 4-8. This normalization has no affect on the IFR per Equation 4-8 except when the Alloy 600 and Alloy 690 specimens in the same study were tested at different temperatures. The average IFR of Alloy 690 relative to Alloy 600 in Table 4-2 and Table 4-4 per Equation 4-8 is 27.1, which is about the same as the 26.5 relative to Alloy 600MA from the Weibull analysis per Equation 4-6. Figure 4-11 plots the duration of the test for Alloy 690 vs. the IFR listed in Table 4-2 and Table 4-4. The IFR is seen to increase with increasing test time, indicating that the IFR is limited by the test duration.

4-18

PWSCC of Thin-Walled SG Tubing Table 4-4 Summary of Alloy 690 hydrogenated and doped hydrogenated steam stress corrosion test data to 2004
Ref. # in MRP-111 [2] Test Temp (F) Alloy 690 Heat Cond. Total Speci (a) No. Test Time at Test Temp. (Hour) 13824 12600 12600 9720
(c)

Test

Test Environ.

Alloy 690 Heat Number

Equation Test Time at (b) 600F (year) 107.4 97.9 97.9 237.4

Time to First Alloy 600 failure (hour)

Improve. Factor (IF)

690 (A) 52, 53 RUB H2 Steam 716 690 (B)

TT, TT TT+aged

1 1 1 3

25.0 552 22.8 22.8 336 28.9

(c)

Framatome, Germany (new data in [2])

RUB RUB surface scored H2 Steam 752 754380 TT

9720

237.4

336

28.9

(a) Total number of specimens of similar heat treatment and test condition and duration. (b) The equivalent test time at 600 F for Alloy 690 is calculated based a PWSCC crack initiation time of 50 kcal/mole (Equation 4). (c) Two of the three Alloy 690 specimens in Suis investigation cracked after 13824 hours of testing. To be conservative, it is assumed here that these two Alloy 690 specimens developed crack soon after the previous examination made after accumulating 12600 hours. None of the other Alloy 690 specimens in this table failed during the test duration.
o

160 140
Improvement Fact, IF

120 100 80 60 40 20 0 0 20,000 40,000 60,000 80,000 100,000 120,000

Stress Corrosion Test Duration, hour

Primary Water H2 Steam Doped H2 Steam H2 Steam, cracked Linear (Primary Water)

Figure 4-11 Improvement factors listed in Table 4-2 and Table 4-4 per Equation 4-8 versus test duration

4-19

PWSCC of Thin-Walled SG Tubing

Since MRP-111 was published in 2004, there have been at least two major additional studies of initiation behavior in Alloy 600 and 690 SG tubing. The first was carried out by Westinghouse in order to ascertain the possible effects on PWSCC of operating at a slightly higher pH value in the primary circuit [22]. A single heat of Alloy 690TT was included together with multiple heats of Alloy 600 and the testing was carried out under different chemical conditions with overall exposures of up to ~10,000 h. None of the Alloy 690 RUB specimens showed any signs of PWSCC in these tests, nor did two heats of Alloy 600TT (out of a total of 7 heats in all). Depending upon the exact Alloy 600 heat considered as a reference (and showing cracking), the IFR factors calculated according to Equation 4-8 would range from 2.6 to 13.3 vs. Alloy 600MA and 1.6 to 4.0 vs. Alloy 600TT in the most susceptible (pH 7.2) test environment. These factors appear low because of the relatively short test durations used. More recently, the experiences from Swedish RUB testing of Alloy 690 have been summarized [23]. The testing was performed in high-purity water with 50 ml/kg H2 at 365C as shown in Table 4-5. No cracking was observed in either the mill-annealed (MA) or thermally-treated (TT) conditions for testing times up to 33000 h, whereas Alloy 600 cracked under these conditions, even in the TT condition (see Figure 4-12 from [24]). In this case, the IFR factors calculated according to Equation 4-8 would be ~ 49 vs. Alloy 600MA and ~ 5 vs. Alloy 600TT.

Table 4-5 Swedish RUB testing for crack initiation in Alloys 600 and 690 [23]

4-20

PWSCC of Thin-Walled SG Tubing

Figure 4-12 Swedish RUB testing for crack initiation in Alloys 600 and 690 from [24]

In addition, some limited testing of Alloy 600 and Alloy 690 steam generator tubing material in supercritical water (SCW) was carried out by Westinghouse as a precursor to a major study of CGRs in these materials (see Section 5.1.2.1.8). This program, carried out for the EPRI MRP, was designed to accelerate the IGSCC process in SCC-resistant, Ni-base alloys by testing at elevated temperatures and pressures [25]. Specifically, this study involved testing at temperatures beyond the critical point of pressurized aqueous environments in an attempt to thermally accelerate corrosion and produce measurable effects from relatively minor SCC processes. High-pressure/temperature test conditions (385C/33 MPa) were chosen to have water-like fluid properties with additions of Li/B and H2. An initial series of crack initiation tests showed that SCC initiation occurred in one Alloy 600 heat in this supercritical water (SCW) environment after a period of only 200 h. Including data from similar testing at 325C suggested an activation energy for crack initiation of ~146 kJ/mole over the temperature range from 325C to 385C. This is less than the 209 kJ/mole often quoted for Alloy 600 in primary water, but consistent with the most recent KAPL data reported by Richey et al. [26]. No cracking was found in any of the 14 Alloy 690TT RUB specimens after 500 h.

4.2

Field Experience

Over the past thirty years, PWSCC of Alloy 600/182/82 material has become an increasingly significant problem in PWRs. As shown in Table 1-1, PWSCC was first detected in Alloy 600 steam generator tubing in 1971. The first report of PWSCC in a pressure boundary penetration was the leak from a pressurizer level nozzle in 1986. Both of these instances occurred after approximately 2 years of service. These initial reports have subsequently been followed by similar occurrences at other PWRs using Alloy 600/182/82 materials for the same component items. These problems have led to significantly increased inspection efforts, have contributed to the early replacement of many steam generators, have led to long forced outages to repair problems at some units, and have resulted in the expenditure of a significant amount of engineering effort to determine the root cause of the problems, develop inspection and repair methods, and develop strategic plans to manage the problems. 4-21

PWSCC of Thin-Walled SG Tubing

In addition to the initial occurrences of Alloy 600/182/82 PWSCC given in Table 1-1, the following steam generator tubing failures are highlighted: 1. Cracking of Obrigheim steam generator tubes (1971) after approximately 2 years of service [27]. This was the first instance of PWSCC of Alloy 600MA. 2. Cracking of Kori-2 steam generator tubes (1986) after nearly 3 years of service. This was the first occurrence of PWSCC in a steam generator fabricated with Alloy 600TT tubing [28]. 3. Cracking of steam generator mechanical tube plug at North Anna 1 (1989) after just over 3 years of service [29]. This was the first occurrence of PWSCC of Alloy 600TT plugging material. 4. Cracking was identified at Oconee Nuclear Station-1 (1995) after about 18 years of operation [30]. This was the first instance of PWSCC in stress relieved OTSG (Once-Through Steam Generator) tubing. It seems quite clear that a PWR steam generator is a very demanding material application, involving high temperature, high stress and strain at roll transitions and at tight U-bends, and a primary water environment that facilitates stress corrosion cracking (at a corrosion potential close to the potential of the Ni/NiO equilibrium). In addition, the number of SG tubes per PWR varies between 8,000 and 30,000. Hence, steam generator tubing is statistically favored to be the first component item in a PWR to show PWSCC. Operating experience indicates that a PWSCC-susceptible Alloy 600 material (low temperature mill annealed) can crack after approximately 2 years of operation in this application. Even improved Alloy 600 materials (high temperature mill annealed and thermally treated or high temperature mill annealed and stress relieved) may also be susceptible to cracking in as little as 3 to 8 years of operation at very highly stressed and/or cold-worked locations (such as tube bulges, kiss rolls, etc.). Nevertheless, Figure 4-13 (taken from reference [31]) shows that there has been a marked decrease in repairs to Alloy 600TT tubing in the last 2 years. As shown in Section 4.1, Alloy 690 material appears to be highly resistant to PWSCC when laboratory tested in conditions where Alloy 600 material exposed under the same conditions routinely showed comparatively rapid cracking. Also, as shown in Table 4-6, a number of steam generators manufactured with Alloy 690 tubing material have now been in service for significant times without crack indications due to PWSCC. This unblemished record with regard to the SCC resistance of Alloy 690 tubing has continued through 2008, even though a small number of repairs have been necessary for other reasons (see Figure 4-14). The oldest steam generators with Alloy 690 tubing have been operating now for approximately 18 years with no reported signs of PWSCC degradation. As of December 2008, a total of 89 PWR plants have replaced steam generators originally tubed with non-thermally-treated Alloy 600 and the vast majority of new staem generators have been tubed with Alloy 690TT.

4-22

PWSCC of Thin-Walled SG Tubing

Figure 4-13 Worldwide causes of Alloy 600TT SG tube repair by degradation mechanism from [31]

Figure 4-14 Worldwide causes of Alloy 690TT SG tube repair by degradation mechanism from [31]

4-23

PWSCC of Thin-Walled SG Tubing Table 4-6 List of operating steam generators manufactured with Alloy 690 tubing (as of December 2008) Installation Date 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Dampierre-1 Ohi-3 Penly-2 Millstone-2, North Anna-1, Beznau-1, Ohi-4 Mihama-2, Takahama-2, Genkai-1, V.C. Summer, Gravelines-1, Daya Bay-1, Daya Bay-2, Golfech-2, Genkai-3, Ikata-3 Ohi-1, Tihange-1, North Anna-2, Ringhals-3, Dampierre-3, St. Laurent-B1, Sizewell-B Mihama-1, Takahama-1, Doel-4, Catawba-1, Ginna, Gravelines-2 McGuire-1, McGuire-2, Point Beach-2, Mihama-3, Ohi-2, Tricastin-2, Genkai-4 Byron-1, St. Lucie-1, Tihange-3, Braidwood-1, Kori-1, Ikata-1, Tricastin-1 Beznau-2 Farley-1, D.C. Cook-1, Krsko, South Texas Project-1, Arkansas Nuclear One-2, Chooz-B1, Civaux-1, Civaux-2, Gravelines-4 Genkai-2, Ikata-2, Kewaunee, Shearon Harris, Farley-2, Tihange-2 Tricastin-3, Fessenheim-1, Chooz-B2, Calvert Cliffs-1, South Texas Project-2 Calvert Cliffs-2, Palo Verde-2, Sequoyah-1, Oconee-1, St.-Laurent Des Eaux B Tricastin 4, Oconee 2, Prairie Island 1, Oconee 3, Dampierre 2, Callaway, ANO 1 Beaver Valley 1, Fort Calhoun, Watts Bar 1 Commanche Peak 1, St. Lucie 2 Diablo Canyon 2 PWR D.C. Cook-2, Indian Point-3, Ringhals-2

Some other replacement and original equipment component items utilize relatively thin-walled Alloy 690 material. E.g., the replacement pressurizer heater sleeves at Calvert Cliffs-2 have been in operation for about 12 years and visual inspections have not identified any problems to date. Obviously, a number of additional improvements to design and fabrication practice have also been made to reduce the likelihood of initiating PWSCC with thin-walled Alloy 690 material. These include optimizing component item construction practices, optimizing material melting and manufacturing practices, and reducing residual stresses. However, the excellent operating experience to date provides assurance that Alloy 690TT is highly resistant to PWSCC. The overall factor of improvement relative to Alloy 600 MA was estimated in 2006 from field data as >20x [3]. However, it is clear that this number is actually very conservative and efforts are ongoing to justify higher values [32].

4-24

PWSCC of Thin-Walled SG Tubing

4.3

References

1. W. Bamford and J. Hall, Cracking of Alloy 600 Nozzles and Welds in PWRs: Review of Cracking Events and Repair Service Experience, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp. 959 to 966. 2. H. Xu et al., Materials Reliability Program (MRP), Resistance to Primary Water Stress Corrosion Cracking of Alloys 690, 52, and 152 in Pressurized Water Reactors (MRP-111). EPRI, Palo Alto, CA, 2004. 1009801. 3. C. Marks, Alloy 690 Improvement Factor Update: Application of Improvement Factor Data to the Analysis of a Secondary System Chemistry Upset at Ginna. EPRI, Palo Alto, CA and Constellation Energy Group, Inc., Baltimore, MD: 2006. 1013640. 4. A.J. Sedricks, J.W. Schultz, and M.A. Cordovi, Inconel Alloy 690 A New Corrosion Resistant Material, Corrosion Engineering (Boshoku Gijutsu), vol. 28, pp. 82-95, 1979, Japan Society of Corrosion Engineering. 5. G.P. Airey, The Stress Corrosion Cracking (SCC) Performance of Inconel Alloy 600 in Pure and Primary Water Environments, Proceedings of the (1st) International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, pp. 462-478, NACE, 1983. 6. Nakayama, H. Tomari, K. Fujiwara, K. Shimogori, H. Hamada, and K. Takaishi, GC/IGSCC and General Corrosion Behavior of Alloy 800 As a PWR S/G Tube Material, Corrosion 87, Paper No. 82, NACE, March 1987. 7. A. Smith and R. Stratton, Relationship between Composition, Microstructure and Corrosion Behavior of Alloy 690 Steam Generator Tubing for PWR Systems, Proceedings of 4th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Jekyll Island, GA, 1989. pp.5-33 to 5-46. 8. A. Smith and R. Stratton, Thermal Treatment, Grain Boundary Composition and Intergranular Attack Resistance of Alloy 690, Proceedings of 5th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1991. pp. 855-860. 9. T.M. Angeliu, J.K. Sung, and G.S. Was, The Role of Carbon and Chromium on the Mechanical and Oxidation Behavior of Nickel-Based Alloys in High Temperature Water, Proceedings of 5th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Monterey, CA, 1991. pp. 475-481. 10. J.K. Sung and G.S. Was, The role of Grain Boundary Chemistry in Pure Water Intergranular Stress Corrosion Cracking of Ni-16Cr-Fe Alloys, Proceedings of 4th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Jekyll Island, GA, 1989. pp.6-25 to 6-37. 11. J.M. Boursier, F. Vaillant, P. Saulay, Y. Brechet, and G. Zacharie, Effect of the Strain Rate on the Stress Corrosion Cracking in High Temperature Primary Water: Comparison Between the Alloys 690 and 600, Proceedings of the 11th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, American Nuclear Society, 2003. 4-25

PWSCC of Thin-Walled SG Tubing

12. Materials Reliability Program Crack Growth Rates for Evaluating Primary Water Stress Corrosion Cracking (PWSCC) of Thick-Wall Alloy 600 Materials (MPR-55) Revision 1. Final Report, November 2002. EPRI Report 1006695. 13. Crack Growth of Alloy 182 Weld Metal in PWR Environments. Final Report, January 1999. EPRI Report TR-111993. 14. G. Sui, G.B. Heys, and J. Congleton, Stress Corrosion Cracking of Alloy 600 and Alloy 690 in Hydrogen/Steam and Primary Water Side Water, Proceedings of 8th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Amelia Island, FL, 1997. pp. 274 to 281. 15. G. Sui, J.M. Titchmarsh, G.B. Heys, and J. Congleton, Stress Corrosion Cracking of Alloy 600 and Alloy 690 in Hydrogen/Steam at 380oC, Corrosion Science, Vol. 39, No.3, pp. 565-587, 1997. 16. The New Weibull Handbook, 2nd Ed., Dr. Robert .B. Abernethy, Author and Publisher, July 1996. 17. K. Norring, J. Engstrom, and P. Norberg, Intergranular Stress Corrosion Cracking in Steam Generator Tubing, Testing of Alloy 690 and Alloy 600 Tubes, Proceedings of 3rd International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Traverse City, MI, 1987. pp. 587-593. 18. F. Vaillant, EDF Report HT-44/95/013/A, 1996, Rsistance a la corrosion sous contrainte en milieu primaire des alliages 690 et 800 Point des rsultats en Dcembre 1995, (English translation of the title, Resistance of Alloys 690 and 800 to Stress Corrosion Cracking in PWR Primary Water Status of Results Available to December 1995). 19. F. Vaillant et al., Influence of chromium content and microstructure on creep and PWSCC resistance of nickel base alloys, 9th Int Symp on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Newport Beach (CA), August 1999, Published by TMS. 20. F. Vaillant et al., Assessment of PWSCC Resistance of Alloy 690: Overview of Laboratory Results and Field Experience. EPRI Workshop Alloy 600, Santa Ana Pueblo (NM), March 2005. 21. F. Vaillant et al., Assessment of PWSCC resistance of Alloy 690: an overview of laboratory results and field experience EPRI MRP PWSCC Expert Panel Meeting, Saint Petersburg, FL, November 2007. 22. R.J. Jacko and R.E. Gold, Crack Initiation in Alloy 600 SG Tubing in Elevated pH PWR Primary Water, Proceedings of 12th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, TMS, 2005, pp. 925-936. 23. K. Norring and P. Efsing, Influence of Material Parameters on Initiation of PWSCC in Nickel Base Alloys in Primary PWR Environment, Workshop on Detection, Avoidance, Mechanisms, Modeling, and Prediction of Stress Corrosion Cracking Initiation in WaterCooled Nuclear Plants. EPRI, Palo Alto, CA; TEPCO R&D Center, Yokohama, Japan; AREVA NP, Technical Center, Le Creusot, France; Institut de Radioprotection et de Sret Nuclaire (IRSN), Fontenay-aux-Roses, France; Institute of Nuclear Safety System, Incorporated (INSS), Fukui, Japan; The Materials Aging Institute (MAI), Moret Sur Loing Cedex, France; and EDF R&D, Moret Sur Loing Cedex, France: 2009. 1018908. 4-26

PWSCC of Thin-Walled SG Tubing

24. P. Efsing, Alloy 690 Issue from a Utility Perspective, EPRI MRP PWSCC Expert Panel Meeting, St. Petersburg, Florida (Nov. 2007). See also K. Norring and J. Engstrom, Initiation of SCC in nickel base alloys in primary PWR environment: studies at Studsvik since mid 1980s, Energy Materials Vol. 3 (2008) No. 2, pp. 113-118. 25. R. Jacko, Materials Reliability Program: Testing the Resistance to Stress Corrosion Cracking of Alloy 690 and its Weld Metal in Supercritical Boron/Lithium/H2 Solutions (MRP-225). EPRI, Palo Alto, CA: 2007. 1015004. 26. E. Richey, D.S. Morton and R.A. Etien: SCC Initiation Testing of Nickel-Based Alloys in High Temperature Water, Proc. 13th Int. Conf. On Environmental Degradation of Materials in Nuclear Power System Water Reactors, CNS (The Canadian Nuclear Society), 2007. 27. Steam Generator Reference Book. December 1994. EPRI TR-103824s-V1R1. 28. Experience of U.S. and International Steam Generators with Alloy 600TT and Alloy 690TT Tubes and Sleeves. 2002. EPRI Document No. 1003589. 29. NRC Bulletin No. 89-01: Failure of Westinghouse Steam Generator Tube Mechanical Plugs, U.S. Nuclear Regulatory Commission, Washington, D.C. May 15, 1989. 30. Experience of U.S. and International Steam Generators with Alloy 600TT and Alloy 690TT Tubes and Sleeves. 2002. EPRI Document No. 1003589. 31. T. Kaul, Steam Generator Management Program: Steam Generator Progress Report: Revision 16. EPRI, Palo Alto, CA: 2008. 1016561. 32. J. Benson, Research on Material Improvement Factors for PWSCC Resistance in Steam Generator Tubes. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008).

4-27

5
PWSCC OF THICK-WALLED ALLOY 690 MATERIAL
The predominant use of thick-walled Alloy 690 material as a replacement for Alloy 600 has been in RPV head penetrations, the vast majority of which are CRDM or CEDM nozzles. The first instance of this followed PWSCC cracking and coolant leakage at the Bugey 3 PWR in France in 1992 and led to the replacement of vessel heads at 33 further plants there up to 2000 (see [1]). In contrast, the first U.S. plant to undergo RPV head replacement following PWSCC problems was North Anna 2 in 2002, but the pace picked up rapidly after that and nearly 30 plants had been equipped with new heads containing Alloy 690 penetrations by the end of 2008 (see Table 5-1). As mentioned in Section 2.1.2 and discussed in more detail in reference [2], a variety of melting and manufacturing steps have been used to produce extruded pipe (with a typical wall thickness of ~25 mm) that has usually then been subjected to a subsequent thermal anneal (Alloy 690TT), although some cold straightening may also have been carried out as a final step.
Table 5-1 Examples of some replaced PWR RPV heads with CRDM penetrations in Alloy 690 In-Service Date 1992 1994 1995 1996 1997 1998 1999 2000 2002 2003 2004 2005 2006 2007 Bugey 3 Bugey 5, Blaiyais 1, Gravelines 4, Bugey 2 Blayais 2, St. Alban 1, Flamanville 1, Gravelines 3, Blaiyais 3, Tricastin 1 Tricastin 4, Paluel 4, St. Laurent B2, Blaiyais 4, Dampierre 1, Fessenheim 1, St. Alban 2 Bugey 4, Dampierre 2, Dampierre 4, Belleville 2, Cruas 4, Gravelines 5 Flamanville 2, Dampierre 3, Paluel 3, Cattenom 2, Fessenheim 2, Cruas 2 Chooz B1, Chooz B2 Civaux 1, Civaux 2 North Anna 2 North Anna 1 and 2, Crystal River 3, Ginna 1, Oconee 1 and 3, Surry 1 and 2, Three Mile Island 1 Farley 1, Kewaunee 1, Oconee 2, Turkey Point 3 ANO 1, Farley 2, Millstone 2, Point Beach 1 and 2, Prairie Island 2, Robinson 2, Salem 1 and 2, St. Lucie 1, Turkey Point 4 Beaver Valley 1, Calvert Cliffs 1, D.C. Cook 1, Fort Calhoun 1, Prairie Island 1, Calvert Cliffs 2, Commanche Peak 1, D.C. Cook 2, St. Lucie 2, PWR

5-1

PWSCC of Thick-Walled Alloy 690 Material

Apart from CRDM penetrations, relatively thick-walled Alloy 690 has been used in number of other Alloy 600 component replacements in U.S. plants, as indicated in Table 5-2.
Table 5-2 Examples of some relatively thick-walled Alloy 690 reactor coolant system original equipment or replacement component items other than CRDM penetrations Component Item In-Service Date 1990 1992 1993 Pressurizer Components (e.g., Instrument Nozzles, etc.) 1994 1995 1997 1998 1999 2001 2005 2006 RV Lower Head BMI Nozzles 2000 2003 PWR Calvert Cliffs 2 Palo Verde 1,San Onofre 3, San Onofre 2, Palo Verde 1 St. Lucie 2, Palo Verde 2 Palo Verde 3, St. Lucie 2 San Onofre 3, St. Lucie 1 San Onofre 2, San Onofre 3 Calvert Cliffs 2, San Onofre 3 San Onofre 3 Waterford 3 St. Lucie 1 Millstone 2, ANO 2, Fort Calhoun Civaux 1, Civaux 2 South Texas 1 Alloy 690 Tubing and Alloy 152/52 Weld Alloy 690 Tubing and Alloy 52 Weld Alloy 690 Tubing and Alloy 52 Weld Alloy 690 Tubing Material

5.1

Laboratory Testing

Despite the widespread early use of thick-walled Alloy 690 material for CRDM penetrations, only one pre-production and one commercial heat of this product form were apparently tested early on in France (by EdF [3] see Table 5-3). No indications were given that the PWSCC behavior of specimens taken from these differed in any way from that of the majority of thin-walled SG tubing materials examined previously (see Section 4.1).

5-2

PWSCC of Thick-Walled Alloy 690 Material

Table 5-3 Origins and heat treatments of Alloy 690 CRDM nozzles tested by EdF [3]
Solution Anneal and Heat Treatment Temperatures

CRDM Nozzles

Heat Manufacturer Process Number

Extrusion Tube Temperature Maker (C)

Final Finish

Diameter/ Thickness (mm)

Experimental Tecphy

ESR

WJ 151

Valinox 1110-1230

- Prestraighten in a press 1 1050 + 5 h x715C 109/21,5 straightenin g pass with hyperbolic rolls - Prestraighten in a press 2 1080+ 5 h x 715C 110/23 straightenin g passes with VALTI rolls

Industrial

Tecphy

ESR

WJ 172

Valinox 1110-1210

With the exception of the Alloy 690 plate material used to simulate a SG divider plate in corrosion fatigue and SCC tests of various welds (see Section 3.2), no other tests on thick-walled Alloy 690 material appear to have been carried out prior to publication of the original MRP-111 report in 2004. That situation has changed rapidly, however, in the last 5 years, as discussed below. 5.1.1 Crack Initiation Studies 5.1.1.1 Testing in Simulated Primary Water by Mitsubishi Heavy Industries (MHI)

Initiation testing at MHI using uniaxial tensile specimens under constant, active load has been carried out in simulated primary water at 360C for many years using the materials shown in Table 5-4 and Table 5-5. The environmental test conditions are shown in Table 5-6 and details of the test system, including specimen form and loading mechanism, are shown in Figure 5-1 to Figure 5-3. Key results were presented by Asada et al. in June 2007 [4] and November 2007 [5]. Figure 5-4 shows the apparent absence of crack initiation after 58,000 h in Alloy 690TT taken from a bottom mounted instrumentation nozzle and Figure 5-5 shows similar good behavior after 73,000 h for Alloy 690TT taken from a CRDM nozzle. Note, however, that it has not been possible to confirm whether or not these specimens have actually been examined yet for freedom from incipient cracks. I.e., the data shown in the above figures apparently refer to the absence of failure in any of the Alloy 690TT tensile specimens, many of which are still thought to be on test.

5-3

PWSCC of Thick-Walled Alloy 690 Material

Considering the highest stress level at which a direct comparison can appropriately be made, the minimum improvement factors calculated according to Equation 4-8 would be ~48 and ~30 vs. Alloy 600MA for the BMI and the CRDM materials, respectively. Note, however, that these numbers are undoubtedly conservative and expected to rise with increased testing time.
Table 5-4 Chemical composition of materials used in MHI testing for PWSCC initiation [4, 5]

Table 5-5 Heat treatment and mechanical properties of materials used in MHI testing for PWSCC initiation [4, 5]

5-4

PWSCC of Thick-Walled Alloy 690 Material Table 5-6 Environmental test conditions used in MHI testing for PWSCC initiation [4, 5]

Figure 5-1 MHI test loop for uni-axial constant load studies of PWSCC initiation [4, 5]

5-5

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-2 Active loading mechanism for uni-axial constant load studies of PWSCC initiation [4, 5]

Figure 5-3 Test specimens for uni-axial constant load studies of PWSCC initiation [4, 5]

5-6

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-4 Dependence of PWSCC initiation in Alloy 600MA on applied stress and lack of cracking in 690TT BMI material after 58,000 hours of testing [4, 5]

Figure 5-5 Dependence of PWSCC initiation in Alloy 600MA on applied stress and lack of cracking in 690TT CRDM nozzle material after 73,000 hours of testing [4, 5]

5-7

PWSCC of Thick-Walled Alloy 690 Material

5.1.1.2

Testing in Pure Supercritical Water at the University of Michigan

Was and Teysseyre reported in 2005 on an extensive test program, sponsored by the U.S. Department of Energy, to qualify various austenitic alloys for service in the Gen IV SCWR [6]. This included quasi CGR dataw on Alloy 690, derived from cross-sections of CERT specimens after testing at a slow strain rate in pure, de-aerated supercritical water (SCW) at temperatures between 400 and 550C (see Figure 5-6). Alloy 690 showed only very marginal SCC susceptibility under these severe test conditions until reaching temperatures of 500C and above. Even then, it performed better than stainless steels and approximately as well as Alloy 625. In conjunction with the Westinghouse test program in SCW described at the end of Section 4.1.7 and in Section 5.1.2.1.8, a small, amount of focused testing at the University of Michigan was x sponsored by EPRI to investigate the possible cause of differences between the two SCW studies [7]. The approach used involved CERT of tensile specimens machined from several Alloy 690 materials (including one tested in the Westinghouse program) and an Alloy 600 control material in SCW at two temperatures and pressures. The data obtained (see Table 5-7) showed that all alloys were more susceptible to cracking in SCW at 400C and a pressure of 25.4 MPa than in a lower temperature (385C) environment at higher pressure (26.7 MPa). Alloy 600 was clearly SCC susceptible (Figure 5-7); the very marginal cracking susceptibility of Alloy 690 (Figure 5-8) was similar in the three material conditions tested. The decrease in cracking at 385C appeared to be predominantly due to the drop in temperature; i.e., the water density had little effect on the extent of cracking.

Figure 5-6 CGRs derived from a CERT study of various austenitic alloys in pure, de-aerated SCW [6]

Note that attempts to derive CGRs from this type of testing are controversial and subject to different interpretations. Given the small crack depths involved for Alloy 690, these were clearly related more to initiation than to crack growth. Higher cracking susceptibility in the Westinghouse CGR tests, which used additions of boron and lithium to SCW.

5-8

PWSCC of Thick-Walled Alloy 690 Material Table 5-7 Summary of maximum crack depths measured on cross-sectioned samples at the University of Michigan [7] Alloy UM-690 Test Condition 400C 385C UM-TT690 400C 385C EPRI-TT690 400C 385C EPRI-600 400C 385C Max Crack Depth Observed (m) 4.4 3.5 7.0 3.5 5.3 3.4 169 17.6 Number of Cracks Measured 3 2 3 1 2 2 2 2 Number of Cracks Seen >20 <5 >20 <5 >10 <5 >40 >40

Figure 5-7 Cross-section of EPRI Alloy 600 after testing in 400C/25.4MPa deaerated pure SCW [7]

5-9

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-8 Cross-section of EPRI Alloy 690 after testing in 400C/25.4 MPa deaerated pure SCW [7]

5.1.2 Crack Growth Rate Studies The approach taken in the 2004 MRP-111 report was to analyze the data from Alloy 690 laboratory studies according to the type of test environment that had been used and this scheme was retained in the 2008 description of ongoing research in this area prepared for the MRP by the present author [8]. It is increasingly apparent, however, that the product form and detailed thermo-mechanical processing of Alloy 690 material can have a profound influence on its PWSCC behavior. Accordingly, the CGR studies that have now been carried out (and are, in some cases, still ongoing) are reported here according to the type of thick-walled Alloy 690 material that was used. Since the majority of initial CGR tests were performed on plate materialy, this is dealt with first, even though it is of far less relevance in the field than the piping material used for CRDM penetrations. These are dealt with next, whereby a distinction is made between the base material that would actually be delivered to the manufacturer of a replacement RPV head (see Section 2.1.2) and similar material that has been deliberately subjected to additional cold work in the laboratory before testing. Finally, consideration is given to the expected behavior of the heat affected zones (HAZ) formed adjacent to the fusion line in thick-walled Alloy 690 components that have undergone welding. 5.1.2.1 Testing of Alloy 690 Material not directly related to PWR Components

The plate and other materials widely used to date in CGR studies of Alloy 690 are not generally thought to be representative of field components, but represent either a standard alloy from metal vendors, or more seldom an experimental heat of material.

Later also on bar material.

5-10

PWSCC of Thick-Walled Alloy 690 Material

5.1.2.1.1 Feasibility Studies in Simulated Primary Water by General Electric Global Research (GE-GRC)

GE-GRC has been working on the PWSCC behavior of Alloys 690/152/52 for the EPRI MRP program since 2003, initially by carrying out a feasibility study of the application of sophisticated CGR testing methods to study such resistant alloys [9]. Product form The feasibility testing was carried out on 2 Alloy 690 plates supplied by EPRI (Heat NX8244 HK11 with the chemical composition and the mechanical properties shown in Table 5-8), one with a final mill anneal (MA) at 982C (1800F) and the other at 1093C (2000F). The low-temperature MA produced a rather heterogeneous grain size, together with a lower level of grain-boundary precipitates and significant compositional banding (mainly of carbides). Details of the extensive banding in this plate material have recently been reported by Morra et al. [10]. The high-temperature MA resulted in a more uniform microstructure and a higher density of grain-boundary carbides. Both plates were subsequently subjected to uniform cold work (by cross-forging at RT) to a level of approx. 25%. This was intended both to raise residual SCC susceptibility and to cover the maximum level of deformation expected in the HAZ immediately adjacent to the fusion line after weldingz.
Table 5-8 Chemical composition and mechanical properties of all the Alloy 690 materials tested at GE-GRC in the MRP program Source EPRI Plate 982C MA Duke CRDM ANL Plate GRC bar Other Heat # NX8244 HK11 Cr 30.0 Fe 9.20 Al 0.36 Ti 0.20 Mn 0.20 Si 0.14 C 0.018 S <0.001 Cu RT YS MPa 242aa 210bb RT UTS MPa 656aa 595bb

WN415 NX3297HK12 B25K NX2865HK

29.1 29.5 29.3 29.3

10.1 9.9 9.21 9.99 0.26 0.37

0.31 0.20 0.22 0.16

0.29 0.07 0.06 0.03

0.018 0.03 0.034 0.03

<0.001 <0.001 <0.001 <0.001

0.007 0.01 <0.01 0.01

260

514

293 292

699 688

Specimen/crack-growth orientation Compact tension type (0.5T CT) specimens, with 5% side grooves on each side, were machined from the cross-forged plates in the L-T orientation (see Figure 5-9). The crack plane was thus perpendicular to the banded microstructure.

z aa

Note, however, that such deformation in the HAZ is likely to be more uni-directional than homogeneous. For material with MA at 982 C. For material with MA at 1093 C.

bb

5-11

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-9 Convention with regard to specimen orientation and the principle axis of cold-working in Alloy 690 base materialcc

Testing details The MRP program at GE-GRC uses simulated primary water at 340 or 360C with the hydrogen level chosen at 18 and 26 cc/kg, respectively, so as to be around the Ni/NiO phase stability transition. Particular attention is paid to transitioning from the transgranular air fatigue pre-crack to a uniform (and possibly intergranular) SCC crack front by gentle cycling with increasing R ratio and hold time, before attempting to impose constant KI conditions (generally at ~27.5 MPam). Data evaluation The test program relies on highly accurate DCPD measurements of on-line crack depth to assess CGRs in the various phases of testing, with a plausibility check being provided by a comparison of cumulative crack increments from on-line data and the overall length of the final crack on the fracture surface after breaking open the CT specimens, often after extended test periods of many thousands of hours. The feasibility testing described in reference [9] showed stable, sustained crack growth albeit -8 at very low rates of < 10 mm/s over long periods of time at constant K in the 25% cold-worked plates tested at 340C (see Figure 5-10 and Figure 5-11). The highest CGR at ~ 0.25 mm/year was assessed to be almost two orders of magnitude lower than that expected for Alloy 600 under comparable conditions and thus of little or no engineering significance. The crack morphology was primarily transgranular over the very limited increments of crack advance obtained (see Figure 5-12) and no apparent difference was found between low and high temperature mill annealing of the starting material.
cc

Rolling is done in the longitudinal (L) direction, with reduction occurring in the short transverse (S) direction and some broadening in the width (T) direction. Forging reduces the S direction, but the T and L directions are indistinguishable.

5-12

PWSCC of Thick-Walled Alloy 690 Material

SCC#2 - c248 - 690, 25%RA, NX8244HK111, 1800F Anneal

11.38 0.4

11.37

Outlet conductivity x 0.01

2.4 x 10-9 mm/s

0.2

To R=0.7, 0.001 Hz + 85,400s hold @ 665h

To R=0.7, 0.001 Hz + 9000s hold @ 506h

11.36

To Constant K @ 1757h

Crack length, mm

5 x 10-9 mm/s

11.35

END OF TEST @ 5135h

-0.2

11.34

-0.4

11.33

NE Power Outage @ 1247h

At 340C, pH = 7.60. At 300C, pH = 5.93 and potential would be ~155 mV higher

11.32

c248 - 0.5TCT of 690 + 25%RA, 340C 25 ksiin, 550 B / 1.1 Li, 18 cc/kg H2
Pt potential
2000 2500 3000 3500 4000

-0.6

11.31

1.4 x 10-8 mm/s

1000

-0.8

CT potential
-1 4500 5000

11.3 500

1500

Test Time, hours

Figure 5-10 CGR response of cold-worked Alloy 690 plate (with low-temperature mill anneal) tested at GE-GRC for >3000 h at constant stress intensity [9]

Conductivity, S/cm or Potential, V she

5-13

PWSCC of Thick-Walled Alloy 690 Material

SCC#2 - c249 - 690, 20%RA, NX8244HK112, 2000F Anneal
11.43 0.4

Outlet conductivity x 0.01

11.41 0.2

END OF TEST @ 5135h

Crack length, mm

11.39

11.37

8 x 10-9 mm/s At 340C, pH = 7.60. At 300C, pH = 6.93 and potential would be ~155 mV higher

-0.2

-0.4

11.35

11.33

1.4 x 10-8 mm/s

To Constant K @ 1757h

c249 - 0.5TCT of 690 + 20%RA, 340C 25 ksiin, 550 B / 1.1 Li, 18 cc/kg H2
Pt potential
2000 2500 3000 3500 4000

-0.6

-0.8

CT potential
-1 4500 5000

11.31 500

1000

1500

Test Time, hours

Figure 5-11 CGR response of cold-worked Alloy 690 plate (with high-temperature mill anneal) tested at GE-GRC for >3000 h at constant stress intensity [9]

5-14

Conductivity, S/cm or Potential, V she

3.7 x 10-9 mm/s

To R=0.7, 0.001 Hz + 9000s hold @ 506h

To R=0.7, 0.001 Hz + 85,400s hold @ 665h NE Power Outage @ 1247h

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-12 Predominantly transgranular morphology (but with some intergranular facets) within band of PWSCC crack growth in cold-worked Alloy 690 plate initially tested at GE-GRC [9]

5.1.2.1.2

Investigations in Simulated Primary Water at Argonne National Laboratory (ANL)

Data from the ANL test program, sponsored by NRC Research, has been reported at successive EPRI Expert Panel meetings [11, 12, 13] as well as in the context of public NRC review meetings [14, 15]. Product form A 5.4 cm (2.125 in) thick, VIM-ESR plate (Heat NX3297HK12, supplied by ATI Wah Chang, with the chemical composition shown in Table 5-9), annealed for 2h at 1038C and apparently subjected to a subsequent laboratory thermal stabilization treatment [15] at 720C for 10h, was uni-directionally cold-rolled at Special Metals in 3 passes (10, 8 and 8%) to achieve ~26% overall reduction in thickness. The deformation was inhomogeneous, with the near-surface region of the 1.5 in thick final plate being harder than the mid-plane by 12%.
Table 5-9 Chemical composition of Alloy 690TT plate material tested at ANL [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5-15

PWSCC of Thick-Walled Alloy 690 Material

Specimen/crack-growth orientation 1/2T CT specimens were removed as shown in Figure 5-13 and have been tested in both the S-L and S-T orientations, with similar crack growth rates being attained. Testing details The test environment is simulated primary water at 320C, and the chosen H2 level (2 ppm = ~22 cc/kg) is in the region of Ni metal stability, i.e. above the expected level for peak CGR in Alloy 600. Pre-cracking was carried out by fatigue at temperature in the autoclavedd, with a gradual change (via increasing R-ratio and sawtooth loading with longer rise times) to approximately constant mechanical load. ANL analyze these cyclic data according to the superimposition model of Shack and Kassner [16] and consider clear acceleration of cyclic crack growth above the so-called true corrosion fatigue line (CF) as frequency is lowered to be a reliable indicator of subsequent IGSCC behavior. In the case of cold-rolled Alloy 690 tested at 320C, the degree of environmental enhancement observed was actually above the best-fit curve for previous ANL tests on Alloy 600 (see Figure 3-9). During subsequent testing at constant load (at a KI value of 26.7 MPam), steady crack growth was seen in the specimen with S-L orientation. After some 2000h, a brief period of fatigue loading was used to increase the crack length. The final constant load test period at a KI value of 28.0 MPam lasted ~600 h (see Figure 5-14). The crack advance during the two constant load test periods amounted to approx. 130 and 70 m, respectively.

17

ng olli r

n ctio dire

Approx 1.5

12
SL ST
3.5

34

Figure 5-13 Details of sample removal from cold-rolled Alloy 690TT plate tested at ANL [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

dd

ANL regard this step in the procedure as an essential part of successful transitioning from transgranular cracking to IGSCC.

5-16

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-14 On-line data from ANL testing [11 to 15] of cold-rolled Alloy 690TT plate (S-L specimen orientation)
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

For the S-T specimen orientation, steady crack growth of around 150 m was obtained over nearly 1500 h under constant load at a KI value of 31.0 MPam (see Figure 5-15).

Figure 5-15 On-line data from ANL testing [11 to 15] of cold-rolled Alloy 690TT plate (S-T specimen orientation)
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5-17

PWSCC of Thick-Walled Alloy 690 Material

Data evaluation The fracture surface for the S-L specimen orientation showed fairly uniform crack advance overall and predominantly intergranular cracking during the constant load periods (see Figure 5-16), whereas the macroscopic crack front appears somewhat more irregular for the S-T orientation (see Figure 5-17).

Figure 5-16 Macro- and microfractography for the ANL 690TT plate specimen with S-L orientation [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5-18

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-17 Macrofractography for the ANL 690TT plate specimen with S-T orientation [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

The following representative PWSCC CGRs were derived (apparently without any correction of the on-line DCPD data from investigation of the fracture surfaces): 1.5 E-8 mm/s at KI = 26.7 MPam and 2.8 E-8 mm/s at 28.0 MPam for the S-L specimen; 3.3 E-8 mm/s at KI = 31 MPam for the S-T specimen. For comparison purposes, the MRP-55 disposition curve (75th percentile of the data on a heat by heat basis) would predict a CGR of ~ 1E -7 mm/s for Alloy 600 at KI = 31 MPam. Thus the ANL study could be interpreted to predict a factor of improvement of only ~3x for heavily cold-worked th Alloy 690 material on this basis, or even no improvement at all if a comparison is made with the 50 percentile of the Alloy 600 database. During testing of a second S-T specimen, the temperature was changed several times within the range 320 to 300C without any apparent effect on the CGR as measured online (see Figure 5-18).

5-19

PWSCC of Thick-Walled Alloy 690 Material

13.20 60
27

temperature back to 320C, re-initialized

Alloy 690 Specimen # A690WC-ST-2 320C, Simulated PWR Water 18 temperature 310C, re-initialized
25 24c 8h

13.10 Crack Length (mm)

24 24 a 2h 24 b 4h

26 2.5 x 1011 m/s 26 MPa m0.5 CL w/ periodic unloading

50 Kmax (MPa m0.5)

13.00
2.8 x 1011

40
temperature to 300C, re-in itializ ed

m/s

12.90

26 MPa m0.5 Periodic Unloading

Crack Length

30

12.80

Kmax

20

320C
12.70 3000 3500

310C
4000 Time (h)

320C
4500

300C
5000 10

Figure 5-18 ANL test showing that the CGR behavior for cold-worked Alloy 690TT plate was unaffected by changing temperature within the range 320 to 300C [13]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5.1.2.1.3 Additional Studies in Simulated Primary Water by General Electric Global Research (GE-GRC) as part of the MRP Test Program

Testing of Alloy 690 has been carried out after uni-directional (1D) rolling, in one case of the same plate used at ANL (heat NX3297HK12 with approx. 26% cold work, see Section 5.1.2.1.2) and in another of Alloy 690 billet material (heat B25K) supplied by GE-GRC and cold rolled by 20% [17, 18]. A further heat (NX2865HK) of nuclear-grade Alloy 690 from another source has also been tested after undergoing ~19% 1D rolling. The composition and mechanical properties of these materials are shown in Table 5-8 and the basic test procedures used at GE-GRC have already been described in Section 5.1.2.1.1. The first test at GE-GRC of one-dimensionally cold rolled Alloy 690 plate supplied by ANL produced even higheree CGRs (~ 4E-7 mm/s) than those reported by ANL in Section 5.1.2.1.2. Crack advance was seen easily at a constant KI of 27.5 MPam after 700 h test time at 360C and some 0.7 mm of growth then occurred over less than 1000 hours (see Figure 5-19).

ee

Albeit at a higher temperature.

5-20

PWSCC of Thick-Walled Alloy 690 Material

SCC#5 - c372 - Alloy 690, 26%RA 1D, NX3297HK12, ANL

13 0.4

12.9

c372 - 0.5TCT of 690 + 26%RA 1D, 360C 25 ksiin, 600 B / 1 Li, 26 cc/kg H2
Outlet conductivity x 0.01

4 x 10 mm/s

-7

0.2

12.8

0 12.7

12.6

To Constant K @ 647h

~0.7 mm of growth at constant K

To Constant Load @ 1339h

Crack length, mm

3 x 10-7 mm/s

-0.2

-0.4

12.5

-0.6

12.4

12.3

Est. pH at 360C = 8.2 used for c At 340C, pH = 7.53. At 300C, pH = 6.86 Pt potential CT potential

-0.8

12.2 650

-1 750 850 950 1050 1150 1250 1350 1450

Test Time, hours

Figure 5-19 Rapid crack growth in 1D cold-rolled (~26%) Alloy 690 plate supplied by ANL and tested at GE-GRC [17, 18]

After advancing the crack some 1.2 mm in fatigue, a second period of SCC crack growth at constant K gave CGRs that were initially somewhat lower, but gradually increased to ~ 2E-7 mm/s, at which point the test temperature was dropped, first to 325C and then to 290C. These changes ff produced no long-term reduction whatsoever in CGR , but indications of stepwise growth (see Figure 5-20). The macroscopic crack front became less uniform at the lower temperatures (lighter areas in Figure 5-21). The observed high CGRs were duplicated in a separate, repeat test on this ANL plate material. This time, however, instead of changing temperature, the hydrogen concentration in the simulated primary water was increased from 26 to 80 cc/kg over a period of 1500 h (see Figure 5-22). This produced a slight increase in CGR, whereas such a change would have been expected to reduce CGR in Alloy 600 undergoing PWSCC.

ff

For Alloy 600, the expected decrease in CGR upon this reduction in temperature would have been more than one order of magnitude.

Conductivity, S/cm or Potential, Vshe

5-21

PWSCC of Thick-Walled Alloy 690 Material

SCC#7a - c372 - Alloy 690, 26%RA 1D, S-L Orientation, NX3297HK12, ANL
370 15.8

Temperature
360

15.6

To 325C @ 2821h

2.2 x 10-7 mm/s

350

To R=0.7, 0.001 Hz + 9000s hold @ 1540h

15.4

340

Crack length, mm

To Constant K @ 1635h

15.2

330

To 290C @ 3473h

320

15

14.8

2.2 x 10-7 mm/s

After ~1.2 mm of growth by cyclic loading from prior constant K data

310

300

14.6

c372 - 0.5TCT of 690 + 26%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2
2700 2900 3100 3300 3500 3700 3900 4100

290

14.4 2300

2500

280 4300

Test Time, hours

Figure 5-20 Second period of rapid PWSCC in 1D cold-rolled Alloy 690 plate from ANL and apparent lack of a CGR response to reducing temperature from 360 to 325, then to 290C [17, 18]

Figure 5-21 Change in macroscopic appearance of PWSCC region upon reducing temperature (c372) [17, 18]

5-22

Temperature, C

PWSCC of Thick-Walled Alloy 690 Material

SCC#2 - c400 - Alloy 690, 26%RA 1D, S-L Orientation, NX3297HK12, ANL
19.5 0.2 0.1

Outlet conductivity x 0.01

18.5

END OF TEST @ 4276h

17.5

-0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8

16.5

15.5

14.5

13.5

2.5 x 10-7 mm/s

To 80 cc/kg H2 @ 1858h

Est. pH at 360C = 8.2 used for c At 340C, pH = 7.53. At 300C, pH = 6.86 Pt potential CT potential
3400 3900

To 26 cc/kg H2 @ 3391h

7 x 10 mm/s

-7

-0.9 -1 -1.1 4400

12.5 1400

1900

2400

2900

Test Time, hours

Figure 5-22 Repeat test at GE-GRC on 1-D rolled ANL plate material showing a slight increase in CGR upon raising the dissolved hydrogen concentration [17, 18]

Figure 5-23 shows the moderate CGR (4.2E-8 mm/s) measured in the S-L orientation on a different heat of Alloy 690 material that had been 1-D cold-rolled by 20% at GE-GRC. The microstructure of this forged bar material was extremely homogeneous before additional cold-working. As for the ANL plate material, changes in test temperature produced no long-term CGR response (Figure 5-24). Interestingly, a significant drop (from 27.5 to 17.3 MPam) in applied stress intensitygg also failed to slow down cracking at all in this specimen. A repeat test gave an almost identical PWSCC CGR and no response whatsoever to an increase in hydrogen concentration from 26 to 80 cc/kg (Figure 5-25). The same material was tested in the S-T orientation (cf. Figure 5-9) and gave a somewhat slower CGR of ~8E-9 mm/s (Figure 5-26). Furthermore, in contrast to the S-L oriented specimens, this CGR dropped significantly upon reducing temperature to 325C, more-or-less in line with the change that would be expected if the 130 kJ/mole activation energy for Alloy 600 were applicable here. Later on, a different area of this specimen was tested and produced even lower CGRs at constant K, indicating a clear effect of specimen orientation in the 1-D cold rolled material.

gg

Occurring because the other (lead) test-specimen in the daisy chain within the same autoclave showed significantly larger amounts of crack growth.

Conductivity, S/cm or Potential, Vshe

c400 - 0.5TCT of 690 + 26%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2

~5 x 10 mm/s

-7

0 -0.1

Crack length, mm

To Constant K @ 802h

5-23

PWSCC of Thick-Walled Alloy 690 Material

SCC#5 - c373 - Alloy 690, 20%RA 1D, S-L Orientation, Heat B25K
11.69 11.68 11.67 0.4

c373 - 0.5TCT of 690 + 20%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2
Outlet conductivity x 0.01
-8

0.2

11.66

Crack length, mm

11.65 11.64 11.63 11.62 11.61 11.6 11.59 650

To Constant Load @ 1339h

To Constant K @ 647h

K has slowly decreased to 22.9 MPam because this specimen's growth rate is slower

4.2 x 10 mm/s

-0.2

-0.4

~0.09 mm of growth at constant K

-0.6

Est. pH at 360C = 8.2 used for c At 340C, pH = 7.53. At 300C, pH = 6.86 Pt potential CT potential
750 850 950 1050 1150 1250 1350 1450

-0.8

-1

Test Time, hours

Figure 5-23 Moderately rapid PWSCC in 20% 1D cold-rolled Alloy 690 GE-GRC forged bar [17, 18]
SCC#7a - c373 - Alloy 690, 20%RA 1D, S-L Orientation, Heat B25K
12.28 370

Temperature, C
12.26

To 325C @ 2821h

K has slowly decreased to 17.3 MPam because this specimen's growth rate is slower

3 x 10 mm/s

-8

360

350

12.24

To R=0.7, 0.001 Hz + 9000s hold @ 1540h

340

Crack length, mm

To Constant K @ 1635h

330

To 290C @ 3473h

12.22

320

12.2

310

2 x 10-8 mm/s
12.18

After ~0.4 mm of growth by cyclic loading

300

c373 - 0.5TCT of 690 + 20%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2
2500 2700 2900 3100 3300 3500 3700 3900

290

12.16 2300

280 4100

Test Time, hours

Figure 5-24 Continued testing of 20% 1D cold-rolled Alloy 690 GE-GRC forged bar with reduction in both test temperature and applied stress intensity [17, 18]

5-24

Temperature, C

Conductivity, S/cm or Potential, Vshe

PWSCC of Thick-Walled Alloy 690 Material

SCC#2a - c401 - Alloy 690, 20%RA 1D, S-L Orientation, Heat B25K
11.72 0.3 0.2 0.1

11.7

Outlet conductivity x 0.01

11.68

To 80 cc/kg H2 @ 1858h

-0.1 -0.2 -0.3 -0.4

11.66

11.64

11.62

4 x 10 mm/s

-8

-0.5 -0.6

11.6

c401 - 0.5TCT of 690 + 20%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2
Est. pH at 360C = 8.2 used for c At 340C, pH = 7.53. At 300C, pH = 6.86 Pt potential CT potential
2100 2200 2300

-0.7 -0.8 -0.9 -1 -1.1 2400

11.58

11.56 1400

1500

1600

1700

1800

1900

2000

Test Time, hours

Figure 5-25 Lack of response to a major change in dissolved hydrogen during testing of a second specimen from a 20% 1-D cold-rolled Alloy 690 forged bar [17, 18]
SCC#3 - c393 - Alloy 690, 20%RA 1D, S-T Orientation, Heat B25K
11.26 370

Close to Activation Energy for A600

11.25

Change temp to 325C & 10.4 cc/kg H2 @ 2663h

c393 - 0.5TCT of 690 + 20%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2

2 x 10 mm/s

-9

360

350

Crack length, mm

To Constant K @ 941h

11.24

8 x 10 mm/s

-9

340

11.23

330

320

11.22

310

Est. pH at 360C = 8.12 used for c At 340C, pH = 7.53. At 300C, pH = 6.86

11.21 900 300 1400 1900 2400 2900

Test Time, hours

Figure 5-26 Results of testing the same material as in Figure 5-23, but this time in the S-T orientation [17, 18]

Temperature, C

Conductivity, S/cm or Potential, Vshe

To Constant K @ 802h

Crack length, mm

5-25

PWSCC of Thick-Walled Alloy 690 Material

Generally similar CGRs overall were measured for a further heat of Alloy 690 that was also 1-D cold-rolled by ~20% at GE-GRC. This time, however, the S-L oriented specimen did respond to a drop in test temperature (see Figure 5-27).
SCC#3 - c394 - Alloy 690, 19%RA 1D, S-L Orientation, Heat NX2865HK
11.42 370

11.4

Close to Activation Energy for A600

360

11.38

c394 - 0.5TCT of 690 + 19%RA 1D, 360C 27.5 MPam, 600 B / 1 Li, 26 cc/kg H2
To Constant K @ 941h

4.6 x 10-9 mm/s

350

11.36

Crack length, mm

11.32

Change temp to 325C & 10.4 cc/kg H2 @ 2663h

11.34

2.4 x 10-8 mm/s

340

330

11.3

320

11.28

11.26

310

Est. pH at 360C = 8.12 used for c At 340C, pH = 7.53. At 300C, pH = 6.86

300 1400 1900 2400 2900 3400 3900

11.24 900

Test Time, hours

Figure 5-27 Results from a further heat of Alloy 690 tested after 1-D cold-rolling at GE-GRC, this time showing expected response to a drop in temperature even for an S-L oriented specimen [17, 18]

Figure 5-28 shows details of the fractography for the 26% 1-D cold rolled ANL plate that has produced the highest SCC growth rates measured to date for Alloy 690 in simulated primary waterhh. Comparing this picture with Figure 5-29 (from a 20% 1-D cold rolled Alloy 690 specimen from forged bar that was originally very homogeneous and exhibited only moderate CGRs even after deliberate cold working), it would appear that there are clear signs for a difference in fracture morphology/crack path that may be an indication of a different fracture mechanism.

hh

Even higher rates have, however, been measured at the Bettis laboratories in deaerated HT water see Section 5.1.2.1.7.

5-26

Temperature, C

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-28 High-resolution fractography (c372) of the 26% 1D-cold rolled ANL 690 plate tested at GE-GRC [17, 18]

Figure 5-29 High-resolution fractography of a 20% 1-D cold rolled Alloy 690 specimen from forged bar [17, 18]

5-27

PWSCC of Thick-Walled Alloy 690 Material

5.1.2.1.4

Investigations in Simulated Primary Water at the University of Tohoku in Japan

These investigations are being carried out as part of a major research effort, co-sponsored by the EPRI Primary Systems Corrosion Research Program, and are oriented more to obtaining fundamental understanding of the SCC process than to developing engineering data [19]. Although some testing of Alloy 690 has been carried out in simulated primary water at 320 to 340C [8], much attention is also being paid to possible LTCP effects on this material at ~50C [20] and many results have not yet been completely reported. The following represents relevant, preliminary information. Product form Various heats of Alloy 690 are being studied, apparently including both mill-annealed and thermally treated (700C for 15h) plate materials, one of which (plate B) was tested both for possibly enhanced CGRs in the HAZ after being welded to Alloy 52 material (see Section 5.1.2.4.5), as well as separately after undergoing 40% cold work (achieved by cross-rolling the plate eight times) [8]. Specimen/crack-growth orientation A 1T-CT (25mm thick) 5% side-grooved specimen appears to have been used in the first CGR test (plate A), with the notch direction of the specimen parallel to the rolling direction. The hardness in the through-thickness direction of the 40% cold-worked Alloy 690TT-B plate is high (~ 340 HV1) and relatively uniform. A 12.5mm thick circular double cantilevered beam (CDCB) specimen with 5% side groove on each side was used with the notch direction parallel to the final rolling direction when testing this latter material [8]. Testing details SCC testing is being performed in simulated PWR primary water in a low flow rate, refreshed stainless steel autoclave. Gentle cycling has been used in order to facilitate transitioning from the air fatigue pre-crack and some of the SCC test periods appear to have been carried out with trapezoidal loading (at an R-value of 0.7) involving 72h hold periods, rather than under purely constant loading. Data evaluation Testing of Alloy 690TT plate without additional cold work has apparently shown no or only extemely slowii crack growth through SCC. However, relatively long, partially intergranular cracks are thought to have been observed at separate locations on the fracture surface of the 40% CW Alloy 690TT plate B specimen after testing in simulated PWR primary water at 340C for 4050 h [8]. No attempt appears to have been made to derive a CGR from these findings and some of these cracks apparently propagated along a plane perpendicular to the main fracture surface.

ii

A CGR of ~ 6E-10 mm/s at 320C and a K-value of 28.5 MPam is indicated in [20].

5-28

PWSCC of Thick-Walled Alloy 690 Material

5.1.2.1.5 Investigations in Simulated Primary Water at the Institute of Nuclear Safety Systems (INSS) in Japan

These studies are being performed on behalf of the Kansai Electric Power Company (KEPCO) [8]. Product form INSS are known to be testing thermally treated Alloy 690 plate with additional cold-work (20 and 50%) from rolling. Specimen/crack-growth orientation The program uses fatigue pre-cracked 1/2T CT specimens with T-L and T-S orientations. I.e., the crack advance plane is not coincident with the rolling direction, but oriented at 90 to this. Testing details The test environment is simulated primary water (with an unchanging H2 level of 30 cc/kg) over a range of temperatures from 360 down to at least 320C. No attempt is apparently made to use mechanical load transitioning and testing is being carried out under constant load at a KI value of 30 MPam. Data evaluation Non-uniform pockets of crack advance through intergranular PWSCC have been detected along the crack front (Figure 5-30) and these apparently have significant dimensions at 360C. The method of calculating CGRs is unclear, but reported values at 360C are as high as 2 E-7 mm/s (Figure 5-31). Virtually no effect of temperature was seen for the 50% CW material, but CGRs for the 20% CW material apparently dropped to below 2 E-8 mm/s at ~330C.

5-29

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-30 Example of crack growth observed at INSS in 20% cold worked 690TT at 360C (from [8])

5.1.2.1.6

Further Investigations in Simulated Primary Water in Japan

CRIEPI is also known to be carrying out more fundamental studies of Alloy 690 [21], but further details (including the product form) are not yet known.

5-30

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-31 Apparent effect of temperature and degree of cold work (CW) on measured CGR for 690TT (from [8])

5.1.2.1.7

Investigations by Bechtel-Bettis in Deaerated High-Temperature Water

The large Bettis testing program on Alloy 690 was first presented by Paraventi and Moshier [22] at the Atlanta meeting of the EPRI PWSCC Expert Panel in Fall 2006. A 2007 presentation [23] updated this and included some key additional information. A main aim of the program is to identify potential areas of vulnerability for Alloy 690 by varying material heats and fabrication methods. Recently, Burke [24] gave a description of some of the microstructural characteristics of the materials that Bettis have used, paying particular attention to the development of so-called banded microstructures. Product form Bettis have tested Alloy 690 in plate form, representing three different heats from two distinctly different melting techniques: One heat (VIM-1) is vacuum induction melted with subsequent electroslag remelting (VIM/ESR), which gives the cleaner microstructure, and Two heats (AOD-1 and AOD-2) are argon oxygen decarburization with subsequent electroslag remelting (AOD/ESR).

Final annealing temperature (1052 vs. 1093C) is a test parameter for the VIM/ESR plate, as is the effect of omitting the final thermal stabilization heat treatment (TT for 10h at 718C) in one case for the AOD/ESR plate. Detailed material characterization [24], carried out on an entirely different heat of Alloy 690 material that was not used for CGR testing, showed the following general features:

5-31

PWSCC of Thick-Walled Alloy 690 Material

Absence of Cr microsegregation in the as-cast structure implies that the development of carbide banding is related to subsequent thermomechanical processing treatments. Borides, carbides and boro-carbides are present in as-cast and as-homogenized A690. Thermomechanical processing should be designed to promote the dissolution of these coarse precipitates and optimize the controlled precipitation of carbides and nitrides to promote grain refinement in the final product.

A key aspect of the Bettis program is to investigate the effect of various levels of cold work. This is being carried out primarily by uni-directional rolling and was performed in one pass (for a target of 12% work) or two passes (for the target of 24% cold work) for heats VIM-1 and AOD-1. Heat AOD-2 was rolled in multiple passes. Table 5-10 shows the full test matrix and it can be seen that uni-directional tensile prestraining of the VIM material has also been used in 2 cases to obtain the highest levels of deformation (32 33% cold work). Material characterization [24] showed the following: Development of extensive slip band structure with a high proportion of dislocations. Fracture of intergranular and intragranular carbides (more fractured carbides as % deformation increases). Bending (i.e., curvature) of slip bands occurs to accommodate shape change. Alloy 690, which has a lower stacking fault energy (SFE), shows finer slip bands and more planar slip than Alloy 600.

Reference [22] reported representative microstructures for each of the three heats. VIM-1 had some regions near the plate surfaces (edges) where large grains were occasionally observed, but generally throughout the thickness the material was uniform in grain size and free of carbide banding. The AOD heats had different levels of banding, with AOD-2 being the most significantly banded both in terms of grain size and carbides. AOD-1 was much less banded than AOD-2 but still showed moderate grain size and carbide banding. In summary, the relative ranking of banding was characterized as follows: VIM-1: Very little to no banding. AOD-1: Intermediate banding (more grain size than carbides). AOD-2: Severe grain size and carbide banding, with a high concentration of inclusions.

5-32

PWSCC of Thick-Walled Alloy 690 Material Table 5-10 Material test matrix for Bettis Alloy 690 test program [22, 23]

Specimen/crack-growth orientation Three orientations for crack growth in the precracked CT specimensjj were used (cf. Figure 5-9): L-T: this represents growth perpendicular to the rolling/straining direction on a plane also perpendicular to that direction and showed the lowest susceptibility to SCC. S-T: this represents growth in the plane of, but perpendicular to, the rolling/straining direction and produced intermediate susceptibility to SCC. S-L: this represents growth in the plane of, and parallel to, the rolling/straining direction and produced the highest susceptibility to SCC.

Testing details A total of 186 bolt-loaded (i.e. constant displacement) specimens were tested for times between 500 and nearly 15,000 h at 3 test temperatures (316, 338 and 360C). Some load relaxation has to be expected here, especially during heat-up of the autoclave. Furthermore, crack advance in such specimens leads to a decrease in both applied load and resulting crack tip stress intensity value (KI). The CGR for each specimen was calculated by dividing the average crack advance on the fracture surface by the total test time, i.e. ignoring any incubation time. It was then plotted with regard to an average value of applied stress intensity (which also involves the assumption of uniform growth rate). As a result of the above issues, such test procedures usually tend to underestimate SCC susceptibility, both in terms of difficulties nucleating uniform crack fronts from the (transgranular) air fatigue pre-crack and in maintaining representative growth rates during crack advance.
jj

The specimens were without side grooves, so that some tendency for the crack growth plane to wander during testing is to be expected.

5-33

PWSCC of Thick-Walled Alloy 690 Material

Seven instrumented specimens were also tested under constant, active load, apparently to check out the abovementioned influences [22]. None of these showed any evidence of incubation times.The one result shown in reference [23] indicates that cracking started quickly (at least for the medium value of stress intensity used) and it is stated that the CGR data from the actively loaded specimens were consistent with rates from sister bolt-load tests. Nevertheless, it is possible that the constant displacement technique used in the majority of the Bettis tests might tend to underestimate representative CGRs when these are low (e.g. at low KI values, or for annealed materials without deliberate cold work) and may thus have led to bias in some of the dependencies reported below. Data evaluation Table 5-11 shows the qualitative evaluation of SCC susceptibility prepared by Bettis. The following key points emerge with regard to material fabrication issues: Only the more susceptible material (VIM/ESR plate) was tested in the fully annealed condition, and then showed no susceptibility to crack advance through SCC. Uni-directional cold rolling induced some SCC susceptibility in both materials, even at the 12% level, but much more strongly at the 24% level (see Figure 5-32) and in the VIM/ESR plate. Comparison of the highest cold worked materials seems to suggest that AOD heats may be somewhat more resistant; however, this may just reflect heat to heat variability. Uni-directional tensile straining (to around 33%) also resulted in moderate susceptibility of the VIM/ESR plate, but was not as detrimental as 24% cold-rolling (see Figure 5-33), even though it resulted in a higher yield strength (cf. Table 5-10). There was no apparent effect of a lower or higher final annealing temperature (before additional TT treatment) on behavior of the VIM/ESR plate. Results on omission of the additional TT treatment with one lot of the AOD/ESR plate do not appear to have been reported yet.
Table 5-11 Bettis summary of relative SCC susceptibility for Alloy 690 [22, 23]

5-34

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-32 Increase in CGRs with increasing level of uni-directional cold-rolling in Bettis studies [22, 23]

Figure 5-33 Cold-rolling is more detrimental than tensile pre-straining (despite lower yield strength) [22, 23]

The above findings clearly imply that SCC susceptibility is associated with the creation of heterogeneous microstructures in the test materials, in which case strong effects of specimen orientation on CGRs would be expected. These were indeed found: The S-L orientation, with crack advance in the direction and plane of deformation, was much more susceptible even than the S-T direction (see Figure 5-33). The L-T orientation was very resistant to cracking (cf. Table 5-11). 5-35

PWSCC of Thick-Walled Alloy 690 Material

For the most susceptible materials and worst CGR orientation, only a low dependence on stress intensity was found over the tested range of approx. 13 to 42 MPam (see top curve in the left-hand diagram in Figure 5-33). Bettis report KI exponents here in the range of 0.7 to 1.9, which is similar to the known dependency for Alloy 600. Higher values (up to an exponent of 6.2) were quoted for lower susceptibility conditions, but it is likely that these have been biased strongly by the difficulty in initiating uniform crack advance at low KI values, as discussed earlier. The Bettis test program has shown a surprisingly low dependence on temperature for crack advance in cold-worked Alloy 690 material, with apparent activation energies as low as 5 kcal/mole being derived in one case of high SCC susceptibility (see Figure 5-34, right-hand side). This is very different from the values of around 35 kcal/mole usually reported for Alloy 600 and implies that, if cracking were to occur in service with cold-worked Alloy 690, it would not necessarily be limited to components with high operating temperatures. Another apparent difference relates to the effect of increasing dissolved hydrogen in the test medium, where the trend, in fact, appears opposite to that observed for Alloy 600 cracking (see Figure 5-35).

Figure 5-34 Effect of test temperature and degree of cold work on CGRs for the VIM/ESR plate [22, 23]

The biggest concern arising from the Bettis data is that very high CGRs (~ 6E-7 mm/s) have been registered for the most susceptible material at quite moderate stress intensities (~ 30 MPam), as shown by the blue curve on the left-hand side of Figure 5-33. This is all the more surprising given the frequent tendency of constant displacement test techniques to under- rather than overestimate CGRs, as discussed earlier. The fractography shown in reference [23] confirms that these rates have been derived from appreciable increments of crack advance (even for the lower levels of cold work) and the morphologies shown have been predominantly intergranular (see Figure 5-36). In some cases, colonies of cracked carbides have apparently been detected on grain facets of the IGSCC fracture surface (see Figure 5-37), but this finding although somewhat unusual does not in itself provide any direct explanation of the high SCC CGRs that were measured. 5-36

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-35 Apparent increase in CGRs at 50 cc/kg hydrogen (blue symbols) vs. 23 cc/kg (pink symbols) for Alloy 690 (left) and comparison (right) with opposite behavior for Alloy 600 [22, 23]

Figure 5-36 Predominantly intergranular crack advance in an S-L-oriented specimen of VIM/ESR TT plate subjected to only 12% cold-rolling [22, 23]

5-37

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-37 Detail from fractography of an S-T-oriented specimen of VIM/ESR plate subjected to 24% cold-rolling [22, 23]

Bettis themselves have indicated two possible approaches to assessing the relevance of their SCC results to date. The first makes use of a direct comparison with similar data they have obtained in a parallel test program using Alloy 600 material [22]. As indicated in Figure 5-38, this leads to the conclusion that the Alloy 690 CGRs are a factor of 5 to 10 lower than for Alloy 600 material of comparable yield strength. This margin is considerably smaller than industry expectations. 3 However, given the relatively low yield strength (cold work) dependence of Alloy 600 (ys ) 7-8 compared to the apparently very high cold work dependence in Alloy 690(ys ), the margin between the two alloys would be expected to get larger as the extent of cold work decreases.

Figure 5-38 Comparison of Alloy 600 and 690 CGRs measured by Bettis [22, 23]

5-38

PWSCC of Thick-Walled Alloy 690 Material

A second approach assumes that the major role of cold-rolling is simply to increase the yield strength of the 690 material and attempts to extrapolate back (see Figure 5-39) to determine a representative CGR for the annealed VIM/ESR plate (yield strength = ~46 ksi), which did not show any crack advance at all in the actual testing. The resulting rate shown would be acceptably low, but any such extrapolation (on a logarithmic-linear plot) is fraught with uncertainty and the correctness of that chosen in Figure 5-39 can be challenged in the absence of additional data at other cold work levels. Furthermore, as pointed out above, a unique dependency on yield strength is not indicated in the complete test data, because of the differences found when cold work was introduced in a different way (rolling versus tensile pre-straining). Ultimately, a proper assessment of the practical relevance of the Bettis data will require more understanding of the heterogeneity produced in Alloy 690 by uni-directional deformation and more confirmation of some of the key dependencies.

Figure 5-39 Attempted back-extrapolation by Bettis of cold-worked Alloy 690 CGRs on the basis of assumed yield strength dependencies [22, 23]

5.1.2.1.8 Investigations by Westinghouse in Supercritical Water with Additions of Boron, Lithium and Hydrogen

A Westinghouse program, carried out for the EPRI MRP, was designed to accelerate the IGSCC process in SCC-resistant, Ni-base alloys by testing at elevated temperatures and pressures [25]. Specifically, this study involved testing at temperatures beyond the critical point of pressurized aqueous environments in an attempt to thermally accelerate corrosion and produce measurable effects from relatively minor SCC processes. High-pressure/temperature test conditions (385C/33MPa) were chosen to have water-like fluid properties with additions of Li/B and H2. Test materials were selected to be representative of thick-walled Alloy 690 (see Table 5-12) and a limited number of Alloy 600 control samples were also tested. Table 5-13 shows the chemical composition of the materials and Table 5-14 shows their tensile properties as supplied. It can be 5-39

PWSCC of Thick-Walled Alloy 690 Material

seen that the yield strengths are rather low (between ~ 230 and 278 Mpa). Since this would have limited the stress intensity factors that can be achieved without violating ASTM testing requirements and in view of the anticipated difficulty in achieving and sustaining meaningful and measurable CGRs in these crack resistant materials, which concern implies the need for high stress intensities, the decision was taken to elevate the yield strengths by cold-working. The means chosen to effect the required cold work was high speed forging and Table 5-15 shows the properties achieved during trial efforts. It was concluded that engineering strain (in terms of a thickness reduction) of 10-11% would produce the desired strain hardening.
Table 5-12 Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25]

Table 5-13 Chemical composition of Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25]

5-40

PWSCC of Thick-Walled Alloy 690 Material Table 5-14 Tensile properties (as reported by material vendors) of the Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25]

Table 5-15 Tensile properties after trial forging of the Alloy 690 materials used in Westinghouse testing for CGRs in supercritical water [25]

The microstructure of all the alloys was examined in some detail (see [25]) and all the 690 CRDM materials exhibited predominantly intergranular carbides. However, the microstructure of the one plate material used (which had been given a laboratory thermal treatment) was slightly different: although most of the M23C6 carbide precipitation seemed to be intergranular,stringers of what are most likely carbonitrides were aligned along what appears to be the rolling direction. This is not an uncommon feature for heavy section hot rolled plate and the role, or lack thereof, of these stringers in the crack propagation process was deemed to be of interest. Two crack growth rate (CGR) test series were conducted in SCW containing 900 ppm boron and 3.15 ppm lithium using (overall) 2 compact tension (CT) specimens fabricated from Alloys 600, 1 CT specimen fabricated from 690 TT plate (with additional cold-work), 2 CT specimens fabricated from CRDM piping in the as-received condition and 4 CT specimens also from CRDM material, but with deliberate cold working. The hydrogen partial pressure in the test was chosen to be reducing compared to the Ni/NiO stability line and the boron and lithium conditions were close to median concentrations for a U.S. PWR. Stress corrosion crack propagation occurred in every specimen tested in the supercritical environment. The results obtained for the mill annealed Alloy 600 specimens (see Table 5-16) indicated higher CGRs in SCW than are observed in pressurized 5-41

PWSCC of Thick-Walled Alloy 690 Material

primary water; however, the acceleration factor was lower than originally anticipated, possibly due to the chosen hydrogen level. The general features of crack morphology and oxide chemistry were essentially the same as those observed at lower temperatures and pressures. Clear intergranular crack growth was observed in both the as-received (690 TT) and the ~ 10% cold-worked Alloy 690 specimens tested at stress intensities of approximately 32 MPam (Figure 5-40). The slow, but steady, IGSCC measured for both material conditions could be clearly distinguished from the transgranular fatigue that occurred in the pre-cracking phase, or during gentle cycling in the autoclave environment. No meaningful differences were observed between the cold-worked plate material and the three different heats of thick-walled Alloy 690TT (thought to represent a range of thermo-mechanical processing typical for current CRDM nozzles in replacement RPV heads) with subsequent deliberate cold working, so although the results have been tabulated separately, they are shown together in Figure 5-41. Note from this figure, however, that lower CGRs were seen in the as-received Alloy 690TT, as discussed further in Section 5.1.2.2.5.
Table 5-16 Detailed Results for the Alloy 600 control samples and Alloy 690 plate material used in Westinghouse testing for CGRs in supercritical water [25]
da Fatigue 1 Calc. mm 0.037 0.037 0.313 da Fatigue 2 Meas. mm n/m n/m 0.612 da SCC 3 Meas. mm 1.690 1.540 0.359 Total 5 Time Hour 1920 1920 4806 SCC 6 Time Hour 1546 1546 4323 Kfinal MPa m 20.7 20.7 32.0
9

Material

Spec ID

Crack 4 Mode

Kphase 2 MPam

Kaverage MPam

Average 10 CGR m/s 3.0E-10 2.8E-10 2.3E-11

600 MA 600 MA 690TT cw Notes:

510-04a 510-04b F8244-cw3

IG IG IG

16.9 16.7 31.9

19.2 19.3 31.3

1. Calculated fatigue crack growth 2. Average transgranular fatigue crack increment averaged across the crack front 3. Average SCC increment averaged across the crack front based on optical and SEM images 4. IG = intergranular; ID = interdendritic; TG = transgranular 5. Total time exposed to 385C supercritical water under applied load and pressure 6. SCC time is time under conditions with steady loading (e.g. unloading period 3600 s) 7. K phase 2 is the stress intensity after initial rapid cycling complete, (e.g. unloading period 3600 s) 8. Kaverage based on mid-crack a/w ratio and load conditions in the middle of the test (e.g. during Phase 3) 9. Kfinal based on final a/w ratio and load conditions in the end of the test (e.g. end of Phase 4) 10. Average CGR = da SCC based on destructive exam divided by SCC time

5-42

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-40 Example of transition from transgranular cracking during cyclic loading in SCW to intergranular SCC at ~ constant load in a 10% cold-worked specimen of Alloy 690TT [25]
Alloy 600/690 Tested in 385C Supercritical B/Li/H2
1.0E-08

SCC Growth Rate (m/s)

1.0E-09

Alloy 600MA
1.0E-10

Alloy 690TT + cold work

1.0E-11
K Phase 2 Average K Final K

Alloy 690TT

1.0E-12 15

20

25

30

35

40

45

Stress Intensity Factor (MPam)

Figure 5-41 Average SCC growth rate vs. stress intensity factor for Alloy 600 and 690 materials [25]

5-43

PWSCC of Thick-Walled Alloy 690 Material

The cold-worked Alloy 690 TT CGRs are approximately 20 times lower than those measured for Alloy 600 MA in the supercritical environment, after adjusting the latterkk to a common stress intensity factor of 32 MPam (see Figure 5-42). I.e., the material improvement factor here at 385C in SCW is ~20. As mentioned earlier, the thermal acceleration factor for Alloy 600 was lower than originally anticipated. There is presently no established basis for extrapolating the measured CGRs down to subcritical temperatures in order to make detailed comparisons with the MRP database for 600 in primary water and the lower dotted line in Figure 5-42 should be regarded only as one possible representation of actual behavior.
Alloy 600/690 Comparisons - Supercritical vs. Pressurized Water
1.0E-08

CGR (m/s) for K = 32 MPam

Q = 80 KJ/mole

Q = 130 KJ/mole

1.0E-09
Kavg = 37 MPam

93510 - MRP55 93511 - MRP55 & W data 93510 - Supercritical B/Li/H2

1.0E-10

3 tests

93510 - Average Primary H2O 93510 - Anticipated SCrit A690 10%CW in Supercritical B/Li

1.0E-11

Supercritical
1.0E-12 1.40 1.45 1.50 1.55 1.60 1.65

A690 0%CW in Supercritical B/Li

1.70

1.75

1.80

(1000/T)

Figure 5-42 Comparison of CGR measured for Alloy 600 MA and Alloy 690 TT in supercritical and primary water [25]

5.1.2.2

Testing of Alloy 690 CRDM Material without Deliberate Cold Working

The majority of the thick-walled Alloy 690 material currently in service in PWRs worldwide is in the form of CRDM penetrations that are not thought to have significant cold work during fabrication, although some uncertainty remains regarding possible pipe straightening (see Section 2.1.2). Nevertheless, relatively few laboratory tests to date have used material in this condition, since it is expected to show extremely high resistance to PWSCC. This makes reliable testing a challenge requiring extreme patience on the part of both the investigator and the program sponsor.

kk

Stress intensity correction factor for Alloy 600 data = (32/K)2.2

5-44

PWSCC of Thick-Walled Alloy 690 Material

5.1.2.2.1

Investigations in Simulated Primary Water at Argonne National Laboratory (ANL)

The testing approach used at ANL has already been described in Section 5.1.2.1.2. Two 0.5T CT specimens with different orientations of through-thickness crack growth have been prepared from Alloy 690TT material (heat WP142) as shown in Figure 5-43. In the first test at 320C, a CGR in the radial direction of ~ 5E-9 mm/s was reported at constant load early in the test (see Figure 5-44, top diagram). Subsequently, even slower crack growth (~ 2E-9 mm/s) was reported to have occurred during a relatively short period (~ 800 h) under constant load at the end of the experiment, but no clear signs of intergranular cracking were subsequently seen on the fracture surface of this specimen (Figure 5-45).

Figure 5-43 Specimen orientation in ANL testing of Alloy 690 CRDM material [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5-45

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-44 Reported CGRs during brief test periods at constant load in ANL work on CRDM material [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5-46

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-45 Absence of clear intergranular cracking on fracture surface of the first CRDM specimen tested at ANL [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

Testing of a second specimen (with different orientation) gave ~ 3E-9 mm/s under constant load at 320C, but slightly more (~ 8E-9 mm/s) for ~ 500 h at a temperature of 350C (see Figure 5-46).

5-47

PWSCC of Thick-Walled Alloy 690 Material

12.78 12.77 12.76 Crack Length (mm) Increased temperature to 350C 12.75 12.74 12.73 12.72 12.71 12.70
3.2 x 1012 m/s 21.3 MPa m0.5 Constant load 8.0 x 1012 m/s 21.3 MPa m0.5 Constant load

50
Alloy 690 Specimen # C690-LR-2 Simulated PWR Water

45 40 35 30
Crack Length

25 20

Kmax

15 10

1200

1400

1600

1800 Time (h)

2000

2200

2400

Figure 5-46 Reported CGRs during constant-load test periods at two test temperatures in further ANL work on CRDM material without deliberate cold working [11 to 15]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5.1.2.2.2 (PNNL)

Investigations in Simulated Primary Water at Pacific Northwest National Laboratory

Product form Six heats of Alloy 690 CRDM thick-walled piping material were obtained from Valinox, all received in the thermally treated (TT) condition. A variety of thermomechanical treatments are being tested to explore what processing and microstructural conditions may influence SCC resistance in PWR primary water [26]. The first Alloy 690 tests compared the response of as-received TT material to a solution anneal (SA) heat treatment at 1100C (chosen to alter the carbide microstructure) using the following three heats of material: Heat # RE243: 10.4Fe, 28.9Cr, 0.02C, 0.31Mn, 0.35Si, 0.14Al, 0.23Ti, Bal Ni Heat # WP142: 10.5Fe, 29.0Cr, 0.02C, 0.31Mn, 0.35Si, 0.18Al, 0.27Ti, Bal Ni Heat # WP140: 10.4Fe, 29.0Cr, 0.03C, 0.31Mn, 0.33Si, 0.18Al, 0.30Ti, Bal Ni

Specimen/crack-growth orientation 0.5T CT specimens with 5% side grooves are being used for all evaluations. Alloy 690TT and SA specimens were cut from the 690 piping material in the C-L orientation as shown in Figure 5-47. Since the grain size near the OD was only about half that observed at the midwall or ID, all specimens have been taken from one or other of the latter locations.

5-48

Kmax (MPa m0.5)

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-47 Orientations of CT specimens used by PNNL for CGR testing on two Alloy 690 forgings: left is heat RE243; right is heat WP140 [26]

Testing details The test environment is simulated primary water, predominately at 325 or 350C. The chosen H2 level of 29cc/kg is in the region of Ni metal stability, well above the expected level for peak CGR in Alloy 600 at 325C. Evaluations have also been performed on two of the Alloy 690 CRDM heats at 20 cc/kg and 350C, which is near the expected peak CGR for Alloy 600. Final pre-cracking is carried out by fatigue at temperature in the autoclave followed by a very gradual ll transition in periodic partial unloading (see Figure 5-48) to a constant KI value of 30 or 40 MPam. Data evaluation Only very minimal amounts of crack growth at constant K (see Figure 5-49 and Figure 5-50), with occasional signs of intergranular morphology on the fracture surface (see Figure 5-51), have been seen in any of the specimens tested to date at PNNL without deliberate cold working. It is unclear whether genuine CGRs can, in fact, be derived here, but all would lie at or below 1E-9 mm/s (i.e. below the typical diameter of a single grain in the material being tested even after one year of growth) and are thus clearly of no engineering significance.

ll

Modeled after the GE-GRC approach described in Section 5.1.2.1.1.

5-49

PWSCC of Thick-Walled Alloy 690 Material

CT014 & CT015 CGR, 0.5TCT Alloy 690 Valinox, Heat RE243, pipe 2360, sample 1 & 2 325C, 30 MPam, 1000 ppm B, 2.0 ppm Li, 29 cc/kg H 2

0.105 0.100 0.095 0.090 0.085 0.080 0.075 0.070 0.065 0.060 0.055

0.001 Hz + 2.5 h, R = 0.7

0.001 Hz + 24 h, R = 0.7

constant K

60

40 2.1e-09 mm/s 30 ~7e-10 mm/s 20 2.6e-09 mm/s 10 5 month transitioning using cycle + hold steps 0 Pt ECP CT ECP 3500 time (hrs) 4000 4500
corrected for actual crack length

Alloy 690TT

0.050 1000

1500

2000

2500

3000

5000

5500

-10 6000

Figure 5-48 PNNL data showing the patience used in transitioning to constant K loading [26]
CT014 & CT015 CGR, 0.5TCT Alloy 690 Valinox, Heat RE243, pipe 2360, sample 1 & 2 325C, 30 MPam, 1000 ppm B, 2.0 ppm Li, 29 cc/kg H 2

0.104 2 m 0.102 0.100 0.098 0.096 0.094

40

constant K

30

Alloy 690SA

~2e-10 mm/s

20

Extremely slow (~1 m/month) of "stable" propagation

0.092 0.090 0.088 0.086 0.084 5500 10

Alloy 690TT

~3e-10 mm/s 0

Pt ECP CT ECP 5700 5900

corrected for actual crack length -10 6500

6100 time (hrs)

6300

Figure 5-49 PNNL data showing Alloy 690 CGR response for as-received TT versus a carbide-modified SA condition [26]

5-50

outlet conductivity (S/cm) or ECP (V 10)

outlet conductivity (S/cm) or ECP (V 10)

0.001 Hz, R = 0.7

50

Alloy 690CM

~6e-10 mm/s

PWSCC of Thick-Walled Alloy 690 Material

CT026 & CT027 CGR, 0.5T CT Alloy 690 Valinox, WP140, Pipe 2502 & WP142, Pipe 2541 350C, 30 MPam, 1000 ppm B, 2.0 ppm Li, 29 cc/kg H 2

0.467 2 m 0.465 0.463 0.461 0.459 0.457 0.455 0.453 0.451 0.449

constant K

30 26

Alloy 690TT WP140

1.9e-08 mm/s

18 14 10 6

Alloy 690TT WP142

1.9e-08 mm/s

~7e-10 mm/s

~0 mm/s

2 -2 -6

CT ECP 3900 4000 4100 4200 time (hrs) 4300

Pt ECP 4400 4500

0.447 3800

-10 4600

Figure 5-50 PNNL data showing Alloy 690 CGR response for two further heats of material in the asreceived TT condition [26]

Figure 5-51 PNNL fractography showing IG cracking limited to isolated grains in specimens without cold-work [26]

outlet conductivity (S/cm) or ECP (V 10)

0.001 Hz + 9 ks, R=0.5

~1e-9 mm/s

~5e-10 mm/s

22

5-51

PWSCC of Thick-Walled Alloy 690 Material

5.1.2.2.3

Investigations in Simulated Primary Water at Studsvik in Sweden

Two 0.5CT specimens of Alloy 690 taken from archive material for the CRDM penetrations in the Ringhals 2 replacment RPV head were tested at 320C [27]. They were subjected to gentle cyclic loading with increased hold times (Kmax = 25 MPam, R = 0.6, 0.05 Hz, 0 and 1000 s hold times) for about 400 and 800 hours, respectively, without any indications of intergranular cracking being observed. However, transgranular cracking, interpreted as corrosion fatigue, was observed on the fracture surfaces. Studies are continuing to determine FOI values relevant to actual plant components and will include material with additional deformation simulating actual thick section plant components.
5.1.2.2.4 Investigations in Simulated Primary water at MHI in Japan

CGR testing of Alloy 690 CRDM material is also ongoing at the MHI laboratories in Japan in collaboration with the Japan Nuclear Energy Safety Organization (JNES) and, in September 2006, had failed to produce any identifiable SCC after 4000 h in Alloy 690 [28]. Further details are not yet available.
5.1.2.2.5 Investigations by Westinghouse in Supercritical Water with Additions of Boron, Lithium and Hydrogen

The key features of this program were described already in Section 5.1.2.1.8 and the numerical results for the CT specimens of Alloy 690 in the as-received condition are tabulated in Table 5-17. Less cracking, and about 3 times lower growth rates, were observed in comparison with the results for the cold-worked Alloy 690 TT specimens (see Figure 5-41).
Table 5-17 Detailed results for the Alloy 600 control samples and as-received Alloy 690 CRDM materials used in Westinghouse testing for CGRs in supercritical water [25]
da Fatigue 1 Calc. mm 0.037 0.037 0.310 0.370 da Fatigue 2 Meas. mm n/m n/m 0.757 0.675 da SCC 3 Meas. mm 1.690 1.540 0.234 0.098 Total 5 Time Hour 1920 1920 4806 4806 SCC 6 Time Hour 1546 1546 4323 4323 Average 10 CGR m/s 3.0E-10 2.8E-10 1.5E-11 6.3E-12

Material

Spec ID

Crack 4 Mode

Kphase 2 MPam

Kaverage MPam

Kfinal MPam

600 MA 600 MA 690TT 690TT

510-04a 510-04b 415-a 7a12-a

IG IG IG TG/IG

16.9 16.7 32.1 32.2

19.2 19.3 31.3 31.1

20.7 20.7 31.8 31.3

Notes: see Table 5-16

The as-received Alloy 690 TT CGRs are approximately 60 times lower than those measured mm for Alloy 600 MA in the supercritical environment, after adjusting the latter to a common stress intensity factor of 32 MPam (see Figure 5-42). I.e., the material improvement factor here at 385C in SCW is ~60. As mentioned earlier, the thermal acceleration factor for Alloy 600 was lower than
mm

Stress intensity correction factor for Alloy 600 data = (32/K)2.2

5-52

PWSCC of Thick-Walled Alloy 690 Material

originally anticipated. There is presently no established basis for extrapolating the measured CGRs down to subcritical temperatures in order to make detailed comparisons with the MRP database for 600 in primary water and the lower dotted line in Figure 5-42 should be regarded only as one possible representation of actual behavior. Nevertheless, the CGRs measured in the as-received Alloy 690 TT in SCW are sufficiently slow that significant SCC propagation in thick-walled Alloy 690 components exposed to primary water would not be expected, even if SCC initiation were to occur. 5.1.2.3 Testing of Alloy 690 CRDM Material after Deliberate Cold Working

From the beginning, GE-GRC chose to use the introduction of deliberate cold work for the EPRI MRP Test Program in order both to raise the chances of measuring any crack growth at all in a very SCC-resistant material and to simulate the local strains that might be expected in the HAZ of welded Alloy 690 immediately adjacent to the fusion line (see Section 5.1.2.4). Following the publication in 2006 of the Bettis results on plate material (see Section 5.1.2.1.7), many other laboratories have followed suit and much attention is currently directed towards the way in which subsequent cold working of various types might affect the CGR behavior of otherwise very resistant CRDM material.
5.1.2.3.1 Investigations in Simulated Primary Water by General Electric Global Research (GE-GRC)

The basic testing approach at GE-GRC was already described in Section 5.1.2.1.1 for the feasibility tests on EPRI plate material and in Section 5.1.2.1.3 for studies on additional heats of material that are not necessarily directly relevant to actual components. Some further testing of extruded Alloy 690, however, used an archive CRDM penetration supplied by Duke Power (Heat WN415) as the starting material (see Figure 5-52). Deliberate levels of cold work (20 or 41%) were then introduced by uniform forging of rectangular blocks prior to manufacture of CT specimens [29]. Note, however, that the material was not thick enough to align the crack growth plane with the plane of forging deformation. For two specimens (tested together) with 20% homogeneous cold work, relatively little time (only ~ 800h) was spent at constant K (see Figure 5-53) and growth rates were then very low ( 5E-9 mm/s). Nevertheless, clear evidence was found of intergranular cracking (Figure 5-54), much of which must have taken place during earlier periodic partial unloading conditions (PPU, i.e. with an element of trapezoidal, cyclic loading). This was also true of a further specimen subjected to 41% uniform cold work and tested under various loading conditions for over 11,000 h in total. In some cases, cracking tended to arrest already when moving to a hold time of 24h during PPU, or growth rates already remained below the abovementioned 5E-9 mm/s threshhold of no engineering significance (see Figure 5-55). Interestingly, the fracture surface of this specimen revealed fingers of crack growth with much out-of-plane secondary cracking (see Figure 5-56).

5-53

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-52 Specimen location in GE testing of Alloy 690 CRDM material [17, 18]

5-54

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-53 Data from one of two specimens of CRDM material with 20% homogeneous cold work tested at GE-GRC [29]

Figure 5-54 Intergranular crack morphology in 20% cold-worked Alloy 690 CRDM specimen [29]

5-55

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-55 Data from a further specimen of CRDM material with 41% homogeneous cold work showing tendency to crack arrest before reaching constant K conditions (top) and very low CGRs even under periodic partial unloading with a 24h hold time (bottom) [29]

5-56

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-56 Macro and micrfractography from the specimen of CRDM material with 41% homogeneous cold work showing extensive out-of-plane secondary cracking [29]

5-57

PWSCC of Thick-Walled Alloy 690 Material

The very low CGRs measured even after introducing heavy, but homogeneous, cold work into this actual CRDM material can be considered to indicate a factor of improvement of at least 70x in comparison to as-received Alloy 600 and probably more like 400x if the Alloy 600 material had been subjected to similar levels of cold work.
5.1.2.3.2 (PNNL) Investigations in Simulated Primary Water at Pacific Northwest National Laboratory

The approach to CGR testing at PNNL is similar to that at GE-GRC and has already been described in Section 5.1.2.2.2. They have also tested four specimens (2 each in the TT and CMnn conditions) of Alloy 690 CRDM material with deliberate, non-uniform cold working [26]. In one case, bar material was 17% unidirectionally cold rolled and tested in the S-L orientation. These specimens exhibited no enhanced SCC susceptibility. CGRs were already low under PPU and virtually arrested at constant K, even after raising this value from 30 to 40 MPam (Figure 5-57).
CT019 & CT020 CGR, 0.5T CT Alloy 690 Valinox, Heat RE243, Pipe 2360, 17% CW S-L 325C, 30 MPam, 1000 ppm B, 2.0 ppm Li, 29 cc/kg H 2

0.007 1 m 0.006

0.001 Hz + 9 ks R = 0.7

25

constant K 20

0.005

15

0.004 5.1e-09 mm/s 0.003

10

0.002 3.4e-09 mm/s

~4e-10 mm/s

0.001

Alloy 690SA +17% CW S-L

-5

0.000 1775

1875

1975

2075

2175

2275

2375

-10 2475

time (hrs)

Figure 5-57 Data from PNNL testing of CRDM material with deliberate cold working (here 17% in the S-L orientation) [26]

nn

CM = carbide modified (by solution annealing).

5-58

outlet conductivity (S/cm)

Alloy 690TT +17%CW S-L

~2e-10 mm/s

PWSCC of Thick-Walled Alloy 690 Material

In the second case, these same materials were 30% unidirectional cold-rolled and tested in the T-L orientation. This did have had some effect on SCC susceptibility, particularly for the TT specimen at K values of 30 and 40 MPam (Figure 5-58). However, even here the measured CGRs were below an engineering significance threshold of 5E-9 mm/s. Only isolated grains of cracking were seen on the fracture surfaces of all but the TT+30%CW specimen, where more intergranular cracking was reported, consistent with the slightly higher observed CGR.
CT022 & CT023 CGR, 0.5T CT Alloy 690 Valinox, Heat RE243, Pipe 2216, 30% CW T-L 350C, 40 MPam, 1000 ppm B, 2.0 ppm Li, 29 cc/kg H 2

0.020 2 m 0.018 0.016 0.014 0.012

constant K

30

2.9e-09 mm/s

20

Aloy 690TT +30%CW T-L

10

0.010 0.008 0.006 0.004 0.002

9e-10 mm/s

Alloy 690SA +30%CW T-L

2.4e-08 mm/s CT ECP 4300 4400 4500 4600 4700 time (hrs) 4800 4900 Pt ECP 5000 5100

0.000 4200

-10 5200

Figure 5-58 Data from PNNL testing of CRDM material with deliberate cold working (here 30% in the T-L orientation) [26]

5.1.2.3.3 Investigations by Westinghouse in Supercritical Water with Additions of Boron, Lithium and Hydrogen

The key features of this program were described already in Section 5.1.2.1.8 and the numerical results for the CT specimens of Alloy 690 in the cold-worked condition are tabulated in Table 5-18. More cracking, and about 3 times higher growth rates, were observed in comparison with the results for the cold-worked Alloy 690 TT specimens (see Figure 5-41). Note, however, that although most of the cold-worked specimens displayed CGRs of around 3 E-8 mm/s, one specimen cracked appreciably faster (albeit at a somewhat higher stress intensity).

outlet conductivity (S/cm) or ECP (V 10)

0.001 Hz + 9 ks, R = 0.5 3.8e-08 mm/s

1.4e-09 mm/s

5-59

PWSCC of Thick-Walled Alloy 690 Material Table 5-18 Detailed results for the Alloy 600 control samples and Alloy 690 cold-worked CRDM materials used in Westinghouse testing for CGRs in supercritical water [25]
da Fatigue 1 Calc. mm 0.037 0.037 0.308 0.312 0.319 0.454 da Fatigue 2 Meas. mm n/m n/m 0.689 0.561 0.204 0.400 da SCC 3 Meas. mm 1.690 1.540 0.309 0.313 0.290 2.144 Total 5 Time Hour 1920 1920 4806 4806 1920 1920 SCC 6 Time Hour 1546 1546 4323 4323 1546 1546 Average 10 CGR m/s 3.0E-10 2.8E-10 2.0E-11 2.0E-11 5.2E-11 3.9E-10

Material

Spec ID

Crack 4 Mode

Kphase 2 MPam

Kaverage MPam

Kfinal MPam

600 MA 600 MA 690TT cw 690TT cw 690TT cw 690TT cw

510-04a 510-04b F415-c F7a21-c F415-b F7a21-a

IG IG IG IG IG IG

16.9 16.7 31.5 31.7 28.8 30.6

19.2 19.3 31.1 31.1 32.4 37.0

20.7 20.7 31.9 31.8 33.6 41.4

Notes: see Table 5-16

5.1.2.4

Testing of Heat Affected Zones (HAZ) from Welding of Alloy 690 Material

Much of the justification for testing Alloy 690 base material with deliberate cold working relates to the possibility of additional tensile strains being present at relatively high levels adjacent to the fusion line after welding (see [17]). These, together with possible changes in microstucture in the HAZ, are thought to explain the more rapid cracking seen in KAPL testing of Alloy 600 HAZ specimens [30]. Thus testing of actual Alloy 690 HAZ material was recommended to fill in a knowledge gap in the 2004 MRP-111 report. Details of known efforts in this direction and preliminary results are reported below, but it is clear that this is work-in-progress and final conclusions cannot yet be drawn.
5.1.2.4.1 Investigations in High-Temperature Water at KAPL

As mentioned above, KAPL drew attention some years ago to the possibility that the heat-affected zone (HAZ) in Alloy 600 might be significantly more susceptible to SCC than unaffected base material and have confirmed this finding more recently [31]. Detailed investigations revealed, however, that Alloy 690 HAZ microstructures are different from Alloy 600 HAZ microstructures due both to the higher chromium content and higher solvus temperature of Cr23C6. Neither unconstrained, nor constrained welds (with up to 14% plastic strain in the weld metal) exhibited SCC susceptibility of the Alloy 690 HAZ under conditions where Alloy 600 HAZs cracked readily [31].
5.1.2.4.2 Investigations in Simulated Primary Water at Argonne National Laboratory (ANL)

The general ANL testing approach was described Section 5.1.2.1.2. Figure 5-59 shows the geometry of a specimen that was under test in November 2008 [13]. No further details are yet available.

5-60

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-59 Details of an Alloy 690 HAZ specimen (CF690) under test at ANL [13]
Graphic provided by Argonne National Laboratory, managed and operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

5.1.2.4.3

Investigations in Simulated Primary Water at Studsvik in Sweden

Figure 5-60 shows details of a 1T CT specimen on test at Studsvik in November 2008 [27] with the crack growing parallel to, but ~ 1mm away from, the fusion line. Preliminary results suggest no enhancement of CGR, but a detailed comparison with both base metal from this particular heat and with an Alloy 600 HAZ specimen has yet to be carried out.

5-61

PWSCC of Thick-Walled Alloy 690 Material

Figure 5-60 Details of CT specimen being used at Studsvik to examine the CGR behavior in Alloy 690 HAZ material [27]

5.1.2.4.4

Investigations in Simulated Primary Water at CIEMAT in Spain

CIEMAT is carrying out a MRP- and UNESA-sponsored test program on the CGR susceptibility of both Alloy 600 and Alloy 690 HAZs [32]. Two CGR tests on Alloy 600 HAZs have been carried out in simulated primary water at 360C. Specimen evaluation proved complicated, however, due both to fabrication defects in some of the test welds and incomplete linearity of the weld fusion lines [33]. Subsequent testing of Alloy 690 HAZ specimens has been delayed pending clarification of these issues, but further results are expcted shortly.
5.1.2.4.5 Investigations in Simulated Primary Water at Tohoku University in Japan

Information on the main testing program at Tohoku University was reported in Section 5.1.2.1.4. As part of this, a 12.5mm thick CDCB specimen with the notch located 2 mm away from, and parallel to, the fusion line has apparently been used for an Alloy 690TT HAZ test [8]. Further details and current status are not known, but it is believed that SCC was not detected after 4360 h at 340C for this specimen, indicating that this particular HAZ probably retained high PWSCC resistance despite significant hardening from the welding process.

5-62

PWSCC of Thick-Walled Alloy 690 Material

5.2

Field Experience

In contrast to the situation with thin-walled SG tubing described in Section 4.2, new or replacement components made from thick-walled Alloy 690 material have only been in service for much shorter periods. Although no problems whatsoever with cracking have been reported to date, there is also not yet the same body of data available from repeat inspections to demonstrate conclusively the absence of incipient PWSCC. However, the efforts made in France to assess the field performance of CRDM penetrations are of considerable value here and some key points from these were recently described by EDF [1]: Repeated inspection of RPV head penetrations is taking place during every 10-year unit outage for 3 lead units as follows: Bugey 3 (head replaced in September 1994) was first inspected in 2002, i.e. after around 8 years of operation. Gravelines 4 (head replaced in March 1994) and Blayais 2 (head replaced in March 1995) were first inspected in 2003, i.e. after 9 and 8 years of operation, respectively. Next inspections for all 3 lead units will be in 2013.

The primary inspection uses eddy-current techniques and indications are noted if their lengths are 2 mm. Any such indications would then be sized and characterized by further ultrasonic inspection. No flaws have been detected so far.

5.3

References

1. F. Cattant, Inspection of Reactor Pressure Vessel Heads equipped with A690 penetrations at EDF. Presentation to the EPRI Primary Systems Corrosion Research Committee, Long Beach, CA: January 2009. 2. G. Theus et al., Materials Reliability Program: Material Production and Component Fabrication and Installation Practices for Alloy 690 Replacement Components in Pressurized Water Reactor Plants (MRP-245). EPRI, Palo Alto, CA: 2008. 1016608. 3. F. Vaillant, EDF Report HT-44/95/013/A, 1996, Rsistance a la corrosion sous contrainte en milieu primaire des alliages 690 et 800 Point des rsultats en Dcembre 1995, (English translation of the title, Resistance of Alloys 690 and 800 to Stress Corrosion Cracking in PWR Primary Water Status of Results Available to December 1995). 4. S. Asada et al., PWSCC Life Time Evaluation on Alloy 690,52 and 152 for PWR Materials. EPRI PWSCC of Alloy 600 2007 International Conference & Exhibition, June 11-14, 2007, Atlanta, GA. 5. S. Asada et al., PWSCC Life Time Evaluation on Alloy 690,52 and 152 for PWR Materials. EPRI MRP PWSCC Expert Panel Meeting, St. Petersburg, Florida (Nov. 2007). 6. G.S. Was and S. Teysseyre, Challenges and Recent Progress in Corrosion and Stress Corrosion Cracking of Alloys for Supercritical Water ReactorCore Components, Proc. 12th Int. Conf. On Environmental Degradation of Materials in Nuclear Power System Water Reactors, TMS (The Minerals, Metals & Materials Society), 2005. 5-63

PWSCC of Thick-Walled Alloy 690 Material

7. G.S. Was and S. Teysseyre, Materials Reliability Program: Constant Extension Rate SCC Testing of Alloys 600 and 690 in Supercritical Water. EPRI, Palo Alto, CA (2007). Technical Report 1016154 (MRP-233). 8. J. Hickling, EPRI Materials Reliability Program: Resistance of Alloys 690, 152 and 52 to Primary Water Stress Corrosion Cracking (MRP-237, Rev 1): Summary of findings from completed and ongoing test programs since 2004. EPRI, Palo Alto, CA: 2008. 1018130. 9. P.L. Andresen, Development of Advanced Testing Techniques to Quantify the Improved PWSCC Resistance of Alloy 690 and its Weld Metals. EPRI, Palo Alto, CA: (2004). Technical Report 1010269 (MRP-123). 10. M. Morra et al., Quantification of grain size and banding in differently Thermo-MechanicalProcessed (TMP) Heats of Alloy 690 using Image Analysis. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 11. B. Alexandreanu et al., Crack Growth Rates of Alloys 690 and 152 in PWR Environment at 320C. EPRI MRP PWSCC Expert Panel Meeting, Atlanta, GA (Oct. 2006). 12. B. Alexandreanu et al., SCC CGRs of Alloys 690 and 152 Weld in PWR Water. EPRI MRP PWSCC Expert Panel Meeting, St. Petersburg, Florida (Nov. 2007). 13. B. Alexandreanu et al., SCC CGRs of Alloys 690 and 52/152 Welds in PWR Water. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 14. B. Alexandreanu et al., Task 3: Cracking of Nickel Alloys and Welds CGRs of Alloys 600 and 690 in PWR Water, ANL presentation at NRC review meeting, Argonne, IL, September 25-26, 2007. 15. B. Alexandreanu, SCC CGRs of Alloys 690 and 52/152 Welds in PWR Water, Alloys 690/52/152 PWSCC Research Test Materials Meeting, Industry/NRC RES, July 17-18, 2008, Rockville, MD http://www.nrc.gov/reading-rm/adams/web-based.html, ADAMS Accession Number: ML082140693. 16. W.J. Shack and T.F. Kassner, Review of Environmental Effects on Fatigue Crack Growth of Austenitic Stainless Steels, NUREG/CR-6176 ANL-94/1, (May 1994). 17. P. Andresen and M. Morra, SCC Growth Rate Data in Alloy 690 Base and Weld Metals. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 18. P. Andresen and M. Morra, Materials Reliability Program: Laboratory Testing to Determine Resistance of Alloys 690/52/152 to Stress Corrosion Crack Growth in Simulated Primary WaterAn Update (MRP-253). EPRI, Palo Alto, CA: 2008. 1016604. 19. T. Shoji, Program on Technology Innovation: Prediction and Evaluation of Environmentally Assisted Cracking in LWR Structural Materials: Interim Report on PEACE-E, November 2008. EPRI, Palo Alto, CA, Tohoku University, EDF-SEPTEN, Hitachi Ltd., The Japan Atomic Power Company, The Kansai Electric Power Co., Inc., Mitsubishi Heavy Industries, Ltd., Swedish Radiation Safety Authority, Tohoku Electric Power Co., Inc., Tokyo Electric Power Company, Toshiba Corporation, and Ishikawajima- Harima Heavy Industries Co., Ltd.: 2008. 1016549.

5-64

PWSCC of Thick-Walled Alloy 690 Material

20. T. Shoji, Program on Technology Innovation: Prediction and Evaluation of Environmentally Assisted Cracking in LWR Structural Materials: Interim Report on Phase III. EPRI, Palo Alto, CA; Fracture and Reliability Research Institute, Tohoku University, Miyago, Japan; EDF-SEPTEN, Villeurbanne Cedex, France; Hitachi Ltd.; The Japan Atomic Power Company; The Kansai Electric Power Co., Inc.; Mitsubishi Heavy Industries, Ltd.; Swedish Nuclear Power Inspectorate, Stockholm, Sweden; Tohoku Electric Power Co., Inc.; Tokyo Electric Power Company; Toshiba Corporation; and Ishikawajima Harima Heavy Industries Co.: 2006. 1013380. 21. T. Arai: CRIEPI Research Program on PWSCC, EPRI MRP PWSCC Expert Panel Meeting, St. Petersburg, Florida (Nov. 2007). 22. D.J. Paraventi and W.C. Moshier, Alloy 690 SCC Growth Rate Testing. EPRI MRP PWSCC Expert Panel Meeting, Atlanta, GA (2006). 23. D.J. Paraventi and W.C. Moshier, Alloy 690 SCC Growth Rate Testing. EPRI Workshop on Cold Work in Iron and Nickel Base Alloys, Toronto (2007). 24. M.G. Burke, Microstructual Characterization Alloys 690 and 600. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 25. R. Jacko, Materials Reliability Program: Testing the Resistance to Stress Corrosion Cracking of Alloy 690 and its Weld Metal in Supercritical Boron/Lithium/H2 Solutions (MRP-225). EPRI, Palo Alto, CA: 2007. 1015004. 26. M. Toloczko and S. Bruemmer, Crack Growth Response of Alloy 690/152/52 in Simulated PWR Water. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 27. A. Jenssen, K. Norring and P. Efsing, Swedish Activities on Alloy 690 and its Weld Metals. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 28. Y. Yamamoto et al., Development of the Crack Growth Rate Curves for Stress Corrosion Cracking of Nickel Based Alloys in a Simulated Primary Water Environment, Fontevraud 6 (2006). 29. P.L. Andresen, Resistance of Alloy 690/52/152 to Stress Corrosion Crack Growth in Simulated PWR Primary Water. EPRI, Palo Alto, CA: (2006). Technical Report 1013516 (MRP-196). 30. G.A. Young, N. Lewis & D.S. Morton: The Stress Corrosion Crack Growth Rate of Alloy 600 Heat Affected Zones Exposed to High Purity Water, NUREG/CP-0191, The Vessel Penetration Inspection, Crack Growth and Repair Conference, Oct. 2, 2003, Gaithersburg, MD, Vol. 1, pp. 371-386. 31. G.A. Young et al.: Stress Corrosion Cracking of Alloy 600 and Alloy 690 Heat Affected Zone Material in High Purity Deaerated Water. EPRI MRP PWSCC Expert Panel Meeting, Atlanta, GA (2006). 32. D. Gomez-Briceno, Crack Growth Rate Studies of Weld HAZ of Alloys 600 and 690 Materials. EPRI MRP PWSCC Expert Panel Meeting, Atlanta, GA (Oct. 2006). 33. D. Gomez-Briceno and J. Lapena, Materials Reliability Program: Effect of Defects in an Alloy 600/82 Weld on Stress Corrosion Cracking in Testing of Heat Affected Zone Specimens (MRP-254). EPRI, Palo Alto, CA: 2008. 1016607.

5-65

6
DISCUSSION
6.1 Resistance of Alloy 690 to PWSCC

The PWR industry selection of Alloy 690 as the material of choice to replace Alloy 600 components susceptible to PWSCC appears to have been a sound one for three main reasons: Firstly, the field experience to date with Alloy 690 components has been exemplary, with no sign whatsoever of PWSCC. This is true both for thin-walled SG tubing (with many thousands of tubes now having been exposed for up to nearly 20 years) and for thick-walled components such as RPV head penetrations (where service experience without detection of cracks has already been demonstrated in 3 lead units after nearly 10 years). Secondly, Alloy 690 has proved to be almost entirely resistant to PWSCC initiation in a wide variety of laboratory tests (see Sections 4.1 and 5.1.1), the only marginal exceptions being some minor cracking in pre-production heats of SG tubing tested under very severe conditions (see Sections 4.1.3 and 4.1.4). Thirdly, fatigue pre-cracked, thick-walled Alloy 690 CRDM material thought to be generally representative of real components has shown no, or only insignificantoo, amounts of crack growth when tested under virtually static load, i.e. at (nearly) constant, stress intensity in simulated PWR primary water. This has been found in multiple laboratories, even when the experimental conditions (temperature, crack transitioning, periodic partial unloading, test duration, etc.) have been chosen so as to maximise even the smallest degree of inherent SCC susceptibility (see Section 5.1.2.2 and Table 6-1). The only real case to date where as-received Alloy 690TT has shown SCC CGRs > 5E-9 mm/s without deliberate cold working involves testing in supercritical water containing additions of Li and B (see Section 5.1.2.2.5) and the relevance of these results to actual PWR operation remains unclear. Even in this case, however, the CGRs observed were only ~ 1E-8 mm/s, i.e. a factor of 2x higher than the threshold rate suggested as representing complete lack of engineering significance in primary water and some 60x lower than the SCC CGRs for Alloy 600 tested in the same supercritical water environment. Somewhat higher rates were measured, however, in specimens from materials that had been subjected to deliberate cold working see Section 5.1.2.3.3.

oo

It is suggested that complete lack of engineering significance might best be defined as a CGR 5E-9 mm/s (i.e. a maximum of ~0.15 mm/year); this often corresponds to around the average grain diameter in thick-walled Alloy 690 materials.

6-1

Discussion

The absence of any significant SCC growth from a pre-existing crack in the thick-walled Alloy 690 CRDM material specimens tested to date in simulated primary water is obviously very satisfactory. This continues to be the case even after the deliberate introduction of significant amounts of additional cold work (both uniform and non-uniform) to as-received material, as summarized in Section 5.1.2.3 and Table 6-2. Thus, it is somewhat difficult to report a meaningful factor of improvement (FOI) for CGR with regard to Alloy 600, but comparison of all the data reported to th date for Alloy 690 CRDM penetration materials with the 75 percentile curve for Alloy 600 CGRs from MRP-55 [12] would suggest a FOI of at least 100x and probably higher.

6-2

Discussion

Table 6-1 Summary of results of CGR testing in simulated PWR primary water on Alloy 690 CRDM material without deliberate cold working (status: December 2008)
Testing Lab Material Supplier Material Form Heat # Final Condition MicroStructure Specimen # Pre-Crack Orientation Test Temp. (C) H2 Level Eff. K Level (cc/kg) (MPam) Rep. CGR (mm/s) Remarks No clear sign of IGSCC after 1000h at constant load CGR increased to ~8E-9 on raising temperature to 350C No SCC found after 4000h; CGR effectively zero Only very isolated signs of SCC; CGR almost zero Only very isolated signs of SCC; CGR almost zero Tests in progress, DCPD data only. CGRs increased by ~2X at H2 = 20 cc/kg. Tests in progress, DCPD data only. CGRs with time under constant K conditions Tested under PPU with increasing hold times; no signs of IGSCC; CGR ~ 0 Tested under PPU with increasing hold times; no signs of IGSCC; CGR ~ 0

ANL

Valinox

CRDM tubing

WP142

TT

pp

homogeneous

C690-CR1

long/radial?

320

23

24

2E-09

ANL

Valinox

CRDM tubing CRDM tubing?

WP142 Not known

TT

homogeneous

C690-LR-2 circ/radial?

320; 350 Not known

23 Not known

21

3E-09

MHI

TT

homogeneous

Not known

Not known

Not known

1E-10

PNNL

Valinox

CRDM tubing

RE243

TT
qq

homogeneous

CT14

long/radial

325

29

30; 40

3E-10

PNNL

Valinox

CRDM tubing

RE243

CM

homogeneous

CT15

long/radial

325

29

30; 40

2E-10

PNNL

Valinox

CRDM tubing

WP140

TT

homogeneous

CT26

long/radial

350

29; 20

30

5E-10

PNNL

Valinox

CRDM tubing

WP142

TT

homogeneous

CT27

long/radial

350

29; 20

30

7E-10

Studsvik

Sumitomo

archive CRDM penetration D520906 material

TT

homogeneous

182-1

long/circ

320

25-29

25

1E-10

Studsvik
pp qq

Sumitomo

archive CRDM penetration material D520906

TT

homogeneous

182-2

long/circ

320

30-31

25

1E-10

TT = Thermally Treated CM = Carbide Modified (TT plus an additional anneal)

6-3

Discussion Table 6-2 Summary of results of CGR testing in simulated PWR primary water on Alloy 690 CRDM material with deliberate cold working (status: December 2008)
Testing Lab Material Supplier Material Form Heat # Final Condition MicroStructure Type of Extra CW Method CW Level (%) Test Spec. Pre-Crack Temp. # Orientation (C) H2 Level (cc/kg) Eff. K Level (MPam) Rep. CGR (mm/s) Remarks

GEGRC

Duke Energy

extruded CRDM penetrati on WN415

TT

rr

homogeneous

uniform

crossforging

41

c280

T-L

360

18

38.5

1E-09

Long-term test under PPU with increasing hold times; IGSCC CGR effectively zero Tested under PPU with <1000h at constant K; IGSCC CGR very low Tested under PPU with <1000h at constant K; IGSCC CGR very low Only isolated signs of SCC; CGR very low indeed Only isolated signs of SCC; CGR very low indeed

GEGRC

Duke Energy

extruded CRDM penetrati on WN415

TT

homogeneous

uniform

crossforging

20

c285

T-L

360

18

27.5

5E-09

GEGRC

Duke Energy

extruded CRDM penetrati on WN415

TT

homogeneous

uniform

crossforging

20

c286

T-L

360

18

27.5

3E-09

PNNL

Valinox

CRDM tubing

RE243

TT

homogeneous

nonuniform

1-D rolling

17

CT19

S-L

325

29

30; 40

4E-10

PNNL

Valinox

CRDM tubing

RE243

CM

ss

homogeneous

nonuniform

1-D rolling

17

CT20

S-L

325

29

30; 40

4E-10

rr ss

TT = Thermally Treated CM = Carbide Modified (TT plus an additional anneal)

6-4

Discussion

Table 6-2 Summary of results of CGR testing in simulated PWR primary water on Alloy 690 CRDM material with deliberate cold working (status: December 2008) (continued)
Testing Lab Material Supplier Material Form Heat # Final Condition MicroStructure Type of Extra CW Method CW Level (%) Test Spec. Pre-Crack Temp. # Orientation (C) H2 Level (cc/kg) Eff. K Level (MPam) Rep. CGR (mm/s) Remarks

PNNL

Valinox

CRDM tubing

RE243

TT

homogeneous

nonuniform

1-D rolling

30

CT22

T-L

350

29

30; 40

3E-9

Somewhat more IGSCC; CGR very low Only isolated signs of SCC; CGR very low indeed

PNNL

Valinox

CRDM tubing

RE243

CM

homogeneous

nonuniform

1-D rolling

30

CT23

T-l

350

29

30; 40

9E-10

6-5

Discussion

It should be noted, however, that the range of material chemistries, mechanical properties and microstructures of Alloy 690 material used for RPV head penetrations in the field is thought to be larger than is represented by the approx. 10 heats tested worldwide to date in various laboratories. Similarly, the complete spectrum of material variability for long-term exposure of all thick-walled Alloy 690 components to primary water in both existing and new plants has not yet been fully clarified. Of primary concern is the possible presence of inhomogeneous microstructures and significant amounts of unidirectional tensile strain in Alloy 690 material, given the further results of CGR testing discussed below. In an earlier report [4], the present author described a bimodal aspect of Alloy 690 laboratory PWSCC behavior where it has been possible under certain circumstances to demonstrate moderate to relatively rapid intergranular crack growthtt in plate and bar materials, thought to be less representative of actual plant components. With regard to testing in high-temperature water, this has always involved additional forms of non-uniform cold work, as reported in detail in Section 5.1.2.1. Figure 6-1 shows a summary diagram of these results prepared in November 2008 by PNNL [26] and compares the CGRs measured on 1-D cold-rolled specimens with the MRP-55 reference disposition curve for cracking in thick-walled Alloy 600 material [12]. It can be seen that most of the Bettis data, in particular, actually lie above this curve, but that such results have also been reproduced in other laboratories (GE-GRC, INSS). Some (but not alluu) of these points are thought to represent the situation where the additional, non-uniform deformation at RT has been superimposed on an inhomogeneous microstructure in the original material (particularly with regard to banding originating both from segregation (e.g. of carbides) and grains of varying size. If such banding (see, e.g., Figure 6-2) is aligned with the direction of subsequent cold working and crack propagation, as may have been the case with the S-L and S-T orientations of the ANL plate material (see Sections 5.1.2.1.2 and 5.1.2.1.3), it is intuitively plausible that high SCC susceptibility may result. In contrast, extensive banding perpendicular to the direction of crack growth (see Figure 6-3) did not induce high CGRs in the original feasibility studies on EPRI plate material (see Section 5.1.2.1.1). For comparison purposes, Figure 6-4 shows the very uniform, homogeneous microstructure of an extruded CRDM penetration that was extremely resistant to SCC growth, even when subjected to deliberate cold working (see Section 5.1.2.3.1). A further group of data points (from ANL, Bettis, GE-GRC and INSS) lie below the MRP-55 curve, but above 5E-9mm/s, the suggested threshold for CGRs that are of no engineering significance. As discussed above, the CGR results from CRDM material all lie below that threshold value, even for the PNNL tests with 30% 1-D cold rolling and a relatively high stress intensity of 40 MPam. Points plotted at 1E-10 mm/s in Figure 6-1 indicate virtually no SCC growth at all, but all data below 1E-9 mm/s can really be regarded as representing a situation where the CGR under constant load is so slow that it usually corresponds to only very localized pockets of minor SCC along the pre-fatigue crack front. Over long periods of time, such growth is unlikely to be sustained, owing to the restraining influence of adjacent uncracked areas. It thus represents an overall CGR that is also very nearly zero.

tt uu

Up to 1000x higher than in CRDM material without deliberate cold work. Bettis report an apparent absence of significant banding in the one heat of their 3 test materials that showed the highest CGRs.

6-6

Discussion

Figure 6-1 Summary of laboratory SCC CGR data (as of November 2008) prepared by PNNL [26] and showing the possibility of measuring moderate to high rates in Alloy 690 plate material subjected to non-uniform cold work

Figure 6-2 Evidence of microstructural banding in some areas of the 1-D cold-rolled ANL Alloy 690 plate material examined at GE-GRC [18]

6-7

Discussion

Figure 6-3 Microstructural banding perpendicular to the crack plane in Alloy 690 plate material from EPRI orginally tested for the MRP Program by GE-GRC [9]

Figure 6-4 Very uniform, homogeneous microstructure in extruded Alloy 690 CRDM material [18]

The mechanism by which unidirectional cold work (with or without prior microstructural inhomogeneity) can induce SCC susceptibility in an otherwise very resistant material such as Alloy 690 is currently unclear and is the subject of much ongoing research. It is noteworthy that several laboratories (ANL, Bettis, GE-GRC, INSS) have found that the expected dependency of CGR on test temperature (based upon Alloy 600 cracking experience cf. Figure 5-27) appears to disappear almost entirely for the most susceptible materials when tested in the worst crack orientation (see Figure 5-18, Figure 5-20, Figure 5-31 and Figure 5-34). Furthermore, the CGRs for such specimens appear to exhibit either little effect of dissolved hydrogen concentration (Figure 5-25), or the opposite behavior to that known from Alloy 600 cracking (see Figure 5-22 and Figure 5-35), and little effect of varying stress intensity. In the authors opinion, all of these observations suggest bimodal behavior, rather than a continuum of Alloy 690 cracking response from very slow to high CGRs. 6-8

Discussion

There has been lively debate concerning the relevance of these moderate to high laboratory CGR data from unidirectionally deformed Alloy 690 plate and bar materials to the behavior of actual reactor components. On the one hand, such materials do not appear to be representative of what is being used in plants, as discussed in Section 2.1.2, although some further clarification (e.g. with regard to possible cold-straightening of extruded CRDM piping) of the complete range of actual material conditions is definitely required. On the other hand, significant, highly localized plastic strains can be expected immediately adjacent to the fusion line in the HAZ of thick-walled Alloy 690 after welding. Although the resulting local microstructure is likely to be very different from that produced by 1-D cold rolling of plate, it is noteworthy that high CGRs were also found during the Bettis studies on material that had been subjected to RT tensile straining (see Section 5.1.2.1.7). This is thought to be far more representative of what might be present in Alloy 690 base material adjacent to real welds [18]. Attempts are also being made to test actual HAZ material, as reported in Section 5.1.2.4, but this presents real experimental challenges and the data available to date are promising, but inconclusive. For the abovementioned reasons, widespread research [1] is continuing with the emphasis being placed both on protypical CRDM materials (whose extreme resistance to PWSCC requires very long testing times to obtain valid data) and on such materials that have been subjected to deliberate cold working and may be more susceptible. If bimodal behavior is involved here, testing of unidirectionally deformed material even if unrepresentative of field conditions may permit the boundaries of Alloy 690 PWSCC susceptibility to be established more quickly and reliably than concentrating just on good material. Furthermore, a better understanding of the way in which non-uniform deformation can induce significant susceptibility to crack growth during laboratory testing is clearly required, as is clarification of whether or not such materials would undergo crack initiation from a smooth surface in simulated primary water. In addition, efforts are being pursued in at least one laboratory [2] to understand the inconclusive results that have been obtained to date during testing of Alloy 690 materials in supercritical water containing Li and B additions, since this was originally expected to be a reliable accelerated test to demonstrate satisfactory PWSCC resistance in advance of long-term field exposure. Although evidence that a pre-existing (fatigue) crack does not exhibit significant growth at constant load in primary water is clearly the best possible demonstration of Alloy 690 PWSCC resistance, it should be noted that the overwhelming majority of cracking incidents with Alloy 600 have actually involved incipient crack formation at a smooth surface. Thus the initiation data (lab and field) from SG tubing (described in Section 4) and the lab testing of thick-walled Alloy 690 for crack initiation (described in Section 5.1.1) are also of considerable importance. Despite the large variety of test methods used, these show essentially no crack formation through PWSCC for materials in all the metallurgical conditions examined to date. These are thought to be generally representative of most field components, including thick-walled CRDM penetrations, although further clarification is needed of some factors (such as cold work from final straightening operations and high residual strains in weld heat-affected zones). It should also be noted that no initiation testing has yet been carried out on the 1-D cold-rolled Alloy 690 plate material that has shown moderate to high susceptibility to crack growth through PWSCC. In the 2004 MRP-111 report [3], the average FOI from lab-testing for crack initiation in thin-walled Alloy 690TT was derived as 26x relative to Alloy 600MA and 13x relative to Alloy 600TT (see Section 4.1.7), although it was emphasized that these numbers were clearly very conservative 6-9

Discussion

and likely to increase, owing to the absence of cracking in the Alloy 690 specimens. An assessment of the relevant lab data for both thin and thick-walled material in 2008 [4] concluded that minimum factors of 40 to 100x can now be justified. As described in Section 4.2, field experience with SG tubing already justifies a FOI >20x, and the realistic value is expected to lie much higher.

6.2

Other Aspects of Alloy 690 Corrosion Behavior

One of the gaps identified in the 2004 MRP-111 report was the absence of corrosion fatigue data for Alloy 690 in primary water. As reported in Section 3.2, this deficit has largely been rectified. Thick-walled Alloy 690 does show a reduction in fatigue life when tested under exposure to a simulated primary water environment, and the extent of the decrease depends upon a number of factors (especially testing frequency/strain rate). However, this behavior is about the same, or somewhat better, than that of other Ni-base alloys. Similarly, cyclic crack growth rates are sometimes affected (in this case, increasedvv) by simultaneous exposure to a high-temperature water environment, but generally slightly less than for Alloy 600. The only exception to this involves the more accelerated cyclic crack growth in simulated primary water seen for the same 1-D cold rolled plate of Alloy 690 from ANL that showed relatively high susceptibility to PWSCC (see Section 5.1.2.1.2), but - as discussed above - this material is not thought to be of direct relevance to plant components. In contrast to this finding, prototypical Alloy 690 CRDM material (that was highly resistant to PWSCC crack growth) showed no environmental enhancement of fatigue crack growth at all in recent Swedish studies [15]. With regard to low temperature crack propagation (LTCP - see Section 3.4), there is a consensus among experts that Alloy 82 and 182 give rise to more potential concerns than the Alloy 152 and 52 weld metals, and that Alloy 690 base metal is of even less concern. The latter is expected to show adequate fracture resistance even under extreme loading conditions in low-temperature water, particularly if dissolved hydrogen levels are kept low, as long as its metallurgical condition is good. As discussed above for PWSCC resistance, however, this assessment may no longer be valid for non-representative materials that have been subjected to severe unidirectional cold work. Additional clarification of the boundaries to enhanced LTCP susceptibility for Alloy 690 is recommended.

6.3

References

1. A. Ahluwalia and R. Tregoning, Collaborative Research Activities: Performance of Alloy 690/152/52 in Nuclear Environments. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 2. R. Jacko and J.K. McKinley, PWSCC Resistance of Alloy 690 and its Weld Metals: Cold Work, Temperature and H2 Effects. EPRI MRP PWSCC Expert Panel Meeting, Los Angeles, CA (Nov. 2008). 3. H. Xu et al., Materials Reliability Program (MRP), Resistance to Primary Water Stress Corrosion Cracking of Alloys 690, 52, and 152 in Pressurized Water Reactors (MRP-111). EPRI, Palo Alto, CA: 2004. 1009801. 4. J. Hickling, EPRI Materials Reliability Program: Resistance of Alloys 690, 152 and 52 to Primary Water Stress Corrosion Cracking (MRP-237, Rev 1): Summary of findings from completed and ongoing test programs since 2004. EPRI, Palo Alto, CA: 2008. 1018130.
vv

E.g., by some 5 to 10 times in Japanese studies at slow loading frequencies [16].

6-10

7
CONCLUSIONS
Significant progress towards demonstrating satisfactory, long-term, PWSCC resistance of thick-section Alloy 690 by laboratory testing has been made since the publication in 2004 of MRP-111, but complete immunity to cracking should not be assumed: Crack initiation through PWSCC from a smooth surface appears to be very difficult, if not impossible, for Alloy 690TT in the metallurgical conditions tested to date. These are thought to be generally representative of most field components, including thick-walled CRDM penetrations, although further clarification is needed of some factors (such as cold work from final tube straightening operations and possible high residual strains in weld heataffected zones). Minimum relative factors of improvement for initiation of 40 to 100 times for Alloy 690TT can now be justified with regard to PWSCC of Alloy 600MA. Such estimates are conservative and likely to rise further with continued testing, since no actual PWSCC originating at a smooth surface has yet been seen (except for marginal levels of crack initiation in some specimens subjected to severe deformation in CERT). These findings from laboratory testing are in accord with the total absence of detected cracks during in-service inspections after nearly 20 years of plant operation with Alloy 690 steam generator tubing materials and nearly 10 years with CRDM nozzles in replacement RPV heads. For thick-walled Alloy 690TT material in the normal metallurgical condition (e.g. as extruded piping for CRDM penetrations), susceptibility to crack growth through PWSCC appears to be very marginal at the most. Using sophisticated testing methods, it has been possible to produce small amounts of intergranular cracking in some, but not all, of the specimens tested to date. However, the measured CGRs are extremely low (< 5E-9 mm/s or < 0.15 mm/yr) and thus of no engineering significance. Despite the small increase in measured CGRs sometimes seen, this statement remains valid even after the deliberate introduction of significant amounts (i.e., well over 10%) of uniform cold work into CRDM material. To date, even attempts to induce higher SCC susceptibility by introducing non-uniform deformation (e.g. by unidirectional cold rolling) into actual CRDM material have not produced CGRs that are of any concern. The relative factor of improvement for Alloy 690 CRDM material with respect to CGRs in Alloy 600MA is thought to be over 100 times (and may actually be much higher).

7-1

Conclusions

Perhaps inevitably, however, certain conditions have been found that appear to lead to bimodal behavior of some Alloy 690 materials in the sense that a much higher susceptibilty to crack growth through SCC (up to 1000 times faster) is then observed during laboratory testing of pre-cracked specimens in high-temperature environments: Rapid, intergranular crack growth (sometimes even at rates comparable to those for Alloy 600) in both pure water and simulated primary water can be triggered in certain directions within some Alloy 690 plate and bar materials by the introduction of inhomogeneous cold work (particularly by uni-directional rolling, but also by tensile straining). In many cases, this also appears to lead to major changes in the dependence of measured CGRs on stress intensity, test temperature and dissolved hydrogen levels, possibly indicating a different cracking mechanism. The limits of such behavior have not yet been satisfactorily established and there is no real mechanistic understanding of the phenomenon. To date, the Alloy 690 materials exhibiting such high susceptibility to intergranular SCC after unidirectional cold working are not thought to be directly relevant to PWR plant components, but further efforts are required to confirm that this interpretation will remain valid under all circumstances. Significant intergranular crack growth has been observed in Alloy 690TT (including in CRDM material after the introduction of 10 % uniform cold work) during testing in a supercritical environment containing lithium, boron and hydrogen. There is presently no established basis for extrapolating the measured CGRs at 385 oC down to subcritical temperatures in order to make detailed comparisons with the existing databases for primary water.

In terms of the further research needed to reach the original goal of demonstrating long-term absence of PWSCC in all thick-walled plant components made from Alloy 690 base material, work is ongoing in a number of areas and these efforts are being coordinated as part of an international collaboration. The industry and regulatory prioritization of these is seen as follows: Complete planned testing programs to investigate the possibility of enhanced PWSCC susceptibility in the HAZ adjacent to welds in thick-walled Alloy 690. Establish the limits to the apparently bimodal behavior of Alloy 690 (high susceptibility to SCC crack growth in some material that has been subjected to non-uniform cold work) that has been observed to date in laboratory testing and understand why this occurs. Carry out testing for crack initiation on some of the uni-directionally rolled Alloy 690 materials that have shown appreciable susceptibility to crack growth through PWSCC. Investigate rigorously whether such material conditions could have any relevance to real plant components, taking into account that the complete spectrum of material variability for long-term exposure of thick-walled Alloy 690 to primary water in both existing and new plants has not yet been fully clarified.

7-2

Conclusions

Considering other knowledge gaps for Alloy 690 identified in the 2004 MRP-111 report, that regarding corrosion fatigue appears to have been nearly closed. Alloy 690 TT material representative of thick-walled plant components is expected to exhibit a reduction in fatigue life and/or an acceleration in cyclic crack growth rates as a result of exposure to primary water that is comparable with, or better than, the known behavior for other Ni-base alloys (such as Alloy 600). With regard to low temperature crack propagation (LTCP), there is a consensus among experts that Alloy 82 and 182 give rise to more potential concerns than the Alloy 152 and 52 weld metals, and that Alloy 690 base metal is of even less concern. The latter is expected to show adequate fracture resistance even under extreme loading conditions in low-temperature water, particularly if dissolved hydrogen levels are kept low, as long as its metallurgical condition is good. Some additional clarification of the boundaries to enhanced LTCP susceptibility for less-representative Alloy 690 materials is recommended.

7-3

A
TRANSLATED TABLE OF CONTENTS
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM: (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE, (IV) THAT ANY TRANSLATION FROM THE ENGLISH-LANGUAGE ORIGINAL OF THIS DOCUMENT IS WITHOUT ERROR; OR (B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT. THE TRANSLATION OF THIS DOCUMENT FROM THE ENGLISH-LANGUAGE ORIGINAL HAS BEEN PREPARED WITH LIMITED BUDGETARY RESOURCES BY OR ON BEHALF OF EPRI. IT IS PROVIDED FOR REFERENCE PURPOSES ONLY AND EPRI DISCLAIMS ALL RESPONSIBILITY FOR ITS ACCURACY. THE ENGLISH-LANGUAGE ORIGINAL SHOULD BE CONSULTED TO CROSS-CHECK TERMS AND STATEMENTS IN THE TRANSLATION. ORGANIZATION(S) THAT PREPARED THIS DOCUMENT Electric Power Research Institute (EPRI)

A-1

Translated Table of Contents

: Alloy 690 (MRP-258)

1019086

2009

600(Alloy 182Alloy 82) (PWR) 30PWR (PWSCC)Alloy 600

A-2

Translated Table of Contents

PWSCC 690Alloy 152Alloy 5252MPWSCC 20 2004EPRI1009801 (MRP111)Alloy 690PWSCCAlloy 15252 2004(MRP-111)Alloy 690 MRP111PWSCCPWR 20690 Alloy 600Alloy 690PWRAlloy 690 PWSCC 690P WRAlloy 600(

A-3

Translated Table of Contents

/) Alloy 60040100 Alloy 690PWSCC Alloy 690 S CC (5E-9 mm/s0.15 mm/year)Alloy 600MAAlloy 690 CRDM100 Alloy 690 PWR Alloy 690PW R1020 PWRAlloy 600 Alloy 600PWSCC

A-4

Translated Table of Contents

/ EPRI Alloy 690PWSCCAlloy 690PWSCC PWSCCAlloy 690 Alloy 690 Alloy 600 Alloy 690 PWSCC

RPV

A-5

Translated Table of Contents

1 .......................................................................................................................... 1-1 1.1 .................................................................................................................... 1-1 1.2 MRP-111 ........................................................................... 1-2 1.3 Alloy 600 PWSCC .................................................................................... 1-4 1.4 ............................................................................................................. 1-5 2 ALLOY 690........................................................................................ 2-1 2.1 .......................................................................................................... 2-1 2.1.1 2.1.2
Alloy 690 SGPWR ................................................... 2-3 Alloy 690 ............................................................... 2-4

2.2 Alloy 690 .............................................................................................. 2-4 2.3 ............................................................................ 2-5 2.4 .............................................................................. 2-8 2.5 .................................................................................................. 2-12 2.6 ........................................................................................................... 2-13 3 PWSCCALLOY 690 ..................................................................... 3-1 3.1 .......................................................................... 3-1 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 SedricksSG 1979 ............................................................... 3-1 K. SmithSG 1985 .............................................................. 3-2 YonezawaSG 1985 ........................................................... 3-2 Esposito 1991 ..................................................................................... 3-2 PWSCC ............................................ 3-3

3.2 ............................................................................. 3-4 3.3 .................................................................................. 3-15 3.4 (LTCP) ........................................................................... 3-17 3.4.1 3.4.2 ............................................................................................ 3-17 PWRLTCP ...................................... 3-18

A-6

Translated Table of Contents

3.5 ........................................................................................................... 3-23 4 SGPWSCC ............................................................................................. 4-1 4.1 ......................................................................................................... 4-1 4.1.1 4.1.2 4.1.3 4.1.4 4.1.5 4.1.6 4.1.7 ................................................................................................. 4-2 B/LiU.................................................. 4-2 B/LiCERT ........................... 4-5
RUB .......................................................... 4-6

Weibayes ............................................. 4-8

....................................................................... 4-15

Alloy 600.................................................... 4-18

4.2 ............................................................................................................ 4-21 4.3 ........................................................................................................... 4-25 5 ALLOY 690PWSCC................................................................................ 5-1 5.1 ......................................................................................................... 5-2 5.1.1
................................................................................................ 5-3

5.1.1.1 (MHI)5-3 5.1.1.2 .......................................... 5-8 5.1.2 ..................................................................................... 5-10

5.1.2.1 PWRAlloy 690 ..................... 5-10 5.1.2.1.1 5.1.2.1.2 5.1.2.1.3 MRP (GE-GRC)5-20 5.1.2.1.4 5.1.2.1.5 (INSS) ............................................................................................ 5-29 5.1.2.1.6 ............. 5-30 ............. 5-28 (GEGRC) ....... 5-11
(ANL)5-15

A-7

Translated Table of Contents

5.1.2.1.7 5.1.2.1.8

Bechtel Bettis ....................... 5-31

................................................................................... 5-39

5.1.2.2 Alloy 690 CRDM............................. 5-44 5.1.2.2.1 5.1.2.2.2

(PNNL)

(ANL)5-45

.............................................................................. 5-48 5.1.2.2.3 5.1.2.2.4 5.1.2.2.5 ....................................................................... 5-52 5.1.2.3 Alloy 690 CRDM ......................... 5-53 5.1.2.3.1 5.1.2.3.2 (PNNL) ....................................................................... 5-58 5.1.2.3.3 ....................................................................... 5-59 5.1.2.4 Alloy 690(HAZ)...... 5-60 5.1.2.4.1 5.1.2.4.2 5.1.2.4.3 5.1.2.4.4 5.1.2.4.5 KAPL ............................................... 5-60 (ANL)5-60 Studsvik 5-61 CIEMAT....... 5-62 ............. 5-62 (GEGRC) ............................. 5-53 Studsvik 5-52 MHI..................... 5-52

5.2 ........................................................................................................... 5-63

A-8

Translated Table of Contents

5.3 ........................................................................................................... 5-63 6 .......................................................................................................................... 6-1 6.1 PWSCCAlloy 690 ...................................................................... 6-1 6.2 Alloy 690 ................................................................... 6-10 6.3 ........................................................................................................... 6-10 7 .......................................................................................................................... 7-1

2 1 [18]Alloy 690Alloy 600 .............................................. 2-7 2 2 Huey Test--[18]Alloy 690 NX4459HG(0.06%C) .................................................................................. 2-9 2 3 Huey Test--[18]Alloy 690 NX9217H (0.01%C) .................................................................................. 2-10 2 4 Huey Test--[18]Alloy 690 NX9780H (0.01%C) .................................................................................. 2-10 2 5 Alloy 690[2] ........................................................... 2-12 3 1 [11]Ni............................ 3-5 3 2 [11]Ni325CPWR ......... 3-5 33 Ni325CPWR(Fen) [11] ................................................................... 3-6 34 325CLWR(Fen) [11] ...................................................................... 3-7

A-9

Translated Table of Contents

3 5 325CPWRNi [11] ............................................. 3-8 3 6 Alloy 690[12] ................................................................ 3-9 3 7 Alloy 690PWSCCANL [13] ............................. 3-10 3 8 320CPWRAlloy 690 [14]EA C................................................................. 3-10 3 9 320CPWRAlloy 690 .............. 3-11 3 10 PWRAlloys 600690510 [16] ...................................................... 3-12 3 11 PWR Ni() [16]ASME............ 3-13 3 12 Alloy 600690CGR[15]....... 3-14 3 13 Alloy 690CGR ............................ 3-14 3 14 BrownMills[36]LTCP.. 3-19 3 15 RTBrownMills Alloy 690JR[36] ......................................................................................................... 3-20 3 16 RTBrownMills Alloy 690JICT()..... 3-21 3 17 BrownMills[36]J-RAlloy 690()Alloy 600 () .................................................................................................. 3-22 3 18 50CAlloy 690J-RParaventiMoshier[42] .................. 3-23 4 1 [2] %VickersAlloy 600Alloy 690 .................................................... 4-5

A-10

Translated Table of Contents

4 2 360CAlloy 600 RUB[2] Alloy 690(3)Weibayes = 5.0 ......................................................................................... 4-9 4 2 Norring365CRUB [2]
[17]. Alloy 690()Weibayes = 5.04-10

4 4 NorringRUB[2]
365CAlloy 600[17]Alloy 690

Weibayes = 5.0 .......................................................... 4-11 45 NorringRUB[2]365C [17]Alloy 690 Weibayes = 5.0 ........... 4-11 4 6 340C(644F)Alloy 600MA (1) CLT[2]320C (608F)Alloy 600MAAlloy 600TT CLT360C (680F)Alloy 690TT (1) CLT
Weibayes =

5.0 ....................................................................................... 4-12 4 7 320C (608F)20%Alloy 600MAAlloy 600TTRUB360C(680F)Alloy 690TT[2]Alloy 690TT Weibayes = 5.0 ....................................................................................... 4-13 4 8 Vaillant360RUB [2][18]. Alloy 600 RUBMATT454,000 Alloy 600 RUBMATT4Alloy 690 Weibayes = 5.0 .......................................................... 4-14 4 9 Framatome ANP360C(680F)SG[2] Alloy 690TT SG100,000Alloy 690 Weibayes = 5.0 .......................................................... 4-14 4 10 Alloy 6004 3.............. 4-17 4 11 484 24 44-19

A-11

Translated Table of Contents

4 12 [23]Alloys 600690RUB4-21 4 13 [31]Alloy 600TT SG... 4-23 4 14 [31]Alloy 690TT SG... 4-23 5 1 PWSCC[45]MHI ................... 5-5 5 2 PWSCC[45][45] ........................................................................................................................... 5-6 5 3 PWSCC[45] ..................... 5-6 5 4 Alloy 600MAPWSCC58,000 [45]690TT BMI .......................................... 5-7 5 5 Alloy 600MAPWSCC73,000 [45]690TT CRDM ............................ 5-7 5 6 SCW [6]CERTCGR5-8 5 7 400C/25.4MpaSCW [7]EPRI Alloy 600 ........ 5-9 5 8 400C/25.4 MpaSCW [7]EPRI Alloy 690.... 5-10 5 9 Alloy 690 ......................................................................................................................... 5-12 5 10 [9]3000GE-GRCAlloy 690() OGR ............................................. 5-13 5 11 [9]3000GE-GRCAlloy 690() OGR ............................................. 5-14 5 12 GE-GRC [9]Alloy 690PWSCC() ......................................................................................................................... 5-15 5 13 ANL [1115]Alloy 690TT ......................................................................................................................... 5-16 5 14 Alloy 690TT(SL)ANL[1115] ....... 5-17 5 15 Alloy 690TT(ST)ANL[1115] ...... 5-17

A-12

Translated Table of Contents

5 16 S-L[1115]ANL 690TT
............................................................................................ 5-18

5 17 S-T[1115]ANL 690TT ......................................................................................... 5-19 5 18 Alloy 690TTCGR320300C [13]ANL 5-20 5 19 ANLGE-GRC [1718]1D(~26%) Alloy 690 ............................................................. 5-21 5 20 1DAlloy 690ANL2PWSCC 360C325C290C [1718]CGR .......................................... 5-22 5 21 (c372) [1718]PWSCC .......... 5-22 5 22 [1718]CGR1-D ANLGE-GRC....................................... 5-23 5 23 20%1DAlloy 690GE-GRC[1718]PWSCC ......................................................................................................................... 5-24 5 24 [1718]20%1 DAlloy 690GE-GRC .................................................................... 5-24 5 25 20%1DAlloy 690 [178]2 ......................................................................................................................... 5-25 5 26 5 23S-T[1718].................. 5-25 5 27 GE-GRC1DAlloy 690SL[1718] ...................................................................................................................... 5-26 5 28 GE-GRC [1718]26%1DANL 690(c372) ................................................................. 5-27 5 29 20%1DAlloy 690 [1718]5-27

A-13

Translated Table of Contents

5 30 360C ([8])20%690TTINSS ......................................................................................................................... 5-30 5 31 690TT ([8])CGR(CW) ......................................................................................................................... 5-31 5 32 Bettis[2223]CGR5-35 5 33 () [2223] .................................. 5-35 5 34 VIM/ESR[2223]CGR ......................................................................................................................... 5-36 5 35 50 cc/kg()23 cc/kg()CGRAlloy 690 ()Alloy 600 [2223]().............................. 5-37 5 36 12%[2223]VIM/ESR TTSL ...................................................... 5-37 5 37 24%[2223]VIM/ESR TTST ...................................................... 5-38 5 38 BettisAlloy 600690 CGR[2223] ..................... 5-38 5 39 Bettis[2223]All oy 690 CGR ..................................................................................... 5-39 5 40 SCWAlloy 690TT [25]10%SCC ......................................................................................................................... 5-43 5 41 Alloy 600690[25]SCC........... 5-43 5 42 [25]Alloy 600 MAAlloy 690 TTCGR ............................................................................ 5-44 5 43 Alloy 690 CRDM[1115]ANL . 5-45 5 44 CRDM[1115]ANLCGR ......................................................................................................................... 5-46 5 45 ANL [1115]CRDM5-47

A-14

Translated Table of Contents

5 46 [1115]CRDMANL2 CGR................................................. 5-48 5 47 PNNL2Alloy 690 CGRCTRE 243WP140 [26] ............................................................ 5-49 5 48 K[26]PNNL.................... 5-50 5 49 TTSA[26]Alloy 690 CGRPNNL .......................................................................... 5-50 5 50 TT[26]2Alloy 690 CGRPNNL ........................................................................... 5-51 5 51 [26]IGPNNL ......................................................................................................................... 5-51 5 52 Alloy 690 CRDM[1718]GE......................... 5-54 5 53 GE-GRC [29]20%2CRDM1 ........................................................................................................... 5-55 5 54 20%Alloy 690 CRDM [29] ........... 5-55 5 55 K()24() [29]CGR41% CRDM..................... 5-56 5 56 [29]41%CRDM ........................................... 5-57 5 57 (S-L17%)CRDMPNNL[26] ......................................................................................................................... 5-58 5 58 (T-L30%)CRDMPNNL[26]5-59 5 59 ANL [13]Alloy 690 HAZ(CF690) ........................... 5-61 5 60 Alloy 690 HAZ[27]CGRStudsvikCT ......................................................................................................................... 5-62

A-15

Translated Table of Contents

6 1 PNNL [26]Alloy 690SCC CGR(200811) ................................................................. 6-7 6 2 GE-GRC [18]1DANL Alloy 690 ................................................. 6-7 6 3 GE-GRC [9]MRPEPRIAlloy 690 ...................................... 6-8 6 4 Alloy 690 CRDM[18]............ 6-8

1 1 PWRAlloy 600 PWSCC....... 1-2 2 1 Alloys 690600ASME ..................................................................... 2-1 2 2 ASME(wt%) .................................................................................. 2-2 2 3 ASME................................................................................................... 2-2 2 4 Alloy 690Alloy 600ASME....................................................... 2-3 2 5 SarverAlloy 690(wt%) [18] ................. 2-6 2 6 Alloy 690[18] ........................................... 2-11 3 1 [3334] pHT.............. 3-17 3 2 BrownMills [36]Alloys 6006908252 ......................................................................................................................... 3-18 4 1 Alloy 690Alloy 52[2]IG...................................... 4-3 4 2 2004Alloy 690 ......................... 4-7 4 3 Alloy 690 [2]Alloy 600 ............................. 4-15 4 4 2004Alloy 690 .................................................................................................................. 4-19

A-16

Translated Table of Contents

4 5 Alloys 600690[23]RUB................. 4-20 4 6 Alloy 690 (200812)4-24 5 1 Alloy 690CRDMPWR RPV............................... 5-1 5 2 Alloy 690CRDM5-2 5 3 EdF [3]Alloy 690 CRDM ............. 5-3 5 4 MHIPWSCC[45]............... 5-4 5 5 MHIPWSCC[45] ........................................................................................................................... 5-4 5 6 MHIPWSCC[45] .................. 5-5 5 7 [7] ........................................................................................................................... 5-9 5 8 GE-GRCMRPAlloy 690 ....................................................................................................... 5-11 5 9 ANL [1115]Alloy 690TT ................... 5-15 5 10 BettisAlloy 690[2223] ........ 5-33 5 11 Alloy 690 [2223]SCCBettis.................. 5-34 5 12 [25]CGRAlloy 690 ......................................................................................................................... 5-40 5 13 [25]CGRAlloy 690 ................................................................................................ 5-40 5 14 [25]CGRAlloy 690 ()............................................. 5-41 5 15 [25]CGRAlloy 690 ........................................................................ 5-41 5 16 [25]CGRAlloy 600 Alloy 690 ...................................... 5-42 5 17 [25]CGRAlloy 600 Alloy 690 CRDM..... 5-52

A-17

Translated Table of Contents

5 18 [25]CGRAlloy 600 Alloy 690 CRDM .................. 5-60 6 1 Alloy 690 CRDMPWR CGR (:200812) .................................................... 6-3 6 2 Alloy 690 CRDMPWR CGR (:200812) .................................................... 6-4

A-18

Translated Table of Contents

1 Alloy 690 (MRP258)

1019086

, 2009 8

Alloy 600 (Alloy 182 Alloy 82) (PWR) . 30 PWR 1 (PWSCC) Alloy 600 . PWSCC .

A-19

Translated Table of Contents

Alloy 690 (Alloy 152 Alloy 52 52M) PWSCC 20 . . 2004 EPRI 1009801 (MRP-111) Alloy 690 PWSCC . Alloy 152 52 , . Alloy 690 . 2004 (MRP-111) . MRP-111 2 PWR Alloy 690 PWSCC . Alloy 600 Alloy 690 . PWR Alloy 690 . Alloy 690 PWR Alloy 600 . PWSCC . ( /) . Alloy 600 40~100 , Alloy 690 PWSCC . (

A-20

Translated Table of Contents

) Alloy 690 1 SCC . (< 5E-9 mm/s 0.15 mm/) Alloy 600MA Alloy 690 CRDM 100 . Alloy 690 . PWR . Alloy 690 . PWR 1 10 20 . , PWR 1 Alloy 600 , . Alloy 600 PWSCC . , / . EPRI Alloy 690 PWSCC . Alloy 600 PWSCC Alloy 690 . , Alloy 690 . PWSCC Alloy 690 A-21

Translated Table of Contents

. Alloy 690 . Alloy 600 Alloy 690 PWSCC RPV

A-22

Translated Table of Contents

1 ...................................................................................................................... 1-1 1.1 .................................................................................................................... 1-1 1.2 MRP-111 ........................................................................... 1-2 1.3 Alloy 600 PWSCC ................................................................................. 1-4 1.4 ............................................................................................................. 1-5 2 ALLOY 690 ...................................................................................... 2-1 2.1 ........................................................................................................... 2-1 2.1.1 2.1.2 Alloy 690 SG PWR .......................... 2-3 Alloy 690 ................................................... 2-4

2.2 Alloy 690 ............................................................................................. 2-4 2.3 ........................................................................... 2-5 2.4 ................................................................................ 2-8 2.5 .............................................................................................. 2-12 2.6 ........................................................................................................... 2-13 3 PWSCC ALLOY 690 ............................................................ 3-1 3.1 1 .............................................................. 3-1 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 Sedricks (1979) SG ........................................................... 3-1 Smith (1985) SG ................................................................ 3-2 Yonezawa (1985) SG ........................................................ 3-2 Esposito (1991) ...................................................................................... 3-2 PWSCC ...................................................... 3-3

3.2 1 ...................................................................... 3-4 3.3 2 ............................................................................. 3-15

A-23

Translated Table of Contents

3.4 (LTCP) ...................................................................................... 3-17 3.4.1 3.4.2 .............................................................................................. 3-17 LTCP PWR ................................ 3-18

3.5 ........................................................................................................... 3-23 4 SG PWSCC.......................................................................................... 4-1 4.1 ........................................................................................................ 4-1 4.1.1 4.1.2 4.1.3 4.1.4 4.1.5 4.1.6 4.1.7 .................................................................................................... 4-2 B/Li U-Bend ............................. 4-2 B/Li CERT ................................ 4-5 RUB ..................................................................... 4-6 Weibull Weibayes ...................................................... 4-8 Weibull ................................................................. 4-15 Alloy 600 .......................................... 4-18

4.2 .......................................................................................................... 4-21 4.3 ........................................................................................................... 4-25 5 ALLOY 690 PWSCC............................................................................. 5-1 5.1 ........................................................................................................ 5-2 5.1.1 ........................................................................................... 5-3

5.1.1.1 (MHI) 1 .................................. 5-3 5.1.1.2 .................................................. 5-8 5.1.2 ...................................................................................... 5-10

5.1.2.1 PWR Alloy 690 ........... 5-10 5.1.2.1.1 5.1.2.1.2 5.1.2.1.3 5.1.2.1.4 5.1.2.1.5 (GE-GRC) 1 ........................................................................................ 5-11 (ANL) 1 ................ 5-15 MRP (GEGRC) 1 ........................................ 5-20 1 .......................... 5-28 (INSS) 1 5-29

A-24

Translated Table of Contents

5.1.2.1.6 5.1.2.1.7 5.1.2.1.8

1 .......................................... 5-30 Bechtel-Bettis .......................................... 5-31 , . 5-39

5.1.2.2 Alloy 690 CRDM .......................... 5-44 5.1.2.2.1 5.1.2.2.2 5.1.2.2.3 5.1.2.2.4 5.1.2.2.5 (ANL) 1 ................ 5-45 (ANL) 1 5-48 Studsvik 1 ................................ 5-52 MHI 1 ........................................... 5-52 , . 5-52

5.1.2.3 Alloy 690 CRDM .............................. 5-53 5.1.2.3.1 5.1.2.3.2 5.1.2.3.3 (GE-GRC) 1 ............................................................................ 5-53 (ANL) 1 ................................................................................................ 5-58 Westinghouse , 5-59

5.1.2.4 Alloy 690 (HAZ) ...................... 5-60 5.1.2.4.1 5.1.2.4.2 5.1.2.4.3 5.1.2.4.4 5.1.2.4.5 KAPL ....................................................... 5-60 (ANL) 1 ................ 5-60 Studsvik 1 ................................ 5-61 CIEMAT 1 ................................ 5-62 1 ........................... 5-62

5.2 .......................................................................................................... 5-63 5.3 ........................................................................................................... 5-63 6 .......................................................................................................................... 6-1 6.1 Alloy 690 PWSCC ............................................................................... 6-1 6.2 Alloy 690 ...................................................................... 6-10 6.3 ........................................................................................................... 6-10 7 .......................................................................................................................... 7-1

A-25

Translated Table of Contents

2.1 [18] Alloy 690 Alloy 600 ............................ 2-7 2.2 [18] , Alloy 690 Heat NX4459HG (0.06%C) - ................................................................ 2-9 2.3 [18] , Alloy 690 Heat NX9217H (0.01%C) - .............................................................. 2-10 2.4 [18] , Alloy 690 Heat NX9780H (0.01%C) - .............................................................. 2-10 2.5 Alloy 690 . [2] . ............................................................ 2-12 3.1. [11] Ni ........ 3-5 3.2. [11] 325C PWR Ni ................................................................................................................ 3-5 3.3. [11] (Fen) 325C PWR Ni .................................................... 3-6 3.4. [11] (Fen) 325C LWR ................................................ 3-7 3.5 [11] 325C PWR Ni ................................................... 3-8 3.6 [12] Alloy 690 ........................................................................................ 3-9 3.7 ANL Alloy 690 ([13] ) .............................................................. 3-10 3.8 Alloy 690 PWR 1 320C [14] EAC ................................................................................... 3-10 . 3.9 320C PWR 1 Alloy 690 ( ) ([14] ) .......................................................................................... 3-11

A-26

Translated Table of Contents

. 3.10 PWR 1 Alloys 600 690 5 10 ([16] )3-12 3.11 PWR Ni ( ) ASME ................. 3-13 3.12 [15] Alloys 600 690 CGR ........... 3-14 3.13 [15] Alloy 690 1 CGR ..................................................................................................... 3-14 3.14 Brown Mills [36] LTCP ..... 3-19 3.15 [36] RT Alloy 690 Brown Mills J-R. .......................................................................................................... 3-20 3.16 Alloy 690 Brown Mills [36] JIC T ( ). ...................................... 3-21 3.17 Brown Mills[36] Alloy 690 () Alloy 600 () J-R ............................................................ 3-22 3.18 50C Alloy 690 J-R Paraventi Moshier [42] ................. 3-23 4.1 [2] . Alloy 690 Alloy 600 ...................................................................................................... 4-5 4.2 360C 1 Alloy 600 RUB Weibull ([2] ). Alloy 690(3 ) Weibayes = 5.0 ............................................................................................................. 4-9 4.3 Norring [17] 365C RUB Weibull ([2] ).Alloy 690( ) Weibayes = 5.0 . ......................................................................................................................... 4-10 4.4 365C Alloy 600 Norring [17] RUB Weibull ([2] ). Alloy 690 Weibayes = 5.0 ........................................................................................................... 4-11 4.5 365C Norring [17] RUB Weibull ([2] ). Alloy 690 Weibayes = 5.0 . ......................................................................................................................... 4-11

A-27

Translated Table of Contents

4.6. 1 340C (644F) Alloy 600MA ( ) CLT Weibull ([2]). 320C (608F) Alloy 600MA Alloy 600TT CLT 360C (680F) Alloy 690TT ( ) CLT . Weibayes = 5.0 . ...... 4-12 4.7 320C (608F) 1 Alloy 600MA Alloy 600TT 20% RUB 360C (680F) Alloy 690TT . Alloy 690TT Weibayes = 5.0 . ....... 4-13 4.8 Vaillant [18] 360C 1 RUB Weibull ([2] ). Alloy 600 RUB MA TT . MA TT Alloy 690 RUB 54,000 . Alloy 690 Weibayes = 5.0 . ............................................................................................ 4-14 4 9 Framatome ANP 680F SG Weibull ([2] ). Alloy 690TT SG 100,000 . Alloy 690 Weibayes = 5.0 . ................ 4-14 4 10 4.3 Alloy 600 Weibull ................................ 4-17 4 11 4-8 4.2 4.4 ........ 4-19 4 12 Alloys 600 690 RUB ([23] ). 4-21 4.13 Alloy 600TT SG ([31] ) ......................................................................................................................... 4-23 4.14 Alloy 690TT SG ([31] ) ......................................................................................................................... 4-23 5.1 PWSCC MHI [4, 5]................ 5-5 5.2 PWSCC [4, 5]. 5-6 5.3 PWSCC [4, 5] ............................... 5-6 5.4 Alloy 600MA PWSCC 58,000 690TT BMI [4,5].................................................................................. 5-7 5.5 Alloy 600MA PWSCC 73,000 690TT CRDM [4,5] ..................................................................... 5-7 5.6 SCW CERT CGR[6] ............................................................................................................... 5-8

A-28

Translated Table of Contents

5.7 400C/25.4Mpa SCW EPRI Alloy 600 [7].5-9 5.8 400C/25.4 MPa SCW EPRI Alloy 690 [7]5-10 5.9 Alloy 690 ................ 5-12 5.10 3000 GE-GRC ( mill anneal ) Alloy 690 CGR [9] ..................................... 5-13 5.11 3000 GE-GRC ( mill anneal ) Alloy 690 CGR[9] ...................................... 5-14 5.12 GE-GRC Alloy 690 PWSCC [9]......................................................................... 5-15 5.13 ANL Alloy 690TT [11 - 15]5-16 5.14 Alloy 690TT (S-L ) ANL [11 - 15] .............................................................................................................. 5-17 5.15 Alloy 690TT (S-T ) ANL [11 - 15] .............................................................................................................. 5-17 5.16 S-L ANL 690TT [11 - 15]5-18 5.17 S-T ANL 690TT [11 - 15] ............... 5-19 5.18 Alloy 690TT CGR 320C 300C ANL [13] ...................... 5-20 5.19 ANL GE-GRC ID (~26%) Alloy 690 [17, 18] .................................................................................. 5-21 5.20 ANL 1D Alloy 690 PWSCC 360C 325C CGR 290C CGR [17, 18].................................................................................... 5-22 5.21 PWSCC (c372) [17, 18] ............ 5-22 5.22 CGR 1-D ANL GE-GRC [17, 18] ................................................................ 5-23 5.23 20% 1D Alloy 690 GE-GRC PWSCC ......................................................................................................................... 5-24 5.24 20% 1D Alloy 690 GE-GRC ............................................................................... 5-24

A-29

Translated Table of Contents

5.25 20% 1D Alloy 690 [17,18] ....................... 5-25 5.26 5.23 , S-T [17,18] ....................................................................................................... 5-25 5.27 GE-GRC 1-D Alloy 690 . S-L [17, 18].5-26 5.28 GE-GRC 26% 1D ANL 690 (c372) [17, 18]........................................................................................... 5-27 5.29 20% 1D Alloy 690 [17, 18]5-27 5.30 INSS 360C 20% 690T ([8])5-30 5.31 690TT CGR (CW) ([8])............................................................................................................ 5-31 5.32 Bettis CGR [22,23]5-35 5.33 ( ) [22, 23] .................................................................................................. 5-35 5.34 VIM/ESR CGR [22, 23] ...................................................................................................... 5-36 5.35 Alloy 690() 50 cc/kg ( ) 23 cc/kg CGR Alloy 600 () [22, 23].......... 5-37 5.36 12% VIM/ESR TT S-L- [22, 23] .......................................................... 5-37 5.37 24% VIM/ESR S-T [22, 23] ......................................................................................................................... 5-38 5.38 Bettis Alloy 600 690 CGR [22, 23] ............................... 5-38 5.39 Bettis 690 CGR ................................................................................................... 5-39 5.40 Alloy 690TT 10% SCC [25] .................................. 5-43 5.41 Alloy 600 690 SCC [25] ......................... 5-43 5.42 1 Alloy 600 MA Alloy 690 TT CGR .................................................................................................................. 5-44

A-30

Translated Table of Contents

5.43 Alloy 690 CRDM ANL [11 - 15] .......... 5-45 5.44 CRDM ANL CGR[11 - 15] .................................................................................................... 5-46 5.45 ANL CRDM [11 - 15] ......................................................................................................................... 5-47 5.46 CRDM ANL CGR[11 - 15] .................................................... 5-48 5.47 Alloy 690 CGR PNNL CT : heat RE243, heat WP140 [26] .................................................. 5-49 5.48 K PNNL ...... 5-50 5.49 SA TT Alloy 690 CGR PNNL [26] .............................................................................. 5-50 5.50 TT Alloy 690 CGR PNNL [26] .................................................. 5-51 5.51 IG PNNL [26] ......................................................................................................................... 5-51 5.52 GEAlloy 690 CRDM [17,18] .............................. 5-54 5.53 GE-GRC 20% CRDM [29] ................................................................................................ 5-55 5.54 20% Alloy 690 CRDM [29].................... 5-55 5.55 K (), 24 CGR () 41% CRDM [29]............. 5-56 5.56 2 41% CRDM [29] .................................................................... 5-57 5.57 CRDM PNNL (S-L 17%)[26]........................................................................................................... 5-58 5.58 CRDM PNNL (T-L 30%)[26]........................................................................................................... 5-59 5.59 ANL Alloy 690 HAZ (CF690) ................. 5-61

A-31

Translated Table of Contents

5.60 Alloy 690 HAZ CGR Studsvik CT [27].................................................................................................... 5-62 6.1 Alloy 690 PNNL (2008 11 ) SCC CGR .................................................................................................................... 6-7 6.2 GE-GRC 1-D ANL Alloy 690 [18]........................................................................... 6-7 6.3 GE-GRC MRP EPRI Alloy 690 [9] ............................................... 6-8 6.4 Alloy 690 CRDM [18]6-8

1.1 PWR Alloy 600 PWSCC ....... 1-2 2.1 Alloys 690 600 ASME ......................................................................... 2-1 2.2 ASME (wt%)...................................................................... 2-2 2.3 ASME .................................................................................................. 2-2 2.4 Alloy 690 Alloy 600 ASME ....................................... 2-3 2.5 Sarver [18] Alloy 690 (wt%) [18] ........................ 2-6 2.6 Alloy 690 [18] ...................................... 2-11 3.1 pHT [33, 34]....................... 3-17 3.2 Brown Mills Alloys 600, 690, 82 52 .................... 3-18 4.1 Alloy 690 Alloy 52 IG [2] ........................... 4-3 4.2 2004 Alloy 690 1 ........................... 4-7 4.3 Alloy 690 Alloy 600 Weibull [2] .......................... 4-15

A-32

Translated Table of Contents

4.4 2004 Alloy 690 ......................................................................................................................... 4-19 4.5 Alloys 600 690 RUB [23]............ 4-20 4.6 Alloy 690 (2008 12 ) ...................................................................................................... 4-24 5.1 Alloy 690 CRDM PWR RPV .................... 5-1 5.2 CRDM Alloy 690 ............................................................................. 5-2 5.3 EdF Alloy 690 CRDM [3].................................. 5-3 5.4 PWSCC MHI [4,5].................. 5-4 5.5 PWSCC MHI [4,5] 5-4 5.6 PWSCC MHI [4,5]................. 5-5 5.7 [7] ...................... 5-9 5.8 MRP GE-GRC Alloy 690 ......................................................................................................................... 5-11 5.9 ANL Alloy 690TT [11 - 15] .............................. 5-15 5.10 Bettis Alloy 690 [22,23].......................... 5-33 5.11 Alloy 690 SCC Bettis [22, 23]............ 5-34 5.12 CGR Alloy 690 [25]5-40 5.13 CGR Alloy 690 [25] ............................................................................................................ 5-40 5.14 CGR Alloy 690 ( ) [25] ................................................................... 5-41 5.15 CGR Alloy 690 [25]............................................................................................ 5-41 5.16 CGR Alloy 690 Alloy 600 [25] ................................................................... 5-42 5.17 CGR Alloy 690 CRDM Alloy 600 [25] ....................................... 5-52

A-33

Translated Table of Contents

5.18 CGR Alloy 690 CRDM Alloy 600 [25] ....................................... 5-60 6.1 Alloy 690 CRDM PWR 1 CGR (2008 12 )............................................. 6-3 6.2 Alloy 690 CRDM PWR 1 CGR (2008 12 ) ........................................................... 6-4

A-34

Export Control Restrictions Access to and use of EPRI Intellectual Property is granted with the specific understanding and requirement that responsibility for ensuring full compliance with all applicable U.S. and foreign export laws and regulations is being undertaken by you and your company. This includes an obligation to ensure that any individual receiving access hereunder who is not a U.S. citizen or permanent U.S. resident is permitted access under applicable U.S. and foreign export laws and regulations. In the event you are uncertain whether you or your company may lawfully obtain access to this EPRI Intellectual Property, you acknowledge that it is your obligation to consult with your companys legal counsel to determine whether this access is lawful. Although EPRI may make available on a case-by-case basis an informal assessment of the applicable U.S. export classification for specific EPRI Intellectual Property, you and your company acknowledge that this assessment is solely for informational purposes and not for reliance purposes. You and your company acknowledge that it is still the obligation of you and your company to make your own assessment of the applicable U.S. export classification and ensure compliance accordingly. You and your company understand and acknowledge your obligations to make a prompt report to EPRI and the appropriate authorities regarding any access to or use of EPRI Intellectual Property hereunder that may be in violation of applicable U.S. or foreign export laws or regulations.

The

Electric

Power

Research

Institute,

Inc.

(EPRI, www.epri.com) conducts research and development relating to the generation, delivery and use of electricity for the benefit of the public. An independent, nonprofit organization, EPRI brings together its scientists and engineers as well as experts from academia and industry to help address challenges in electricity, including reliability, efficiency, health, safety and the environment. EPRI also provides technology, policy and economic analyses to drive longrange research and development planning, and supports research in emerging technologies. EPRIs members represent more than 90 percent of the electricity generated and delivered in the United States, and international participation extends to 40 countries. EPRIs principal offices and laboratories are located in Palo Alto, Calif.; Charlotte, N.C.; Knoxville, Tenn.; and Lenox, Mass. TogetherShaping the Future of Electricity

Program: Nuclear Power

2009 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power Research Institute, EPRI, and TOGETHER...SHAPING THE FUTURE OF ELECTRICITY are registered service marks of the Electric Power Research Institute, Inc.

1019086

Electric Power Research Institute 3420 Hillview Avenue, Palo Alto, California 94304-1338 PO Box 10412, Palo Alto, California 94303-0813 USA 800.313.3774 650.855.2121 askepri@epri.com www.epri.com